title
stringlengths 1
827
⌀ | uuid
stringlengths 36
36
| pmc_id
stringlengths 5
8
| search_term
stringclasses 44
values | text
stringlengths 8
8.58M
|
|---|---|---|---|---|
Postmortem histological freeze–thaw artifacts: a case report of a frozen infant and literature review | dfd5fa9d-940f-4309-a9d7-e76236e8a14f | 11790699 | Pathology[mh] | Freezing and thawing have the potential to alter the gross and histological appearance of tissues, damaging individual cells and disrupting the overall architecture. However, the exact mechanism and influencing factors of this damage are not yet fully understood . It is believed that the formation and rupture of cell wall pores result from fluid shifts that take place during the freezing and thawing process. As the extracellular water freezes, fluid moves out from the cells and pours into the extracellular space. This movement of liquid leads to increased intracellular solute concentration, causing cell shrinkage and the disruption of cellular junctions. When the tissue thaws, the fluid returns to the cells, resulting in changes in volume that can rupture weakened cell walls. These combined effects are considered to be the main driving force behind the observed changes, rather than physical damage caused by ice crystals . The outcome of the cells is highly dependent on the rate at which the tissue has been frozen. Depending on the freezing rate and the permeability of the cell, water will either flow out of the cell, which will shrink, or stay inside, forming intracellular ice crystals . In forensic investigations, freezing and thawing can play a crucial role in cases of unknown cause of death. Perpetrators may employ freezing preservation to conceal the body or obscure the time of death. Freezing can also occur naturally when a body is exposed to the elements, and it may even be a cause of death itself. Determining whether a body has been frozen can help corroborate or challenge a suspect’s statements and ultimately determine if any laws have been violated . In this paper, we present a case report in which an autopsy was performed on an infant, who had died of natural causes, after undergoing freezing and thawing. The objective of this study was to reveal several histological artifacts in different tissues resulting from the freeze–thaw process. The authors compare and discuss their findings with the literature. Medical history The infant was a 1-month-old male, born at full term with anthropometrical parameters within the standard range for age: weight 3970 g, 51 cm in crown-heel length, and head circumference of 36 cm. The pregnancy was unremarkable, with normal periodic ultrasound scans, and the delivery was spontaneous and uncomplicated. Maternal vaginal and rectal swabs were negative, and she received intrapartum antibiotic prophylaxis. Approximately 1 month after birth, the infant presented persistent cough, and aerosol therapy was promptly administered. However, after 3 days, he progressively became drowsy and unresponsive. The parents quickly took him to the hospital, but he was already in cardiopulmonary arrest. Cardiopulmonary resuscitation (CPR) was immediately started: oxygen was given through a bag valve mask, and then, an injection of adrenaline was necessary. Despite all attempts, CPR was stopped after 35 min due to pulseless electrical activity (PEA) and the infant was pronounced dead. Subsequent nasal and pharyngeal swabs revealed positive results for respiratory syncytial virus A (RSV), Haemophilus influenzae , and Staphylococcus aureus . Autopsy findings An autopsy was requested to better understand the cause of death. However, in that period COVID-19 was initially spreading, and the lack of medical resources made it necessary to temporarily store the infant’s body at − 10 °C in the freezer of the morgue. Then, due to the health-related challenges arising from the COVID-19 pandemic, the autopsy was delayed for 21 days. Before the autopsy was performed, the body was thawed at room temperature (20 °C) for 12 h. External examination showed growth parameters within the normal limits for 1 month of age: weight 4301 g (50th percentile), crown-heel length 55 cm (50th percentile), and head circumference 37 cm (25th percentile) . The face was regular with no dysmorphisms. Grossly, no anomalies were identified. Internal examination evidenced diffuse congestion of the organs, but no anatomical anomalies. The lungs displayed the expected lobes, with three on the right and two on the left. The right lung weighed 61 g and the left 50 g. Segmental sequential analysis of the heart (weight 29 g) revealed atrioventricular and ventriculoarterial concordance and regular systemic and pulmonary venous returns. The coronary arteries presented a regular course. The other organ weights were as follows: liver 173 g, right kidney 20.8 g, left kidney 22.4 g, spleen 16.4 g, brain 489 g. Histological examination of the lungs revealed bilateral extensive acute necrotizing bronchitis and bronchiolitis with focal hyaline membrane formation and pulmonary edema (Fig. ). Scattered cells with enlarged nuclei due to RSV infection were seen (Fig. ). Minimal diffuse ischemic changes were observed in the myocardium and brain. Both kidneys showed acute tubular necrosis, tubular dilatation, and occasional microcalcifications. No significant histopathological findings were observed in the other organs. The integration of clinical, autopsy, and histological data identified the cause of death as acute respiratory failure resulting from bilateral and diffuse acute necrotizing bronchitis and bronchiolitis. This was caused by RSV type A complicated by bacterial over-infection with Haemophilus influenzae and Staphylococcus aureus . Histological freeze–thaw artifacts Evidence of histological freeze–thaw artifacts was present in many organs. The brain disclosed ice crystal artifacts with multiple clefts within the parenchyma (Fig. ). The lungs, apart from histological evidence of infection, exhibited alveolar distension (Fig. ). The heart presented marked extracellular separation (Fig. ). Abnormal dilatation of lymphatic or capillary vessels was excluded by the regular expression of CD31, which is a marker for endothelial cells, in the vascular structures located between the bundles of cardiomyocytes. Moreover, the artifactual clefts were devoid of endothelial lining (Fig. ). The liver showed shrinkage of the hepatocellular laminae with marked expansion of the sinusoidal space (Fig. ). The kidneys presented unusual cystic lacunae, probably due to expanded extracellular spaces, with tubules strained and flattened (Fig. ). The thymus structure was almost fully preserved (Fig. A). The thyroid presented clefts mostly along the septa, and the cell nuclei appeared more basophilic (Fig. B). The acinar architecture of the pancreas was maintained, with a few artifactual lacunae within the septa and the islets were still recognizable (Fig. C). On the whole, the cell nuclei appeared shrunk and strongly basophilic. The spleen displayed some cystic slightly eosinophilic spaces devoid of epithelium (Fig. D). The brown adipose tissue was mainly preserved, but focal cellular shrinkage was seen in the most eosinophilic cells (Fig. ). The white adipose tissue and the skeletal muscle were virtually unaffected by artifacts (Fig. ). The infant was a 1-month-old male, born at full term with anthropometrical parameters within the standard range for age: weight 3970 g, 51 cm in crown-heel length, and head circumference of 36 cm. The pregnancy was unremarkable, with normal periodic ultrasound scans, and the delivery was spontaneous and uncomplicated. Maternal vaginal and rectal swabs were negative, and she received intrapartum antibiotic prophylaxis. Approximately 1 month after birth, the infant presented persistent cough, and aerosol therapy was promptly administered. However, after 3 days, he progressively became drowsy and unresponsive. The parents quickly took him to the hospital, but he was already in cardiopulmonary arrest. Cardiopulmonary resuscitation (CPR) was immediately started: oxygen was given through a bag valve mask, and then, an injection of adrenaline was necessary. Despite all attempts, CPR was stopped after 35 min due to pulseless electrical activity (PEA) and the infant was pronounced dead. Subsequent nasal and pharyngeal swabs revealed positive results for respiratory syncytial virus A (RSV), Haemophilus influenzae , and Staphylococcus aureus . An autopsy was requested to better understand the cause of death. However, in that period COVID-19 was initially spreading, and the lack of medical resources made it necessary to temporarily store the infant’s body at − 10 °C in the freezer of the morgue. Then, due to the health-related challenges arising from the COVID-19 pandemic, the autopsy was delayed for 21 days. Before the autopsy was performed, the body was thawed at room temperature (20 °C) for 12 h. External examination showed growth parameters within the normal limits for 1 month of age: weight 4301 g (50th percentile), crown-heel length 55 cm (50th percentile), and head circumference 37 cm (25th percentile) . The face was regular with no dysmorphisms. Grossly, no anomalies were identified. Internal examination evidenced diffuse congestion of the organs, but no anatomical anomalies. The lungs displayed the expected lobes, with three on the right and two on the left. The right lung weighed 61 g and the left 50 g. Segmental sequential analysis of the heart (weight 29 g) revealed atrioventricular and ventriculoarterial concordance and regular systemic and pulmonary venous returns. The coronary arteries presented a regular course. The other organ weights were as follows: liver 173 g, right kidney 20.8 g, left kidney 22.4 g, spleen 16.4 g, brain 489 g. Histological examination of the lungs revealed bilateral extensive acute necrotizing bronchitis and bronchiolitis with focal hyaline membrane formation and pulmonary edema (Fig. ). Scattered cells with enlarged nuclei due to RSV infection were seen (Fig. ). Minimal diffuse ischemic changes were observed in the myocardium and brain. Both kidneys showed acute tubular necrosis, tubular dilatation, and occasional microcalcifications. No significant histopathological findings were observed in the other organs. The integration of clinical, autopsy, and histological data identified the cause of death as acute respiratory failure resulting from bilateral and diffuse acute necrotizing bronchitis and bronchiolitis. This was caused by RSV type A complicated by bacterial over-infection with Haemophilus influenzae and Staphylococcus aureus . Evidence of histological freeze–thaw artifacts was present in many organs. The brain disclosed ice crystal artifacts with multiple clefts within the parenchyma (Fig. ). The lungs, apart from histological evidence of infection, exhibited alveolar distension (Fig. ). The heart presented marked extracellular separation (Fig. ). Abnormal dilatation of lymphatic or capillary vessels was excluded by the regular expression of CD31, which is a marker for endothelial cells, in the vascular structures located between the bundles of cardiomyocytes. Moreover, the artifactual clefts were devoid of endothelial lining (Fig. ). The liver showed shrinkage of the hepatocellular laminae with marked expansion of the sinusoidal space (Fig. ). The kidneys presented unusual cystic lacunae, probably due to expanded extracellular spaces, with tubules strained and flattened (Fig. ). The thymus structure was almost fully preserved (Fig. A). The thyroid presented clefts mostly along the septa, and the cell nuclei appeared more basophilic (Fig. B). The acinar architecture of the pancreas was maintained, with a few artifactual lacunae within the septa and the islets were still recognizable (Fig. C). On the whole, the cell nuclei appeared shrunk and strongly basophilic. The spleen displayed some cystic slightly eosinophilic spaces devoid of epithelium (Fig. D). The brown adipose tissue was mainly preserved, but focal cellular shrinkage was seen in the most eosinophilic cells (Fig. ). The white adipose tissue and the skeletal muscle were virtually unaffected by artifacts (Fig. ). An electronic search was performed in three databases (PubMed, Scopus, and Web of Science), and keywords related to the study aim and included in the search string were (freezing OR frosty OR thawing) AND (forensic OR autopsy OR histological OR histology. The English language and time interval of publication, from January 1960 to March 2023, were applied as filters and inclusion criteria. The literature review showed that nine articles reported histological features connected to tissue freezing and thawing . All the related data are summarized in Table and discussed in the “ ” section. Histological artifacts in the freezing–thawing process of tissue have not been widely reported in the literature. However, some major histological changes caused by freezing have been noted, including loss of staining, extracellular fluid accumulation, cell shrinkage, fractures, hemolysis, and hematin formation, and minor changes include loss of bronchial cilia, prominence of collagen in alveolar septa and meninges, and intracellular vacuolization of epithelial cells. In the respective studies, despite these artifacts, adequate visualization of the tissues was possible, allowing diagnosis . The most common histological findings in the freeze–thaw process were the expansion of the extracellular spaces and cell shrinkage, and the heart and the liver were the most widely affected organs. In general, in most tissues, strong basophilic nuclear staining and hemolysis were common . In a few reported cases of frozen newborns, cardiac and hepatic artifactual features were the most evident, with marked separation of the myocardiocytes and the sinusoidal spaces, respectively . In the brain, pseudobubbles or linear freeze artifacts were described . In another study, skeletal muscle was reported to be preserved, but in the case of two freeze–thaw cycles, fibers underwent separation due to large crystal formation . A further study showed that if a body is frozen soon after death, autolysis and putrefaction are less evident, as the cold reduces the activity of enteric microorganisms, and anaerobic decomposition is much less strong than in fresh bodies . In forensic investigations, the freezing and thawing of bodies can be critical in cases of unknown cause of death, as histological artifacts are produced in the process. The mechanisms underlying the extended extracellular areas are still unclear . Freezing of organic tissues begins in the extracellular spaces with ice crystal formation and fixing of water. Thawing melts the ice, resulting in tissue damage due to the changes in osmotic pressure between the cells and the extracellular space . We described a case of an infant who died of respiratory failure caused by bronchitis and bronchiolitis in RSV infection. The autopsy was delayed, and the body was kept frozen in the morgue at − 10 °C for 3 weeks. It was then thawed for 12 h at room temperature (20 °C) before dissection. The infant presented the most common histological features mentioned in the literature . However, freezing artifacts in brain tissue were only mentioned in one book in our review and were described as parenchymal clefts, as we found in our case . Extracellular space expansion, tissue clefts, cell shrinkage, and deep staining nuclei were found in almost every organ. Ice crystal artifacts were evident in the brain as parenchymal clefts and in the heart and liver as marked expansion of the extracellular spaces. Freeze–thaw artifacts may also show some features of congenital diseases. The abnormal cardiac structure raised the suspicion of dilatated lymphatic or vascular channels, but immunostaining for CD31 highlighted their regular location among the bundles of cardiomyocytes. The suspicion of ventricular non-compaction was ruled out, as endocardial fibroelastosis was absent . The hepatic sinusoidal dilatation could mimic hepatic peliosis, a feature of X-linked myotubular myopathy, but liver immunohistochemistry for smooth muscle actin was weakly expressed and not increased . The renal lacunar cystic changes could resemble polycystic kidney disease, but these spaces were devoid of epithelial lining . In our study, the adipose tissue and the skeletal muscle were the only two tissues without artifacts. Only one experimental study described artifacts in skeletal muscle, but they appeared after two freeze–thaw cycles . The absence of artifacts in our case aligns with the fact that the body underwent a single freeze–thaw cycle. The effects of freezing and thawing on adipose tissue have not yet been documented. Our case exhibited white and brown adipose tissues virtually unaffected by freezing and thawing, because they are rich in fat and scarce in water, and ice crystal formation is limited. The brown adipose tissue showed minimal cell shrinkage and deep staining nuclei, due to its composition, as the cytoplasm is packed with multiple lipid droplets and water concentration is higher than in white adipose cells . Therefore, the lack of crystals should not be considered a confounding factor in interpreting freeze–thaw artifacts. In forensic investigations, indirect indicators of freezing and thawing can provide valuable circumstantial evidence to aid forensic pathologists in understanding the sequence of events. Freezing of a body can be connected to criminal activities, such as attempts to conceal a body or manipulate the time of death, as well as natural weather conditions when the body is exposed to the elements. Ice crystal identification can offer insights into the rate at which the body froze and, when combined with other circumstantial data, becomes a significant source of information. The findings of our study can have significant implications in the field of forensic pathology. Specifically, when it is known that a body has been frozen, awareness of these potential artifacts can help forensic pathologists avoid misdiagnoses by considering the possibility of their formation. On the other hand, the presence of these alterations indirectly suggests that the body had been previously frozen and might have already thawed at the time of discovery. Freezing and thawing may disrupt the tissue architecture due to ice crystal formation and subsequent melting. Histological artifacts are mostly evident in the brain, heart, liver, and kidneys. Recognition of histological artifacts may help in forensic analysis whether the body is known to have been frozen or not. In infants, histological freeze–thaw artifacts may mimic congenital diseases. White adipose tissue is unaffected by these artifacts as it contains little water. |
Family-related risk indicators and dental attendance in association with dental caries in preschool children | 387ac210-b7dc-4a50-b587-e405ad3ed0a2 | 11401361 | Dental[mh] | According to WHO’s Global Action Plan on Oral Health by 2023 strategic objectives, it is necessary to promote oral health and to prevent oral diseases, by addressing their risk factors. To understand and control dental caries as a disease, the underlying risk factors must be understood. Pitts et al. presented a division of dental caries risk factors into two categories: patient and intraoral level risk factors. At patient/family level, these risk factors are, e.g., recently treated dental caries lesions in parents or siblings , DGA need in the family , and family functioning . At intraoral level, the risk factors include teeth with caries lesions as well as its severe consequences such as extractions or pufa (pulpitis, ulceration, fistula, or abscess caused by dental caries) . Family functioning has been found to play a significant role in children’s well-being. Family-related risk factors have been noted to expose children to a variety of health problems. Impaired family functioning has been associated with psychiatric disorders , overweight and obesity from early to mid-childhood , higher sugar consumption of preschool-aged children and poor eating habits . Moreover, Duijster et al. suggested that family functioning may be a mediator of socio-economic inequalities in children’s oral health. Long-term need for family social assistance has also been found to increase the risk of dental caries and missing appointments at dental health care . Parenting practices and family interaction have also been shown to be associated with dental caries . Duijster et al. showed that children in families without problems in functioning had on average lower dmft-scores compared to children living in more poorly functioning families. This may be explained by oral hygiene behaviors. The same trend was seen in a study of Almutairi et al. where unhealthy behavior control was associated with greater number of dental caries lesions diminishing a child’s oral health-related quality of life (OHRQoL). However, a study by Bilal et al. did not find a direct association concerning family functioning and OHRQoL in association with dental caries. Adolescents’ dental avoidance has been associated with higher dental caries experience (DMFT) and more invasive dental treatment . A Norwegian study noted that sociodemographic load and dental fear or behavior management problems may lead to avoidance of dental care. In addition, caries risk indicators were associated with past oral health problems and missed appointments . In Finland, public health care including oral health care for all children under 18 years, is equal and free of charge in all parts of the country. According to the outpatient notification register of primary health care (Avohilmo), 99.5% of families with children under school age use public child health clinic services . At least three statutory oral health examinations are offered for children under school age (1- to 2-year-olds, 3- to 4-year-olds and 5- to 6-years-olds) and additionally, according to individual need . Generally, dental caries experience is low among Finnish children. However, it has been suggested that caries is still common among certain subpopulations that should be identified and targeted for preventive dental care. So far, the research on the association of family-related indicators with dental caries in preschool children is scarce. The aim was to study the association of family-related risk indicators and dental attendance in association with dental caries prevalence and experience in preschool children considering also the most severe consequences of dental caries. Challenges in family functioning, missing appointments and utilization of oral health care were also targets of interest.
Data collection This data-based study was designed and written according to the STROBE Checklist . The material was collected from the medical records of 206 randomly chosen children under 7 years who had lived in the city of Oulu, Finland, during the entire study period during the years 2014–2020. Data on all three statutory clinical oral examinations were collected of children at the age of 5–6 years (children born in 2014), on two examinations at the age of 3–4-years (children born in 2014 and 2016) and one examination at the age of 1–2 -years (children born in 2014, 2016 and 2018). The total numbers of children in those age groups living in Oulu were 2456, 2249 and 2023, respectively. This study design allowed monitoring caries development longitudinally in the oldest age group (born in 2014). None of the children were excluded due to potential problems with general health, chronic diseases or physical or mental disabilities. Data collection was performed from March 2022 to July 2022. The randomization of the study population was performed using a number generator (Calculator.net). The sample size of the study was confirmed by power calculations assuming that the difference in decay value between the groups d = 1.2─0.7 = 0.5. With the significance level alpha = 0.95 and power 1-Beta = 0.8, the sample size was confirmed to be 54 per age group. Data collection was continued until at least 54 children from each age group with statutory examinations were obtained. Those children who did not have statutory examinations were not excluded from the study, thus, the final number of children in the study was 206. The background information included the child’s year of birth and gender ( f/m/othe r). The information collected from the medical and dental records comprised children’s visits in dental care ( number of visits , both clinical oral examinations and treatments ), the profession of the person performing the examination/treatment ( dentist/oral hygienist/dental nurse ), and clinical findings ( number of teeth with initial or dentinal caries lesions and dmf indices ). The presence of the recorded caries risk indicators was registered ( yes/no ) according to earlier validation and Caries (control) Current Care Guidelines in Finland . The risk indicators related to previous caries history comprised pufa index (number of teeth with signs of pulpitis , ulcerations , fistula and abscesses; oral conditions resulting from untreated caries) and number of extracted teeth due to caries as well as DGAs (dental general anesthesia) treatments ( yes/no ). The information on challenges in family life, cancelled appointments ( n ) and no shows ( n ) in public child health clinic of all included children was collected from the records of public health nurses ( n = 206) as well as oral health care, including those who did not have statutory clinical dental examinations. The data of possible challenges in family life were recorded as follows: parental exhaustion , sleeping or behavioral problems of a child , death or grief in the family , need for support from social services or child living in out-of-home care , guardian’s unemployment or self-reported problems with mental health , divorce of parents or problems in the parents’ relationship . Challenges were recorded ( yes/no ) every time when mentioned. Statistics Cumulative, descriptive information of the participants at the age 1–2, 3–4 and 5–6 years in the background information and oral health and dental attendance was given. The results of the analyses were described as frequencies and distributions by mean values (standard deviation, SD), proportions and graphically. Associations between the initial and dentinal caries lesions (i, d) and other variables were analyzed using the Chi-square or Fisher’s exact test and t-test. A statistically significant difference between the groups was described by p < 0.05. The statistical analyzes were performed using IBM SPSS Statistics for Windows, Version 29.0 (Armonk, NY: IBM Corp). Ethical considerations Permission for the study was obtained from the register holder, the City of Oulu (Nov 19, 2021, registration authorization number OUKA/12798/07.01.04.02/2021). According to Finnish legislation (Data Protection Act 1050/2018), a statement from the regional medical research ethics committee of the Wellbeing Services County of North Ostrobothnia was not required. The data were stored and analyzed protecting the anonymity of the participants by creating ID numbers, the key for which was held by the first author.
This data-based study was designed and written according to the STROBE Checklist . The material was collected from the medical records of 206 randomly chosen children under 7 years who had lived in the city of Oulu, Finland, during the entire study period during the years 2014–2020. Data on all three statutory clinical oral examinations were collected of children at the age of 5–6 years (children born in 2014), on two examinations at the age of 3–4-years (children born in 2014 and 2016) and one examination at the age of 1–2 -years (children born in 2014, 2016 and 2018). The total numbers of children in those age groups living in Oulu were 2456, 2249 and 2023, respectively. This study design allowed monitoring caries development longitudinally in the oldest age group (born in 2014). None of the children were excluded due to potential problems with general health, chronic diseases or physical or mental disabilities. Data collection was performed from March 2022 to July 2022. The randomization of the study population was performed using a number generator (Calculator.net). The sample size of the study was confirmed by power calculations assuming that the difference in decay value between the groups d = 1.2─0.7 = 0.5. With the significance level alpha = 0.95 and power 1-Beta = 0.8, the sample size was confirmed to be 54 per age group. Data collection was continued until at least 54 children from each age group with statutory examinations were obtained. Those children who did not have statutory examinations were not excluded from the study, thus, the final number of children in the study was 206. The background information included the child’s year of birth and gender ( f/m/othe r). The information collected from the medical and dental records comprised children’s visits in dental care ( number of visits , both clinical oral examinations and treatments ), the profession of the person performing the examination/treatment ( dentist/oral hygienist/dental nurse ), and clinical findings ( number of teeth with initial or dentinal caries lesions and dmf indices ). The presence of the recorded caries risk indicators was registered ( yes/no ) according to earlier validation and Caries (control) Current Care Guidelines in Finland . The risk indicators related to previous caries history comprised pufa index (number of teeth with signs of pulpitis , ulcerations , fistula and abscesses; oral conditions resulting from untreated caries) and number of extracted teeth due to caries as well as DGAs (dental general anesthesia) treatments ( yes/no ). The information on challenges in family life, cancelled appointments ( n ) and no shows ( n ) in public child health clinic of all included children was collected from the records of public health nurses ( n = 206) as well as oral health care, including those who did not have statutory clinical dental examinations. The data of possible challenges in family life were recorded as follows: parental exhaustion , sleeping or behavioral problems of a child , death or grief in the family , need for support from social services or child living in out-of-home care , guardian’s unemployment or self-reported problems with mental health , divorce of parents or problems in the parents’ relationship . Challenges were recorded ( yes/no ) every time when mentioned.
Cumulative, descriptive information of the participants at the age 1–2, 3–4 and 5–6 years in the background information and oral health and dental attendance was given. The results of the analyses were described as frequencies and distributions by mean values (standard deviation, SD), proportions and graphically. Associations between the initial and dentinal caries lesions (i, d) and other variables were analyzed using the Chi-square or Fisher’s exact test and t-test. A statistically significant difference between the groups was described by p < 0.05. The statistical analyzes were performed using IBM SPSS Statistics for Windows, Version 29.0 (Armonk, NY: IBM Corp).
Permission for the study was obtained from the register holder, the City of Oulu (Nov 19, 2021, registration authorization number OUKA/12798/07.01.04.02/2021). According to Finnish legislation (Data Protection Act 1050/2018), a statement from the regional medical research ethics committee of the Wellbeing Services County of North Ostrobothnia was not required. The data were stored and analyzed protecting the anonymity of the participants by creating ID numbers, the key for which was held by the first author.
Challenges in a child’s family life were present among as many as 20.3% of the five-year-olds (Table ) and 73.7% of the children in this age group had challenges in their family life recorded at least twice during the follow-up period. The most common reasons for challenges in family life were parental exhaustion and problems in the parents’ relationship or divorce (Table ). Missed appointments were found in 19.9% of the participants in child welfare care and 16.8% in dental care. Dental general anesthesia (DGA) was performed on average for 1.5% of the study population (Table ). The risk indicators associated with at least initial caries lesions are presented in Table . There was no statistically significant difference in dental caries manifestations between the genders (Table ). In the oldest age group, skewness in dmf distribution was seen if d < 2. Prevalence of teeth with dentinal and initial caries lesions > 1 was significantly ( p < 0.001) associated with previous caries history, pufa index > 0 and prevalence of extracted teeth due to dental caries (> 0) in all age groups. In the age group of 1–2 years, missing appointments at dental health care clinic was associated ( p < 0.05) with DGA, pufa index > 0, prevalence of caries lesions and extracted teeth due to dental caries (Table ; Fig. ). At that age, children with dentinal caries lesions had more symptom-driven dental attendance and pufa index > 0 than those with sound teeth ( p < 0.05). Missing appointments were also associated with prevalence of caries lesions, pufa index > 0 and DGA in the age group of 3–4 years ( p < 0.05) (Table ; Fig. ). At the age of 3–4 years, children who had symptomatic-driven dental attendance had more often pufa index > 0 ( p < 0.05). (Table ; Fig. ). In the age group of 5–6 years, 52.2% of the children with initial or dentinal caries lesions had a history of missing dental health care appointments. In the age group of 5-6-years, there was a significant association between the number of caries risk indicators and the prevalence of initial or dentinal caries lesions ( p < 0.001). All children in the age group of 5–6 years having at least three risk indicators ( n = 3) had manifested caries lesions (Fig. ). Challenges in family life were associated with severe consequences of dental caries. Statistically significant associations ( p < 0.05) between variables studied are presented in Fig. showing the multifactorial character of dental caries.
There is still need for new aspects and findings in the fight against one of the world’s most widespread, polarizing, and economically burdening diseases that impairs the quality of life of children and their families . Here, as many as around 20% of the children had challenges in family life; in some families, challenges persisted for years. All the analyzed family-related risk indicators and information about dental attendance (challenges in family life, missing appointments at child welfare clinic or dental health care clinic and symptom-driven dental care attendance) were found to be associated with dental caries. In addition, previous caries history measured as severe consequences of dental caries (pufa index > 0 and the number of extracted teeth due to dental caries and DGAs) was found to associate significantly with caries prevalence. Our finding is in line with earlier studies [ , , ]. Of course, pufa, extractions and DGA can be considered as consequences of untreated caries, but they can also tell about patient´s future caries risk. This study highlighted the connections between different risk indicators. All these risk indicators were clearly connected to each other, which may have a cumulative effect on dental caries risk, as well as on other threats to the child’s well-being. Along with the traditional ones, risk indicators related to challenges of family life, family functioning [ , , , , , , ] and general life management have emerged, manifesting, for example, as the need for DGA and missing health care appointments. Dental attendance and the family-related risk indicators in question are also in many ways linked to dental caries [ , , ]. This was shown here and in the study of Wigen et al. in 5-year-old children. It is especially important for oral health care professionals to be aware of this association. So far, this perspective has received less attention when studying the causes of dental caries in children. However, it should be kept in mind that also dental fear can be a reason for missing appointments and genetics (epigenetics) can play a role in caries risk . Families having challenges in family life are vulnerable in many ways. Stressful periods and situations in life threaten children’s wellbeing . Challenges in family life can mean that parental skills or resources are not in balance with children’s needs. This loading situation can lead to loss of control over life and among other things, to neglect of adequate oral hygiene, eating habits and missed dental appointments thus threatening the overall well-being of families. In this study, we were not able to measure family functioning in a validated way, but the challenges in family life recorded by health and dental health professionals and dropping out of health care services gave clear indications of it. In Finland, maternity and child health clinic employees must determine whether the family’s resources are sufficient but assessing this requires the presence of a professional with appropriate skills and training. In this study, according to child’s age, some challenges in family life were recorded in as many as one fifth of the families which means that this burden is very common. Literature on this topic is scarce. One of the main findings of this study was the clear relationship between the number of risk indicators and the prevalence of manifested caries lesions. This connection seems to be obvious, but despite this, the importance of determining family-related and dental attendance related risk indicators is still undervalued. It is important for oral health care professionals to record potential risk factors in patient records and consider the risk in determining, e.g., the length of the recall interval and a plan for preventive care. In Finland, data related to oral health are collected comprehensively and monitoring data are available for research. One of the strengths of the study is that Finland has an oral health care service system covering the entire population. For the different age groups included in the study, information was obtained about the oral health status of children of different ages, although some of the study groups remained rather small and conclusions should be drawn with caution. In addition to the cross-sectional study design, this study allowed monitoring caries development longitudinally in the oldest age group. Due to the comprehensive data set and randomisation, the study population can be considered a representative sample of children under school age in Northern Finland. Recordings on the family-related risk indicators are important but unfortunately, these were found to be more incomplete in dental registers than in the registers of public health nurses. This can be considered as a limitation of the study. A larger research population would enable more advanced analyses to be carried out from the perspective of polarization of dental caries, for example. Based on the findings of this study, it would be important to assess the situation of all families in more detail when determining the risk for dental caries. Oral health care professionals could assume a more significant role in supporting the well-being of families and in providing more child-centered dental care.
|
The Effectiveness of Virtual Training on the MiniMed™ 670G System in People with Type 1 Diabetes During the COVID-19 Pandemic | 9d07d32d-988e-45e3-8782-3235b7641357 | 7868572 | Patient Education as Topic[mh] | The coronavirus disease 2019 (COVID-19) pandemic has upended the nonmedical and medical worlds in many ways. Overall, 39% of people in the United States have lost income and/or suffered layoffs in their household. For people with diabetes, there are numerous additional concerns ranging from the possibility of having an increased risk of contracting COVID-19 and the potential worsening of their prognosis to how they can optimize, if not adequately maintain, their current treatment. Although supply chains have remained intact, 55% of people with diabetes are concerned about obtaining their diabetes supplies. For those managing their diabetes with closed-loop systems, many of their health care providers (HCPs) expect that their patients would stop using continuous glucose monitoring (CGM) before stopping pump therapy if they lost their job. During this unprecedented time, HCPs face several challenges. Although 74% of endocrinologists continue to see patients, 20% of them are seeing only existing patients. Overall, there has been a 44% decline in patient visit volumes due to reluctance by many to have face-to-face encounters. In addition, many HCPs may not have access to their office notes, patient logbooks, and other tools such as fax machines that may be required to transmit prescriptions for advanced insulin management technology securely. In situations wherein prescription transmission is not an issue, some people with diabetes may have interrupted or have no access to insulin or consumables. Office staff and personnel who assist in preparing the information for pump therapy authorization may, also, not have access to the requisite clinical information. Even in the event that insulin pumps and supplies are received, many device users have concerns about how they will learn to use their new therapy. Pump trainers and HCPs may question the feasibility of therapy training in a virtual environment, especially since training materials have not been specifically designed for remote training. In addition, the virtual one-on-one interface may not guarantee the undivided attention of the trainee at home, where there may be numerous distractions. Medtronic Diabetes has provided virtual trainings on a case-by-case basis to a limited degree in the past. However, in mid-March 2020, all insulin delivery device trainings shifted to a virtual environment and Medtronic Diabetes has since trained thousands of individuals on insulin pumps with and without CGM devices. Herein, we report the results of the first 33 days of the virtual training approach on the MiniMed 670G™ hybrid closed-loop system from the standpoint of the patient trainee, HCP, and trainer and compare it with retrospective data before the COVID-19 era of sheltering-in-place and social distancing.
This is a retrospective study of both clinical and process metrics comparing face-to-face training (January 20–February 22, 2020, i.e., pre-COVID-19) with virtual training (March 20 to April 22, 2020, i.e., intra-COVID-19). Glycemic outcomes data from CareLink™ Personal were collected on May 11, 2020, and on June 8, 2020, for all patients (with at least 10 days of SmartGuard™ auto mode usage) who were new to MiniMed™ 670 G system use during the pre-COVID-19 and the intra-COVID-19 eras. No IRB approval was sought. The baseline therapy of the trainees was characterized as (1) MDI and SMBG; (2) MDI and sensor users; (3) pump and sensor users—upgrade of pump; and (4) pump and sensor users—upgrade of both. Training on the MiniMed 670 G system was conducted by Medtronic employees using the Zoom Enterprise Version of the Zoom video conferencing application (Zoom Video Communications, San Jose, California). The level of trainee and educator satisfaction with the application was captured using a postsession questionnaire. Medtronic employees delivered ∼96% of all training in the pre-COVID-19 era and 94% in the intra-COVID-19 era with the remainder of training accomplished by individual clinic staff. In-person training on the MiniMed 670 G system is usually done in three sessions of 90 min (pump), 60 min (CGM), and 60 min (SmartGuard technology) depending on the patient's current therapy and knowledge of diabetes management technology. The teaching materials during virtual training were the same as those used during the in-person training sessions. We used two metrics as a proxy for the efficiency and effectiveness of the virtual training program. The first was the number of days between pump shipment and completion of the first training and the number of days between therapy start and the final training, and the second was the percentage of calls to the Medtronic 24-h technical support team that requested educational or software assistance and/or help with HCP telehealth visits. Those calculated from March 20, 2020, to April 22, 2020, (intra-COVID-19), were compared with those during pre-COVID in-person training days performed from January 20, 2020, to February 22, 2020, (pre-COVID-19). We also compared the percentage of uploads to CareLink Personal versus CareLink Professional in the pre-COVID-19 era to the intra-COVID-19 era, as a proxy for the ability of HCPs and/or their staff to upload insulin pump data to professional accounts. The number of unique page views to the CareLink product page was measured during 2 weeks in the pre-Covid-19 era and compared with 2 weeks in the intra-COVID-19 period. Patient experience was assessed with the Net Promoter ® Score (NPS ® ), a single-question management tool. The NPS is graded on a scale of −100 to +100 with values ≥50 regarded as excellent and ≥70 as “world class.” The two most relevant of the questions asked were (1) “based solely upon your recent training experience, how likely might you recommend Medtronic to another person who themselves are insulin-dependent (if they were considering pump therapy)”; and (2) “based upon all of your product and service experiences to-date, how likely might you recommend Medtronic to another person who themselves are insulin-dependent (if they were considering pump therapy)?” Statistical analysis used the independent Student's t -test. Data are reported in percentages and/or percentage change rather than absolute numbers to protect trade secrets. However, the data are based on several thousand pump starts in each time period.
There were 6% of patients in the pre-COVID-19 era and 4% of patients in the intra-COVID era who did not receive training after the pump was shipped to them. The reasons were that the patients self-started, declined the training, or were unavailable and did not differ between time periods. The percentage of patients in each age range for the pre-COVID-19 and intra-COVID-19 time periods was 7–13 years—1% for both; 14–21 years—5% and 6%, respectively; and 22–80+ years—93% and 94%, respectively. Mean age for both cohorts was 47 years with 51% female and 49% male. The baseline therapy for each age group for the pre-COVID-19 and intra-COVID-19 by age is given in . The glycemic outcomes for the pre-COVID-19 cohort and the intra-COVID-19 cohort are shown in . Times-in-ranges (TIRs) data were obtained on May 11, 2020, which represented an overall time from final training to data collection of 3 and 1 months, respectively, and represent all data once the auto mode feature has been turned on for the first time. The data were retrieved again on June 11, 2020, to determine whether there was a difference with another month having elapsed ( ). Overall, the glycemic results were similar between those getting in-person training compared with those receiving virtual training, although marginally better in the former cohort. There were minor but no clinically significant differences in the TIRs between the May and June data extractions. The TIR of 70–180 mg/dL increased progressively from the youngest to oldest group and the difference between the pre-COVID-19 and intra-COVID-19 era TIR narrowed from youngest to oldest. The time in auto mode was 94.9% and 95.5% in the pre-COVID-19 cohort and the intra-COVID-19 era, respectively. There were a slightly higher percentage of patients ≥65 years (8.9% vs. 6.6%) who received training in the intra- versus pre-COVID-19 eras ( ). The number of days (mean ± standard deviation) between the pump shipment to patients (“start date”) and the first training (on the pump) was shorter in the intra-COVID-19 cohort than in the pre-COVID-19 cohort: 11 ± 5 days versus 14 ± 7 days ( P < 0.001) ( ). The number of days between the pump shipment to patients (“start date”) and the third/final training (inclusive of the time to the first training) was 15 ± 15 days versus 19 ± 7 days ( P < 0.01), respectively. The difference between the first and final training was 4 and 5 days in the pre-COVID-19 and intra-COVID-19 eras, respectively. There were no differences in the shortening of training completion (start date to third training) based on age ( ). The videoconferencing survey showed that the platform received a 98% satisfaction rating of “good.” The reasons for the 2% “not good” rating included poor video quality (41.1%), “they could not hear us” (31.3%), “we could not hear them” (8.9%), “we could not see them” (16.9%); and “other” (1.9%). The NPS for question 1 was 78 in the pre-COVID-19 cohort (82.7% promoters, 12.3% passive, and 5% detractors) and 84 in the intra-COVID-19 cohort (87.6% promoters, 9% passive; 3.4% detractors). The NPS for question 2 was 74 in the pre-COVID-19 cohort (79% promoters, 15.9% passive, and 5.1% detractors) and 83 in the intra-COVID-19 cohort (86.2% promoters, 10.4% passive; 3.4% detractors). In the first 2 months of 2020, ∼9% of calls to the 24-h technical support team were for educational and/or software support compared with 10.4% of calls between March 20, 2020, and April 22, 2020. The top five reasons for these calls were system feature inquiry, software assistance, and education on SmartGuard™ features and sensor glucose versus blood glucose differences. The details are shown in . There was a 0–19% decrease in the percentage of calls for educational support in the intra-COVID-19 month (e.g., inquiries about system or sensor features and SmartGuard technology education). However, there was a 187% increase in software support inquiries ranging from installation to login assistance. There was a 92% increase in unique CareLink product page views observed, with >27,000 views occurring during 2 weeks in the intra-COVID-19 era compared with 2 weeks in the pre-COVID-19 era.
Recently imposed social distancing potentially impairs the ability of all stakeholders to initiate and utilize advanced insulin management technologies. However, since early March 2020, the data suggest that videoconferencing-enabled virtual training has been successfully employed by Medtronic employees. Operationally, the number of days from MiniMed 670 G system shipping to training completion has shortened, the subsequent follow-up calls for educational support to the technical support team have decreased, and patient satisfaction has increased. Such outcomes were not intuitively obvious at the onset of this virtual journey. Delays in shipping due to limitations of package carriers, preoccupation of both patients with childcare, education, increase in time spent in virtual versus in-person work in those who were still employed, and schedule disruption may have made virtual training more difficult. Similar potential limitations apply to the trainers themselves. In hindsight, the success of the virtual training may be, in part, due to the use of a comprehensive curriculum and the availability of a 24-h technical support team but also related to patients not having to come to an office for training that may entail time away from their work or home obligations, fighting traffic, finding a parking space, and sitting in a waiting room. In addition, the favorable results may be due to the extra time that patients may have had to focus on the pump training tasks given their home sequestration. Clinically, the glycemic metrics are only slightly lower in the intra-COVID-19 era with the use of the same curriculum in virtual training as is used in face-to-face training. One of the concerns of telehealth visits in general is that it may disadvantage those patients who are older because they are not as comfortable with technology. The data showing that the percentage of older patients who underwent virtual training was slightly higher in the intra- versus pre-COVID-19 era suggest that no substantive barriers to virtual training occurred in a potentially less technology savvy cohort although these patients may represent a higher overall comfort with technology given their desire to use a hybrid closed-loop system. Although not directly related to the results of the switch from face-to-face to virtual training, use of CareLink has dramatically changed in the COVID-19 pandemic era. There was an overall 37% decrease in uploads to CareLink software in March–April compared with the same time period 1 year earlier. Eighty-one percent of this decrease was due to a reduction in uploads to CareLink Professional. There was a 12.6% increase in uploads to CareLink Personal in the same time period. This is most likely due to fewer in-clinic visits, where HCPs and staff are relied upon to perform software uploads. This is mirrored by an increase in CareLink Personal use and a sharp increase in requests for software assistance and number of unique visits to the CareLink product page. This may be due to unfamiliarity with CareLink Personal software or because help was required with telemedicine needs unrelated to CareLink software. This also suggests that additional education may be needed to better prepare system users to perform regular uploads from home; an important practice even after the pandemic recedes. The TIR and time-above-range (TAR) were marginally better in the pre-COVID-19 era cohort than in the intra-COVID-19 era cohort. This may be due to the greater number of days of use of the MiniMed 670 G system after face-to-face training, which may have afforded these patients more time to use the system and adjust their settings. We plan to re-evaluate data from these cohorts in 3 to 6 months that will allow the differences in the durations of time from training to CGM data collection on May 11 and June 7, 2020, to washout. There was also a high time in auto mode of 95% in both groups. This is higher than real-world data of 87% as reported by Stone et al. The higher time in auto mode may reflect the newness of the therapy, enthusiasm for its use, and initial results as well as the correction of a recurrent alert called “BG loop” by introducing a new sensor transmitter in the past 2 years that eliminated this issue that forced recurrent exits from auto mode to calibrate the sensor with a blood glucose. Another possible reason for these differences in TIR and TAR may be the fact that training materials were not changed for a virtual environment and that the time duration of each training session remained the same. Although multiple videos about the MiniMed 670 G system are available on the Medtronic website, there are plans to modify and optimize training materials for a virtual environment. It will also be important to assess the satisfaction level of prescribing HCPs with the virtual training. Supporting this speculation is the large increase in unique CareLink product page views. In the post-COVID-19 world, there may be a new “normal” for how and where patients are treated. The telehealth market is expected to grow by 80% this year. This will likely include fewer in-person visits to clinics and more telemedicine visits, as long as payer reimbursement supports this switch. Since insulin pumps and CGM data can be uploaded from home and automatically accessed anywhere by an HCP, it is feasible for practically anyone to start a diabetes technology to manage their disease. This also affords an opportunity to help people living with diabetes in rural areas or endocrinology “deserts.” There are other consequences of moving to a more virtual medical technology world. For example, it is likely there will be fewer device manufacturer representatives coming to clinics having been replaced by virtual HCP education and training sessions. The implications of job loss for device manufacturer employees and/or clinic employees are, as yet, to be determined. At the very least, there will be a change in the priorities and the requisite skills of medical professionals including improved competency in a wide range of telehealth technologies or platforms (e.g., Clarity ® software [Dexcom, Inc., San Diego, CA], t:connect ® Diabetes Management Application [Tandem ® Diabetes Care, Inc., San Diego, CA], Tidepool Uploader [Tidepool Project, San Francisco, CA], electronic medical record systems, and even DocuSign ® eSignature [DocuSign ® , Inc., San Francisco, CA]). The strengths of this study are the large data set used for analysis and the timeliness of the analysis. Limitations of the study include the inability to provide absolute numbers for the analysis, because they would reveal confidential business material or disclose proprietary information. However, as already noted, there were thousands of training sessions in each time frame that provide reassurance that the data are meaningful.
The data demonstrate that satisfaction, training effectiveness, and early glycemic outcomes are comparable with and in some cases better than those observed in the pre-COVID era. Virtual training for new users of advanced diabetes management technology is a viable method for the post-COVID-19 world.
Supplemental data Supplemental data
|
Long-Term Evaluation of Ocular Surface and Meibomian Gland Function after Laser-Assisted in situ Keratomileusis Surgery | 0a64da25-aa94-4f1d-9d45-417a6f52ea09 | 11844694 | Surgical Procedures, Operative[mh] | Laser-assisted in situ keratomileusis (LASIK) is one of the most common refractive surgeries. It applies an excimer laser to create a hinged corneal flap from the epithelium, Bowman’s membrane, and the superficial stromal layer . Corneal sensitivity is reduced following LASIK due to the damage of the subbasal nerve plexus and might result in neurotrophic keratitis. Dry eye symptoms following LASIK are related to impaired corneal blink reflex, lacrimal gland secretion, and reduced tear secretion . Despite intracorneal nerve regeneration usually occurring between 3 and 6 months postoperatively, chronic dry eye symptoms are reported in post-LASIK patients . Ocular surface characteristics during preoperative examination might predict chronic dry eye development in LASIK. The pathophysiology of long-term dry eye symptoms in post-LASIK eyes remains imcompletely understood . Dry eye disease is a multifactorial disease involving the tear film and ocular surface that result in ocular discomfort, visual disturbance, and tear film instability . It imposes disruptive effects on the quality of life if not properly treated and managed . There are different hypotheses for chronic dry eye symptoms following LASIK. Corneal hypoesthesia resulting from the damage to the sub-basilar never plexus can contribute to reduced tear production from sensory input disruption . Post-LASIK eyes were found to have a decreased blinking rate of up to 40% and thus prone to developing dry eyes . The risk of chronic dry eye after LASIK was found higher among Asian than Caucasian patients . A reduction of functional meibomian was observed in post-LASIK eyes . Despite the intriguing relationships that have been observed, the mechanism of chronic dry eye symptoms in post-LASIK eyes is not completely understood. In this study, we evaluate and report the ocular surface changes, partial blinking (PB) rate, and both the functional and structural meibomian gland (MG) statuses in patients who underwent LASIK for at least 48 months. This is a cross-sectional, matched case-control comparison study. This study was approved by the Institutional Review Board (CUHK-NTEC CREC 2020.34). Written informed consent was obtained from every individual participant, and all study procedures conformed to the tenets of the Declaration of Helsinki. We included adult patients (18 years old or above) who underwent LASIK for myopia at least 48 months ago. All healthy controls were sex-, age-, smoking-, and axial length-matched. Both post-LASIK patients and healthy controls with a history of inflammatory disease, neoplastic disease, ophthalmic disease, and ocular treatment, including any surgery, topical eyedrops, or contact lenses in the past 12 months, were excluded from this study. Post-myopic LASIK patients who fulfilled the inclusion criteria and healthy controls were recruited from a community eye screening program at The Chinese University of Hong Kong between July 2022 and December 2022 . Ocular Surface Evaluation The slit-lamp examination was performed by one of the 2 masked ophthalmologists on 9 MG dysfunction items: corneal fluorescein staining, lid telangiectasia, lid margin thickness, lid margin irregularity, lid-parallel conjunctival folds, papillae, MG plugging, MG expressibility, and meibum quality . Oxford Scheme was used to grade corneal fluorescein staining . Lid margin telangiectasia was graded on a scale from 0 to 3: 0 = no lid margin redness or telangiectasia, 1 = redness in the lid margin and no telangiectasia crossing the MG orifices, 2 = redness in lid margin and telangiectasia crossing MG orifices, moderate telangiectasia or redness, and 3 = severe telangiectasia or redness . Lid margin thickness was graded on a scale from 0 to 2, i.e., 0 = no thickening, 1 = lid margin thickening with or without localized rounding, and 2 = lid margin thickening with diffuse rounding . Eyelid margin irregularity was graded from 0 to 2: 0 = no irregularity, 1 = fewer than 3 lid margin irregularities and shallow notching, and 2 = three or more lid margin irregularities or deep notching . Lid-parallel bulbar conjunctival fold was graded on a scale from 0 to 3: 0 = without fold, 1 = single small fold, 2 = more than 2 folds and not higher than the tear meniscus height (TMH), and 3 = multiple folds and higher than the tear meniscus . Papillae were graded using the papillae-limbal corneal epithelial score . MG plugging was assessed using the scale from 0 to 3: 0 = no plugging, 1 = fewer than 3 pluggings of gland orifices, 2 = three or more pluggings of gland orifices with a distribution of less than half of the full length of the eyelid, and 3 = three or more pluggings of gland orifices with a distribution of half or more of the full length of the lid . MG expressibility was graded from 0 to 3 based on the central 5 MGs of the lower eyelid: 0 = all glands expressible, 1 = 3–4 glands expressible, 2 = 1–2 glands expressible, and 3 = no gland expressible . Meibum quality was graded on a scale of 3: 0 = clear, 1 = cloudy, 2 = cloudy with debris, and 3 = inspissated, toothpaste-like . Subjective dry eye symptom was measured using the ocular surface disease index (OSDI) 12-item questionnaire, which had been translated into Chinese version . Baseline tear production was measured by Schirmer’s test without topical anesthetia . The noninvasive tear break-up time and TMH were measured using a corneal topographer with a built-in real-time keratometer (Oculus Keratograph 5M, Oculus Inc., Arlington, WA, USA) . Lipid layer thickness (LLT), PB rates, and meibography were obtained using the tear interferometer (LipiViewII, TearScience Inc., Morrisville, NC, USA). The MG dropout was measured from the meibograph using the ImageJ software. The area covered by tarsal conjunctiva and MG dropout was segmented using a freehand tool by two independently trained masked investigators, and the average was used for data analysis. The degree of MG dropout was measured by dividing the MG dropout area by the total area of the tarsal conjunctiva . Schirmer test was performed without anesthesia. Numerical results were presented as mean ± standard deviation and range unless otherwise stated. The comparison of ocular surface evaluation was made only on the right eye between post-LASIK patients and healthy controls, using the paired t test. All statistical analyses were performed using SPSS statistical software package (Windows version 24.0; IBM Corp., Armonk, NY, USA). The slit-lamp examination was performed by one of the 2 masked ophthalmologists on 9 MG dysfunction items: corneal fluorescein staining, lid telangiectasia, lid margin thickness, lid margin irregularity, lid-parallel conjunctival folds, papillae, MG plugging, MG expressibility, and meibum quality . Oxford Scheme was used to grade corneal fluorescein staining . Lid margin telangiectasia was graded on a scale from 0 to 3: 0 = no lid margin redness or telangiectasia, 1 = redness in the lid margin and no telangiectasia crossing the MG orifices, 2 = redness in lid margin and telangiectasia crossing MG orifices, moderate telangiectasia or redness, and 3 = severe telangiectasia or redness . Lid margin thickness was graded on a scale from 0 to 2, i.e., 0 = no thickening, 1 = lid margin thickening with or without localized rounding, and 2 = lid margin thickening with diffuse rounding . Eyelid margin irregularity was graded from 0 to 2: 0 = no irregularity, 1 = fewer than 3 lid margin irregularities and shallow notching, and 2 = three or more lid margin irregularities or deep notching . Lid-parallel bulbar conjunctival fold was graded on a scale from 0 to 3: 0 = without fold, 1 = single small fold, 2 = more than 2 folds and not higher than the tear meniscus height (TMH), and 3 = multiple folds and higher than the tear meniscus . Papillae were graded using the papillae-limbal corneal epithelial score . MG plugging was assessed using the scale from 0 to 3: 0 = no plugging, 1 = fewer than 3 pluggings of gland orifices, 2 = three or more pluggings of gland orifices with a distribution of less than half of the full length of the eyelid, and 3 = three or more pluggings of gland orifices with a distribution of half or more of the full length of the lid . MG expressibility was graded from 0 to 3 based on the central 5 MGs of the lower eyelid: 0 = all glands expressible, 1 = 3–4 glands expressible, 2 = 1–2 glands expressible, and 3 = no gland expressible . Meibum quality was graded on a scale of 3: 0 = clear, 1 = cloudy, 2 = cloudy with debris, and 3 = inspissated, toothpaste-like . Subjective dry eye symptom was measured using the ocular surface disease index (OSDI) 12-item questionnaire, which had been translated into Chinese version . Baseline tear production was measured by Schirmer’s test without topical anesthetia . The noninvasive tear break-up time and TMH were measured using a corneal topographer with a built-in real-time keratometer (Oculus Keratograph 5M, Oculus Inc., Arlington, WA, USA) . Lipid layer thickness (LLT), PB rates, and meibography were obtained using the tear interferometer (LipiViewII, TearScience Inc., Morrisville, NC, USA). The MG dropout was measured from the meibograph using the ImageJ software. The area covered by tarsal conjunctiva and MG dropout was segmented using a freehand tool by two independently trained masked investigators, and the average was used for data analysis. The degree of MG dropout was measured by dividing the MG dropout area by the total area of the tarsal conjunctiva . Schirmer test was performed without anesthesia. Numerical results were presented as mean ± standard deviation and range unless otherwise stated. The comparison of ocular surface evaluation was made only on the right eye between post-LASIK patients and healthy controls, using the paired t test. All statistical analyses were performed using SPSS statistical software package (Windows version 24.0; IBM Corp., Armonk, NY, USA). A total of 48 right eyes of 48 post-LASIK Chinese patients (39 females, 2 smokers) were analyzed with 48 right eyes of 48 matched healthy controls. The age on ocular surface examination was 50.0 ± 11.0 years for post-LASIK patients and 51.1 ± 11 years for healthy controls. The duration between LASIK and examination was 177 ± 68 (range, 48–240) months. The axial length was 26.0 ± 1.0 mm for post-LASIK eyes and 25.4 ± 1.2 mm for healthy eyes. The demographic and clinical data of post-LASIK patients and healthy controls are summarized in . Grading of the 9 MG dysfunction items on anterior segment examination was compared between post-LASIK eyes and healthy eyes. Post-LASIK eyes were found to have a higher grade of meibum quality (1.3 vs. 0.9, p = 0.008) compared to healthy controls. Other anterior segment data were comparable between the 2 groups . Post-LASIK eyes were associated with a shorter Schirmer test (10.8 vs. 14.1, p = 0.03). The OSDI score was higher in post-LASIK patients (20.8 vs. 7.5, p = 0.00001), with 75% of post-LASIK patients scored over 12. PB rate, average noninvasive tear break-up time, MG dropout, LLT, and TMH were comparable between the 2 groups . In this cross-sectional, matched case-control comparison study, the post-LASIK patients were associated with a greater degree of dry eye symptoms, and up to 75% of them scored over 12 on the OSDI questionnaire, which indicates the presence of dry eye symptoms. Despite the post-LASIK eyes being found to have a reduction of aqueous production on the Schirmer test score, the TMH and average tear break-up time were not reduced. LLT and MG dropouts were comparable with the healthy controls. The percentage of PB was similar between the 2 groups. Dry eye disease is the most common complication following LASIK, and several hypotheses were proposed in the pathophysiology of post-LASIK dry eye including the loss of corneal sensitivity, impaired blinking mechanism, altered tear film biomarkers, and MG dysfunction . The sensory automatic neural reflex loop between the corneal surface and lacrimal gland is important to maintain the tear film stability, as lacrimal glands secrete a mixture of aqueous, mucus, lipids, antibodies, and enzymes. A longitudinal study reported that the Schirmer test score was reduced at 3 months following LASIK . Another Greece study showed that the Schirmer test was decreased at 1 and 3 months following LASIK but restored by the 6th month . Lee et al. reported that corneal nerve regeneration in post-LASIK eyes is inferior to that in laser-assisted epithelial keratomileusis eyes and does not return to the preoperative level even after 6 months. Lee et al. reported that the number of subbasal and stromal nerve fiber bundles decreased by 90% after LASIK and gradually returned at 1 year despite remaining less than half of the baseline. Our study showed a reduced Schirmer test in post-LASIK eyes. However, both the tear break-up time and TMH were not significantly affected. A long-term longitudinal study would be warranted to study the regeneration of corneal nerves in LASIK patients. A normal blinking motion protects the ocular surface and serves as a pumping force to secrete the meibum to form the lipid tear layer . Jung et al. reported that a reduction in LLT was associated with the history of refractive surgery, and a subsequent study showed a reduction of the functional MG in post-LASIK eyes . An impaired blinking mechanism may result from the disruption of corneal sensation that leads to obstructive MG obstruction . The impaired corneal nerves may affect the sensory feedback to the lacrimal gland and also reduce blinking in post-LASIK patients . The reduction of blinking can lead to increase of tear hyperosmolarity and upregulation of tear inflammatory cytokines . The meibum quality is worse with more cloudy debris in post-LASIK eyes, and we postulate that the changes in tear film dynamics may lead to an increased MG stress, causing the meibum to stagnate, become thicker, and more turbid . Notably, both the patrial blinking rates and MG dropout were comparable between post-LASIK eyes and healthy controls in our study. It is known that the Chinese population has a higher percentage of baseline MG dysfunction when compared to other ethnic groups, and no exaggeration of MG dropout was observed from our patients in contrast to Caucasian patients . Subjective dry eye symptoms are commonly reported in post-LASIK patients, and they usually improve 6 months after LASIK . Heightened parasympathetic tone and prolonged pain sensitivity in patients measured before LASIK predicted greater dry eye symptoms after the surgery . A higher OSDI was found in the post-LASIK patients in our study, which was consistent with most reported studies and not explained by the ocular surface parameters. Review study suggested the corneal nerve damage produced by LASIK may resemble the pathologic neuroplasticity associated with other forms of persistent postoperative pain . Spierer et al. recently reported that clinically significant asymptomatic preoperative MG dysfunction was not found to have a significant impact on LASIK outcomes, especially the dry eye complaints at 1 month. In contrary, our findings indicate the need for periodic comprehensive ocular surface evaluation and appropriate treatment in post-LASIK patients with chronic dry eye symptoms. There are several limitations in the current study. First, this is a cross-sectional study without follow-up data. The longitudinal development of this study cohort will have to be monitored. Second, the LASIK procedures were not standardized in every patient, and the surgical techniques may confound our data. Third, the associations that may affect the degree of ocular surface parameters, such as corneal sensation, nerve evaluation, corneal biomechanics, and tear composition, were not evaluated. Fourth, the history of MG dysfunction treatment was not documented. Our data serve as a reference for future prospective longitudinal studies to understand the chronic dry eye characteristics in post-LASIK patients. In summary, up to 75% of post-LASIK patients had symptomatic dry eye disease on the OSDI questionnaire. Long-term post-LASK eyes showed reduction of aqueous tear production, but without difference in the ocular surface parameters. Post-LASIK patients with chronic dry eye symptoms are therefore recommended to have comprehensive ocular surface evaluation in regular follow-up. Treatment for dry eye disease should be continued in patients with poor tear film stability. Further studies are warranted to evaluate and understand the ocular surface changes in long-term post-LASIK eyes. This study was approved by the Institutional Review Board (Chinese University of Hong Kong – New Territories East Cluster Clinical Research Ethics Committee 2020.34). Written informed consent was obtained from any adults participating in this study. No conflicting relationship exists for any author. The work in this paper was supported in part by a research grant from the Food and Health Bureau (FHB) – Health and Medical Research Fund (HMRF) (Project No. 08190266 to K.K.L.C.). K.K.H.L., Z.H., F.A., J.T.C., and K.K.L.C. designed the study. J.U.S., G.P.M.C., W.W.K.Y., A.L.Y., and K.K.L.C. collected recruited patients and recorded clinical data. K.K.H.L., F.M.A.A.A., J.T.C., and Z.H. reviewed clinical data. K.K.H.L. wrote the first draft of the manuscript. K.K.H.L, F.M.A.A.A., J.T.C., Z.H., G.P.M.C., W.W.K.Y., A.L.Y., C.C.Y.T., C.C.P.P., and K.K.L.C. interpreted the data and reviewed the manuscript. All authors have read and approved the final manuscript. |
Immunization coverage and timeliness of vaccination in young patients with inborn errors of metabolism: a French multicentric study | f0a791e1-1d58-4f6d-af61-8a4d68b6bd2c | 11959845 | Vaccination[mh] | Inborn errors of metabolism (IEMs) are rare disorders caused by genetic deficiencies of an enzyme or transporter. The clinical presentation and severity of these diseases vary widely. Individually, each IEM is rare, but its overall prevalence is estimated to be approximately 1 per 2,000 live births . Some (mostly deficits in amino acid catabolism and certain abnormalities in energy metabolism) are at risk of acute metabolic crisis, especially in catabolic states. Others, such as lysosomal storage diseases, lead to multisystem degradation with progressive cardiac or respiratory failure. These children are therefore particularly vulnerable to infectious diseases and must be protected by vaccination. A recent review of literature summarizes the safety and recommendations for vaccinations in patients with IEMs, focusing on the risk of decompensation after any degree of metabolic stress associated to vaccine preventable diseases . However, several studies have shown vaccination gaps within the population of children with IEMs. In 2011, a multicenter study conducted in California reported lower vaccination coverage among children classified as having stable IEM than among control subjects . A monocentric French study dating from 2015 confirmed lower vaccination coverage among children followed for IEM than among a control group . Beyond vaccination coverage, the concept of vaccination delay is also an important indicator, reflecting the quality of vaccination during vulnerable ages for vaccine-preventable diseases. Delaying the first vaccine doses in infants increases the period of vulnerability to these diseases. For example, pertussis outbreaks occur regularly, with particularly high morbidity and mortality rates among young infants . Definitions of potentially harmful vaccination delays have been established by expert opinions via the Delphi process . A multicenter study conducted in France in 2014 reported vaccination delays in 47% of children attending primary care . Several studies have assessed vaccination delays in children with chronic diseases , but none have specifically focused on patients with IEMs. The primary objective of this multicenter study was to evaluate vaccination coverage and delays in a population of young patients with IEMs. The secondary objectives were to assess vaccination coverage and delay according to the stability of the disease (stable IEM versus a high risk of metabolic crisis or cardiorespiratory failure), compare vaccination coverage to national data, and describe vaccination coverage for specifically indicated vaccines not included in the standard vaccination schedule, such as influenza vaccination.
Patients We conducted a retrospective multicentric study of vaccination coverage and delay in young patients with IEMs. Patients were enrolled in 7 French centers for IEMs (Angers, Brest, Bordeaux, Limoges, Rennes, Toulouse, and Tours) between February 2021 and May 2022. The inclusion criteria were an age between 6 months and 20 years, a diagnosis of IEM for at least 6 months, a follow-up at a specialized center for IEM, and consultation at that center during the inclusion period. Patients without available vaccination data (unavailable health records), those vaccinated according to a non-French vaccination schedule, or those with a contraindication to vaccination were excluded. Procedure and data collection The data collected for each patient included age, type of IEM and date of diagnosis, potential risk of acute metabolic decompensation, presence of respiratory and/or heart failure, and finally, the presence of a contraindication to a vaccine and any adverse event considered serious and associated with a previous vaccination. The vaccination data were collected from the patients’ health records and included the type of vaccine and injection date. All the data were anonymized at each center and transferred via email via a secure messaging system. Definitions, vaccination schedule and delay IEMs were classified according to their classical pathophysiology into IEMs related to small-molecule disorders (accumulation or deficiency), complex-molecule disorders (accumulation, deficiency, or cell processing and trafficking defects) and energy defects (membrane transporters and cytosolic and mitochondrial energy defects) . Patients were separated into two groups according to their health condition: stable or at-risk. At-risk conditions were defined by cardiorespiratory failure or by an IEM with a serious risk of metabolic crisis. The analysis of vaccination data considered the yearly French vaccination schedules published since 2002 (available at the site of the French Ministry of Health, https://sante.gouv.fr/ ), accounting for the evolution of vaccination recommendations. The timeframe corresponding to detrimental vaccination delay was defined according to the study by Gras et al. . This study, which uses a 3-round Delphi process, obtained an expert consensus to define a potentially dangerous vaccination delay for each dose of each vaccine recommended for children younger than 2 years of age, according to the 2015 French immunization schedule. These definitions are summarized in Table . For other vaccinations (schedules before 2015 and subsequent modifications), vaccination delay was empirically defined as a timeframe relative to the recommended injection date, exceeding 15 days for injections recommended in the first 6 months of life, exceeding 2 months for injections from 6 months to 6 years, and exceeding 1 year for vaccines after the age of 6 years. The delay durations presented in the results corresponded to the timeframe beyond the detrimental delay. When a vaccine had not yet been administered, the child was considered unvaccinated. We also compared the vaccination coverage of our group of patients with an IEM to that of the French population at the age of 2 years. National vaccination coverage data at 24 months were extrapolated from national surveys by Santé Publique France (available on the site https://www.santepubliquefrance.fr/ ), weighted for each vaccine by the number of patients in the corresponding birth year. Ethics approval and consent to participate A written information letter was provided to legal representatives, and non-opposition to participation in the study was obtained. Patients or their parents who refused to participate were not included. No nominative, sensitive or personal data were collected from the patients. This study was registered locally at our institution and was approved by our local ethics committee (n°2024–018). Statistical analysis The results were expressed as the median [min‒max], mean ± SD or number of subjects (%), as appropriate. The quantitative data were compared via the Mann‒Whitney test. Qualitative data were compared via Fisher’s exact test. All the statistical analyses were carried out via GraphPad Prism version 6.0 (GraphPad software, Inc.). A p value < 0.05 was considered to indicate statistical significance.
We conducted a retrospective multicentric study of vaccination coverage and delay in young patients with IEMs. Patients were enrolled in 7 French centers for IEMs (Angers, Brest, Bordeaux, Limoges, Rennes, Toulouse, and Tours) between February 2021 and May 2022. The inclusion criteria were an age between 6 months and 20 years, a diagnosis of IEM for at least 6 months, a follow-up at a specialized center for IEM, and consultation at that center during the inclusion period. Patients without available vaccination data (unavailable health records), those vaccinated according to a non-French vaccination schedule, or those with a contraindication to vaccination were excluded.
The data collected for each patient included age, type of IEM and date of diagnosis, potential risk of acute metabolic decompensation, presence of respiratory and/or heart failure, and finally, the presence of a contraindication to a vaccine and any adverse event considered serious and associated with a previous vaccination. The vaccination data were collected from the patients’ health records and included the type of vaccine and injection date. All the data were anonymized at each center and transferred via email via a secure messaging system.
IEMs were classified according to their classical pathophysiology into IEMs related to small-molecule disorders (accumulation or deficiency), complex-molecule disorders (accumulation, deficiency, or cell processing and trafficking defects) and energy defects (membrane transporters and cytosolic and mitochondrial energy defects) . Patients were separated into two groups according to their health condition: stable or at-risk. At-risk conditions were defined by cardiorespiratory failure or by an IEM with a serious risk of metabolic crisis. The analysis of vaccination data considered the yearly French vaccination schedules published since 2002 (available at the site of the French Ministry of Health, https://sante.gouv.fr/ ), accounting for the evolution of vaccination recommendations. The timeframe corresponding to detrimental vaccination delay was defined according to the study by Gras et al. . This study, which uses a 3-round Delphi process, obtained an expert consensus to define a potentially dangerous vaccination delay for each dose of each vaccine recommended for children younger than 2 years of age, according to the 2015 French immunization schedule. These definitions are summarized in Table . For other vaccinations (schedules before 2015 and subsequent modifications), vaccination delay was empirically defined as a timeframe relative to the recommended injection date, exceeding 15 days for injections recommended in the first 6 months of life, exceeding 2 months for injections from 6 months to 6 years, and exceeding 1 year for vaccines after the age of 6 years. The delay durations presented in the results corresponded to the timeframe beyond the detrimental delay. When a vaccine had not yet been administered, the child was considered unvaccinated. We also compared the vaccination coverage of our group of patients with an IEM to that of the French population at the age of 2 years. National vaccination coverage data at 24 months were extrapolated from national surveys by Santé Publique France (available on the site https://www.santepubliquefrance.fr/ ), weighted for each vaccine by the number of patients in the corresponding birth year.
A written information letter was provided to legal representatives, and non-opposition to participation in the study was obtained. Patients or their parents who refused to participate were not included. No nominative, sensitive or personal data were collected from the patients. This study was registered locally at our institution and was approved by our local ethics committee (n°2024–018).
The results were expressed as the median [min‒max], mean ± SD or number of subjects (%), as appropriate. The quantitative data were compared via the Mann‒Whitney test. Qualitative data were compared via Fisher’s exact test. All the statistical analyses were carried out via GraphPad Prism version 6.0 (GraphPad software, Inc.). A p value < 0.05 was considered to indicate statistical significance.
Patients Two hundred seventy-nine patients were enrolled in this study (see Flow chart Fig. ) between February 2021 and May 2022; 4 were excluded because of a non-French vaccination schedule (n = 2), unavailable health records (n = 1), or contraindications to vaccination related to recent hematopoietic stem cell transplantation (n = 1). Finally, 275 patients were analyzed and divided into stable (n = 162) and at-risk (n = 113) patients. The main characteristics of the patients are reported in Table . Briefly, the age of the patients at inclusion varied between 6 months and 19.8 years, with a median age of 8 years, whereas the median age at diagnosis was 0.6 months. A majority of patients had an IEM related to a defect in small molecule metabolism (64%), whereas disorders of complex molecules and energy defects represented 17% and 19% of the patients, respectively. Compared with at-risk patients, stable patients were significantly younger at diagnosis and had a greater proportion of patients with small molecule disorders, which was probably related to the greater number of patients with phenylketonuria (n = 94). In contrast, complex molecule disorders were more common in at-risk patients (33% vs 6%, p < 0.001). Age at inclusion was not different between the two groups. Vaccination coverage and delays for all patients No serious adverse events associated with vaccination were reported. The results of vaccination coverage are presented in Table . Briefly, immunization against diphtheria, tetanus, acellular pertussis, inactivated poliomyelitis and Haemophilus influenzae b (DTaP-IPV and Hib) was satisfactory, with primary vaccination coverage ranging between 98 and 100%. The first booster dose was almost systematically administered (vaccination coverage between 96 and 100%), whereas the next booster doses were given to 87% to 95% of the patients. The immunization against pneumococcus was also satisfactory, with primary vaccination coverage ranging between 91 and 99%, and a booster dose was given to 92% of the patients. Ninety-seven% of the patients had received 2 doses of the measles-mumps-rubella (MMR) vaccine. Most of the patients (91%) received 3 doses of the hepatitis B vaccine. As expected, the vaccination coverage for the Meningococcal C vaccine was lower (84% for the first dose and 74% for the second dose). Finally, a complete vaccination schedule was used for only 60% of the patients. Taken together, these results were unsatisfactory given the insufficient vaccination coverage of patients with IEMs. Furthermore, vaccination delay was frequent for a majority of the vaccines. Almost half of the patients received their primary vaccinations for DTaP-IPV, Hib and PCV with a delay. This delay was longer than 1 month in one-third of the cases, with maximal values of approximately 4–5 years. A vaccination delay was less frequent and shorter for booster doses. A vaccination delay was also frequent for other vaccines (concerning 30 to 40% of the patients for MenC and MMR vaccines), except for the HBV vaccine (for whom vaccinations are recommended during the first year of life, but vaccination delay is defined by the absence of vaccination after the age of 11 years). Taken together, a vaccination delay was reported in 83% of the patients. Compared with national data at the age of 2 years, the vaccination coverage of patients with an IEM was in the same range as that of the French population (Table S1 in supplementary data files), at least for the primary vaccinations, whereas it seems to be slightly lower for booster doses. Again, a vaccination delay appears to be the explanation because the coverage of these booster doses was satisfactory in the results irrespective of patient age (Table ). Stable versus at-risk patients Insufficient vaccination coverage was detected in both groups (Table S2 in the supplementary data), with no difference between stable and at-risk patients, except for the percentage of patients receiving the HBV vaccine, which was lower for at-risk patients (86% vs 95%, p < 0.05). The vaccination schedule was completed in 64% of stable patients and in 54% of at-risk patients (NS). However, a vaccination delay was significantly more frequent in at-risk patients for the primary doses and for the first booster doses of DTaP-IPV and Hib (Table ). The difference did not reach significance for the second and third booster doses. A vaccination delay for the 1st dose of the MMR vaccine was also more frequent in at-risk patients (54% vs 34%, p < 0.01). More in details, differences existed between the various types of IEMs, with a minimal rate of vaccinations with delay in PKU patients (median 16%), contrasting with others IEMs with a maximal rate of vaccinations with delay in patients with others aminoacidopathies (Maple Syrup Urine Disease, type 1 tyrosinemia, etc., median 53%, p < 0.05, Table S3 in the supplementary data). The rate of delay also increased significantly with age, but this effect remained of limited importance (linear regression, p < 0.0001, R2 = 0.078). However, the duration of vaccination delay was mostly the same between stable and at-risk patients (Table S4 in the supplementary data). Specifically indicated vaccines The seasonal flu vaccine was administered to a limited number of patients (approximately one-third of at-risk patients, whereas it was recommended for all these patients and less than 10% of stable patients; Table ). Patients vaccinated with varicella vaccine were also rare (18% of at-risk patients and 3% of stable patients), but our collected data did not include the notion of a past history of chickenpox. The BCG vaccine has been used in 21% of patients, with no difference between stable and at-risk conditions. This vaccine remains recommended in France, but only for children with an identified risk factor for tuberculosis. The others optional vaccinations remained very limited in patients with IEMs.
Two hundred seventy-nine patients were enrolled in this study (see Flow chart Fig. ) between February 2021 and May 2022; 4 were excluded because of a non-French vaccination schedule (n = 2), unavailable health records (n = 1), or contraindications to vaccination related to recent hematopoietic stem cell transplantation (n = 1). Finally, 275 patients were analyzed and divided into stable (n = 162) and at-risk (n = 113) patients. The main characteristics of the patients are reported in Table . Briefly, the age of the patients at inclusion varied between 6 months and 19.8 years, with a median age of 8 years, whereas the median age at diagnosis was 0.6 months. A majority of patients had an IEM related to a defect in small molecule metabolism (64%), whereas disorders of complex molecules and energy defects represented 17% and 19% of the patients, respectively. Compared with at-risk patients, stable patients were significantly younger at diagnosis and had a greater proportion of patients with small molecule disorders, which was probably related to the greater number of patients with phenylketonuria (n = 94). In contrast, complex molecule disorders were more common in at-risk patients (33% vs 6%, p < 0.001). Age at inclusion was not different between the two groups.
No serious adverse events associated with vaccination were reported. The results of vaccination coverage are presented in Table . Briefly, immunization against diphtheria, tetanus, acellular pertussis, inactivated poliomyelitis and Haemophilus influenzae b (DTaP-IPV and Hib) was satisfactory, with primary vaccination coverage ranging between 98 and 100%. The first booster dose was almost systematically administered (vaccination coverage between 96 and 100%), whereas the next booster doses were given to 87% to 95% of the patients. The immunization against pneumococcus was also satisfactory, with primary vaccination coverage ranging between 91 and 99%, and a booster dose was given to 92% of the patients. Ninety-seven% of the patients had received 2 doses of the measles-mumps-rubella (MMR) vaccine. Most of the patients (91%) received 3 doses of the hepatitis B vaccine. As expected, the vaccination coverage for the Meningococcal C vaccine was lower (84% for the first dose and 74% for the second dose). Finally, a complete vaccination schedule was used for only 60% of the patients. Taken together, these results were unsatisfactory given the insufficient vaccination coverage of patients with IEMs. Furthermore, vaccination delay was frequent for a majority of the vaccines. Almost half of the patients received their primary vaccinations for DTaP-IPV, Hib and PCV with a delay. This delay was longer than 1 month in one-third of the cases, with maximal values of approximately 4–5 years. A vaccination delay was less frequent and shorter for booster doses. A vaccination delay was also frequent for other vaccines (concerning 30 to 40% of the patients for MenC and MMR vaccines), except for the HBV vaccine (for whom vaccinations are recommended during the first year of life, but vaccination delay is defined by the absence of vaccination after the age of 11 years). Taken together, a vaccination delay was reported in 83% of the patients. Compared with national data at the age of 2 years, the vaccination coverage of patients with an IEM was in the same range as that of the French population (Table S1 in supplementary data files), at least for the primary vaccinations, whereas it seems to be slightly lower for booster doses. Again, a vaccination delay appears to be the explanation because the coverage of these booster doses was satisfactory in the results irrespective of patient age (Table ).
Insufficient vaccination coverage was detected in both groups (Table S2 in the supplementary data), with no difference between stable and at-risk patients, except for the percentage of patients receiving the HBV vaccine, which was lower for at-risk patients (86% vs 95%, p < 0.05). The vaccination schedule was completed in 64% of stable patients and in 54% of at-risk patients (NS). However, a vaccination delay was significantly more frequent in at-risk patients for the primary doses and for the first booster doses of DTaP-IPV and Hib (Table ). The difference did not reach significance for the second and third booster doses. A vaccination delay for the 1st dose of the MMR vaccine was also more frequent in at-risk patients (54% vs 34%, p < 0.01). More in details, differences existed between the various types of IEMs, with a minimal rate of vaccinations with delay in PKU patients (median 16%), contrasting with others IEMs with a maximal rate of vaccinations with delay in patients with others aminoacidopathies (Maple Syrup Urine Disease, type 1 tyrosinemia, etc., median 53%, p < 0.05, Table S3 in the supplementary data). The rate of delay also increased significantly with age, but this effect remained of limited importance (linear regression, p < 0.0001, R2 = 0.078). However, the duration of vaccination delay was mostly the same between stable and at-risk patients (Table S4 in the supplementary data).
The seasonal flu vaccine was administered to a limited number of patients (approximately one-third of at-risk patients, whereas it was recommended for all these patients and less than 10% of stable patients; Table ). Patients vaccinated with varicella vaccine were also rare (18% of at-risk patients and 3% of stable patients), but our collected data did not include the notion of a past history of chickenpox. The BCG vaccine has been used in 21% of patients, with no difference between stable and at-risk conditions. This vaccine remains recommended in France, but only for children with an identified risk factor for tuberculosis. The others optional vaccinations remained very limited in patients with IEMs.
This study demonstrated insufficient immunization coverage among French patients with IEMs. Moreover, a significant number of vaccinations were performed with potentially harmful delays. The percentage of patients vaccinated with delay was greater in the at-risk group than in the stable group. Furthermore, the coverage of specifically indicated vaccines remained very limited in patients with IEMs, including seasonal flu vaccines, which were rarely used even in at-risk patients. Vaccination is one of the most cost-effective public health interventions, as it prevents the costs associated with treating and caring for people who become ill. To achieve community immunization, the objective of the World Health Organization (WHO) Expanded Program on Immunization is to reach vaccination coverage of 90% . In our study, vaccination coverage among the population of young patients with IEM exceeded 90% for the first vaccine doses. However, it fell below this threshold for later doses, including the third DT-IPV booster, the second and third pertussis vaccine boosters, and the second dose of the meningococcal C vaccine. Furthermore, for an individual benefit, and given that these patients had a chronic illness and received close medical follow-up, vaccination coverage higher than that of the whole population was expected, ideally close to 100% . A majority of patients with IEMs are at increased risk of vaccine-preventable diseases and hence require additional strategies to maximize protection against these diseases . In our study, 83% of patients with IEMs received at least one vaccination from the vaccination schedule late. In fact, measuring standard vaccination coverage does not reflect whether children received vaccine doses at the recommended ages or whether vaccines were given concomitantly per the schedule. Quantifying vaccination timeliness, i.e., comparing when children receive vaccine doses relative to age recommendations, is also important for the success of an expanded program on immunization . Vaccination delay seems to be frequent in the whole population. Two studies conducted in the United States involving 16,211 infants aged 24–35 months and 9,223 children aged 25–72 months reported vaccination delay rates of 58% and 48%, respectively. Similar results were also recently reported in France, with primary care pediatricians reporting that 47% (among 443 children) had at least one potentially dangerous immunization delay . A timely vaccination is even more expected for patients with chronic diseases. However, a recent French study, which used the same definitions of vaccination delay as ours, highlights a vaccination delay rate of 26% to 75% (depending on the vaccine) among children with chronic diseases . Since children with chronic diseases represent a primary target for immunization strategies, it is important that their immunization coverage and the timeliness of vaccination are optimal. Improving timeliness and minimizing missed opportunities to vaccinate individuals with these special risk conditions will also optimize protection from vaccine-preventable diseases . Our recommendation is that children with IEMs should be strictly monitored for routine and recommended vaccinations and that health care providers and families should be properly informed to avoid false contraindications (such as a risk of inducing a metabolic crisis). Thus, Klein et al . reported that vaccination of children with IEM (n = 271) was not associated with any significant increase in emergency-department visits or hospitalizations during the 30 days after vaccination . Protection can also include ‘cocooning’ (i.e., ensuring appropriate immunizations within the immediate family; e.g., varicella, influenza and pertussis vaccination). Compared with national data at the age of 2 years, the vaccination coverage rates for primary vaccinations (DTaP-IPV, Hib, and PCV) among patients with IEMs were globally similar to those reported for the whole population, according to data from Santé Publique France. However, for booster doses, vaccination coverage appears to be lower among patients with IEMs. Moreover, vaccination coverage for the first dose of MMR was similar, whereas for the second dose, it was lower among patients with IEM than among the whole population. Given that the overall vaccination coverage of both populations was almost similar, these results suggest that in our patients with IEM, subsequent doses are administered but delayed. Pandolfi et al . also reported that vaccination coverage among children with chronic diseases increased between the ages of 12 and 24 months, indicating that vaccines in the first year of life are often administered late . However, these data should be interpreted prudently because we extrapolated vaccination coverage data from the general population. This methodology did not allow for statistical comparison. Using Northern California Kaiser Permanente’s electronic medical records, Klein et al . compared vaccination coverage data at 24 months from 77 infants with IEM to 1540 matched control subjects and reported no difference . In further analyses, we found similar vaccination coverage between patients with at-risk IEM and patients with stable IEM, except for hepatitis B vaccination. In their study, Cerutti et al . reported some differences in vaccination coverage within a population of 132 patients with IEMs, which were analyzed in three groups (sickest, chronic, and stable). Patients defined as chronic had lower vaccination coverage for pneumococcus, and those defined as sickest had lower vaccination coverage for meningococcus . There was a greater rate of vaccination delay among at-risk patients than among stable patients. These results are particularly concerning, as they indicate that the most vulnerable patients are protected with delay against vaccine-preventable diseases. Cerutti et al . also reported that patients with IEMs in the sickest group were vaccinated at a later age than healthy controls and had lower vaccination coverage than patients with chronic and stable IEMs . Furthermore, in our study, the coverage rate for specifically indicated vaccines was low in the at-risk group of IEM patients. Indeed, 30 to 35% of these patients were vaccinated against influenza during the winters of 2018, 2019, and 2020. This rate is similar to that reported by Cerutti et al . in their population of patients with IEMs (35 to 50%) . Diallo et al . reported an influenza vaccination coverage of 15% among a population of 146 patients with various chronic diseases . Immunization with the 23-valent pneumococcal polysaccharide vaccine remained exceptional (6% of at-risk patients), whereas the presence of a chronic disease also predisposes children to invasive pneumococcal diseases, with a high mortality rate that has been reported to be 25 times higher among children with chronic disease . Finally, it was not possible to determine the lower rate of patients vaccinated with the varicella vaccine because we did not include the notion of a past history of chickenpox. Overall, our study revealed inadequate vaccine coverage for specifically indicated vaccines among young patients with IEMs. Our study was not designed to investigate the reasons of this insufficient vaccination coverage with significant delays, but we can formulate some hypotheses on the reasons for this vaccination delay. Factors contributing to vaccination delays have been largely reported in the whole population , and a “Catalogue of interventions addressing vaccine hesitancy” is available on the website of the European Centre for Disease Prevention and Control (ECDC) and proposed various tools for helping healthcare workers. Additional factors specific to our group of patients also exist . First, there may be parental or medical fear of exacerbating chronic IEM, which is linked to potential adverse events of vaccines, such as fever, increased energy demands or a catabolic state . Cases of metabolic crisis following vaccination have been described in the literature, revealing IEMs . However, when the IEM is well known, vaccination is not associated with serious adverse events . The safety of vaccines in patients with IEMs has been largely demonstrated in recent literature, including those with live attenuated viruses . The fear from metabolic crises has been refuted, even in high-risk groups, such as urea cycle disorders or mitochondrial diseases. Some experts still recommend special precautions and careful monitoring after vaccine administration to patients with IEMs prone to rapid metabolic decompensation or immunodeficiency . In any case, vaccination prevention in children with IEMs must therefore be ensured according to the vaccination schedule, such as healthy children in the whole population and those receiving the COVID-19 vaccine . The only exception is the rotavirus vaccine (containing a source of fructose), which is contraindicated for hereditary fructose intolerance . Individuals with IEMs need to be up-to-date with their immunizations. These children may also warrant specific vaccination indications, such as seasonal influenza vaccination or enhanced pneumococcal vaccination, for example, in cases of respiratory failure . Vaccinations may also be delayed due to an intercurrent event related or unrelated to the chronic illness. For vaccinations not recommended for the whole population, such as influenza or expanded pneumococcal vaccination, there may be a lack of awareness, both parental and medical, of eligibility for these vaccines. Finally, these children are often followed up by several doctors (pediatricians and specialists), and the responsibility for updating the vaccination schedule is sometimes not assured, with one doctor assuming that another will do it. Additionally, as in the general population, parental vaccine refusal due to vaccine hesitancy may exist, particularly in France, where declining confidence and acceptance in vaccines and immunization programs have been recently reported . These different causes are identified in the literature for children in the whole population, such as those with IEMs . We recognize that our study had several limitations. First, we did not have a control group from the whole population. However, the use of national data from Santé Publique France provides a rough estimate of differences with the whole population. Another limitation is that we did not collect data on the patients’ sex, which hinders the interpretation of HPV vaccination. Finally, the design of our study did not allow us to evaluate the causes of vaccination absence or delays. Many patients with IEMs are especially vulnerable if unprotected against vaccine preventable diseases (intoxication or energy metabolism disorders, presence of cardiac and/or respiratory failure, etc.). Moreover, several IEMs are associated with some level of immunodeficiency (ex: congenital disorder of glycosylation, glycogen storage disease type Ib), which can require a personalized vaccine schedule. National or international guidelines exist for the most frequent IEMs. A first suggestion to improve vaccination coverage in these patients is to include in guidelines recommendations for vaccine schedule, including if necessary personalized vaccine schedule. A second suggestion is to include in the annual visit to check for vaccination delay. In presence of vaccination delay, the reason for this delay must be questioned, with the possibility to use the ECDC questionnaires, and resolved. Finally, the general practitioner and all the healthcare providers involved in the patient’s care must be associated in the recommendations to vaccinate the patient (Fig. ).
The present study demonstrated inadequate vaccine coverage and significant delays among young patients with IEMs, especially at-risk patients with cardiorespiratory failure or with a serious risk of metabolic crisis. It is essential to communicate with parents and healthcare providers involved in the follow-up of patients with IEMs about the importance of administering vaccinations at the ages recommended by the vaccination schedule. Indeed, achieving timely vaccination coverage against vaccine-preventable diseases is crucial, especially among children with at-risk IEMs. These conditions can indeed decompensate during infections, leading to a catabolic state. It can also increase the vulnerability of these patients to infection complications. Finally, it is also necessary to optimize vaccinations with specific recommendations, such as annual influenza or varicella vaccination, for patients with particular vulnerabilities among those with IEMs.
Supplementary material 1
|
Exercise-based cardiac rehabilitation for patients with atrial fibrillation receiving catheter ablation: protocol for a feasibility randomised controlled trial (RCT) with embedded process evaluation | 58470fe0-beef-45d8-8dd4-26208a3518a8 | 11784165 | Surgical Procedures, Operative[mh] | Atrial fibrillation (AF) is the most common cardiac dysrhythmia, with an estimated global prevalence of 46.3 million. AF occurs due to disorganised electrical activity in the atrial chambers of the heart and can present with symptoms such as heart palpitations, breathlessness and fatigue. While AF is not a direct cause of death, it is associated with substantial morbidity and mortality from stroke, heart failure, ans impaired quality of life (QoL). It can be classified into three primary clinical subtypes, indicative of its duration and severity: paroxysmal AF, persistent AF and permanent AF. Interventions to reduce AF burden, prevent disease progression and reduce symptoms typically include pharmacological treatments such as antiarrhythmic drug therapy and catheter ablation. This latter, more invasive therapy, uses heat or freeze cauterisation methods to destroy the abnormal heart tissue causing the erratic electrical activity, thus restoring sinus rhythm. Its main function is to prolong the duration of sinus rhythm, as well as reduce the number of acute AF episodes. It has an established value in AF care, evidenced by its superiority over antiarrhythmic drug therapy for long-term outcomes such as stroke, hospitalisation, higher exercise capacity and higher QoL. Evidence from both randomised clinical trials (RCTs) and observational studies indicates that exercise-based interventions, including exercise-based cardiac rehabilitation (ExCR), can improve cardiorespiratory fitness, positively contribute to heart rate regulation, reduce symptom burden, decrease depression and anxiety and increase the QoL for patients with AF. There is emerging evidence that lifestyle modifications, including exercise and physical activity, have the potential to improve outcomes following catheter ablation for patients with AF. There is limited but promising evidence for the role of ExCR in patients with AF receiving catheter ablation leading to favourable outcomes, including improvements in physical capacity and reduced anxiety. Due to limited available data, ExCR is not part of usual care for people with AF. A recent Cochrane review of ExCR for people with AF included 20 trials (n=2039 patients) demonstrated improved AF-specific outcomes and QoL for patients with AF in general, but only six trials included patients receiving catheter ablation, none of which were based in the UK. Collectively, the catheter ablation and ExCR trials were underpowered to investigate long-term clinical outcomes for patients with AF, and the feasibility of adding ExCR to usual care for patients referred for catheter ablation is unknown. In light of these preliminary findings, investigation into ExCR for patients with AF awaiting catheter ablation is needed to determine if it is acceptable and feasible within UK infrastructure; this will then inform a future definitive trial. Study aims The purpose of the study is to determine the feasibility and acceptability of delivering an existing ExCR programme versus usual care to aid the treatment of patients with AF receiving catheter ablation through a feasibility RCT. The feasibility RCT with embedded process evaluation will be conducted to inform the design and conduct of a future larger-scale effectiveness trial for which separate funding would be sought. The overall aims of this essential preliminary work are: To test the feasibility and acceptability of an existing ExCR programme when delivered to symptomatic patients with AF being referred for catheter ablation. Assess the feasibility and acceptability of the proposed trial methods and outcome measurements. Study objectives To explore the facilitators and challenges of delivering an existing ExCR programme that focuses on exercise training, physical activity and self-management of AF-related cardiovascular risk factors to people with AF receiving catheter ablation. To determine the feasibility and acceptability of the cardiac rehabilitation programme for patients with AF receiving catheter ablation. Investigate the mechanisms and processes that explain the implementation and preliminary effectiveness of the rehabilitation programme (intervention group) and usual care (control group). Investigate preliminary effectiveness through assessing improvements in (1) symptom severity and burden, (2) AF recurrence, (3) QoL, (4) exercise capacity and (5) cardiac structure and function, and whether this is maintained following the ExCR programme for patients with AF. To provide data for sample size calculation for a RCT.
The purpose of the study is to determine the feasibility and acceptability of delivering an existing ExCR programme versus usual care to aid the treatment of patients with AF receiving catheter ablation through a feasibility RCT. The feasibility RCT with embedded process evaluation will be conducted to inform the design and conduct of a future larger-scale effectiveness trial for which separate funding would be sought. The overall aims of this essential preliminary work are: To test the feasibility and acceptability of an existing ExCR programme when delivered to symptomatic patients with AF being referred for catheter ablation. Assess the feasibility and acceptability of the proposed trial methods and outcome measurements.
To explore the facilitators and challenges of delivering an existing ExCR programme that focuses on exercise training, physical activity and self-management of AF-related cardiovascular risk factors to people with AF receiving catheter ablation. To determine the feasibility and acceptability of the cardiac rehabilitation programme for patients with AF receiving catheter ablation. Investigate the mechanisms and processes that explain the implementation and preliminary effectiveness of the rehabilitation programme (intervention group) and usual care (control group). Investigate preliminary effectiveness through assessing improvements in (1) symptom severity and burden, (2) AF recurrence, (3) QoL, (4) exercise capacity and (5) cardiac structure and function, and whether this is maintained following the ExCR programme for patients with AF. To provide data for sample size calculation for a RCT.
Trial design This is a two-arm feasibility RCT comprising intervention testing, feasibility work and process evaluation. Participants will be randomised to either ExCR (intervention group) or standard care (control group). Outcome assessments will be completed three times: (1) preintervention, (2) after ExCR and (3) at 6-month follow-up . Investigations into whether an economic evaluation will be feasible are based on the completion of the five-level EuroQol-five dimensions (EQ-5D). Aligned with the Medical Research Council (MRC) guidance, this feasibility study will include a qualitative process evaluation in order to ensure that both the intervention and trial procedures are optimised, thus increasing the likelihood that a full-scale trial will generate the desired outcomes, and it can be successfully incorporated into routine care. The trial is described in accordance with the current Standard Protocol Items: Recommendations for Interventional Trials guidelines. Results will be reported following the CONsolidated Standard Of Reporting Trials guidelines for non-pharmacological interventions. Study setting The study will be conducted at a single National Health Service (NHS) centre in North-West England: Liverpool Heart and Chest Hospital, where clinical exercise physiologists deliver an existing cardiac rehabilitation service. Participants Consecutive patients on the waiting list for catheter ablation at Liverpool Heart and Chest will be screened for inclusion and approached for trial participation. We will recruit from catheter ablation waiting lists; however, the ExCR programme may be initiated pre-procedure or post-procedure, depending on patient preference. For example, although patients will consent prior to their catheter ablation procedure, they may choose to initiate the rehabilitation programme post-procedure. The feasibility of ExCR delivery pre versus post-ablation will be investigated. Patients aged >18 years, able to provide verbal and written informed consent, will be eligible for participation. Patients who are unable to understand trial instructions, have reduced ability to follow the planned programme due to other physical illnesses, or who do not wish to participate, as well as patients already enrolled in clinical trials that prohibit participation in additional trials, will be excluded. Detailed inclusion criteria The participant is willing and able to give informed consent for participation in the study. Aged ≥18 years. Diagnosed with AF and on a waiting list or referred for medical treatment for symptomatic AF (eg, catheter ablation). Is eligible and willing to take part in an ExCR programme. Detailed exclusion criteria Blood pressure >180/100. Unstable angina. Severe valvular heart disease as diagnosed by echocardiography. Heart failure New York Heart Association class 4. <6 months post-transplant. Resting/uncontrolled tachycardia. Stroke in the last 6 weeks. Cardiac sarcoidosis. Injury or disability preventing exercise. Inability to understand trial procedures, for example, difficulties with speaking and understanding the English language. Recruitment, consent, randomisation and data collection Recruitment and consent Recruitment is planned for between August 2024 and March 2025. Patients with AF The following recruitment methods will be employed for patients with AF. The primary care team will advertise the study in face-to-face consultations and via a poster sent to all patients on the waiting list. Verbal consent will be taken for the researcher to contact patients who are interested. Patients who decline to participate after having read the study information will be asked if they would be willing to share their reasons for declining to participate. Participants who agree to take part will be asked to provide informed written consent. A member of the cardiac multidisciplinary team will contact patients with AF and invite them to attend a cardiac rehabilitation clinical/risk stratification assessment to determine whether the patient will be able to safely exercise and plan physical activity goals tailored to individual patient needs. Patients who are deemed safe to exercise will be invited to attend cardiac rehabilitation classes. Patients who decline to attend a clinical/risk stratification assessment will be asked if they would be willing to share their reasons for declining to participate. Clinical exercise physiologists All clinical exercise physiologists involved in screening, recruitment and delivery of the intervention will be approached by a researcher and invited to attend a semistructured, face-to-face interview about their experiences of the trial procedures and the intervention. Data collection A screening and recruitment log will be completed by a researcher to document all patients considered for the study and subsequently included or excluded at each stage of the recruitment process and the reasons given. This will include information such as when the patient was given information about the study, referred to cardiac rehabilitation, attended for clinical/risk assessment and received an offer to attend rehabilitation classes. On completion of rehabilitation, semistructured face-to-face interviews will be conducted with a subsample of patients and clinical exercise physiologists involved in delivering the intervention in order to gather responses about the acceptability of the intervention and trial procedures. A recruitment flow chart will be produced to identify patient numbers throughout the recruitment process. Outcome measures will be taken at baseline, upon immediate completion of the intervention and at a 6-month follow-up. Outcome measures will be collected at baseline, on completion of the intervention and at 6-month follow-up. Randomisation and blinding Participants will be randomised to the intervention or control group after they have consented to participate in the study and after baseline measures have been collected. Participants will be randomised to the intervention (cardiac rehabilitation) or control (standard care) on a 1:1 basis using a computer-generated random allocation sequence. To ensure allocation concealment, researchers will request randomisation on completion of all baseline assessments. Due to the nature of the intervention, blinding of the participants or healthcare practitioners delivering the interventions is not possible. Outcome measures Primary outcome Throughout the study, information will be collected on (1) the total number of patients screened, eligible and approached, (2) the percentage of patients who (a) decline ExCR (including reasons for declining), (b) agree to ExCR and (c) consent to being part of the study; (3) the percentage of patients who take up ExCR and reasons for dropping out of ExCR before the end of the intervention (if provided) and (4) the percentage of participants who complete outcome assessments and reasons for dropping out (if provided). Age, sex, ethnicity, reason for enrolment into ExCR, the centre referred to, education and employment status will also be collected via an initial screening telephone/video call. Secondary outcomes The proposed outcomes in a future definitive trial will likely include AF recurrence, AF severity/burden, QoL, and exercise capacity, comparing the intervention with usual care. Data on these will be collected in the proposed feasibility trial. All outcomes will be assessed in all participants at baseline, post-intervention and at 6-month follow-up and the respective data for these will be collected in the proposed feasibility trial. AF recurrence AF recurrence will be measured using the AliveCor KardiaMobile, which is a hand-held, one-lead ECG device. The device provides three potential outcomes once the reading is complete: ‘possible AF’, ‘normal’ and ‘unclassified’. Participants will be asked to measure their heart rate three times a day for a 7-day period, regardless of symptoms and, additionally, if they develop any symptoms. They will be asked to do this at three separate time points: at baseline (preintervention), immediately postintervention and at 6-month follow-up. Feasibility of the time frame of device use will be assessed along with the percentage of ‘uncertain’ ECG recordings. If AF recurrence is identified, the participant will be referred to secondary care for a 12-lead ECG or 14-day patch as necessary. Transthoracic echocardiography (echo) A cardiac (heart) ultrasound will be performed by a clinically accredited echocardiographer (J Maxwell BSE 32820). Cardiac structure and function will be assessed non-invasively with the participant lying on their left side. Standard two-dimensional and three-dimensional Doppler, tissue-Doppler and M-mode scans will be performed using a commercially available ultrasound system (Vivid iQ, GE Medical, Horton, Norway) with a 1.5–4 MHz phased array transducer applied to the participant’s chests. All participants will be offered a chaperone for the scan and provided with a gown. Images will be stored and analysed offline on password-protected monitors. Exercise capacity Exercise capacity (peak oxygen consumption (VO 2 peak)) will be measured using cardiopulmonary exercise testing (CPET). The test is performed according to current guidelines for ergo spirometry testing and by an ergometer bicycle, simultaneously monitoring heart rhythm, blood pressure, ECG and measuring gas exchange during workload and in the following recovery period. The average test duration is 10–15 min, including the pretest and post-test phases without workload. Before each session, calibration is performed to address changes in room temperature, humidity and air oxygen content. A standardised ramp protocol is used with an initial workload of 25/50 watts, increasing gradually by 25 W load increasing for every 2 min until peak exhaustion is reached or the participant can no longer continue. Peak exhaustion will be evaluated by several variables (eg, respiratory exchange ratio ≥T 1.10, heart rate and subjective exhaustion of the patient). During the test period, clinical manifestations, ECG abnormalities (ST depression, ST elevation, Q wave and T wave changes, supraventricular or ventricular arrhythmias), blood pressure response and several physiological variables are observed and documented. VO 2 peak will be defined as the peak VO 2 reached during the test. The test will be performed by two members of the research team. For safety reasons, preset criteria for initiation and/or termination of the test have been defined. If the CPET is not possible, that is, the participant is unable to complete it for whatever reason, the maximum walking distance (in metres) within 6 minutes will be used to assess exercise capacity, using standardised instructions, while subjective exhaustion with regard to fatigue and dyspnoea using the Borg Scale will be administered. Questionnaires Quality of life All patients will complete two QoL-related questionnaires and their perception of treatment. The internationally validated Atrial Fibrillation Effect on Quality-of-Life (AFEQT; http://www.afeqt.org ) questionnaire. The AFEQT questionnaire is a 20-item questionnaire that quantifies four domains of AF-related QoL, including symptoms, daily activities, treatment concern and treatment satisfaction, by using seven-point Likert response scales. An overall summary score can be calculated from the first three domains, which range from 0 to 100 (100, best possible health status (no impairment); 0, worst health status). The EQ-5D is a parsimonious measure of health-related QoL consisting of five dimensions: mobility, self-care, usual activities, pain/discomfort and anxiety/depression. The EQ-5D will also be used to inform economic evaluation. AF severity and burden AF burden will be measured using the University of Toronto Atrial Fibrillation Severity Scale (AFSS), which is a 19-item self-administered questionnaire developed to obtain both objective and subjective ratings of AF-related symptoms, AF disease burden, including frequency, duration and severity of episodes, as well as healthcare utilisation and AF disease burden, including frequency, duration and severity of episodes. The AF symptom burden score is derived from the AFSS summary score that averages the frequency, duration and patient-perceived severity of AF episodes. A higher score indicates a greater AF burden. Quantitative analysis The proportion of eligible patients who consent to participate will be presented, along with the proportions in each intervention group completing each follow-up assessment and the reasons for withdrawal. Descriptive characteristics and outcome data will be summarised overall and by intervention group, as mean (SD) for normally distributed continuous variables, median (IQR) for non-normally distributed continuous variables and number (percentage) for categorical variables. Quantitative data (recruitment logs) will be analysed using descriptive statistics. Uptake will be measured as the percentage of eligible participants agreeing to participate, retention will be measured as the percentage of participants remaining until the close of the study and compliance will be measured by the percentage of compliance to the ExCR programme up until the close of the study. All statistical analysis of quantitative data will be conducted in IBM SPSS Statistics for Windows (V.28.0) and R. Semistructured interviews and focus groups Process evaluation A concurrent mixed-methods process evaluation using the MRC framework for complex interventions will be used to identify and explain the mechanisms and processes that enabled/acted as barriers to the implementation of the ExCR programme. We will determine how the intervention was delivered and identify ‘active ingredients’ by examining referral and recruitment rates, compliance with the ExCR and loss to follow-up. This will aid in identifying the mechanisms and processes that enable or challenge the delivery of the ExCR for patients with AF. Postintervention all clinical exercise physiologists will be interviewed, and a focus group will be conducted with a small subsample of patients. A topic guide for both will be used with questions relating to the acceptability of the rehabilitation programme and barriers/facilitators of uptake and completion. All interviews will be audio-recorded and transcribed verbatim. Patient focus groups For participants who consented, focus groups will be conducted immediately postservice. Participants will complete focus groups with a member of the Liverpool John Moores University (LJMU) team, guided by a topic guide. Postservice-guided discussions will aim to learn about experiences (barriers, facilitators, receptivity to the clinical exercise physiologists, perceived appropriateness and suggestions for improvement) of the ExCR programme. Guided discussions at follow-up (6 months) will aim to learn about the engagement of physical activity both during the cardiac rehabilitation sessions and self-directed exercises at home, as well as the continuity of exercise and physical activity after the service concluded. This will aid in identifying any causal mechanisms of the ExCR programme related to self-directed activity. Focus groups will be conducted via in-person/phone/video call per patient preference. Clinical exercise physiologists interviews Clinical exercise physiologists will be invited to attend interviews after the evaluation period. Clinical exercise physiologists will be given a participant information sheet and have the opportunity to ask questions before providing written informed consent. Interviews will be led by a member of the research team and guided by a topic guide. Guided discussion will aim to learn about experiences of delivering the service, including barriers and enablers to intervention delivery, adaptations made to the service framework and ways to improve the service. Interviews will be in-person or via telephone or video call depending on preference. Qualitative analysis Interviews will be audio-recorded and transcribed verbatim, deductively coded and analysed using the theoretical domains framework, enabling challenges and facilitators within the intervention to be identified. Transcriptions will be thematically analysed and coded using NVivo V.12 software. Data will be thematically analysed using reflexive thematic analysis recommendations such as data familiarisation, generating initial themes, coding and finalising patterns of shared meanings underpinned by a central concept and writting them up using data extracts interspersed with researcher interpretations. Although the data themes will be created deductively, the patterns of shared meaning will be inductively generated from the data themselves, allowing interpretation and researcher contextual awareness to be discussed. Member checking will be the final step in the analysis, ensuring that interviewed participants have the opportunity to confirm the researcher’s interpretation and add comments that will be incorporated into the final analysis. Our aim is to develop a comprehensive understanding of the intervention’s acceptability, implementation and mechanisms of impact. Economic evaluation We will explore the feasibility of an economic evaluation alongside the trial to assess the cost-utility of cardiac rehabilitation compared with treatment as usual. The economic evaluation will compare the costs to quality-adjusted life years (QALYs) and take a societal perspective, as recommended nationally. QALYs will be estimated using the self-reported EQ-5D instrument, which is a standardised instrument assessing five dimensions of self-reported health status (mobility, self-care, usual activities, pain/discomfort and anxiety/depression). The EQ-5D is a useful tool for facilitating the calculation of QALYs that are in turn used to inform economic evaluations of healthcare interventions or policies on health. During the interventions, researcher time per participant will also be recorded in both groups. Intervention Cardiac rehabilitation intervention group The intervention is a comprehensive cardiac rehabilitation programme delivered to patients with AF receiving catheter ablation. The structured exercise service will consist of one supervised exercise session per week for 8 weeks at the chosen centre. In addition, patients will be encouraged to undertake 1–2 home-based exercise sessions per week. Patients will also be invited to a weekly education session and encouraged to increase physical activity levels outside of the exercise sessions. Exercise sessions are circuit-based, cardiovascular and strength exercises of light to moderate intensity (monitored via the Borg rating of perceived effort scale). Classes start with a 15-min warm-up session, followed by the main exercise session lasting 20–30 min. Exercises are designed to improve cardiovascular fitness and include strength training using weights as well as walking, squats, shoulder raises, bicep curls and tricep extensions. Patients will be advised to exercise at a moderate intensity. The classes conclude with a 10-min cooldown. Participants will set individual physical activity goals with advice and support from the clinical exercise physiologist. To optimise intervention fidelity, professionals delivering the intervention will be expected to follow their site-specific rehabilitation programme. Usual care control group Participants randomised to usual care will not receive any intervention but continue with usual medical treatment for their AF as determined by their healthcare team. Study withdrawal Each participant has the right to withdraw from the study at any time with no obligation to provide a reason. In addition, participants may be withdrawn from the study by the research team at any time if the research team considers it necessary for any reason, including: Ineligibility (either arising during the study or retrospectively having been overlooked at screening). Significant protocol deviation. Significant non-compliance with treatment regimen or study requirements. Withdrawal of consent. Loss to follow-up. Withdrawal from the study will not result in the exclusion of the participant’s data from analysis, including audio recordings that have already been transcribed, as all of this data will be pseudonymised. Withdrawn participants will not be replaced. Participants will be asked the reasoning behind withdrawal either via email or phone call. The reason will be recorded in the study file. Participants are free to give no reason. Serious adverse event reporting and monitoring All adverse events will be reported and assignment of the severity/grading (mild, moderate, severe, life-threatening and death) will be made by the investigator responsible for the care of the participant. The assignment of causality will be made by an independent clinician in the UK. All non-serious adverse events, whether expected or not, will be recorded and updated at each visit. All new SAEs will be reported from the point of consent until follow-up. Investigators will report SAEs to the sponsor within 24 hours of the local site becoming aware of the event. All adverse events will be followed until satisfactory resolution. Data management Data will be collected and stored in accordance with the Data Protection Act 1998/General Data Protection Regulation 2018 in the UK. All data will be entered electronically. Outcomes and questionnaire data collected by participants will be reported using online survey software (Microsoft Forms; www.forms.office.com ). The administrative database (ie, participant information) and trial data will be managed by the research teams. Random checks will be performed on the entered data against online records. All errors will be logged and corrected. All data will be stored on password-protected and encrypted computers. Participant files will be maintained in storage for a period of 15 years after completion of the trial, with access granted to the local research team only. Our intended policy is that the research team should have exclusive use of the data for a period of 12 months or until the data are published. Data will be shared with named collaborators during this time. Following this period, data will be made publicly available through the LJMU Data Repository and published under a permissive reuse license. Sample size calculation A power calculation is not appropriate as the study does not aim to provide a definitive estimate of the treatment effect. Rather, the aim is to provide robust estimates of the likely rates of recruitment and retention and to yield estimates of the variability of the primary and secondary outcomes to inform power calculations for a future full-scale trial. Our aim is to recruit 60 participants, and this is based on recommendations for pilot/feasibility studies and an audit reporting pilot and feasibility trials registered in the UK clinical research network. Trial oversight The quality of the study will be assured through the series of management groups. The trial will be overseen by a trial steering committee (TSC) and operated on a day-to-day basis by a trial delivery group (TDG). The TSC will comprise experienced academic experts (the research team), clinicians and patients but does not require and, therefore, will not have an independent chair. The TSC will meet quarterly to discuss progress. The role of the TSC is to provide overall supervision of the trial. In particular, the TSC will concentrate on the progress of the trial, adherence to the protocol, participant safety and consideration of new information. The TSC must be in agreement with the final protocol and, throughout the trial, will take responsibility for major decisions, such as the need to change the protocol for any reason, monitoring and supervising the progress of the trial, reviewing relevant information from other sources and informing and advising the TDG on all aspects of the trial. The TDG will comprise the same research team and will hold monthly meetings to discuss progress. The responsibilities of the TDG will include: Report to the TSC. Maintain the trial master file. Confirm all approvals are in place before the start of the trial at a site. Provide study materials. Data management centre. Give collaborators regular information about the progress of the study. Respond to any questions (eg, from collaborators) about the trial. Ensure data security and quality and observe data protection laws. Safety reporting. Ensure the trial is conducted in accordance with Good Clinical Practice. Statistical analysis. Publication of trial results. Patient and public involvement National Institute for Health and Care Research North West Coast-funded patient and public involvement work with patients with AF has been completed. A patient involvement workshop (n=24) exploring their views of incorporating AF rehabilitation into usual care was conducted. All had positive responses regarding the potential to incorporate AF rehabilitation into usual care. Developing an intervention that supports patients with AF to become more physically active in the long term is therefore wanted by patients, is of great potential NHS benefit, and thus fills an important clinical and scientific ‘gap’. Workshops with healthcare practitioners (including clinical exercise physiologists and electrophysiologist consultants) were also conducted to determine any facilitators and barriers to pathway referral and rehabilitation delivery to patients with AF. Moving forward, two patient representatives have been invited to the TSC. They will advise on study information materials to recruit participants to the study. At the end of the project, our patient representatives will contribute to the reporting of the study by reading and reviewing the ‘lay’ sections of the report. They will also be involved in the dissemination of research findings through reviewing literature outlining the results before they are circulated. Ethics and dissemination The trial has received a favourable ethical opinion from the Preston NHS Research Ethics Committee (22/NW/0061) in the UK. On study completion, the CI owns the data. On completion of the study, the data will be analysed, and results will be disseminated via publication in clinical and physiological journals, presented at national and international conferences and in the form of feedback sheets or perhaps local articles. Participants will not be identifiable from the results of the study. Pseudonymised data from this study will be made available for sharing with other investigators after the publication of the study’s key papers. Data will be shared through the LJMU Data Repository ( http://opendata.ljmu.ac.uk/ ). This is a secure institutional data repository, which is searchable on the www, and it is managed by Library Services. A DOI will be generated for datasets as they are deposited to the repository. Data will be stored in this repository for a minimum of 10 years or for 10 years from the last date of access. It is our intention to present our research findings to all our research participants in a written lay summary and hold an open feedback session where the results will be presented in a lay-friendly manner. We plan to present the scientific findings as oral communications and abstracts at regional, national and international scientific meetings related to AF and preventative cardiology. We also intend to publish our findings in peer-reviewed journals.
This is a two-arm feasibility RCT comprising intervention testing, feasibility work and process evaluation. Participants will be randomised to either ExCR (intervention group) or standard care (control group). Outcome assessments will be completed three times: (1) preintervention, (2) after ExCR and (3) at 6-month follow-up . Investigations into whether an economic evaluation will be feasible are based on the completion of the five-level EuroQol-five dimensions (EQ-5D). Aligned with the Medical Research Council (MRC) guidance, this feasibility study will include a qualitative process evaluation in order to ensure that both the intervention and trial procedures are optimised, thus increasing the likelihood that a full-scale trial will generate the desired outcomes, and it can be successfully incorporated into routine care. The trial is described in accordance with the current Standard Protocol Items: Recommendations for Interventional Trials guidelines. Results will be reported following the CONsolidated Standard Of Reporting Trials guidelines for non-pharmacological interventions.
The study will be conducted at a single National Health Service (NHS) centre in North-West England: Liverpool Heart and Chest Hospital, where clinical exercise physiologists deliver an existing cardiac rehabilitation service.
Consecutive patients on the waiting list for catheter ablation at Liverpool Heart and Chest will be screened for inclusion and approached for trial participation. We will recruit from catheter ablation waiting lists; however, the ExCR programme may be initiated pre-procedure or post-procedure, depending on patient preference. For example, although patients will consent prior to their catheter ablation procedure, they may choose to initiate the rehabilitation programme post-procedure. The feasibility of ExCR delivery pre versus post-ablation will be investigated. Patients aged >18 years, able to provide verbal and written informed consent, will be eligible for participation. Patients who are unable to understand trial instructions, have reduced ability to follow the planned programme due to other physical illnesses, or who do not wish to participate, as well as patients already enrolled in clinical trials that prohibit participation in additional trials, will be excluded.
The participant is willing and able to give informed consent for participation in the study. Aged ≥18 years. Diagnosed with AF and on a waiting list or referred for medical treatment for symptomatic AF (eg, catheter ablation). Is eligible and willing to take part in an ExCR programme.
Blood pressure >180/100. Unstable angina. Severe valvular heart disease as diagnosed by echocardiography. Heart failure New York Heart Association class 4. <6 months post-transplant. Resting/uncontrolled tachycardia. Stroke in the last 6 weeks. Cardiac sarcoidosis. Injury or disability preventing exercise. Inability to understand trial procedures, for example, difficulties with speaking and understanding the English language.
Recruitment and consent Recruitment is planned for between August 2024 and March 2025. Patients with AF The following recruitment methods will be employed for patients with AF. The primary care team will advertise the study in face-to-face consultations and via a poster sent to all patients on the waiting list. Verbal consent will be taken for the researcher to contact patients who are interested. Patients who decline to participate after having read the study information will be asked if they would be willing to share their reasons for declining to participate. Participants who agree to take part will be asked to provide informed written consent. A member of the cardiac multidisciplinary team will contact patients with AF and invite them to attend a cardiac rehabilitation clinical/risk stratification assessment to determine whether the patient will be able to safely exercise and plan physical activity goals tailored to individual patient needs. Patients who are deemed safe to exercise will be invited to attend cardiac rehabilitation classes. Patients who decline to attend a clinical/risk stratification assessment will be asked if they would be willing to share their reasons for declining to participate. Clinical exercise physiologists All clinical exercise physiologists involved in screening, recruitment and delivery of the intervention will be approached by a researcher and invited to attend a semistructured, face-to-face interview about their experiences of the trial procedures and the intervention.
Recruitment is planned for between August 2024 and March 2025. Patients with AF The following recruitment methods will be employed for patients with AF. The primary care team will advertise the study in face-to-face consultations and via a poster sent to all patients on the waiting list. Verbal consent will be taken for the researcher to contact patients who are interested. Patients who decline to participate after having read the study information will be asked if they would be willing to share their reasons for declining to participate. Participants who agree to take part will be asked to provide informed written consent. A member of the cardiac multidisciplinary team will contact patients with AF and invite them to attend a cardiac rehabilitation clinical/risk stratification assessment to determine whether the patient will be able to safely exercise and plan physical activity goals tailored to individual patient needs. Patients who are deemed safe to exercise will be invited to attend cardiac rehabilitation classes. Patients who decline to attend a clinical/risk stratification assessment will be asked if they would be willing to share their reasons for declining to participate. Clinical exercise physiologists All clinical exercise physiologists involved in screening, recruitment and delivery of the intervention will be approached by a researcher and invited to attend a semistructured, face-to-face interview about their experiences of the trial procedures and the intervention.
The following recruitment methods will be employed for patients with AF. The primary care team will advertise the study in face-to-face consultations and via a poster sent to all patients on the waiting list. Verbal consent will be taken for the researcher to contact patients who are interested. Patients who decline to participate after having read the study information will be asked if they would be willing to share their reasons for declining to participate. Participants who agree to take part will be asked to provide informed written consent. A member of the cardiac multidisciplinary team will contact patients with AF and invite them to attend a cardiac rehabilitation clinical/risk stratification assessment to determine whether the patient will be able to safely exercise and plan physical activity goals tailored to individual patient needs. Patients who are deemed safe to exercise will be invited to attend cardiac rehabilitation classes. Patients who decline to attend a clinical/risk stratification assessment will be asked if they would be willing to share their reasons for declining to participate.
All clinical exercise physiologists involved in screening, recruitment and delivery of the intervention will be approached by a researcher and invited to attend a semistructured, face-to-face interview about their experiences of the trial procedures and the intervention.
A screening and recruitment log will be completed by a researcher to document all patients considered for the study and subsequently included or excluded at each stage of the recruitment process and the reasons given. This will include information such as when the patient was given information about the study, referred to cardiac rehabilitation, attended for clinical/risk assessment and received an offer to attend rehabilitation classes. On completion of rehabilitation, semistructured face-to-face interviews will be conducted with a subsample of patients and clinical exercise physiologists involved in delivering the intervention in order to gather responses about the acceptability of the intervention and trial procedures. A recruitment flow chart will be produced to identify patient numbers throughout the recruitment process. Outcome measures will be taken at baseline, upon immediate completion of the intervention and at a 6-month follow-up. Outcome measures will be collected at baseline, on completion of the intervention and at 6-month follow-up.
Participants will be randomised to the intervention or control group after they have consented to participate in the study and after baseline measures have been collected. Participants will be randomised to the intervention (cardiac rehabilitation) or control (standard care) on a 1:1 basis using a computer-generated random allocation sequence. To ensure allocation concealment, researchers will request randomisation on completion of all baseline assessments. Due to the nature of the intervention, blinding of the participants or healthcare practitioners delivering the interventions is not possible.
Primary outcome Throughout the study, information will be collected on (1) the total number of patients screened, eligible and approached, (2) the percentage of patients who (a) decline ExCR (including reasons for declining), (b) agree to ExCR and (c) consent to being part of the study; (3) the percentage of patients who take up ExCR and reasons for dropping out of ExCR before the end of the intervention (if provided) and (4) the percentage of participants who complete outcome assessments and reasons for dropping out (if provided). Age, sex, ethnicity, reason for enrolment into ExCR, the centre referred to, education and employment status will also be collected via an initial screening telephone/video call. Secondary outcomes The proposed outcomes in a future definitive trial will likely include AF recurrence, AF severity/burden, QoL, and exercise capacity, comparing the intervention with usual care. Data on these will be collected in the proposed feasibility trial. All outcomes will be assessed in all participants at baseline, post-intervention and at 6-month follow-up and the respective data for these will be collected in the proposed feasibility trial.
Throughout the study, information will be collected on (1) the total number of patients screened, eligible and approached, (2) the percentage of patients who (a) decline ExCR (including reasons for declining), (b) agree to ExCR and (c) consent to being part of the study; (3) the percentage of patients who take up ExCR and reasons for dropping out of ExCR before the end of the intervention (if provided) and (4) the percentage of participants who complete outcome assessments and reasons for dropping out (if provided). Age, sex, ethnicity, reason for enrolment into ExCR, the centre referred to, education and employment status will also be collected via an initial screening telephone/video call.
The proposed outcomes in a future definitive trial will likely include AF recurrence, AF severity/burden, QoL, and exercise capacity, comparing the intervention with usual care. Data on these will be collected in the proposed feasibility trial. All outcomes will be assessed in all participants at baseline, post-intervention and at 6-month follow-up and the respective data for these will be collected in the proposed feasibility trial.
AF recurrence will be measured using the AliveCor KardiaMobile, which is a hand-held, one-lead ECG device. The device provides three potential outcomes once the reading is complete: ‘possible AF’, ‘normal’ and ‘unclassified’. Participants will be asked to measure their heart rate three times a day for a 7-day period, regardless of symptoms and, additionally, if they develop any symptoms. They will be asked to do this at three separate time points: at baseline (preintervention), immediately postintervention and at 6-month follow-up. Feasibility of the time frame of device use will be assessed along with the percentage of ‘uncertain’ ECG recordings. If AF recurrence is identified, the participant will be referred to secondary care for a 12-lead ECG or 14-day patch as necessary.
A cardiac (heart) ultrasound will be performed by a clinically accredited echocardiographer (J Maxwell BSE 32820). Cardiac structure and function will be assessed non-invasively with the participant lying on their left side. Standard two-dimensional and three-dimensional Doppler, tissue-Doppler and M-mode scans will be performed using a commercially available ultrasound system (Vivid iQ, GE Medical, Horton, Norway) with a 1.5–4 MHz phased array transducer applied to the participant’s chests. All participants will be offered a chaperone for the scan and provided with a gown. Images will be stored and analysed offline on password-protected monitors.
Exercise capacity (peak oxygen consumption (VO 2 peak)) will be measured using cardiopulmonary exercise testing (CPET). The test is performed according to current guidelines for ergo spirometry testing and by an ergometer bicycle, simultaneously monitoring heart rhythm, blood pressure, ECG and measuring gas exchange during workload and in the following recovery period. The average test duration is 10–15 min, including the pretest and post-test phases without workload. Before each session, calibration is performed to address changes in room temperature, humidity and air oxygen content. A standardised ramp protocol is used with an initial workload of 25/50 watts, increasing gradually by 25 W load increasing for every 2 min until peak exhaustion is reached or the participant can no longer continue. Peak exhaustion will be evaluated by several variables (eg, respiratory exchange ratio ≥T 1.10, heart rate and subjective exhaustion of the patient). During the test period, clinical manifestations, ECG abnormalities (ST depression, ST elevation, Q wave and T wave changes, supraventricular or ventricular arrhythmias), blood pressure response and several physiological variables are observed and documented. VO 2 peak will be defined as the peak VO 2 reached during the test. The test will be performed by two members of the research team. For safety reasons, preset criteria for initiation and/or termination of the test have been defined. If the CPET is not possible, that is, the participant is unable to complete it for whatever reason, the maximum walking distance (in metres) within 6 minutes will be used to assess exercise capacity, using standardised instructions, while subjective exhaustion with regard to fatigue and dyspnoea using the Borg Scale will be administered.
Quality of life All patients will complete two QoL-related questionnaires and their perception of treatment. The internationally validated Atrial Fibrillation Effect on Quality-of-Life (AFEQT; http://www.afeqt.org ) questionnaire. The AFEQT questionnaire is a 20-item questionnaire that quantifies four domains of AF-related QoL, including symptoms, daily activities, treatment concern and treatment satisfaction, by using seven-point Likert response scales. An overall summary score can be calculated from the first three domains, which range from 0 to 100 (100, best possible health status (no impairment); 0, worst health status). The EQ-5D is a parsimonious measure of health-related QoL consisting of five dimensions: mobility, self-care, usual activities, pain/discomfort and anxiety/depression. The EQ-5D will also be used to inform economic evaluation. AF severity and burden AF burden will be measured using the University of Toronto Atrial Fibrillation Severity Scale (AFSS), which is a 19-item self-administered questionnaire developed to obtain both objective and subjective ratings of AF-related symptoms, AF disease burden, including frequency, duration and severity of episodes, as well as healthcare utilisation and AF disease burden, including frequency, duration and severity of episodes. The AF symptom burden score is derived from the AFSS summary score that averages the frequency, duration and patient-perceived severity of AF episodes. A higher score indicates a greater AF burden.
All patients will complete two QoL-related questionnaires and their perception of treatment. The internationally validated Atrial Fibrillation Effect on Quality-of-Life (AFEQT; http://www.afeqt.org ) questionnaire. The AFEQT questionnaire is a 20-item questionnaire that quantifies four domains of AF-related QoL, including symptoms, daily activities, treatment concern and treatment satisfaction, by using seven-point Likert response scales. An overall summary score can be calculated from the first three domains, which range from 0 to 100 (100, best possible health status (no impairment); 0, worst health status). The EQ-5D is a parsimonious measure of health-related QoL consisting of five dimensions: mobility, self-care, usual activities, pain/discomfort and anxiety/depression. The EQ-5D will also be used to inform economic evaluation.
AF burden will be measured using the University of Toronto Atrial Fibrillation Severity Scale (AFSS), which is a 19-item self-administered questionnaire developed to obtain both objective and subjective ratings of AF-related symptoms, AF disease burden, including frequency, duration and severity of episodes, as well as healthcare utilisation and AF disease burden, including frequency, duration and severity of episodes. The AF symptom burden score is derived from the AFSS summary score that averages the frequency, duration and patient-perceived severity of AF episodes. A higher score indicates a greater AF burden.
The proportion of eligible patients who consent to participate will be presented, along with the proportions in each intervention group completing each follow-up assessment and the reasons for withdrawal. Descriptive characteristics and outcome data will be summarised overall and by intervention group, as mean (SD) for normally distributed continuous variables, median (IQR) for non-normally distributed continuous variables and number (percentage) for categorical variables. Quantitative data (recruitment logs) will be analysed using descriptive statistics. Uptake will be measured as the percentage of eligible participants agreeing to participate, retention will be measured as the percentage of participants remaining until the close of the study and compliance will be measured by the percentage of compliance to the ExCR programme up until the close of the study. All statistical analysis of quantitative data will be conducted in IBM SPSS Statistics for Windows (V.28.0) and R.
Process evaluation A concurrent mixed-methods process evaluation using the MRC framework for complex interventions will be used to identify and explain the mechanisms and processes that enabled/acted as barriers to the implementation of the ExCR programme. We will determine how the intervention was delivered and identify ‘active ingredients’ by examining referral and recruitment rates, compliance with the ExCR and loss to follow-up. This will aid in identifying the mechanisms and processes that enable or challenge the delivery of the ExCR for patients with AF. Postintervention all clinical exercise physiologists will be interviewed, and a focus group will be conducted with a small subsample of patients. A topic guide for both will be used with questions relating to the acceptability of the rehabilitation programme and barriers/facilitators of uptake and completion. All interviews will be audio-recorded and transcribed verbatim. Patient focus groups For participants who consented, focus groups will be conducted immediately postservice. Participants will complete focus groups with a member of the Liverpool John Moores University (LJMU) team, guided by a topic guide. Postservice-guided discussions will aim to learn about experiences (barriers, facilitators, receptivity to the clinical exercise physiologists, perceived appropriateness and suggestions for improvement) of the ExCR programme. Guided discussions at follow-up (6 months) will aim to learn about the engagement of physical activity both during the cardiac rehabilitation sessions and self-directed exercises at home, as well as the continuity of exercise and physical activity after the service concluded. This will aid in identifying any causal mechanisms of the ExCR programme related to self-directed activity. Focus groups will be conducted via in-person/phone/video call per patient preference. Clinical exercise physiologists interviews Clinical exercise physiologists will be invited to attend interviews after the evaluation period. Clinical exercise physiologists will be given a participant information sheet and have the opportunity to ask questions before providing written informed consent. Interviews will be led by a member of the research team and guided by a topic guide. Guided discussion will aim to learn about experiences of delivering the service, including barriers and enablers to intervention delivery, adaptations made to the service framework and ways to improve the service. Interviews will be in-person or via telephone or video call depending on preference.
A concurrent mixed-methods process evaluation using the MRC framework for complex interventions will be used to identify and explain the mechanisms and processes that enabled/acted as barriers to the implementation of the ExCR programme. We will determine how the intervention was delivered and identify ‘active ingredients’ by examining referral and recruitment rates, compliance with the ExCR and loss to follow-up. This will aid in identifying the mechanisms and processes that enable or challenge the delivery of the ExCR for patients with AF. Postintervention all clinical exercise physiologists will be interviewed, and a focus group will be conducted with a small subsample of patients. A topic guide for both will be used with questions relating to the acceptability of the rehabilitation programme and barriers/facilitators of uptake and completion. All interviews will be audio-recorded and transcribed verbatim.
For participants who consented, focus groups will be conducted immediately postservice. Participants will complete focus groups with a member of the Liverpool John Moores University (LJMU) team, guided by a topic guide. Postservice-guided discussions will aim to learn about experiences (barriers, facilitators, receptivity to the clinical exercise physiologists, perceived appropriateness and suggestions for improvement) of the ExCR programme. Guided discussions at follow-up (6 months) will aim to learn about the engagement of physical activity both during the cardiac rehabilitation sessions and self-directed exercises at home, as well as the continuity of exercise and physical activity after the service concluded. This will aid in identifying any causal mechanisms of the ExCR programme related to self-directed activity. Focus groups will be conducted via in-person/phone/video call per patient preference.
Clinical exercise physiologists will be invited to attend interviews after the evaluation period. Clinical exercise physiologists will be given a participant information sheet and have the opportunity to ask questions before providing written informed consent. Interviews will be led by a member of the research team and guided by a topic guide. Guided discussion will aim to learn about experiences of delivering the service, including barriers and enablers to intervention delivery, adaptations made to the service framework and ways to improve the service. Interviews will be in-person or via telephone or video call depending on preference.
Interviews will be audio-recorded and transcribed verbatim, deductively coded and analysed using the theoretical domains framework, enabling challenges and facilitators within the intervention to be identified. Transcriptions will be thematically analysed and coded using NVivo V.12 software. Data will be thematically analysed using reflexive thematic analysis recommendations such as data familiarisation, generating initial themes, coding and finalising patterns of shared meanings underpinned by a central concept and writting them up using data extracts interspersed with researcher interpretations. Although the data themes will be created deductively, the patterns of shared meaning will be inductively generated from the data themselves, allowing interpretation and researcher contextual awareness to be discussed. Member checking will be the final step in the analysis, ensuring that interviewed participants have the opportunity to confirm the researcher’s interpretation and add comments that will be incorporated into the final analysis. Our aim is to develop a comprehensive understanding of the intervention’s acceptability, implementation and mechanisms of impact.
We will explore the feasibility of an economic evaluation alongside the trial to assess the cost-utility of cardiac rehabilitation compared with treatment as usual. The economic evaluation will compare the costs to quality-adjusted life years (QALYs) and take a societal perspective, as recommended nationally. QALYs will be estimated using the self-reported EQ-5D instrument, which is a standardised instrument assessing five dimensions of self-reported health status (mobility, self-care, usual activities, pain/discomfort and anxiety/depression). The EQ-5D is a useful tool for facilitating the calculation of QALYs that are in turn used to inform economic evaluations of healthcare interventions or policies on health. During the interventions, researcher time per participant will also be recorded in both groups.
Cardiac rehabilitation intervention group The intervention is a comprehensive cardiac rehabilitation programme delivered to patients with AF receiving catheter ablation. The structured exercise service will consist of one supervised exercise session per week for 8 weeks at the chosen centre. In addition, patients will be encouraged to undertake 1–2 home-based exercise sessions per week. Patients will also be invited to a weekly education session and encouraged to increase physical activity levels outside of the exercise sessions. Exercise sessions are circuit-based, cardiovascular and strength exercises of light to moderate intensity (monitored via the Borg rating of perceived effort scale). Classes start with a 15-min warm-up session, followed by the main exercise session lasting 20–30 min. Exercises are designed to improve cardiovascular fitness and include strength training using weights as well as walking, squats, shoulder raises, bicep curls and tricep extensions. Patients will be advised to exercise at a moderate intensity. The classes conclude with a 10-min cooldown. Participants will set individual physical activity goals with advice and support from the clinical exercise physiologist. To optimise intervention fidelity, professionals delivering the intervention will be expected to follow their site-specific rehabilitation programme.
The intervention is a comprehensive cardiac rehabilitation programme delivered to patients with AF receiving catheter ablation. The structured exercise service will consist of one supervised exercise session per week for 8 weeks at the chosen centre. In addition, patients will be encouraged to undertake 1–2 home-based exercise sessions per week. Patients will also be invited to a weekly education session and encouraged to increase physical activity levels outside of the exercise sessions. Exercise sessions are circuit-based, cardiovascular and strength exercises of light to moderate intensity (monitored via the Borg rating of perceived effort scale). Classes start with a 15-min warm-up session, followed by the main exercise session lasting 20–30 min. Exercises are designed to improve cardiovascular fitness and include strength training using weights as well as walking, squats, shoulder raises, bicep curls and tricep extensions. Patients will be advised to exercise at a moderate intensity. The classes conclude with a 10-min cooldown. Participants will set individual physical activity goals with advice and support from the clinical exercise physiologist. To optimise intervention fidelity, professionals delivering the intervention will be expected to follow their site-specific rehabilitation programme.
Participants randomised to usual care will not receive any intervention but continue with usual medical treatment for their AF as determined by their healthcare team.
Each participant has the right to withdraw from the study at any time with no obligation to provide a reason. In addition, participants may be withdrawn from the study by the research team at any time if the research team considers it necessary for any reason, including: Ineligibility (either arising during the study or retrospectively having been overlooked at screening). Significant protocol deviation. Significant non-compliance with treatment regimen or study requirements. Withdrawal of consent. Loss to follow-up. Withdrawal from the study will not result in the exclusion of the participant’s data from analysis, including audio recordings that have already been transcribed, as all of this data will be pseudonymised. Withdrawn participants will not be replaced. Participants will be asked the reasoning behind withdrawal either via email or phone call. The reason will be recorded in the study file. Participants are free to give no reason.
All adverse events will be reported and assignment of the severity/grading (mild, moderate, severe, life-threatening and death) will be made by the investigator responsible for the care of the participant. The assignment of causality will be made by an independent clinician in the UK. All non-serious adverse events, whether expected or not, will be recorded and updated at each visit. All new SAEs will be reported from the point of consent until follow-up. Investigators will report SAEs to the sponsor within 24 hours of the local site becoming aware of the event. All adverse events will be followed until satisfactory resolution.
Data will be collected and stored in accordance with the Data Protection Act 1998/General Data Protection Regulation 2018 in the UK. All data will be entered electronically. Outcomes and questionnaire data collected by participants will be reported using online survey software (Microsoft Forms; www.forms.office.com ). The administrative database (ie, participant information) and trial data will be managed by the research teams. Random checks will be performed on the entered data against online records. All errors will be logged and corrected. All data will be stored on password-protected and encrypted computers. Participant files will be maintained in storage for a period of 15 years after completion of the trial, with access granted to the local research team only. Our intended policy is that the research team should have exclusive use of the data for a period of 12 months or until the data are published. Data will be shared with named collaborators during this time. Following this period, data will be made publicly available through the LJMU Data Repository and published under a permissive reuse license.
A power calculation is not appropriate as the study does not aim to provide a definitive estimate of the treatment effect. Rather, the aim is to provide robust estimates of the likely rates of recruitment and retention and to yield estimates of the variability of the primary and secondary outcomes to inform power calculations for a future full-scale trial. Our aim is to recruit 60 participants, and this is based on recommendations for pilot/feasibility studies and an audit reporting pilot and feasibility trials registered in the UK clinical research network.
The quality of the study will be assured through the series of management groups. The trial will be overseen by a trial steering committee (TSC) and operated on a day-to-day basis by a trial delivery group (TDG). The TSC will comprise experienced academic experts (the research team), clinicians and patients but does not require and, therefore, will not have an independent chair. The TSC will meet quarterly to discuss progress. The role of the TSC is to provide overall supervision of the trial. In particular, the TSC will concentrate on the progress of the trial, adherence to the protocol, participant safety and consideration of new information. The TSC must be in agreement with the final protocol and, throughout the trial, will take responsibility for major decisions, such as the need to change the protocol for any reason, monitoring and supervising the progress of the trial, reviewing relevant information from other sources and informing and advising the TDG on all aspects of the trial. The TDG will comprise the same research team and will hold monthly meetings to discuss progress. The responsibilities of the TDG will include: Report to the TSC. Maintain the trial master file. Confirm all approvals are in place before the start of the trial at a site. Provide study materials. Data management centre. Give collaborators regular information about the progress of the study. Respond to any questions (eg, from collaborators) about the trial. Ensure data security and quality and observe data protection laws. Safety reporting. Ensure the trial is conducted in accordance with Good Clinical Practice. Statistical analysis. Publication of trial results.
National Institute for Health and Care Research North West Coast-funded patient and public involvement work with patients with AF has been completed. A patient involvement workshop (n=24) exploring their views of incorporating AF rehabilitation into usual care was conducted. All had positive responses regarding the potential to incorporate AF rehabilitation into usual care. Developing an intervention that supports patients with AF to become more physically active in the long term is therefore wanted by patients, is of great potential NHS benefit, and thus fills an important clinical and scientific ‘gap’. Workshops with healthcare practitioners (including clinical exercise physiologists and electrophysiologist consultants) were also conducted to determine any facilitators and barriers to pathway referral and rehabilitation delivery to patients with AF. Moving forward, two patient representatives have been invited to the TSC. They will advise on study information materials to recruit participants to the study. At the end of the project, our patient representatives will contribute to the reporting of the study by reading and reviewing the ‘lay’ sections of the report. They will also be involved in the dissemination of research findings through reviewing literature outlining the results before they are circulated.
The trial has received a favourable ethical opinion from the Preston NHS Research Ethics Committee (22/NW/0061) in the UK. On study completion, the CI owns the data. On completion of the study, the data will be analysed, and results will be disseminated via publication in clinical and physiological journals, presented at national and international conferences and in the form of feedback sheets or perhaps local articles. Participants will not be identifiable from the results of the study. Pseudonymised data from this study will be made available for sharing with other investigators after the publication of the study’s key papers. Data will be shared through the LJMU Data Repository ( http://opendata.ljmu.ac.uk/ ). This is a secure institutional data repository, which is searchable on the www, and it is managed by Library Services. A DOI will be generated for datasets as they are deposited to the repository. Data will be stored in this repository for a minimum of 10 years or for 10 years from the last date of access. It is our intention to present our research findings to all our research participants in a written lay summary and hold an open feedback session where the results will be presented in a lay-friendly manner. We plan to present the scientific findings as oral communications and abstracts at regional, national and international scientific meetings related to AF and preventative cardiology. We also intend to publish our findings in peer-reviewed journals.
10.1136/bmjopen-2024-088460 online supplemental file 1 10.1136/bmjopen-2024-088460 online supplemental file 2
|
Relationship Between Sarcopenia, Femoral Cartilage Thickness, and Knee Osteoarthritis: Case–Control Study | 6000fbf8-afe4-40ea-8e45-3691416a3bdf | 11931267 | Musculoskeletal System[mh] | Introduction Sarcopenia, a condition with prevalence increasing with age, is characterized by the loss of muscle mass and function, leading to diminished physical capability and reduced quality of life . All body muscles can be affected by sarcopenia, but the lower extremity muscles are particularly prone to be impacted . Some studies have shown a relationship between quadriceps muscle strength, thickness, and Femoral Cartilage Thickness (FCT) [ , , ]. A study investigating patients with knee osteoarthritis (KO) found a positive correlation between knee extensor muscle strength and FCT, showing that decreased muscle strength is associated with reduced FCT . Additionally, it has been reported that FCT decreases in various conditions that could lead to muscle weakness . Furthermore, an increased prevalence of KO, a disease associated with thinning of the femoral cartilage, has been demonstrated in sarcopenia patients [ , , , , ]. Existing studies have primarily focused on the relationship between FCT and knee extensor muscle strength or the association between sarcopenia and KO. However, as far as we can tell, no study has examined the relationship between sarcopenia, FCT, and KO. Understanding these relationships is crucial to uncover potential common mechanisms underlying these conditions and improving clinical outcomes. This study aims to investigate FCT differences between sarcopenia patients and healthy controls and to examine the relationship between sarcopenia, FCT, and KO. Additionally, it explores the potential of FCT as a predictive marker for sarcopenia and evaluates the association between sarcopenia and KO, focusing on the role of FCT in this relationship. In this study, the aim was to contribute to understanding potential risk factors for KO in sarcopenia patients and to guide early interventions for the prevention or management of KO.
Materials and Methods 2.1 Study Population This cross‐sectional study included a total of 80 individuals. 2.1.1 Inclusion Criteria Individuals − presenting to the Physical Medicine and Rehabilitation Clinic of the university hospital between February and May 2021. − aged 40–75 years. − 23 pre‐sarcopenia, 21 sarcopenia patients, and 36 healthy controls. − equal number of control subjects selected based on similar characteristics in age, gender, and BMI. 2.1.2 Exclusion Criteria Individuals − With knee surgery, joint deformities that may affect knee function. − With any inflammatory rheumatic diseases, malignancies, neurological diseases, and diseases that may cause muscle weakness. − Using medications that could impact muscle strength or function (e.g., steroids). − With cognitive impairments or conditions that affect the ability to provide informed consent. − With uncontrolled chronic conditions, such as diabetes or cardiovascular diseases. − Who declined to participate in the study. 2.2 Measurements HGS was measured using a Jamar hand dynamometer (Lafayette, IN, USA). HGS was calculated as the average of two measurements obtained from the dominant hand. Bioelectrical impedance analysis (BIA) was evaluated using the Body Composition Analyzer, model bc ‐418 (Tanita Corp, Tokyo, Japan). Appendicular Skeletal Muscle Mass (ASMM) is calculated as the total muscle mass of all four extremities. The diagnoses of pre‐sarcopenia and sarcopenia were made according to the updated diagnostic criteria set by the European Working Group on Sarcopenia in the Elderly (EWGSOP2) . Pre‐sarcopenia was defined as HGS values of < 16 N for women and < 27 N for men. Sarcopenia was identified with HGS values of < 16 N and ASMM values of < 15 kg for women and HGS values of < 27 N and ASMM values of < 20 kg for men. The same ultrasound specialist performed FCT measurements on all individuals using a B‐mode ultrasound with a linear transducer at a frequency range of 6–18 MHz (e‐soate, My Lab 70) while they were lying supine with their knees flexed at a 90° angle. FCT measurements were taken twice from the medial femoral condyle, intercondylar area, and lateral femoral condyle in both knees, and the average was calculated. The specialist was blinded to the groups (Figure ). FCT; − L‐IFC, Left İntercondylar Femoral Cartilage thicknesses. − L‐LFC, Left Lateral Femoral Cartilage Thicknesses. − L‐MFC, Left Medial Femoral Cartilage thicknesses. − Mean‐IFC, Right and Left İntercondylar Femoral Cartilage thicknesses mean. − Mean‐LFC, Right and Left Lateral Femoral Cartilage thicknesses mean. − Mean‐MFC, Right and Left Medial Femoral Cartilage thicknesses mean. − R‐IFC, Right İntercondylar Femoral Cartilage thicknesses. − R‐LFC, Right Lateral Femoral Cartilage thicknesses. − R‐MFC, Right Medial Femoral Cartilage thicknesses. All participants were diagnosed with clinical or radiological OA according to the ACR diagnostic criteria , based on history, physical examination, and knee X‐ray imaging from their medical records. Knee X‐rays of patients with radiological OA were staged according to the Kellgren‐Lawrence (KL) grading system (0–5 grading). 2.3 Statistical Analysis The effect size was calculated using the G Power 3.1 software program for the difference between FCT (left medial condylar area) in the paired t test, which was 0.65 (Cohen's d = 0.38) . Considering a margin of error of 0.05 ( α = 0.05) and a power of 0.80, the minimum required sample size was 72. A total of 80 participants were included in this study. Statistical analyses were conducted using SPSS Statistics version 21, released in 2012 by IBM, and Jamovi version 2.2, released in 2021 by Jamovi Project. ANOVA was used for normally distributed data comparisons between the three groups, with homogeneity checked using Levene's test. In contrast, the Kruskal–Wallis test was applied for non‐normally distributed values. The post hoc Tukey test was used for pairwise group comparisons. A logistic regression model was created to predict KO using group, age, gender, BMI, and medial FCT as predictors (EPV = 60/4 = 15). Multinomial Logistic Regression analysis was conducted to predict Pre‐Sarcopenia and Sarcopenia using KO, age, gender, BMI, and FCT (EPV = 42/7 = 6). Subsequently, a second model was created by removing non‐significant variables, retaining only KO and FCT (EPV = 42/4 = 10.5). These EPV values surpass the recommended minimum of 10, indicating that the sample size and model design are adequate for this logistic regression analysis. Receiver Operating Characteristic (ROC) analysis was conducted to assess the performance of FCT models in predicting the diagnosis of pre‐sarcopenia and sarcopenia. 2.4 Ethics Statement This study protocol was reviewed and approved by the ethics committee of Akdeniz University Faculty of Medicine (KAEK‐60, 27.01.2021). It was conducted according to the ethical standards of the 2000 Declaration of Helsinki. All subjects were informed about the study and obtained their written informed consent. We used the STROBE case–control checklist when writing our report .
Study Population This cross‐sectional study included a total of 80 individuals. 2.1.1 Inclusion Criteria Individuals − presenting to the Physical Medicine and Rehabilitation Clinic of the university hospital between February and May 2021. − aged 40–75 years. − 23 pre‐sarcopenia, 21 sarcopenia patients, and 36 healthy controls. − equal number of control subjects selected based on similar characteristics in age, gender, and BMI. 2.1.2 Exclusion Criteria Individuals − With knee surgery, joint deformities that may affect knee function. − With any inflammatory rheumatic diseases, malignancies, neurological diseases, and diseases that may cause muscle weakness. − Using medications that could impact muscle strength or function (e.g., steroids). − With cognitive impairments or conditions that affect the ability to provide informed consent. − With uncontrolled chronic conditions, such as diabetes or cardiovascular diseases. − Who declined to participate in the study.
Inclusion Criteria Individuals − presenting to the Physical Medicine and Rehabilitation Clinic of the university hospital between February and May 2021. − aged 40–75 years. − 23 pre‐sarcopenia, 21 sarcopenia patients, and 36 healthy controls. − equal number of control subjects selected based on similar characteristics in age, gender, and BMI.
Exclusion Criteria Individuals − With knee surgery, joint deformities that may affect knee function. − With any inflammatory rheumatic diseases, malignancies, neurological diseases, and diseases that may cause muscle weakness. − Using medications that could impact muscle strength or function (e.g., steroids). − With cognitive impairments or conditions that affect the ability to provide informed consent. − With uncontrolled chronic conditions, such as diabetes or cardiovascular diseases. − Who declined to participate in the study.
Measurements HGS was measured using a Jamar hand dynamometer (Lafayette, IN, USA). HGS was calculated as the average of two measurements obtained from the dominant hand. Bioelectrical impedance analysis (BIA) was evaluated using the Body Composition Analyzer, model bc ‐418 (Tanita Corp, Tokyo, Japan). Appendicular Skeletal Muscle Mass (ASMM) is calculated as the total muscle mass of all four extremities. The diagnoses of pre‐sarcopenia and sarcopenia were made according to the updated diagnostic criteria set by the European Working Group on Sarcopenia in the Elderly (EWGSOP2) . Pre‐sarcopenia was defined as HGS values of < 16 N for women and < 27 N for men. Sarcopenia was identified with HGS values of < 16 N and ASMM values of < 15 kg for women and HGS values of < 27 N and ASMM values of < 20 kg for men. The same ultrasound specialist performed FCT measurements on all individuals using a B‐mode ultrasound with a linear transducer at a frequency range of 6–18 MHz (e‐soate, My Lab 70) while they were lying supine with their knees flexed at a 90° angle. FCT measurements were taken twice from the medial femoral condyle, intercondylar area, and lateral femoral condyle in both knees, and the average was calculated. The specialist was blinded to the groups (Figure ). FCT; − L‐IFC, Left İntercondylar Femoral Cartilage thicknesses. − L‐LFC, Left Lateral Femoral Cartilage Thicknesses. − L‐MFC, Left Medial Femoral Cartilage thicknesses. − Mean‐IFC, Right and Left İntercondylar Femoral Cartilage thicknesses mean. − Mean‐LFC, Right and Left Lateral Femoral Cartilage thicknesses mean. − Mean‐MFC, Right and Left Medial Femoral Cartilage thicknesses mean. − R‐IFC, Right İntercondylar Femoral Cartilage thicknesses. − R‐LFC, Right Lateral Femoral Cartilage thicknesses. − R‐MFC, Right Medial Femoral Cartilage thicknesses. All participants were diagnosed with clinical or radiological OA according to the ACR diagnostic criteria , based on history, physical examination, and knee X‐ray imaging from their medical records. Knee X‐rays of patients with radiological OA were staged according to the Kellgren‐Lawrence (KL) grading system (0–5 grading).
Statistical Analysis The effect size was calculated using the G Power 3.1 software program for the difference between FCT (left medial condylar area) in the paired t test, which was 0.65 (Cohen's d = 0.38) . Considering a margin of error of 0.05 ( α = 0.05) and a power of 0.80, the minimum required sample size was 72. A total of 80 participants were included in this study. Statistical analyses were conducted using SPSS Statistics version 21, released in 2012 by IBM, and Jamovi version 2.2, released in 2021 by Jamovi Project. ANOVA was used for normally distributed data comparisons between the three groups, with homogeneity checked using Levene's test. In contrast, the Kruskal–Wallis test was applied for non‐normally distributed values. The post hoc Tukey test was used for pairwise group comparisons. A logistic regression model was created to predict KO using group, age, gender, BMI, and medial FCT as predictors (EPV = 60/4 = 15). Multinomial Logistic Regression analysis was conducted to predict Pre‐Sarcopenia and Sarcopenia using KO, age, gender, BMI, and FCT (EPV = 42/7 = 6). Subsequently, a second model was created by removing non‐significant variables, retaining only KO and FCT (EPV = 42/4 = 10.5). These EPV values surpass the recommended minimum of 10, indicating that the sample size and model design are adequate for this logistic regression analysis. Receiver Operating Characteristic (ROC) analysis was conducted to assess the performance of FCT models in predicting the diagnosis of pre‐sarcopenia and sarcopenia.
Ethics Statement This study protocol was reviewed and approved by the ethics committee of Akdeniz University Faculty of Medicine (KAEK‐60, 27.01.2021). It was conducted according to the ethical standards of the 2000 Declaration of Helsinki. All subjects were informed about the study and obtained their written informed consent. We used the STROBE case–control checklist when writing our report .
Results The mean age of the 80 individuals (55 females and 25 males) between 40 and 75 was 62.22 ± 7.56 years (Table ). Medial Femoral Cartilage (MFC) and Lateral Femoral Cartilage (LFC) thicknesses, KO prevalence, and KL grading values were found to be different between the groups ( p < 0.05) (Table ) (Figure ). No significant difference was found between females and males in FCT values and pre‐sarcopenia and sarcopenia prevalences (all p ≥ 0.05). A weak positive correlation was found between MFC thickness and BMI, ASMM, and HGS (respectively, p = 0.037/Spearman r = 0.233, p = 0.027/Spearman r = 0.248, p = 0.003/Spearman r = 0.333). In the logistic regression analysis conducted for KO prediction, age and sarcopenia were found to be significant (all p < 0.01, odds ratio 0.276 for age, 3.248 for sarcopenia) (Table ) (Figure ). In the multinomial logistic regression analysis for predicting pre‐sarcopenia and sarcopenia, KO and MFC thickness were significant for sarcopenia ( p < 0.05). In contrast, LFC thickness was significant for pre‐sarcopenia ( p < 0.05) (Table ). ROC analysis demonstrated the ability of FCT to discriminate between pre‐sarcopenia and sarcopenia. MFC thickness was statistically significant for sarcopenia ( p = 0.001, AUC = 0.736), while MFC, LFC, and intercondylar femoral cartilage thickness (IFC) were significant for pre‐sarcopenia (all p < 0.05) (Figures and ) (Table ). For Predicting Pre‐Sarcopenia ; − Mean‐IFC model, Accuracy = 62.5%, F1 = 69.41%, AIC = −198.986. − Mean‐LFC model, Accuracy = 72.5%, F1 = 72.3%, AIC = −196.822. − Mean‐MFC model, Accuracy = 65%, F1 = 71.4%, AIC = −210.768. − For predicting pre‐sarcopenia, the mean LFC model is the best. For Predicting Sarcopenia ; − Mean‐IFC model, Accuracy = 48.75%, F1 = 47.6%. − Mean‐LFC model, Accuracy = 66.25%, F1 = 55.3%. − Mean‐MFC model, Accuracy = 52.5%, F1 = 51.1%. − For predicting sarcopenia, the mean MFC model is the best.
Discussion This study's findings, which investigated the relationship between sarcopenia and KO and the role of FCT in this context, showed a decrease in medial and lateral FCT and an increase in KO prevalence in both pre‐sarcopenia and sarcopenia patients compared to healthy controls. The decrease in FCT was shown to be a predictor of sarcopenia, while the presence of sarcopenia was also a predictor of KO. These findings support the idea that the reduction in FCT in sarcopenia may predispose individuals to KO and that sarcopenia may contribute to the coexistence of KO. In the literature, studies have shown a positive correlation between quadriceps muscle strength and FCT, as well as a reduction in FCT in diseases accompanied by muscle weakness [ , , ]. Based on this information, a decrease in FCT is a plausible outcome in sarcopenia, and our study results support this theory by showing a decrease in FCT in sarcopenia patients. The literature has extensively studied the coexistence of sarcopenia and KO, and the relationship between these two conditions has been demonstrated in numerous studies [ , , , , , ]. Consistent with the literature, our study found an increased prevalence of KO in sarcopenia patients. Furthermore, the study demonstrated that sarcopenia may be an indicator of KO, and KO may be an indicator of sarcopenia. Aging, obesity, diabetes, and vitamin D deficiency have been reported as common risk factors for both sarcopenia and KO. Reduced muscle strength, pro‐inflammatory cytokines, and irisin are believed to play roles in their pathogenesis . However, a definitive conclusion has not been reached regarding a cause‐and‐effect relationship. The notion of a vicious cycle between the two conditions is widely accepted . A decrease in knee extensor muscle strength and FCT in patients with KO has been demonstrated in some studies [ , , , ]. The relationship between quadriceps muscle strength and FCT could be one of the mechanisms explaining the association between sarcopenia and KO. However, the cause‐and‐effect relationship between muscle strength and cartilage damage is unclear. In an animal experiment to investigate whether cartilage damage occurs due to muscle weakness or if it emerges first, rabbits' quadriceps muscles were weakened by injecting Botulinum toxin type A into the muscles. After 4 weeks, significant changes were observed in the patellofemoral region. The results indicated that the initial signs of joint cartilage deterioration appeared with muscle weakness, suggesting that muscle weakness may be a risk factor for cartilage damage and osteoarthritis . Another study investigating the relationship between knee muscle strength and cartilage thickness in individuals with KO found a positive correlation, and after a 1 month quadriceps strengthening program, an increase in quadriceps strength was correlated with an increase in cartilage thickness. This led to the interpretation that cartilage thinning might be attributed to muscle strength loss . It is an accepted concept that a decrease in knee muscle strength leads to increased load on the knee and destabilization, causing cartilage damage. However, it has been reported that mechanical effects alone cannot solely explain this relationship. Muscle cells may play a significant role in cartilage homeostasis and regulating cartilage gene expression, indicating a more complex interplay beyond mechanical factors . The finding of reduced FCT in sarcopenia patients in this study is consistent with the literature. It indirectly supports the relationship between muscle strength and cartilage thickness and suggests that this relationship plays a role in the coexistence of sarcopenia and KO. Another finding of this study is that while a reduction in medial FCT indicates sarcopenia, it is not an indicator of KO. Instead, the significant association of sarcopenia presence with KO supports the view that sarcopenia may be a risk factor for KO. The literature generally indicates a positive relationship between FCT and BMI [ , , ]. However, one study has reported a positive correlation between FCT and muscle mass and a negative relationship with fat mass . Consistent with the literature, this study found a weak positive correlation between BMI, appendicular muscle mass, and HGS with FCT. This positive relationship suggests that muscle strength, muscle mass, and BMI influence FCT through mechanical factors and biochemical signals regulating cartilage metabolism. Studies investigating the relationship between gender and FCT have indicated that FCT tends to be thinner in females compared to males . Males' higher muscle strength can explain their thicker femoral cartilage. In this study, no difference in FCT was found between females and males, which could be attributed to the similar rates of sarcopenia in both groups. The limitations of this study include the small sample size for regression analyses, despite sufficient patient numbers and a large effect size (> 0.14) for group comparisons, leading to wide confidence intervals in the regression models. Additionally, ethical and financial constraints prevented knee radiographs from being performed on all participants, requiring them to rely on knee radiographs retrieved from patient records instead. The study also did not evaluate factors such as obesity, exercise habits, and nutrition, which play a shared role in the pathogenesis of sarcopenia and KO. Another limitation of this study is its cross‐sectional design, lack of prognostic insight, inability to establish causality, and reliance on binary outcomes, which may overlook important associations. Further prospective studies with larger patient populations and more comprehensive data are needed in this regard. This study showed a decrease in FCT and an increase in the prevalence of KO in sarcopenia patients. It has been observed that a decrease in medial FCT may predict sarcopenia, while sarcopenia may also predict KO. Moreover, the reduction in FCT may play an important role in explaining the relationship between KO and sarcopenia. A weak positive correlation was found between HGS and muscle mass with FCT, highlighting the importance of mechanical and biochemical factors affecting cartilage thickness. Sarcopenia may be a risk factor for KO, and these two conditions might share common pathophysiological mechanisms. Prospective studies with larger sample sizes could contribute to a better understanding of the relationship between sarcopenia, KO, and FCT.
S.T., N.B., E.K. investigation, data curation, methodology, design, formal analysis, writing. S.T. performing an ultrasound. S.T., E.K., N.B. writing, supervision, review and editing. All authors have read and approved the final manuscript.
The ethics committee of Akdeniz University Faculty of Medicine (KAEK‐60, 27.01.2021) reviewed and approved this study protocol. All subjects were informed about the study, and their written informed consent was obtained. Written and verbal consent was obtained from all participants.
The authors declare no conflicts of interest.
Data S1. Figure S1. Ultrasound image (Femoral Cartilage Thickness measurements) 1. Medial Femoral Cartilage Thickness, 2. İntercondylar Femoral Cartilage Thickness, 3. Lateral Femoral Cartilage Thickness. Figure S2. ROC Curve for Logistic Regression Model 2 in Predicting Knee. Figure S3. ROC Curve for Femoral Cartilage Thickness in Predicting Pre‐Sarcopenia. Figure S4. ROC Curve for Femoral Cartilage Thickness in Predicting Sarcopenia.
|
cRGD peptide-conjugated polyethylenimine-based lipid nanoparticle for intracellular delivery of siRNA in hepatocarcinoma therapy | 1c56deaa-922e-42be-a2db-9520708cf407 | 8168781 | Pharmacology[mh] | Introduction Hepatocarcinoma is one of the most common tumors with high mortality all around the world. Chemotherapy is a common treatment. However, the survival of the patients is dismal (Wu et al., ; Yang & Heimbach, ). RNA interference (RNAi) technology provides challenges and therapeutic opportunities for hepatocarcinoma (Hajiasgharzadeh et al., ), because RNAi can make intracellular mRNA-specific degradation occurs, leading to silencing of target gene expression (Bhattacharjee et al., ; Setten et al., ). It has been confirmed that survivin is highly expressed in hepatocellular carcinoma, so survivin is a potential target of RNAi for the treatment of hepatocarcinoma (Su, ). In recent years, more and more researches have focused on the preclinical research of small interfering RNA (siRNA) of cancer (Scarabel et al., ; Subhan & Torchilin, 2019). As a kind of nucleic acid drug, siRNA showed rapid clearance in the circulation and poor transport across biological barriers which resulting in limited treatment efficiency (Kulkarni et al., ). So effective delivery systems are promising to protect siRNA and be capable to release siRNA into the cytoplasm (Barba et al., ; Dong et al., ). During the past years, many efforts have been made to construct efficient nano-drug delivery systems (Khan et al., ). Suitable delivery systems are promising to have the appropriate size and electric charge, meanwhile, they should have high biocompatibility and low toxicity to cells (Han et al., ). The controllable release of siRNA is also a difficult point in the design of the delivery systems (Ma, ; Mainini & Eccles, ). To improve the biocompatibility of the delivery systems, many kinds of nanocarriers were used such as liposomes (Diao et al., ), microemulsions (Wang et al., ), nanocapsules (Jiang et al., ), and cationic nanoparticles (Jin et al., ). It is noteworthy that intracellular environment-sensitive delivery systems are efficient and controllable for the special microenvironment of tumor tissue (Kozielski et al., ). In recent years, redox-responsive nanoparticles have received tremendous attention for triggered siRNA release in response to the higher concentration of reductive matters (7- to 10-fold) in tumor cells compared to that in normal cells (Pezzoli et al., ; Chi et al., ). In order to further improve the accumulation of siRNA in the tumor cells, targeting components, such as receptor-recognizing peptides and antibodies, which were connected to various species of nanoparticles. It was reported that a targeting peptide, cyclic arginine-glycine-aspartic acid (cRGD)-modified nanoparticle, enhanced the tumor accumulation through the targeting of α v β 3 integrin, which is over expressed in various types of malignant tumors such as melanoma, glioblastoma, hepatocarcinoma, breast, pancreatic and ovarian carcinoma (Cheng & Ji, ). In our previous study, a reduction-sensitive cationic polymer (PEI-SS-HA, PSH) was synthesized and the nanoparticle contained PSH was designed to deliver an antisense oligonucleotide, which showed suitable properties and high transfection efficiency (Zheng et al., ). In order to further design a system with safety, low toxicity, high siRNA loading efficiency, hepatocarcinoma targeting, and rapid intracellular drug release, here we report a multifunctional nanoparticle based on logical design. Egg phosphatidylcholine (ePC), cholesterol, DSPE-PEG were incorporated into a supported bilayer. Cationic PSH was incorporated into the nanoparticle to give the ability of reduction-sensitive and the ‘proton pump effect’ based on PEI to release siRNA into the nucleus rapidly. In order to increase circulation time and target the hepatocarcinoma cells precisely, DSPE-PEG 2000 -cRGD was used to constitute the outermost structure as a targeting ligand of the nanoparticles. The properties of the nanoparticles were characterized, the in vitro and in vivo transfection activity was evaluated.
Materials and methods 2.1. Materials Egg phosphatidylcholine (ePC), cholesterol, and DSPE-PEG 2000 were purchased from Avanti Polar Lipids, Inc (Shanghai, China). DSPE-PEG 2000 -cRGD was purchased from Xi’an Ruixi Biological Technology Co., Ltd (Xi’an, China). Survivin siRNA (5′-mGCAGGUUCCUmUAUCUGUCATT-3′) and 5′-Cy3 labeled survivin siRNA were synthesized by Biomics Biotechnologies (Jiangsu, China). PEI (branched, 25 kDa) and hexadecyl amine were purchased from Sigma-Aldrich (St. Louis, MO, USA). Dulbecco’s modified Eagle’s medium (DMEM), Roswell Park Memorial Institute 1640 (RPMI 1640), and fetal bovine serum (FBS) were purchased from HyClone (Logan, UT, USA). Anti-survivin, anti-GAPDH antibody and horseradish peroxidase (HRP)-conjugated goat anti-rabbit IgG were obtained from Abcam Inc. (Cambridge, MA, USA). HepG2 cells and LO2 cells were purchased from American Type Culture Collection (ATCC). All other analytical reagents were commercially obtained in reagent grade. 2.2. Preparation and optimization of the formulation of the nanoparticles PEI-SS-HA (PSH) was synthesized as described in our previous report (Zheng et al., ). The structural formulae of DSPE-PEG 2000 -cRGD and PSH were showed in . The ethanol dilution method was used for preparing the lipid nanoparticles (NP) or the lipid nanoparticles containing survivin siRNA (NP/S). In order to verify that cRGD-PSH-NP was superior to other nanoparticles as a carrier of siRNA, PEI-NP, PSH-NP, and cRGD-NP were prepared and used to combine with survivin siRNA. Briefly, for the NP containing PSH (PSH-NP), ePC, cholesterol, PSH, and DSPE-PEG 2000 were dissolved in ethanol with homogeneous mixing to obtained the lipid solution, the molar ratios of PSH/total lipids were from 10% to 50%. The lipid solution was then injected into PBS solution (pH 6.8) under rapid mixing at the volume ratios of 1:9 to obtain PSH-NP with different concentrations of PSH. For the PSH-NP containing siRNA (PSH-NP/S), survivin siRNA was dissolved in pre-cooled DEPC water and added into the PSH-NP solution followed by mixing for 60 s at various PSH/siRNA molar ratios from 12:1 to 2:1. Meanwhile, PEI-NP/S was prepared as the group of control to confirm the low toxicity and high cellular uptake efficiency of modified PEI (PSH), and to prove the necessity of adding PSH instead of PEI into nanoparticles. In the process of preparation, PSH was is replaced by PEI (25 kDa) in the group of PEI-NP/S, other processes were consistent with the preparation method of PSH-NP/S. For the NP containing cRGD (cRGD-NP), ePC, cholesterol, PEI, DSPE-PEG 2000 , and DSPE-PEG 2000 -cRGD were dissolved in ethanol, which was injected into PBS solution (pH 6.8) under rapid mixing. Survivin siRNA was then dissolved in pre-cooled DEPC water and added into the cRGD-NP followed by mixing for 60 s to obtain cRGD-NP/S. cRGD-NP/S was used for the comparison to show whether PSH could increase the transfection efficiency in nanoparticles. Finally, cRGD-PSH-NP, which represented the NP containing cRGD and PSH was designed and prepared. Briefly, ePC, cholesterol, PSH, DSPE-PEG 2000 , and DSPE-PEG 2000 -cRGD were dissolved in ethanol. Survivin siRNA was then dissolved in the obtained solution at different molar ratios of cRGD/total lipids from 1% to 9%, the final system was named cRGD-PSH-NP/S. The diagram of the composition and drug release process in vivo of cRGD-PSH-NP/S is shown in . The particle size and zeta potential of PSH-NP at various PSH/total lipids molar ratios, PSH-NP/S at various PSH/siRNA molar ratios and cRGD-PSH-NP/S at different molar ratios of cRGD/total lipids were determined on a Zetasizer Nano ZS 90 instrument (Malvern Instruments, Ltd., Malvern, UK). Each sample was measured averaging three times. Meanwhile, the morphology of cRGD-PSH-NP/S was observed under the field emission scanning electron microscope (FE-SEM) (JSM-6700F, JEOL, Tokyo, Japan) at 3.0 kV accelerating voltage. 2.3. Determination of the nanoparticles combined with siRNA by gel retardation assay PSH-NP/S at various ratios of PSH:siRNA (12:1 to 2:1) were prepared and confirmed by agarose gel retardation assay to investigate the ability of PSH-NP to form nanocomplexes with survivin siRNA. Then PEI-NP/S, PSH-NP/S, cRGD-NP/S, and cRGD-PSH-NP/S were prepared at the optimal formulation and confirmed by agarose gel retardation assay. Ten microliters of the samples were incubated with 10× sample buffer and then loaded onto a 2% agarose gel containing 5% (v/v, μl/ml) Goldview. After electrophoresis (120 V, 15 min), the gels were photographed under UV-illumination. The stability of siRNA and cRGD-PSH-NP/S was investigated by agarose gel retardation assay. siRNA and cRGD-PSH-NP/S were mixed with equal-volume fetal bovine serum (FBS) and incubated at 37 °C for 2, 4, 6, 8, 10, and 12 h, respectively. TritonX-100 was added into the solution until the final concentration was 1%, and undegradable siRNA was released for electrophoresis and image development. The same amount of siRNA was used as control. 2.4. Cell culture HepG2 cells and LO2 cells were propagated in DMEM and RPM1640 supplemented with 10% FBS and 1% antibiotics (100 units/ml penicillin and 100 μg/ml streptomycin), respectively. The cells were cultured at 37 °C in a humidified 5% CO 2 incubator. 2.5. Cytotoxicity assay of the nanoparticles HepG2 cells and normal liver cells LO2 were seeded at a density of 1 × 10 4 cells/well in a 96-well plate for 24 h. Cells were treated with PEI-NP, PSH-NP, cRGD-NP, or cRGD-PSH-NP at various concentrations for 4 h, the medium was removed and incubated in fresh DMEM or RPM1640 medium for another 20 h. Then each well was incubated with 20 μl MTT (5 mg/ml in PBS solution) stock solution for 4 h. The culture medium was removed and then 100 μl/well of DMSO was added to dissolve formazan crystals that formed, the absorbance was measured at OD 490 nm in a plate reader. 2.6. In vitro treatment and cellular uptake of the nanoparticles The cellular uptakes of 5′-Cy3-labeled survivin siRNA which were wrapped in nanoparticles were determined by an EPICS XL flow cytometer (Beckman Coulter Corp., Tokyo, Japan) in HepG2 cells. Specifically, HepG2 cells were plated on 24 well plates (1 × 10 5 cells/well) and cultured for 24 h in DMEM medium with 10% FBS. Then, the cells were washed three times with PBS solution and a fresh serum-free DMEM medium was added. Then cells were treated with naked siRNA, PSH-NP/S, cRGD-NP/S, or cRGD-PSH-NP/S for 4 h without or with FBS at 37 °C, the siRNA was labeled with 5′-Cy3 in the above component. The localization and the quantitative fluorescence intensity of the nanoparticles after uptaking by cells were realized by tracking the fluorescence signal of Cy3. Then the HepG2 cells were washed with PBS solution three times and fixed in 4% paraformaldehyde. The cells were collected and analyzed by flow cytometry and untreated HepG2 cells were collected as a negative control. The internalizations of the nanoparticles were further observed by Zeiss 710 LSMNLO Confocal Microscope (Carl Zeiss; Jena, Germany). HepG2 cells were cultured in a glass-bottom cell culture dish (1 × 10 5 cells/well) for 24 h. Naked siRNA, PSH-NP/S, cRGD-NP/S or cRGD-PSH-NP/S were added into the cells and incubated without or with FBS for 4 h at 37 °C, the siRNA was labeled with 5′-Cy3. Then the cells were washed three times with PBS solution and fixed in 4% paraformaldehyde for 15 min. Cellular nuclei were stained with DAPI (3 μg/ml) for 3 min. The cells were washed with PBS solution three times to remove excess dye. 2.7. Determination of the survivin protein expression HepG2 cells were plated on 6 well plates (1 × 10 5 cells/well) overnight. Then, the medium was removed and cells were treated with PEI-NP/S, PSH-NP/S, cRGD-NP/S, or cRGD-PSH-NP/S for 4 h at 37 °C. Cells were cultured for another 44 h with a fresh medium. RIPA buffer containing 2% PMSF and 1% protease inhibitor cocktail was added to lyse the cells for 10 min at 4 °C. The lysed solution was collected and centrifuged at 10,000 rpm for 5 min to obtain the total proteins. Bradford protein assay was used to determine the concentration of protein. Then 40 μg total proteins were separated by 10% SDS-PAGE gel and transferred onto polyvinylidene fluoride (PVDF) membrane. The PVDF membrane was blocked with 5% bovine serum albumin for 3 h and immunoblotted against GAPDH or survivin antibody at 4 °C overnight. The membrane was then incubated with HRP-conjugated goat anti-rabbit IgG for 4 h at 4 °C. ECL detection kits (GE Healthcare Life science, Waukesha, WI, USA) were used for detecting chemiluminescence. 2.8. In vivo antitumor efficacy To investigate the in vivo antitumor activity of the nanoparticles, HepG2 cells were transplanted into BALB/C nude mice. The nude mice were randomly divided into three treatment groups with three mice: the control group (normal saline), cRGD-NP/S group, and cRGD-PSH-NP/S group. When the tumor volume reached approximately 100 mm 3 , the nanoparticles were injected via the tail vein. Briefly, the control group was injected with 0.2 ml 0.9% saline solution, the cRGD-NP/S group or cRGD-PSH-NP/S group was injected with 0.2 ml cRGD-NP/S or cRGD-PSH-NP/S (2.5 mg/kg siRNA per injection). The drug loading of each batch of nanoparticles was determined by agarose gel retardation assay before administration. Triton X-100 was added to the nanoparticle solution until the final concentration was 1%, siRNA was released and agarose gel retardation assay was performed to calculate the drug loading before administration. All the groups were dosed every 3 days for 21 days. The size of the tumor and the bodyweight were monitored every 3 days. The antitumor activity was evaluated in terms of tumor volume ( V ), which was estimated by the equation V = a × b 2 /2, where a and b are the largest and smallest diameter of the tumor, respectively. At the same time, weights of mice were taken. After the last treatment, the xenograft tumors were harvested and photographed, and the protein levels of survivin in tumors were analyzed by Western blot. 2.9. In vivo safety evaluation After continuous injection for 21 days, euthanasia was conducted on all mice and major organs including heart, liver, spleen, lung, and kidney were collected and fixed in 4% paraformaldehyde, sliced, and then stained with hematoxylin and eosin (H&E). Images of tissue sections were detected by ImageXpress ® Pico (Molecular Devices Corporation, CA, USA). 2.10. Statistical analysis All the data in the above experiments were expressed as mean ± standard deviation (SD). The significant differences between different groups were carried out with t -test. p < .05 represents a significant difference, and p < .01 indicates a highly significant difference.
Materials Egg phosphatidylcholine (ePC), cholesterol, and DSPE-PEG 2000 were purchased from Avanti Polar Lipids, Inc (Shanghai, China). DSPE-PEG 2000 -cRGD was purchased from Xi’an Ruixi Biological Technology Co., Ltd (Xi’an, China). Survivin siRNA (5′-mGCAGGUUCCUmUAUCUGUCATT-3′) and 5′-Cy3 labeled survivin siRNA were synthesized by Biomics Biotechnologies (Jiangsu, China). PEI (branched, 25 kDa) and hexadecyl amine were purchased from Sigma-Aldrich (St. Louis, MO, USA). Dulbecco’s modified Eagle’s medium (DMEM), Roswell Park Memorial Institute 1640 (RPMI 1640), and fetal bovine serum (FBS) were purchased from HyClone (Logan, UT, USA). Anti-survivin, anti-GAPDH antibody and horseradish peroxidase (HRP)-conjugated goat anti-rabbit IgG were obtained from Abcam Inc. (Cambridge, MA, USA). HepG2 cells and LO2 cells were purchased from American Type Culture Collection (ATCC). All other analytical reagents were commercially obtained in reagent grade.
Preparation and optimization of the formulation of the nanoparticles PEI-SS-HA (PSH) was synthesized as described in our previous report (Zheng et al., ). The structural formulae of DSPE-PEG 2000 -cRGD and PSH were showed in . The ethanol dilution method was used for preparing the lipid nanoparticles (NP) or the lipid nanoparticles containing survivin siRNA (NP/S). In order to verify that cRGD-PSH-NP was superior to other nanoparticles as a carrier of siRNA, PEI-NP, PSH-NP, and cRGD-NP were prepared and used to combine with survivin siRNA. Briefly, for the NP containing PSH (PSH-NP), ePC, cholesterol, PSH, and DSPE-PEG 2000 were dissolved in ethanol with homogeneous mixing to obtained the lipid solution, the molar ratios of PSH/total lipids were from 10% to 50%. The lipid solution was then injected into PBS solution (pH 6.8) under rapid mixing at the volume ratios of 1:9 to obtain PSH-NP with different concentrations of PSH. For the PSH-NP containing siRNA (PSH-NP/S), survivin siRNA was dissolved in pre-cooled DEPC water and added into the PSH-NP solution followed by mixing for 60 s at various PSH/siRNA molar ratios from 12:1 to 2:1. Meanwhile, PEI-NP/S was prepared as the group of control to confirm the low toxicity and high cellular uptake efficiency of modified PEI (PSH), and to prove the necessity of adding PSH instead of PEI into nanoparticles. In the process of preparation, PSH was is replaced by PEI (25 kDa) in the group of PEI-NP/S, other processes were consistent with the preparation method of PSH-NP/S. For the NP containing cRGD (cRGD-NP), ePC, cholesterol, PEI, DSPE-PEG 2000 , and DSPE-PEG 2000 -cRGD were dissolved in ethanol, which was injected into PBS solution (pH 6.8) under rapid mixing. Survivin siRNA was then dissolved in pre-cooled DEPC water and added into the cRGD-NP followed by mixing for 60 s to obtain cRGD-NP/S. cRGD-NP/S was used for the comparison to show whether PSH could increase the transfection efficiency in nanoparticles. Finally, cRGD-PSH-NP, which represented the NP containing cRGD and PSH was designed and prepared. Briefly, ePC, cholesterol, PSH, DSPE-PEG 2000 , and DSPE-PEG 2000 -cRGD were dissolved in ethanol. Survivin siRNA was then dissolved in the obtained solution at different molar ratios of cRGD/total lipids from 1% to 9%, the final system was named cRGD-PSH-NP/S. The diagram of the composition and drug release process in vivo of cRGD-PSH-NP/S is shown in . The particle size and zeta potential of PSH-NP at various PSH/total lipids molar ratios, PSH-NP/S at various PSH/siRNA molar ratios and cRGD-PSH-NP/S at different molar ratios of cRGD/total lipids were determined on a Zetasizer Nano ZS 90 instrument (Malvern Instruments, Ltd., Malvern, UK). Each sample was measured averaging three times. Meanwhile, the morphology of cRGD-PSH-NP/S was observed under the field emission scanning electron microscope (FE-SEM) (JSM-6700F, JEOL, Tokyo, Japan) at 3.0 kV accelerating voltage.
Determination of the nanoparticles combined with siRNA by gel retardation assay PSH-NP/S at various ratios of PSH:siRNA (12:1 to 2:1) were prepared and confirmed by agarose gel retardation assay to investigate the ability of PSH-NP to form nanocomplexes with survivin siRNA. Then PEI-NP/S, PSH-NP/S, cRGD-NP/S, and cRGD-PSH-NP/S were prepared at the optimal formulation and confirmed by agarose gel retardation assay. Ten microliters of the samples were incubated with 10× sample buffer and then loaded onto a 2% agarose gel containing 5% (v/v, μl/ml) Goldview. After electrophoresis (120 V, 15 min), the gels were photographed under UV-illumination. The stability of siRNA and cRGD-PSH-NP/S was investigated by agarose gel retardation assay. siRNA and cRGD-PSH-NP/S were mixed with equal-volume fetal bovine serum (FBS) and incubated at 37 °C for 2, 4, 6, 8, 10, and 12 h, respectively. TritonX-100 was added into the solution until the final concentration was 1%, and undegradable siRNA was released for electrophoresis and image development. The same amount of siRNA was used as control.
Cell culture HepG2 cells and LO2 cells were propagated in DMEM and RPM1640 supplemented with 10% FBS and 1% antibiotics (100 units/ml penicillin and 100 μg/ml streptomycin), respectively. The cells were cultured at 37 °C in a humidified 5% CO 2 incubator.
Cytotoxicity assay of the nanoparticles HepG2 cells and normal liver cells LO2 were seeded at a density of 1 × 10 4 cells/well in a 96-well plate for 24 h. Cells were treated with PEI-NP, PSH-NP, cRGD-NP, or cRGD-PSH-NP at various concentrations for 4 h, the medium was removed and incubated in fresh DMEM or RPM1640 medium for another 20 h. Then each well was incubated with 20 μl MTT (5 mg/ml in PBS solution) stock solution for 4 h. The culture medium was removed and then 100 μl/well of DMSO was added to dissolve formazan crystals that formed, the absorbance was measured at OD 490 nm in a plate reader.
In vitro treatment and cellular uptake of the nanoparticles The cellular uptakes of 5′-Cy3-labeled survivin siRNA which were wrapped in nanoparticles were determined by an EPICS XL flow cytometer (Beckman Coulter Corp., Tokyo, Japan) in HepG2 cells. Specifically, HepG2 cells were plated on 24 well plates (1 × 10 5 cells/well) and cultured for 24 h in DMEM medium with 10% FBS. Then, the cells were washed three times with PBS solution and a fresh serum-free DMEM medium was added. Then cells were treated with naked siRNA, PSH-NP/S, cRGD-NP/S, or cRGD-PSH-NP/S for 4 h without or with FBS at 37 °C, the siRNA was labeled with 5′-Cy3 in the above component. The localization and the quantitative fluorescence intensity of the nanoparticles after uptaking by cells were realized by tracking the fluorescence signal of Cy3. Then the HepG2 cells were washed with PBS solution three times and fixed in 4% paraformaldehyde. The cells were collected and analyzed by flow cytometry and untreated HepG2 cells were collected as a negative control. The internalizations of the nanoparticles were further observed by Zeiss 710 LSMNLO Confocal Microscope (Carl Zeiss; Jena, Germany). HepG2 cells were cultured in a glass-bottom cell culture dish (1 × 10 5 cells/well) for 24 h. Naked siRNA, PSH-NP/S, cRGD-NP/S or cRGD-PSH-NP/S were added into the cells and incubated without or with FBS for 4 h at 37 °C, the siRNA was labeled with 5′-Cy3. Then the cells were washed three times with PBS solution and fixed in 4% paraformaldehyde for 15 min. Cellular nuclei were stained with DAPI (3 μg/ml) for 3 min. The cells were washed with PBS solution three times to remove excess dye.
Determination of the survivin protein expression HepG2 cells were plated on 6 well plates (1 × 10 5 cells/well) overnight. Then, the medium was removed and cells were treated with PEI-NP/S, PSH-NP/S, cRGD-NP/S, or cRGD-PSH-NP/S for 4 h at 37 °C. Cells were cultured for another 44 h with a fresh medium. RIPA buffer containing 2% PMSF and 1% protease inhibitor cocktail was added to lyse the cells for 10 min at 4 °C. The lysed solution was collected and centrifuged at 10,000 rpm for 5 min to obtain the total proteins. Bradford protein assay was used to determine the concentration of protein. Then 40 μg total proteins were separated by 10% SDS-PAGE gel and transferred onto polyvinylidene fluoride (PVDF) membrane. The PVDF membrane was blocked with 5% bovine serum albumin for 3 h and immunoblotted against GAPDH or survivin antibody at 4 °C overnight. The membrane was then incubated with HRP-conjugated goat anti-rabbit IgG for 4 h at 4 °C. ECL detection kits (GE Healthcare Life science, Waukesha, WI, USA) were used for detecting chemiluminescence.
In vivo antitumor efficacy To investigate the in vivo antitumor activity of the nanoparticles, HepG2 cells were transplanted into BALB/C nude mice. The nude mice were randomly divided into three treatment groups with three mice: the control group (normal saline), cRGD-NP/S group, and cRGD-PSH-NP/S group. When the tumor volume reached approximately 100 mm 3 , the nanoparticles were injected via the tail vein. Briefly, the control group was injected with 0.2 ml 0.9% saline solution, the cRGD-NP/S group or cRGD-PSH-NP/S group was injected with 0.2 ml cRGD-NP/S or cRGD-PSH-NP/S (2.5 mg/kg siRNA per injection). The drug loading of each batch of nanoparticles was determined by agarose gel retardation assay before administration. Triton X-100 was added to the nanoparticle solution until the final concentration was 1%, siRNA was released and agarose gel retardation assay was performed to calculate the drug loading before administration. All the groups were dosed every 3 days for 21 days. The size of the tumor and the bodyweight were monitored every 3 days. The antitumor activity was evaluated in terms of tumor volume ( V ), which was estimated by the equation V = a × b 2 /2, where a and b are the largest and smallest diameter of the tumor, respectively. At the same time, weights of mice were taken. After the last treatment, the xenograft tumors were harvested and photographed, and the protein levels of survivin in tumors were analyzed by Western blot.
In vivo safety evaluation After continuous injection for 21 days, euthanasia was conducted on all mice and major organs including heart, liver, spleen, lung, and kidney were collected and fixed in 4% paraformaldehyde, sliced, and then stained with hematoxylin and eosin (H&E). Images of tissue sections were detected by ImageXpress ® Pico (Molecular Devices Corporation, CA, USA).
Statistical analysis All the data in the above experiments were expressed as mean ± standard deviation (SD). The significant differences between different groups were carried out with t -test. p < .05 represents a significant difference, and p < .01 indicates a highly significant difference.
Results 3.1. Formulation optimization of cRGD-PSH-NP/S The particle sizes and zeta potentials of PSH-NP with various concentrations of PSH were firstly measured to explore the appropriate concentration of PSH . As the ratios of PSH:total lipids increased, both the particle sizes and zeta potentials gradually increased. When the ratio of PSH:total lipids was greater than or equal to 40%, the standard deviations of both particle sizes and zeta potentials became obviously larger. So too much PSH was not conducive to the formation of nanoparticles. When the ratio of PSH:total lipids was 30%, the particle size, and zeta potential were 131.87 ± 8.45 nm and 15.09 ± 0.70 mV, which were both suitable for encapsulating siRNA. In order to load siRNA efficiently, the ratio of siRNA in nanoparticles was necessary to be controlled. As shown in , the change of particle size was not obvious when the value of PSH:siRNA was 12:1 to 8:1. Unfortunately, as the value of PSH:siRNA changed to 6:1 or less, the particle size increased obviously. Meanwhile, as the concentration of siRNA increased, the zeta potential decreased. In order to further verify the combination of PSH-NP and siRNA at various ratios of PSH:siRNA (12:1 to 2:1), agarose gel retardation assay was used. showed that the bright electrophoretic band disappeared as the ratio of PSH:siRNA increased, which showed that PSH-NP could completely combine with siRNA when the ratio of PSH:siRNA was greater than 6:1. Hence, PSH:siRNA = 8:1 was the appropriate ratio for the preparation of PSH-NP/S from the result of the agarose gel retardation assay, which was consistent with the result of particle size and zeta potential measurement. The concentration of the targeting peptide had an important influence on the physical and chemical properties of the nanoparticles. For the group of cRGD-PSH-NP/S, the particle sizes and zeta potentials of the nanoparticles with various concentrations of cRGD were tested . As the ratio of cRGD:total lipids changed from 1% to 9%, the zeta potentials showed decreased slightly, which might result from the presence of aspartic acid in cRGD. However, with the gradual increase of cRGD, the particle sizes gradually increased, possibly due to the agglomeration of nanoparticles. Especially, when the ratio of cRGD:total lipids was greater than 5%, the particle size of cRGD-PSH-NP/S increased obviously. Based on the above optimization results, the optimal formulation of cRGD-PSH-NP/S was as follows: PSH:total lipids = 30%, PSH:siRNA = 8:1, cRGD:total lipids = 5%. 3.2. Characterization of cRGD-PSH-NP/S FE-SEM was used to observe the morphology and size of cRGD-PSH-NP/S. The SEM image was shown in . The shape of cRGD-PSH-NP/S was mostly spherical and the surface was smooth. The size distribution of cRGD-PSH-NP/S was 100–400 nm within the field of view under SEM. The results of the particle size and the zeta potential of cRGD-PSH-NP/S measured by the laser particle size analyzer were shown in , the mean values were 169.60 ± 5.43 nm and 4.45 ± 0.19 mV. In order to investigate the combination of cRGD-PSH-NP and the other nanoparticles with survivin siRNA, an agarose gel retardation assay was implemented. As shown in , naked survivin siRNA showed bright electrophoretic bands, and the electrophoretic bands of PEI-NP/S, PSH-NP/S, cRGD-NP/S, and cRGD-PSH-NP/S disappeared, which showed PEI-NP, PSH-NP, cRGD-NP, and cRGD-PSH-NP could completely combine with survivin siRNA at the tested ratios. PEI-NP and PSH-NP could bind to siRNA, cRGD-NP, and cRGD-PSH-NP could still bind tightly to siRNA after adding DSPE-PEG 2000 -cRGD, there was no separation of the vectors and siRNA within a certain concentration range. In order to investigate the stability of cRGD-PSH-NP/S in FBS, an agarose gel retardation assay was conducted. As shown in , the electrophoretic band of the naked siRNA group disappeared completely after incubating with FBS for 6 h, while the electrophoretic band of the cRGD-PSH-NP/S group could still be seen for 12 h. The result showed that cRGD-PSH-NP could effectively protect siRNA from being degraded by nucleases in serum. 3.3. Cytotoxicity tests Excellent nanoparticles loaded with siRNA should be safe and have low toxicity, so MTT assay was used to test the viability of HepG2 and LO2 cells after they were treated with PEI-NP, PSH-NP, cRGD-NP, or cRGD-PSH-NP, the results were shown in . For the group of PEI-NP, the cell viability of HepG2 cells and LO2 cells decreased gradually with the increase of the concentration of PEI-NP. PEI-NP showed toxicity because of the large numbers of aminos. Fortunately, PSH-NP, cRGD-NP, and cRGD-PSH-NP showed lower toxicity compared with PEI-NP. The cell viability of HepG2 and LO2 cells remained above 80% at the experimental concentrations, the addition of the nanoparticles did not have major impacts on cell viability, which showed PSH-NP, cRGD-NP, and cRGD-PSH-NP were safe as the vehicles for delivering survivin siRNA. 3.4. Cellular uptakes of the nanoparticles The cellular uptakes of 5′-Cy3-labeled survivin siRNA delivered by PSH-NP, cRGD-NP and cRGD-PSH-NP in HepG2 cells were firstly determined by flow cytometry. showed the fluorescence migration curve of HepG2 cells treated with naked siRNA, PSH-NP/S, cRGD-NP/S, or cRGD-PSH-NP/S. The migration diagram on the left showed that no FBS involved in the administration process, and the one on the right had FBS. The fluorescence curve of the group of naked siRNA did not significantly migrate compared with the control group because the siRNA was degraded by nuclease or repelled from the cell membrane due to the strong negative charge. By contrast, the curves of PSH-NP/S, cRGD-NP/S, and cRGD-PSH-NP/S were significantly migrated, especially the group of cRGD-PSH-NP/S with or without FBS, which showed that in the presence of FBS, cRGD-PSH-NP could also efficiently transfect survivin siRNA into HepG2 cells. The mean fluorescence intensities of HepG2 cells were presented to evaluate the transfection efficiency of each group more specifically . The mean fluorescence intensity values of the group of cRGD-PSH-NP/S were 18.00 ± 0.96 and 19.03 ± 1.57 in HepG2 cells with or without FBS, which were more than 2.5 times that of cRGD-NP/S and 3 times that of PSH-NP/S. For the cells treated with cRGD-PSH-NP/S, the mean fluorescence intensity was significantly increased compared with PSH-NP/S and cRGD-NP/S ( p < .01), whether FBS existed or not. The results of cellular uptakes determined by flow cytometry of the nanoparticles showed cRGD-PSH-NP was effective in improving the transfection efficiency of survivin siRNA. 3.5. Intracellular distribution of the nanoparticles by confocal laser microscopy Confocal laser microscopy was also used to observe the transfection of naked siRNA, PSH-NP/S, cRGD-NP/S, and cRGD-PSH-NP/S more directly, siRNA was labeled with 5′-Cy3. showed the intracellular distribution of naked siRNA, PSH-NP/S, cRGD-NP/S, and cRGD-PSH-NP/S in HepG2 cells without FBS and with FBS, respectively. The images showed that only weak red fluorescence in the field of vision of the group of naked siRNA which indicated naked siRNA could not effectively close to the cell membrane and then internalize into the cell to perform RNAi without the help of the vectors. And as expected, the red fluorescence signals of the groups of PSH-NP/S, cRGD-NP/S, and cRGD-PSH-NP/S were enhanced in the field of vision compared with the naked siRNA, especially the group of cRGD-PSH-NP/S. Results showed the uptake of survivin siRNA by HepG2 cells was obviously enhanced with the help of cRGD-PSH-NP even in the presence of FBS. showed the internalization of cRGD-PSH-NP/S under high magnification (63×). Within the field of view, a lot of red fluorescence was observed around or inside the blue nucleus, which showed 5′-Cy3-labeled siRNA delivered by cRGD-PSH-NP entered HepG2 cells in large quantities, cRGD-PSH-NP/S could cross the cell membrane and be distributed in the cytoplasm. The results of the confocal laser microscope in further illustrated that cRGD-PSH-NP was an effective vector for loading and delivering survivin siRNA in HepG2 cells. 3.6. Determination of survivin protein expression The expression level of survivin protein was an important indicator to evaluate whether the above vectors transfect siRNA into HepG2 cells efficiently. The survivin protein levels in HepG2 cells treated with PEI-NP/S, PSH-NP/S, cRGD-NP/S, or cRGD-PSH-NP/S were determined by Western blot, the image and analysis result were shown in . The levels of survivin protein in PSH-NP/S, cRGD-NP/S, and cRGD-PSH-NP/S groups were all down-regulated at varying degrees, especially of the group of cRGD-PSH-NP/S. More specifically, the relative survivin protein expression of PEI-NP/S, PSH-NP/S, cRGD-NP/S, and cRGD-PSH-NP/S were 79.42%, 73.83%, 55.23%, and 32.01%, respectively. Correspondingly, the survivin down-regulations were 20.58%, 26.17%, 44.77%, and 67.99%, respectively. The survivin down-regulation of cRGD-PSH-NP/S in HepG2 cells was significantly improved compared with the other groups, which showed the adornment of cRGD and PSH was effective for increasing the transfection efficiency of siRNA in HepG2 cells. 3.7. In vivo antitumor efficacy of the nanoparticles In vivo antitumor efficacy of cRGD-NP/S and cRGD-PSH-NP/S were evaluated in BALB/C nude mice. The relative body weights and the tumor volumes are shown in . During the administration period, the weight of the nude mice did not decrease significantly in the group of cRGD-NP/S and cRGD-PSH-NP/S, however, since the 12th day of administration, the weight of the nude mice in the control group decreased, which was closely related to the rapid growth of the tumors. The tumor growth curve after administration demonstrated that the tumor growth in cRGD-NP/S and cRGD-PSH-NP/S groups was significantly slower than that in the normal saline-treated control group. It was obvious that the cRGD-NP/S and cRGD-PSH-NP/S inhibited the tumor growth effectively, especially the group of cRGD-PSH-NP/S. The tumor inhibition rates of cRGD-NP/S and cRGD-PSH-NP/S were 44.30%, and 74.71%, respectively. The inhibition rate of cRGD-PSH-NP/S was significantly higher than that of the saline-treated control ( p < .01) and of cRGD-NP/S ( p < .01). The photograph of the harvested tumors further confirmed the tumor inhibition and verified cRGD-PSH-NP/S was effective for delivering survivin siRNA into HepG2 cells. Meanwhile, there was no obvious functional abnormality in eating, behavior, response to external stimulus in the group of cRGD-NP/S and cRGD-PSH-NP/S. Further, the Western blot analysis result showed that the expression levels of survivin in tumors treated with cRGD-NP/S and cRGD-PSH-NP/S were lower than that of the control group, and the level of survivin in tumors treated with cRGD-PSH-NP/S was significantly lower than that treated with cRGD-NP/S . These results confirmed that cRGD and PSH could help the siRNA enter tumor cells and enhance the efficiency of RNAi in tumor-bearing mice. 3.8. In vivo safety evaluation The in vivo safety of cRGD-NP/S and cRGD-PSH-NP/S in mice was further evaluated through H&E staining. There were no obvious histological differences between the major organs including heart, liver, spleen, lungs, and kidney of cRGD-NP/S, cRGD-PSH-NP/S, and saline-treated mice . Images proved that the major organs did not show any significant damage or inflammation to the mice in the groups of cRGD-NP/S and cRGD-PSH-NP/S.
Formulation optimization of cRGD-PSH-NP/S The particle sizes and zeta potentials of PSH-NP with various concentrations of PSH were firstly measured to explore the appropriate concentration of PSH . As the ratios of PSH:total lipids increased, both the particle sizes and zeta potentials gradually increased. When the ratio of PSH:total lipids was greater than or equal to 40%, the standard deviations of both particle sizes and zeta potentials became obviously larger. So too much PSH was not conducive to the formation of nanoparticles. When the ratio of PSH:total lipids was 30%, the particle size, and zeta potential were 131.87 ± 8.45 nm and 15.09 ± 0.70 mV, which were both suitable for encapsulating siRNA. In order to load siRNA efficiently, the ratio of siRNA in nanoparticles was necessary to be controlled. As shown in , the change of particle size was not obvious when the value of PSH:siRNA was 12:1 to 8:1. Unfortunately, as the value of PSH:siRNA changed to 6:1 or less, the particle size increased obviously. Meanwhile, as the concentration of siRNA increased, the zeta potential decreased. In order to further verify the combination of PSH-NP and siRNA at various ratios of PSH:siRNA (12:1 to 2:1), agarose gel retardation assay was used. showed that the bright electrophoretic band disappeared as the ratio of PSH:siRNA increased, which showed that PSH-NP could completely combine with siRNA when the ratio of PSH:siRNA was greater than 6:1. Hence, PSH:siRNA = 8:1 was the appropriate ratio for the preparation of PSH-NP/S from the result of the agarose gel retardation assay, which was consistent with the result of particle size and zeta potential measurement. The concentration of the targeting peptide had an important influence on the physical and chemical properties of the nanoparticles. For the group of cRGD-PSH-NP/S, the particle sizes and zeta potentials of the nanoparticles with various concentrations of cRGD were tested . As the ratio of cRGD:total lipids changed from 1% to 9%, the zeta potentials showed decreased slightly, which might result from the presence of aspartic acid in cRGD. However, with the gradual increase of cRGD, the particle sizes gradually increased, possibly due to the agglomeration of nanoparticles. Especially, when the ratio of cRGD:total lipids was greater than 5%, the particle size of cRGD-PSH-NP/S increased obviously. Based on the above optimization results, the optimal formulation of cRGD-PSH-NP/S was as follows: PSH:total lipids = 30%, PSH:siRNA = 8:1, cRGD:total lipids = 5%.
Characterization of cRGD-PSH-NP/S FE-SEM was used to observe the morphology and size of cRGD-PSH-NP/S. The SEM image was shown in . The shape of cRGD-PSH-NP/S was mostly spherical and the surface was smooth. The size distribution of cRGD-PSH-NP/S was 100–400 nm within the field of view under SEM. The results of the particle size and the zeta potential of cRGD-PSH-NP/S measured by the laser particle size analyzer were shown in , the mean values were 169.60 ± 5.43 nm and 4.45 ± 0.19 mV. In order to investigate the combination of cRGD-PSH-NP and the other nanoparticles with survivin siRNA, an agarose gel retardation assay was implemented. As shown in , naked survivin siRNA showed bright electrophoretic bands, and the electrophoretic bands of PEI-NP/S, PSH-NP/S, cRGD-NP/S, and cRGD-PSH-NP/S disappeared, which showed PEI-NP, PSH-NP, cRGD-NP, and cRGD-PSH-NP could completely combine with survivin siRNA at the tested ratios. PEI-NP and PSH-NP could bind to siRNA, cRGD-NP, and cRGD-PSH-NP could still bind tightly to siRNA after adding DSPE-PEG 2000 -cRGD, there was no separation of the vectors and siRNA within a certain concentration range. In order to investigate the stability of cRGD-PSH-NP/S in FBS, an agarose gel retardation assay was conducted. As shown in , the electrophoretic band of the naked siRNA group disappeared completely after incubating with FBS for 6 h, while the electrophoretic band of the cRGD-PSH-NP/S group could still be seen for 12 h. The result showed that cRGD-PSH-NP could effectively protect siRNA from being degraded by nucleases in serum.
Cytotoxicity tests Excellent nanoparticles loaded with siRNA should be safe and have low toxicity, so MTT assay was used to test the viability of HepG2 and LO2 cells after they were treated with PEI-NP, PSH-NP, cRGD-NP, or cRGD-PSH-NP, the results were shown in . For the group of PEI-NP, the cell viability of HepG2 cells and LO2 cells decreased gradually with the increase of the concentration of PEI-NP. PEI-NP showed toxicity because of the large numbers of aminos. Fortunately, PSH-NP, cRGD-NP, and cRGD-PSH-NP showed lower toxicity compared with PEI-NP. The cell viability of HepG2 and LO2 cells remained above 80% at the experimental concentrations, the addition of the nanoparticles did not have major impacts on cell viability, which showed PSH-NP, cRGD-NP, and cRGD-PSH-NP were safe as the vehicles for delivering survivin siRNA.
Cellular uptakes of the nanoparticles The cellular uptakes of 5′-Cy3-labeled survivin siRNA delivered by PSH-NP, cRGD-NP and cRGD-PSH-NP in HepG2 cells were firstly determined by flow cytometry. showed the fluorescence migration curve of HepG2 cells treated with naked siRNA, PSH-NP/S, cRGD-NP/S, or cRGD-PSH-NP/S. The migration diagram on the left showed that no FBS involved in the administration process, and the one on the right had FBS. The fluorescence curve of the group of naked siRNA did not significantly migrate compared with the control group because the siRNA was degraded by nuclease or repelled from the cell membrane due to the strong negative charge. By contrast, the curves of PSH-NP/S, cRGD-NP/S, and cRGD-PSH-NP/S were significantly migrated, especially the group of cRGD-PSH-NP/S with or without FBS, which showed that in the presence of FBS, cRGD-PSH-NP could also efficiently transfect survivin siRNA into HepG2 cells. The mean fluorescence intensities of HepG2 cells were presented to evaluate the transfection efficiency of each group more specifically . The mean fluorescence intensity values of the group of cRGD-PSH-NP/S were 18.00 ± 0.96 and 19.03 ± 1.57 in HepG2 cells with or without FBS, which were more than 2.5 times that of cRGD-NP/S and 3 times that of PSH-NP/S. For the cells treated with cRGD-PSH-NP/S, the mean fluorescence intensity was significantly increased compared with PSH-NP/S and cRGD-NP/S ( p < .01), whether FBS existed or not. The results of cellular uptakes determined by flow cytometry of the nanoparticles showed cRGD-PSH-NP was effective in improving the transfection efficiency of survivin siRNA.
Intracellular distribution of the nanoparticles by confocal laser microscopy Confocal laser microscopy was also used to observe the transfection of naked siRNA, PSH-NP/S, cRGD-NP/S, and cRGD-PSH-NP/S more directly, siRNA was labeled with 5′-Cy3. showed the intracellular distribution of naked siRNA, PSH-NP/S, cRGD-NP/S, and cRGD-PSH-NP/S in HepG2 cells without FBS and with FBS, respectively. The images showed that only weak red fluorescence in the field of vision of the group of naked siRNA which indicated naked siRNA could not effectively close to the cell membrane and then internalize into the cell to perform RNAi without the help of the vectors. And as expected, the red fluorescence signals of the groups of PSH-NP/S, cRGD-NP/S, and cRGD-PSH-NP/S were enhanced in the field of vision compared with the naked siRNA, especially the group of cRGD-PSH-NP/S. Results showed the uptake of survivin siRNA by HepG2 cells was obviously enhanced with the help of cRGD-PSH-NP even in the presence of FBS. showed the internalization of cRGD-PSH-NP/S under high magnification (63×). Within the field of view, a lot of red fluorescence was observed around or inside the blue nucleus, which showed 5′-Cy3-labeled siRNA delivered by cRGD-PSH-NP entered HepG2 cells in large quantities, cRGD-PSH-NP/S could cross the cell membrane and be distributed in the cytoplasm. The results of the confocal laser microscope in further illustrated that cRGD-PSH-NP was an effective vector for loading and delivering survivin siRNA in HepG2 cells.
Determination of survivin protein expression The expression level of survivin protein was an important indicator to evaluate whether the above vectors transfect siRNA into HepG2 cells efficiently. The survivin protein levels in HepG2 cells treated with PEI-NP/S, PSH-NP/S, cRGD-NP/S, or cRGD-PSH-NP/S were determined by Western blot, the image and analysis result were shown in . The levels of survivin protein in PSH-NP/S, cRGD-NP/S, and cRGD-PSH-NP/S groups were all down-regulated at varying degrees, especially of the group of cRGD-PSH-NP/S. More specifically, the relative survivin protein expression of PEI-NP/S, PSH-NP/S, cRGD-NP/S, and cRGD-PSH-NP/S were 79.42%, 73.83%, 55.23%, and 32.01%, respectively. Correspondingly, the survivin down-regulations were 20.58%, 26.17%, 44.77%, and 67.99%, respectively. The survivin down-regulation of cRGD-PSH-NP/S in HepG2 cells was significantly improved compared with the other groups, which showed the adornment of cRGD and PSH was effective for increasing the transfection efficiency of siRNA in HepG2 cells.
In vivo antitumor efficacy of the nanoparticles In vivo antitumor efficacy of cRGD-NP/S and cRGD-PSH-NP/S were evaluated in BALB/C nude mice. The relative body weights and the tumor volumes are shown in . During the administration period, the weight of the nude mice did not decrease significantly in the group of cRGD-NP/S and cRGD-PSH-NP/S, however, since the 12th day of administration, the weight of the nude mice in the control group decreased, which was closely related to the rapid growth of the tumors. The tumor growth curve after administration demonstrated that the tumor growth in cRGD-NP/S and cRGD-PSH-NP/S groups was significantly slower than that in the normal saline-treated control group. It was obvious that the cRGD-NP/S and cRGD-PSH-NP/S inhibited the tumor growth effectively, especially the group of cRGD-PSH-NP/S. The tumor inhibition rates of cRGD-NP/S and cRGD-PSH-NP/S were 44.30%, and 74.71%, respectively. The inhibition rate of cRGD-PSH-NP/S was significantly higher than that of the saline-treated control ( p < .01) and of cRGD-NP/S ( p < .01). The photograph of the harvested tumors further confirmed the tumor inhibition and verified cRGD-PSH-NP/S was effective for delivering survivin siRNA into HepG2 cells. Meanwhile, there was no obvious functional abnormality in eating, behavior, response to external stimulus in the group of cRGD-NP/S and cRGD-PSH-NP/S. Further, the Western blot analysis result showed that the expression levels of survivin in tumors treated with cRGD-NP/S and cRGD-PSH-NP/S were lower than that of the control group, and the level of survivin in tumors treated with cRGD-PSH-NP/S was significantly lower than that treated with cRGD-NP/S . These results confirmed that cRGD and PSH could help the siRNA enter tumor cells and enhance the efficiency of RNAi in tumor-bearing mice.
In vivo safety evaluation The in vivo safety of cRGD-NP/S and cRGD-PSH-NP/S in mice was further evaluated through H&E staining. There were no obvious histological differences between the major organs including heart, liver, spleen, lungs, and kidney of cRGD-NP/S, cRGD-PSH-NP/S, and saline-treated mice . Images proved that the major organs did not show any significant damage or inflammation to the mice in the groups of cRGD-NP/S and cRGD-PSH-NP/S.
Discussion As a potential nucleic acid drug, siRNA needs efficient transport vectors to assist it to exert the RNA interference effect (Khalil et al., ; Jiao et al., ). In this study, a reduction-sensitive cationic polymer PSH and a targeting ligand cRGD were used for preparing the nanoparticle named cRGD-PSH-NP to deliver survivin siRNA . The formulations and characteristics of cRGD-PSH-NP loaded with siRNA (cRGD-PSH-NP/S) were investigated, meanwhile a series of the in vitro and in vivo transfection activity were evaluated. The results of the formulation optimization of cRGD-PSH-NP/S showed the molar ratios of various components of cRGD-PSH-NP/S had an important influence on its particle size and zeta potential. Excessive PSH might cause instability or even aggregation of nanoparticles (Yang et al., ). The additional siRNA could cause the nanoparticles to approach neutrality and thus decrease the interaction with the cell membrane. DSPE-PEG 2000 -cRGD could form a dense structure on the surface of nanoparticles which will help to prolong the circulation time in the blood and provide targeting ligand, however, excessive DSPE-PEG 2000 -cRGD could also make the nanoparticles unstable. Finally, based on the results of particle size, zeta potential, and agarose gel retardation assay, the optimized molar ratios of various components were: PSH:total lipids = 30%, PSH:siRNA = 8:1, and cRGD:total lipids = 5% . The particle size and zeta potential of cRGD-PSH-NP/S prepared by the optimal formulation were 169.60 ± 5.43 nm (<200 nm) and 4.45 ± 0.19 mV, under this condition , which were suitable for enhanced permeability and retention effect of solid tumor. Other studies on the properties of the nanoparticles show that cRGD-PSH-NP could bind tightly to siRNA and cRGD-PSH-NP could effectively protect siRNA from being degraded by nucleases in serum . The cell viabilities of HepG2 and LO2 cells treated with PSH-NP, cRGD-NP, and cRGD-PSH-NP remained above 80% at the experimental concentrations in MTT assay, which proved PSH and DSPE-PEG 2000 -cRGD based on PEI and cRGD had been modified to reduce toxicity. The above information showed cRGD-PSH-NP were safe as the vehicles for survivin siRNA in HepG2 cells and normal liver cells . The related experiments in vitro showed that the cellular uptakes of 5′-Cy3-labeled survivin siRNA delivered by cRGD-PSH-NP in HepG2 cells were significantly increased compared with PSH-NP/S and cRGD-NP/S, which were 3 times that of PSH-NP despite the presence of FBS . The cellular uptakes determined by flow cytometry of the nanoparticles showed cRGD-PSH-NP was effective in improving the transfection efficiency of survivin siRNA. The intracellular distribution of the nanoparticles was observed by confocal laser microscopy. The fluorescence signal of the group of cRGD-PSH-NP/S was obviously enhanced compared with the naked siRNA even in the presence of FBS. A large number of nanoparticles carrying siRNA surround and enter the cell, or even enter the nucleus which was more intuitive to observe that cRGD-PSH-NP/S was efficiently transfected into the tumor cells ( and ). The expression level of survivin protein was determined to further evaluate the RNAi effect of siRNA. The survivin down regulations of cRGD-PSH-NP/S was 67.99%, which was significantly improved compared with PEI-NP/S and PSH-NP/S in HepG2 cells . In vivo antitumor efficacy of cRGD-PSH-NP/S showed that the tumor inhibition rate of cRGD-PSH-NP/S was 74.71% , and the level of survivin in tumors treated with cRGD-PSH-NP/S was significantly lower than that treated with cRGD-NP/S . Fortunately, the treatment did not affect the weight and safety of the nude mice ( and ). The in vivo and in vitro test results indicated that cRGD-PSH-NP was an effective and safe vector for loading and delivering siRNA into tumor cells to exert RNAi effect. The increase of the antitumor activity of cRGD-PSH-NP might be mainly due to the role of cRGD as a targeting ligand for the α v β 3 integrin over-expression on the surface of hepatocarcinoma cells (Tang et al., ). At the same time, the reduction response characteristics and ‘proton pump effect’ of PSH were conducive to the disintegration of nanoparticles and the rapid release of siRNA into the cytoplasm (Yang et al., ). The long circulation characteristic of PEG-DSPE also played a certain role in the improvement of antitumor activity (Sasayama et al., ).
Conclusion In conclusion, the nanoparticles cRGD-PSH-NP containing a cationic polymer based on PEI (PSH) and a targeting ligand (DSPE-PEG 2000 -cRGD) were efficient to deliver survivin siRNA. We showed that the cRGD-PSH-NP was low in toxicity and be able to bind siRNA efficiently. In vitro and in vivo experiments demonstrated that the cRGD-PSH-NP/S was safe and had high antitumor activity. We suggested that cRGD-PSH-NP might be suitable and effective for further research for the delivery of other nucleic acid drugs.
|
The Measurement of Vital Signs in Pediatric Patients by Lifelight Software in Comparison to the Standard of Care: Protocol for the VISION-Junior Observational Study | 8850da34-b4b2-4c7b-aec8-50a5e0047319 | 11953603 | Medicine[mh] | Measurement of vital signs (VS) is an essential part of clinical evaluation, providing important information about a patient’s health; importantly, a change in VS may herald a deterioration in clinical condition and may precede the emergence of physical symptoms during the early stages of an infection . High pulse rate (PR) and respiratory rate (RR) have been shown to predict hospital admission in children . Infections such as influenza, scarlet fever, and chickenpox are common causes of morbidity and mortality in immunocompromised children . Three-quarters of children who develop severe sepsis from infection have a comorbid condition such as asthma, diabetes, cystic fibrosis, or cancer. Sepsis carries a mortality rate of almost 50% in the most vulnerable children . Treatment of infection in children can be expensive, especially if intensive care is required. Given that monitoring of VS is important in clinically vulnerable children to aid early detection of infection , a robust method that parents can use at home to measure VS is likely to be beneficial and would support remote initial assessment . The measurement of VS is time-consuming and requires trained staff and multiple pieces of equipment, some of which need regular calibration. Accurate measurement of blood pressure (BP) using a sphygmomanometer requires the use of the correct size cuff for the patient. The physical contact of equipment can be uncomfortable and distressing for patients, particularly those with learning disabilities. Stress can affect BP, PR, and RR, such that the measurements may not accurately reflect the patient’s state of health . Standard of care (SOC) medical equipment is not suitable for regular measurement of VS in the home or community setting. Adherence to frequent VS monitoring using traditional technologies is low because of the inconvenience and stigmatizing nature of specialist hardware equipment (eg, thermometers, BP cuffs, and finger-clip oximeters) . The COVID-19 pandemic has also highlighted the need for methods suitable for the remote or contactless measurement of VS that can be operated by people without specific medical training. The growing use of telemedicine since the pandemic also points to the need for easy but accurate measurement of VS. Photoplethysmography (PPG) is an optical measurement technique that records changes in the light reflected from the skin surface due to volumetric changes in the facial blood vessels; small variations in perfusion provide valuable information about the cardiovascular system . PPG has been used to measure PR , oxygen saturation , BP , and RR and to detect atrial fibrillation . Lifelight is an app being developed for the contactless measurement of VS using remote PPG via the camera on smart devices such as phones and tablets. The app captures the average color of regions of interest on the face 30 times every second for 60 seconds and sends these as red, green, and blue color values to a central server for processing, yielding VS values. The VISION-D (Measurement of Vital Signs by Lifelight Software in Comparison to the Standard of Care—Development) study in 2018-2019 measured VS in 8585 patients and healthy volunteers simultaneously using the app and SOC methods. The data were used for machine learning to develop the algorithms used to calculate VS. The subsequent VISION-V (Validation) study demonstrated the accuracy of the app compared with SOC methods for measuring PR, RR, and diastolic BP , providing the basis for the current Class I Conformité Européenne (CE) mark . The app also received ISO 13485:2016 certification in August 2023 , and the accuracy of the app for measuring PR and BP against ISO standards in adults with stage 1 hypertension has recently been reported . The VISION-MD (multisite development) study evaluates the app versus SOC in patients with a wide range of clinically relevant VS values, including critically ill patients, and across the full range of skin tones . In VISION-MD, high-resolution full-face videos of participants’ faces have been recorded, which has proved instrumental in improving the accuracy of VS measurement compared with the VISION-D and -V studies. The VISION-Junior study explores the measurement of VS using a Lifelight app developed for pediatric patients (aged 18 years or younger) attending the pediatric emergency department. As in VISION-MD in adults, high-resolution full-face videos will be recorded. In younger children (younger than 5 years of age), the video will also include the torso to increase the surface area from which the signal is recorded, and because the torso is likely to move less than the face. Audio data will also be recorded, as crying, wheezing, coughing, and other sounds can also predict illness in children . The aim of the VISION-Junior study is to assess whether the app can be used to accurately measure VS in pediatric patients. This protocol describes the collection of data that will initially be used to develop the algorithms for the measurement of VS in children. Later data will be used to assess the accuracy of the app across the range of VS values encountered in children of different ages. The objectives are as follows. The primary objective is to collect data to support the development of the app for the accurate measurement of PR, RR, and oxygen saturation in pediatric patients and to assess the accuracy of measurements against SOC methods. Secondary objectives are: (1) to assess—using the collected PPG data—whether other (direct or indirect) measurements can be made using the app, including temperature, BP, PR variability, and audio analysis; (2) to evaluate the impact of subject-specific variables on the PPG VS measurements (eg, medication, cosmetics, facial or body hair, and skin tone); (3) to obtain data to understand patient and parent or guardian or carer habits and preferences for VS measurement using questionnaires ( and ); and (4) to assess the usability of Lifelight Junior for measuring VS in pediatric patients through questionnaires completed by participating health care professionals ( ). Most of the data collected in the study will be used to train the app algorithms to measure VS in pediatric patients. The remaining data will be kept separate and used to assess the accuracy of the algorithms at regular intervals during the data collection period. The study delivery process will be responsive to this learning curve to increase the likelihood that the predetermined accuracy target for each VS estimation is achieved by the study end.
The design and protocol for this study have been informed by the VISION-MD study and earlier VISION studies . Participants and Recruitment Approximately 500 pediatric patients are being recruited at the Sunderland Royal Hospital pediatric emergency department of the South Tyneside and Sunderland National Health Service (NHS) Foundation Trust. All pediatric patients (younger than 18 years of age) attending the pediatric emergency department are potentially eligible for the study. The exclusion criteria are critical illness (as judged by the treating physician), including unconscious patients, patients who were noncompliant in terms of excessive movement during the VS measurements, and patients whose parents, guardians, or carers do not speak English (required to give informed consent). The study aims to recruit according to age, sex, Fitzpatrick skin tone, and medical history criteria, as subsequently described. The research nurse approaches the parents, guardians, or carers of potential participants identified by the clinical team about taking part in the study. Potential participants aged 17 or 18 years or the parents, guardians, or carers of younger children are provided with an ethics-approved information sheet explaining the study aims, what is involved, and the requirements for participation. Younger children also watch ethics-approved age-appropriate videos (for ages 5-10 and 11-16 years) explaining the study on a tablet device ( and ). The information sheet and videos were developed in collaboration with the Young People Advisory Group North England and the South Tyneside and Sunderland NHS Foundation Trust Young Persons Group. These groups also provided feedback on the design of the study, consent/assent forms, and other study documents at 2 in-person meetings. Study Procedures Demographic data (age, sex, ethnicity, skin tone [Fitzpatrick 6-point scale ranging from type 1 for skin that always burns to type 6 for skin that never burns]) are recorded on an electronic case report form using the Castor electronic platform. Medical history questions are limited to the presence or absence of conditions that might affect skin perfusion and pigmentation, cardiovascular processes, and any prescription medicines for these conditions. The study staff records a set of premeasurement observations and the presence or absence of sweat on the participant’s face; any facial hair on the cheeks; tattoos, jewelry, birthmarks, scars, or other features on the face; the use of foundation or concealer; and the position of the participant (seated, prone, supine or lying on one side). Patients may also be recruited into 1 of 3 subprotocols depending on premeasurement observations ( ). The study staff ensures that participants have been at rest for at least 10 minutes before VS measurement starts (which can include the period of data collection described previously). In each study session, VS is measured as per the subprotocol using the SOC equipment, while simultaneously capturing a video of the participant’s face using the Lifelight Data Collect app running on a tablet (standard iPad). This app is a research tool that captures and uploads video data to secure cloud storage for research purposes only and is distinct from the CE-marked Lifelight medical device developed from data collected from previous studies. The Data Collect app does not display any VS estimates and is not used for clinical interpretation or analysis. In each study session, at least 2 VS (PR, oxygen saturation, RR) and temperature are measured twice over a 60-second period using SOC equipment and methods. A standard clinical finger clip sensor for the measurement of oxygen saturation and PR is placed on a finger. Oxygen saturation and PR are measured at 0, 30, and 60 seconds of the recording period and averaged. RR is determined by counting chest rises throughout the 60-second period. The nurse may place their hand on the participant’s chest to improve the accuracy of counting while being mindful not to obscure the camera’s line of sight. At the same time, a video of the participant’s face (and torso in children younger than 5 years of age) and audio is captured for up to 3 minutes using the Data Collect app running on a tablet positioned approximately 1 meter away and angled toward the participant’s face. Background luminosity is measured automatically. Recording is started and stopped by the research nurse via the iPad. A maximum of 3 attempts at recording measurements is made. Once measurements have been completed, the study staff fill in postmeasurement questions relating to how much the participant moved, their position, whether they were wearing glasses, any hairstyle or other item (eg, face covering) that obscured any part of their face during the recording, and whether the app reported “face not found” at any point during the recording. A questionnaire is also available to garner feedback on the technology from the clinical user’s point of view (ie, the nurses who take the VS measurements). Questionnaire data are fully anonymized and recorded without any identifiable information. Each study session, including approaching the participant, obtaining consent, and performing the study procedure takes about 30 minutes. Participants were not followed up after the VS measurements but continued to receive clinical care as appropriate for their condition. Subprotocols Participants are recruited into 3 subprotocols as appropriate ( ). Subprotocols 1 and 2 may be run concurrently with the main protocol if the research nurse considers it practical to also record BP and an electrocardiogram ( ). Ethical Considerations The study is being performed in accordance with the spirit and the letter of the Declaration of Helsinki, the Good Clinical Practice Guidelines, and the protocol and applicable local regulatory requirements and laws. The VISION-Junior protocol was approved by the Health Research Authority and Health and Care Research Wales (IRAS reference 321956) and the Newcastle North Tyneside Research Ethics Committee on March 30, 2023. Informed consent is obtained electronically using Castor eConsent, a secure NHS-compliant platform. Informed consent is provided by participants aged 16 years and older and by the parent, guardian, or carer of younger children. Children younger than 16 years of age are also given the opportunity to sign an assent form. As the study is noninterventional, participants will not receive financial compensation for participation but may be offered small incentives such as stickers or certificates. Participation in the study is entirely voluntary, and participants may withdraw from the study at any time. Specific, explicit written consent was obtained from all individuals featured in the videos provided in Multimedia appendices 4 and 5, permitting use for publication purposes. Privacy and Data Collection Each study participant is assigned a unique sequential patient identifier; no identifiable data are stored. All documents are stored securely and can only be accessed by study staff and authorized personnel. The code linking the patient identifier to their personal information is kept within the hospital study site and can only be accessed by the research team within that site. For each reading, a high-quality video of the face (and torso in young children) and an audio file are saved to the internal storage of the iPad in encrypted form and transmitted directly to the sponsor’s NHS-compliant cloud server. Files were automatically deleted following upload to the server. All data collected about participants during the study will be kept strictly confidential and handled according to the General Data Protection Regulations. Videos and audio recordings collected during the study constitute personal data as it may be possible to identify participants. Full-resolution, full-face, or torso video data are required to develop Lifelight Junior into a clinically useful device; the analysis cannot be performed using blurred videos. The data collected for research purposes only (not clinical care) are processed within the legitimate interests of Xim Ltd. Data Handling Analysis of the videos uses the encrypted files, which, in most cases, are processed automatically within a secure cloud infrastructure (ie, without any person viewing the videos); in rare cases, it may be necessary to analyze a video manually if the recording is not as expected. The analysis results in anonymized aggregate datasets. No decrypted files are stored at any point in the processing. The electronic case report forms are stored and managed using the Castor electronic cloud. Performance Targets While Lifelight is certified as a class I CE medical device , the performance of Lifelight Junior needs to be appropriate for routine clinical practice in pediatric patients. lists the performance targets for Lifelight Junior for measuring VS in pediatric patients; training data collected during VISION-Junior will support progress toward these targets. Sample Size An estimated 500 pediatric patients (18 years old or younger) will be recruited into the study. This sample size is informed by the previous VISION-D, VISION-MD, and VISION-Acute studies . However, the sample size cannot be formally calculated because it depends on the incremental improvement in the performance of the app achieved through machine learning using the training data generated early in the study. The split between training and testing data will be determined during the study according to the quality of the data collected. The study management team monitors the progress of data collection and accuracy and updates the study teams weekly. The study will continue until the accuracy of the app for measuring VS in pediatric patients is sufficient for various clinical use cases. In addition, recruitment targets according to age, sex, Fitzpatrick skin tone, and medical history have been defined, as shown in , to ensure accuracy in a broad range of clinically relevant patients. Data Analysis All statistical analyses will be performed using Python (Python Software Foundation). Training data will be used for machine learning to improve the performance of Lifelight Junior for measuring VS in pediatric patients. Appropriate machine learning models will be developed according to the data collected. This is likely to include signal processing and measurement algorithms for PR and RR, and multilayer neural network-based approaches for BP, similar to those used in the adult version of the app. The separate set of test data will subsequently be used to determine the performance of the app for measuring VS in pediatric patients. Correspondence between VS measurements predicted by the app and the measurements taken manually using SOC methods will be compared using regression coefficients, mean error, and SD, and compared against the performance targets set out in . An accuracy target will be considered met if the mean error and SD for the app measurements are at least equal to the target ( ). Heat maps or scatter plots will also be developed. Linear regression will be used to assess the impact of skin tone on the accuracy of the app for measuring each VS, using the Fitzpatrick skin tones as the explanatory variable. Quantitative data from the questionnaires will be summarized. Thematic analysis will be used to analyze free-text responses to identify perceptions of the app and variations in the way patients, their parents or carers, and health care professionals interact with it and any challenges they encounter. All analyses will be completed per protocol since there is no intention to treat. There will be no imputation of missing or implausible data, and any missing, implausible, or problematic readings will be excluded from analysis.
Approximately 500 pediatric patients are being recruited at the Sunderland Royal Hospital pediatric emergency department of the South Tyneside and Sunderland National Health Service (NHS) Foundation Trust. All pediatric patients (younger than 18 years of age) attending the pediatric emergency department are potentially eligible for the study. The exclusion criteria are critical illness (as judged by the treating physician), including unconscious patients, patients who were noncompliant in terms of excessive movement during the VS measurements, and patients whose parents, guardians, or carers do not speak English (required to give informed consent). The study aims to recruit according to age, sex, Fitzpatrick skin tone, and medical history criteria, as subsequently described. The research nurse approaches the parents, guardians, or carers of potential participants identified by the clinical team about taking part in the study. Potential participants aged 17 or 18 years or the parents, guardians, or carers of younger children are provided with an ethics-approved information sheet explaining the study aims, what is involved, and the requirements for participation. Younger children also watch ethics-approved age-appropriate videos (for ages 5-10 and 11-16 years) explaining the study on a tablet device ( and ). The information sheet and videos were developed in collaboration with the Young People Advisory Group North England and the South Tyneside and Sunderland NHS Foundation Trust Young Persons Group. These groups also provided feedback on the design of the study, consent/assent forms, and other study documents at 2 in-person meetings.
Demographic data (age, sex, ethnicity, skin tone [Fitzpatrick 6-point scale ranging from type 1 for skin that always burns to type 6 for skin that never burns]) are recorded on an electronic case report form using the Castor electronic platform. Medical history questions are limited to the presence or absence of conditions that might affect skin perfusion and pigmentation, cardiovascular processes, and any prescription medicines for these conditions. The study staff records a set of premeasurement observations and the presence or absence of sweat on the participant’s face; any facial hair on the cheeks; tattoos, jewelry, birthmarks, scars, or other features on the face; the use of foundation or concealer; and the position of the participant (seated, prone, supine or lying on one side). Patients may also be recruited into 1 of 3 subprotocols depending on premeasurement observations ( ). The study staff ensures that participants have been at rest for at least 10 minutes before VS measurement starts (which can include the period of data collection described previously). In each study session, VS is measured as per the subprotocol using the SOC equipment, while simultaneously capturing a video of the participant’s face using the Lifelight Data Collect app running on a tablet (standard iPad). This app is a research tool that captures and uploads video data to secure cloud storage for research purposes only and is distinct from the CE-marked Lifelight medical device developed from data collected from previous studies. The Data Collect app does not display any VS estimates and is not used for clinical interpretation or analysis. In each study session, at least 2 VS (PR, oxygen saturation, RR) and temperature are measured twice over a 60-second period using SOC equipment and methods. A standard clinical finger clip sensor for the measurement of oxygen saturation and PR is placed on a finger. Oxygen saturation and PR are measured at 0, 30, and 60 seconds of the recording period and averaged. RR is determined by counting chest rises throughout the 60-second period. The nurse may place their hand on the participant’s chest to improve the accuracy of counting while being mindful not to obscure the camera’s line of sight. At the same time, a video of the participant’s face (and torso in children younger than 5 years of age) and audio is captured for up to 3 minutes using the Data Collect app running on a tablet positioned approximately 1 meter away and angled toward the participant’s face. Background luminosity is measured automatically. Recording is started and stopped by the research nurse via the iPad. A maximum of 3 attempts at recording measurements is made. Once measurements have been completed, the study staff fill in postmeasurement questions relating to how much the participant moved, their position, whether they were wearing glasses, any hairstyle or other item (eg, face covering) that obscured any part of their face during the recording, and whether the app reported “face not found” at any point during the recording. A questionnaire is also available to garner feedback on the technology from the clinical user’s point of view (ie, the nurses who take the VS measurements). Questionnaire data are fully anonymized and recorded without any identifiable information. Each study session, including approaching the participant, obtaining consent, and performing the study procedure takes about 30 minutes. Participants were not followed up after the VS measurements but continued to receive clinical care as appropriate for their condition.
Participants are recruited into 3 subprotocols as appropriate ( ). Subprotocols 1 and 2 may be run concurrently with the main protocol if the research nurse considers it practical to also record BP and an electrocardiogram ( ).
The study is being performed in accordance with the spirit and the letter of the Declaration of Helsinki, the Good Clinical Practice Guidelines, and the protocol and applicable local regulatory requirements and laws. The VISION-Junior protocol was approved by the Health Research Authority and Health and Care Research Wales (IRAS reference 321956) and the Newcastle North Tyneside Research Ethics Committee on March 30, 2023. Informed consent is obtained electronically using Castor eConsent, a secure NHS-compliant platform. Informed consent is provided by participants aged 16 years and older and by the parent, guardian, or carer of younger children. Children younger than 16 years of age are also given the opportunity to sign an assent form. As the study is noninterventional, participants will not receive financial compensation for participation but may be offered small incentives such as stickers or certificates. Participation in the study is entirely voluntary, and participants may withdraw from the study at any time. Specific, explicit written consent was obtained from all individuals featured in the videos provided in Multimedia appendices 4 and 5, permitting use for publication purposes.
Each study participant is assigned a unique sequential patient identifier; no identifiable data are stored. All documents are stored securely and can only be accessed by study staff and authorized personnel. The code linking the patient identifier to their personal information is kept within the hospital study site and can only be accessed by the research team within that site. For each reading, a high-quality video of the face (and torso in young children) and an audio file are saved to the internal storage of the iPad in encrypted form and transmitted directly to the sponsor’s NHS-compliant cloud server. Files were automatically deleted following upload to the server. All data collected about participants during the study will be kept strictly confidential and handled according to the General Data Protection Regulations. Videos and audio recordings collected during the study constitute personal data as it may be possible to identify participants. Full-resolution, full-face, or torso video data are required to develop Lifelight Junior into a clinically useful device; the analysis cannot be performed using blurred videos. The data collected for research purposes only (not clinical care) are processed within the legitimate interests of Xim Ltd.
Analysis of the videos uses the encrypted files, which, in most cases, are processed automatically within a secure cloud infrastructure (ie, without any person viewing the videos); in rare cases, it may be necessary to analyze a video manually if the recording is not as expected. The analysis results in anonymized aggregate datasets. No decrypted files are stored at any point in the processing. The electronic case report forms are stored and managed using the Castor electronic cloud.
While Lifelight is certified as a class I CE medical device , the performance of Lifelight Junior needs to be appropriate for routine clinical practice in pediatric patients. lists the performance targets for Lifelight Junior for measuring VS in pediatric patients; training data collected during VISION-Junior will support progress toward these targets.
An estimated 500 pediatric patients (18 years old or younger) will be recruited into the study. This sample size is informed by the previous VISION-D, VISION-MD, and VISION-Acute studies . However, the sample size cannot be formally calculated because it depends on the incremental improvement in the performance of the app achieved through machine learning using the training data generated early in the study. The split between training and testing data will be determined during the study according to the quality of the data collected. The study management team monitors the progress of data collection and accuracy and updates the study teams weekly. The study will continue until the accuracy of the app for measuring VS in pediatric patients is sufficient for various clinical use cases. In addition, recruitment targets according to age, sex, Fitzpatrick skin tone, and medical history have been defined, as shown in , to ensure accuracy in a broad range of clinically relevant patients.
All statistical analyses will be performed using Python (Python Software Foundation). Training data will be used for machine learning to improve the performance of Lifelight Junior for measuring VS in pediatric patients. Appropriate machine learning models will be developed according to the data collected. This is likely to include signal processing and measurement algorithms for PR and RR, and multilayer neural network-based approaches for BP, similar to those used in the adult version of the app. The separate set of test data will subsequently be used to determine the performance of the app for measuring VS in pediatric patients. Correspondence between VS measurements predicted by the app and the measurements taken manually using SOC methods will be compared using regression coefficients, mean error, and SD, and compared against the performance targets set out in . An accuracy target will be considered met if the mean error and SD for the app measurements are at least equal to the target ( ). Heat maps or scatter plots will also be developed. Linear regression will be used to assess the impact of skin tone on the accuracy of the app for measuring each VS, using the Fitzpatrick skin tones as the explanatory variable. Quantitative data from the questionnaires will be summarized. Thematic analysis will be used to analyze free-text responses to identify perceptions of the app and variations in the way patients, their parents or carers, and health care professionals interact with it and any challenges they encounter. All analyses will be completed per protocol since there is no intention to treat. There will be no imputation of missing or implausible data, and any missing, implausible, or problematic readings will be excluded from analysis.
The study started on June 12, 2023, and as of December 20, 2023, had recruited 303 participants. The study was extended until the end of March 2024 to enable the target recruitment (500) to be met; 532 participants had been recruited as of April 5, 2024. Data are currently being used for machine learning to refine the algorithms to improve clinical accuracy. We anticipate that final analyses to determine the performance of the app against the targets set out in will be complete by mid-August 2024. The performance of the app will be determined across the range of VS values encountered in children in clinical practice by comparing the VS values recorded using PPG versus standard methods.
Principal Findings This paper reports the protocol for obtaining data to develop algorithms for the measurement of VS in children using the app, and to evaluate the performance of the app against SOC measurement of VS in pediatric patients of all ages. Monitoring of VS in children is challenging because existing methods (eg, BP measurement via an arterial line or cuff; PR and RR via electrodes, adhesive pulse oximeters, and adhesive temperature probes and thermometers) are either invasive or contact-based and can cause discomfort, pain or distress, may disturb sleep, and may damage paper-thin skin, potentially increasing the risk of infection. Frequent VS measurement using different techniques is also resource-intensive. The app being developed will provide contactless noninvasive measurement of VS. Data collected in this and other studies will be used to train algorithms to develop the app for the measurement of VS in pediatric patients. The app aims to provide a risk-measured way for nonexperts to measure VS quickly and nonintrusively, and therefore, detect any early signs of deteriorating health or infection. This will make it easier for clinically vulnerable or immunocompromised children to attend school and interact with their peer groups, meeting their mental, physical, social, and educational needs, whilst providing reassurance to parents, guardians, or carers. Early detection of infection before any significant clinical deterioration occurs means that treatment can start sooner, improving the likelihood of a full recovery whilst also potentially avoiding hospital admission and intensive care, which are expensive for the child’s family (lost income, expenses) and health care providers. Lifelight can also be an effective tool for at-home monitoring and triage, reducing the burden on carers, emergency services, and hospitals. It may also be useful in nonspecialist health care settings such as primary care and prehospital care where specialist pediatric equipment is not available. Several other devices for the measurement of VS are currently being developed but all require physical contact, whether through a finger, the wrist, or the chest, and only one (Biospectral) is able to measure BP. Limitations This research may have some limitations. BP is inherently more complex to measure than PR and RR, in terms of the data form and machine learning and because reference measurements are less accurate. As with any recording device, signal quality may be compromised if the participant moves excessively or if light levels are insufficient. However, we expect the use of high-resolution video recording to overcome such issues. In addition, measurements can be easily repeated if a recording is of poor quality. Skin type is a potential source of error with PPG devices, as melanin absorbs green light, potentially increasing errors in measurements in dark-compared with light-skinned individuals . However, we have already shown that skin type does not affect the accuracy of Lifelight . Although the Fitzpatrick Skin Type Scale is the current gold standard , its use has been criticized because of racial bias, weak correlation with skin color, and broad within-group variations in skin tone. Spectrocolorimetry, which uses multiple variables to categorize skin tone objectively, has been proposed as an alternative , which may be incorporated into later studies to confirm our findings. Conclusions The results of this study will be disseminated via submissions to relevant journals and conferences. The study team will explore different avenues and formats for the dissemination of the study findings with the Young People Advisory Groups and other groups to ensure a wide public audience, for example, through coproduced public talks, studies in community communications, and social media.
This paper reports the protocol for obtaining data to develop algorithms for the measurement of VS in children using the app, and to evaluate the performance of the app against SOC measurement of VS in pediatric patients of all ages. Monitoring of VS in children is challenging because existing methods (eg, BP measurement via an arterial line or cuff; PR and RR via electrodes, adhesive pulse oximeters, and adhesive temperature probes and thermometers) are either invasive or contact-based and can cause discomfort, pain or distress, may disturb sleep, and may damage paper-thin skin, potentially increasing the risk of infection. Frequent VS measurement using different techniques is also resource-intensive. The app being developed will provide contactless noninvasive measurement of VS. Data collected in this and other studies will be used to train algorithms to develop the app for the measurement of VS in pediatric patients. The app aims to provide a risk-measured way for nonexperts to measure VS quickly and nonintrusively, and therefore, detect any early signs of deteriorating health or infection. This will make it easier for clinically vulnerable or immunocompromised children to attend school and interact with their peer groups, meeting their mental, physical, social, and educational needs, whilst providing reassurance to parents, guardians, or carers. Early detection of infection before any significant clinical deterioration occurs means that treatment can start sooner, improving the likelihood of a full recovery whilst also potentially avoiding hospital admission and intensive care, which are expensive for the child’s family (lost income, expenses) and health care providers. Lifelight can also be an effective tool for at-home monitoring and triage, reducing the burden on carers, emergency services, and hospitals. It may also be useful in nonspecialist health care settings such as primary care and prehospital care where specialist pediatric equipment is not available. Several other devices for the measurement of VS are currently being developed but all require physical contact, whether through a finger, the wrist, or the chest, and only one (Biospectral) is able to measure BP.
This research may have some limitations. BP is inherently more complex to measure than PR and RR, in terms of the data form and machine learning and because reference measurements are less accurate. As with any recording device, signal quality may be compromised if the participant moves excessively or if light levels are insufficient. However, we expect the use of high-resolution video recording to overcome such issues. In addition, measurements can be easily repeated if a recording is of poor quality. Skin type is a potential source of error with PPG devices, as melanin absorbs green light, potentially increasing errors in measurements in dark-compared with light-skinned individuals . However, we have already shown that skin type does not affect the accuracy of Lifelight . Although the Fitzpatrick Skin Type Scale is the current gold standard , its use has been criticized because of racial bias, weak correlation with skin color, and broad within-group variations in skin tone. Spectrocolorimetry, which uses multiple variables to categorize skin tone objectively, has been proposed as an alternative , which may be incorporated into later studies to confirm our findings.
The results of this study will be disseminated via submissions to relevant journals and conferences. The study team will explore different avenues and formats for the dissemination of the study findings with the Young People Advisory Groups and other groups to ensure a wide public audience, for example, through coproduced public talks, studies in community communications, and social media.
|
What Happens Inside the Germinating Grain After Microbial Decontamination by Pulsed Electric Field? Data-Driven Multi-Omics Helps Find the Answer | 1066b99b-07a4-4c7b-b57f-9fb059cdee80 | 11858265 | Biochemistry[mh] | The research on plant metabolic pathways has advanced significantly in recent decades, driven by breakthroughs in genomic, transcriptomic, proteomic, and metabolomic studies. These comprehensive investigations have yielded vast amounts of information on the biochemical and physiological processes in plants and microorganisms. The genome sequencing of numerous plant species has facilitated the identification of genes encoding key enzymes involved in essential biosynthetic pathways. Meanwhile, the rise of metabolomics, powered by cutting-edge analytical techniques, has enabled the detailed analysis of a broad spectrum of plant metabolites. A wide array of bioinformatic tools and databases such as KEGG , YMDB , PlantCyc , MetaCyc , GNPS , and Cytoscape , among others, have been developed to help scientists explore and understand the intricate biochemical pathways in plant organisms . However, despite the wealth of data available in these resources, challenges remain in covering specific aspects of secondary metabolism, particularly in relation to responses to abiotic stresses (such as drought, salinity, or extreme temperatures) or biotic stresses caused by pathogenic microbial interactions . These databases often fall short in providing clear and definitive answers to these complex questions. The complexity of the plant metabolome , coupled with the insufficiently mapped knowledge databases, has been a topic of ongoing discussion . This lack of understanding often hinders the accurate interpretation of metabolomic studies, which are crucial for improving agricultural practices, such as crop cultivation and plant resistance, together with minimizing fungal diseases, as well as for improving the processing technologies of plant-based foods and minimizing the occurrence of undesirable fungal metabolites. One of the food processing technologies particularly susceptible to the spread of microbial infections, especially toxinogenic micromycetes of the Fusarium genus, is barley malting. During the malting process, the barley is soaked and germinated for several days under humid conditions, which create an environment conducive to the spread of fungal pathogens naturally present on the grain surface, followed by toxin production . However, as demonstrated in recent studies, treating the barley grains with short, intensive electric pulses (food processing technology called pulsed electric field, PEF) prior to malting disrupts the viability of these fungi through electroporation, significantly reducing the spread of fungal pathogens . Tackling the issue of minimizing the development of Fusarium micromycetes during malting using PEF holds great potential. Consequently, it is crucial to investigate how this intervention influences the physiology and metabolism of germinating grains, as well as the interactions between the grain and the pathogen. Although some information on the mutual interaction between plants and Fusarium species is available, most studies have been conducted on cereals under field cultivation conditions . Field or greenhouse experiments have shown, for instance, that metabolites closely associated with pathogenesis, which are upregulated in cereal varieties resistant to Fusarium diseases, include phenylpropanoids, flavonoids, terpenoids, hydroxycinnamic acid amides, and lignans . Similarly, at the genomic and transcriptomic levels, numerous pathogenesis-related genes involved in the development of field Fusarium diseases have been identified . There are also publications utilizing metabolomics and other “omics” disciplines to investigate the effects of biotic and abiotic stress on plant physiology . However, the processes occurring during barley growth in the field may differ significantly from those taking place during barley germination in malt houses. To the best of our knowledge, no studies currently exist that explore biochemical changes in barley (or other cereals) germinated under malting or similar food processing technologies and their relationship with Fusarium pathogens and/or abiotic stress. This study fills this knowledge gap by providing novel insights into gene expression and metabolic changes in germinated barley subjected to PEF intervention. Using a combined metabolomics–transcriptomics approach, it contextualizes these findings within the framework of Fusarium diseases. 2.1. Analysis of Fusarium Fungi In this study, malting barley grown in the field under the standard agricultural practice was used. The initial analysis revealed natural infection by Fusarium fungi in the grains at low levels, with specific concentrations measured as follows: 0.2423 ± 0.1524 ng/g for F. graminearum , 0.0014 ± 0.0011 ng/g for F. sporotrichioides , and 1.3307 ± 0.2613 ng/g for F. poae . After 72 h of germination, with and without the PEF pre-treatment, the barley samples were analyzed using RT-PCR to quantify pathogen DNA content. As shown in , despite the fact that the fungal DNA content increased during germination for all species studied, the content of F. poae and F. sporotrichioides DNA was significantly lower in the germinated barley samples treated with PEF compared to the untreated control. In contrast, the DNA levels of F. culmorum and F. graminearum remained relatively unchanged. Although a slight increase in mean DNA concentration for these two species was observed following PEF treatment, the changes were not statistically significant. These trends align with those observed by Karabin et al. and Prusova et al., who conducted similar experiments using barley artificially inoculated with Fusarium species . 2.2. Multi-Omics Analysis and Data Processing The level of expressed genes in the germinated barley, with and without prior PEF treatment, was analyzed by the next-generation sequencing (NGS), and changes in metabolite production were recorded by ultra-high-performance liquid chromatography coupled with high-resolution mass spectrometry (U-HPLC-HRMS/MS). As mentioned in the Introduction, although significant advances have been made in metabolomics and transcriptomics in plant research in recent years, there are still significant knowledge gaps, particularly in understanding plant-pathogen crosstalk. This is primarily due to the inherent complexity of plant and microbial metabolism and the still-incomplete mapping of these processes. Such limitations significantly hinder the application of the so-called “knowledge-driven” approach to data processing, which relies on integrating information from databases and libraries and establishing cross-connections across various levels of omics disciplines . Unlike humans and experimental mammals, whose metabolism is comparatively well studied and described, plants and fungi possess unique and highly complex biosynthetic pathways. These pathways are not only species-specific but often vary significantly even between different varieties of the same species. The influence of biotic or abiotic stress adds further layers of complexity, introducing additional unknowns into this area of research. Following that, for data processing in this study, a new multi-omics approach utilizing the ‘data-driven’ correlation method described by Ewald et al. was employed. This approach involved an initial selection of statistically significant variables from each single-omics level prior to the correlation analysis. Unlike the original protocol by Ewald et al. , the pre-selection of statistically significant variables (as described in , Variable Filtration at the Single-Omics Level, of ) was implemented to limit the number of variables entering the correlation analysis and to filter out correlations with low significance. The complete lists of output features from the transcriptomics and metabolomics analyses are provided in and , respectively, with all relevant details about the identified genes and metabolites available in the . These output lists were subsequently correlated using the OmicsAnalyst software (v. 2.0), following the methodology of Ewald et al. . The results of the data-driven correlation between transcriptomics and metabolomics data are presented in . For better clarity, all positive and negative correlation relationships illustrated by a relatively uncluttered ‘multi-omics spider’ are also listed in the . As shown in and , the only two barley genes whose expression was upregulated by PEF were the genes encoding vegetative gp1-like protein (gene ID LOC123411962), a late embryogenesis abundant protein (LEA) EMB564-like (gene ID LOC123439637). The vegetative gp1-like protein is known as a part of the plant cell wall acting as a mechanical barrier against negative external influences, including the penetration of fungal pathogens . The expression of the gene encoding this protein positively correlates with several metabolites, including amentoflavone 7,4′-dimethyl ether (met. ID neg_3828). According to the literature, flavonoids are known to protect plant cell wall integrity upon fungal infection by inhibiting the activity of specific cell wall degrading enzymes that are secreted by fungal pathogens to penetrate plant tissues . There is also a positive correlation between the vegetative gp1-like protein gene overexpression and the upregulation of the compound from triterpenoid glycosides class, called (2,3,4-trihydroxy-5-{[11-hydroxy-1-(2-hydroxy-6-methylhept-5-en-2-yl)-3a,3b,6,6,9a-pentamethyl-hexadecahydro-1H-cyclopenta[a]phenanthren-7-yl]oxy}cyclohexyl)methyl acetate (met. ID pos_7652). Terpene glycosides generally belong to the plant secondary metabolites that act as protective factors against fungal pathogens , by inhibiting the fungal (1→3)-β-D-glucan synthase, a key enzyme for rigid fungal cell wall . Another metabolite positively correlated with vegetative gp1-like protein gene expression was (2-methoxyphenyl)methyl 3-(methylsulfanyl)prop-2-enoate (met. ID neg_1070). Specifically, this compound itself has not been described in barley before, but related substances, derivatives of (methylsulfanyl)prop-2-enoate, have been identified as important components of the extract of other plants for which antifungal activity has been demonstrated . The above-described cluster of PEF-upregulated gene expression and metabolite production mutually correlated with coefficient ≥ 0.8 allows for the formulation of a hypothesis that PEF stimulates the biosynthesis of defensive biotic factors as a defense against Fusarium micromycetes, which could theoretically explain the reduction in the incidence of Fusarium species, specifically F. sporotrichioides and F. poae in PEF-treated barley (see ). Another factor upregulated after the PEF treatment is the gene encoding the LEA protein EMB564-like. In plants, these hydrophilic LEA proteins are highly produced in embryos during the late embryogenesis stage and in vegetative organs in response to the plant tissue desiccation and drought stress conditions . As can be seen in E, the expression of the LEA-encoding gene positively correlates with the upregulation of metabolite 6,7-dimethoxy-N-methyl-2-[3-(1H-pyrrol-1-yl)-3-(thiophen-3-yl)propanoyl]-1,2,3,4-tetrahydroisoquinoline-3-carboxamide (met ID pos_4675), the isoquinoline alkaloid derivative. Despite the fact that this metabolite has not been described in any previous barley-associated study, its positive correlation with LEA protein production makes good sense in the context of the results of the study performed on callus plants , where the metabolites upregulated in response to drought stress were identified as intermediates of isoquinoline alkaloid biosynthetic pathways . Given that drought stress cannot be taken into account during barley germination during malting (the usual relative humidity is between 90% and 100%), our data-driven correlation study suggests that the germinating grain accumulates LEA in response to another type of abiotic stress, most likely that caused by PEF treatment. In the recent literature, there are many references cross-connecting the expression of LEA genes and plant hormones, e.g., abscisic acid (ABA), as one of the key players in the adaptive response to drought stress; under the stress conditions, the biosynthesis and accumulation of ABA is enhanced, which triggers LEA gene expression . At the same time, it is known that ABA also triggers key defense mechanisms relevant to Fusarium infection . Despite the fact that ABA was not among the metabolites identified in our study, we can indirectly generalize that the increased production of specific signaling molecules in reaction to PEF-induced abiotic stress triggers not only the accumulation of LEA but also the Fusarium defensive factors. This theory is supported by the PEF-induced upregulation of two metabolites found in our study, in particular 4,12-dihydroxy-6-(hydroxymethyl)-16-methyl-17-[(N’-methylcarbamimidamido)methyl]-2-oxa-25-azapentacyclo [22.3.1.1 3,7 .1 13,17 .0 9,20 ]triaconta-3(30),4,6-trien-18-yn-10-yl acetate (met ID neg_4876), from the group of pentacyclic methylguanidine derivatives, and methyl 15-ethyl-10-[15-ethyl-18-(methoxycarbonyl)-17-methyl-10,17-diazatetracyclo [12.3.1.0 3,11 .0 4,9 ]octadeca-3(11),4,6,8-tetraen-12-yl]-12-hydroxy-17-methyl-10,17-diazatetracyclo [12.3.1.0 3,11 .0 4,9 ]octadeca-3(11),4,6,8-tetraene-18-carboxylate (met ID neg_5133), the derivative of indole alkaloids. As shown in previous studies, some of the polycyclic guanidine derivatives isolated from the natural resources show antifungal activities . Simple polyamines then play a positive role in the response of plants to various stress conditions, including drought , and indole alkaloids have been previously described as substances enhancing tolerance of plants to drought- or Fusarium -induced stress . Neither the methylguanidine derivative nor the indole alkaloid characterized in this study have been previously reported as constituents of barley, and their direct association with LEA has not been demonstrated in the literature before. Nevertheless, the relationships revealed by the data-driven multi-omics allow for the raising of new theories that need to be tested in the future and would never have emerged without the methodology used here. On the other hand, several genes associated with barley’s ability to defend itself against fungal infection were overexpressed also in the PEF untreated (control) variant. One such gene is the gene encoding the ethylene response factor 3-like (ERF3) protein (gene ID LOC123416694), which induces the expression of pathogen-related proteins that help alleviate pathogen infection ; in addition to this role, proteins from the ERF family are involved also in plant growth and fruit maturation . Another significant gene upregulated in untreated control is that encoding putrescine hydroxycinnamoyltransferase 3-like protein (gene ID LOC123405092), which belongs to the hydroxycinnamoyltransferase (HCT) family of enzymes. These enzymes catalyze the transfer of hydroxycinnamoyl groups in the biosynthesis of phenylpropanoids, from which monolignols and lignin, forming a mechanical barrier of the cell walls against infection, are further biosynthesized . Similarly, another overexpression in the control variant was observed for the gene encoding the dirigent protein 21-like (gene ID LOC123451201), which has also been previously described as an important factor in relation to lignin biosynthesis in response to biotic and abiotic stress . These trends may be related to the slightly lower levels of F. culmorum and F. graminearum in ‘control’ germinated barley without PEF intervention and illustrate well the complexity and ambiguity of the events occurring in a living plant organism. The overexpression of all of these control-related genes positively correlates with the upregulation of some lipid species in control samples, in particular with lysophospholipid LysoPE(0:0/22:5(7Z,10Z,13Z,16Z,19Z)), metabolite ID neg_3480, and ceramide N-(1,3-dihydroxy-16-methyloctadec-4-en-2-yl)-2-hydroxyhexadecanamide, metabolite ID neg_4323, (see E). In the current literature, the association of these metabolites with ERF has been described. The recent study by Wang et al. discusses the effects of ethylene and ERF on changes in the lipidome composition of peaches and presents the increased levels of ceramides and phospholipids , which is in agreement with our study. There are also studies in which the lipid profile was directly monitored in the context of resistance to Fusarium disease , and lysophospholipids have been identified as important signaling molecules involved in plant responses to stress . The above discussion represents the interpretation of several correlations that explain well the ‘germinating grain-pathogen’ crosstalk and find support in the existing literature. At the same time, there are still many correlations that have not been elucidated in this study. It is evident that the multi-omics data-driven correlation of barley metabolomics and transcriptomics is very complicated, even though only a relatively small number of statistically significant variables from each omics layer entered the correlation analysis. Although some of the correlations make good sense, others cannot be confirmed on the basis of the available knowledge. The difficulty in interpreting complex biological relationships between expressed genes and metabolites is undoubtedly also related to the fact that correlated variables may not always be the end points of mutual interactions but may be the intermediates of various biosynthetic or signaling pathways that are not yet fully described. Although this study does not provide a comprehensive explanation of all the correlated relationships, it is undoubtedly useful material for any follow-up studies dealing with the issue of plant-pathogen crosstalk, even completely independent of the PEF technology, since all the positively correlated relationships in control-related genes and metabolites actually correspond to normal PEF-unaffected barley. In this study, malting barley grown in the field under the standard agricultural practice was used. The initial analysis revealed natural infection by Fusarium fungi in the grains at low levels, with specific concentrations measured as follows: 0.2423 ± 0.1524 ng/g for F. graminearum , 0.0014 ± 0.0011 ng/g for F. sporotrichioides , and 1.3307 ± 0.2613 ng/g for F. poae . After 72 h of germination, with and without the PEF pre-treatment, the barley samples were analyzed using RT-PCR to quantify pathogen DNA content. As shown in , despite the fact that the fungal DNA content increased during germination for all species studied, the content of F. poae and F. sporotrichioides DNA was significantly lower in the germinated barley samples treated with PEF compared to the untreated control. In contrast, the DNA levels of F. culmorum and F. graminearum remained relatively unchanged. Although a slight increase in mean DNA concentration for these two species was observed following PEF treatment, the changes were not statistically significant. These trends align with those observed by Karabin et al. and Prusova et al., who conducted similar experiments using barley artificially inoculated with Fusarium species . The level of expressed genes in the germinated barley, with and without prior PEF treatment, was analyzed by the next-generation sequencing (NGS), and changes in metabolite production were recorded by ultra-high-performance liquid chromatography coupled with high-resolution mass spectrometry (U-HPLC-HRMS/MS). As mentioned in the Introduction, although significant advances have been made in metabolomics and transcriptomics in plant research in recent years, there are still significant knowledge gaps, particularly in understanding plant-pathogen crosstalk. This is primarily due to the inherent complexity of plant and microbial metabolism and the still-incomplete mapping of these processes. Such limitations significantly hinder the application of the so-called “knowledge-driven” approach to data processing, which relies on integrating information from databases and libraries and establishing cross-connections across various levels of omics disciplines . Unlike humans and experimental mammals, whose metabolism is comparatively well studied and described, plants and fungi possess unique and highly complex biosynthetic pathways. These pathways are not only species-specific but often vary significantly even between different varieties of the same species. The influence of biotic or abiotic stress adds further layers of complexity, introducing additional unknowns into this area of research. Following that, for data processing in this study, a new multi-omics approach utilizing the ‘data-driven’ correlation method described by Ewald et al. was employed. This approach involved an initial selection of statistically significant variables from each single-omics level prior to the correlation analysis. Unlike the original protocol by Ewald et al. , the pre-selection of statistically significant variables (as described in , Variable Filtration at the Single-Omics Level, of ) was implemented to limit the number of variables entering the correlation analysis and to filter out correlations with low significance. The complete lists of output features from the transcriptomics and metabolomics analyses are provided in and , respectively, with all relevant details about the identified genes and metabolites available in the . These output lists were subsequently correlated using the OmicsAnalyst software (v. 2.0), following the methodology of Ewald et al. . The results of the data-driven correlation between transcriptomics and metabolomics data are presented in . For better clarity, all positive and negative correlation relationships illustrated by a relatively uncluttered ‘multi-omics spider’ are also listed in the . As shown in and , the only two barley genes whose expression was upregulated by PEF were the genes encoding vegetative gp1-like protein (gene ID LOC123411962), a late embryogenesis abundant protein (LEA) EMB564-like (gene ID LOC123439637). The vegetative gp1-like protein is known as a part of the plant cell wall acting as a mechanical barrier against negative external influences, including the penetration of fungal pathogens . The expression of the gene encoding this protein positively correlates with several metabolites, including amentoflavone 7,4′-dimethyl ether (met. ID neg_3828). According to the literature, flavonoids are known to protect plant cell wall integrity upon fungal infection by inhibiting the activity of specific cell wall degrading enzymes that are secreted by fungal pathogens to penetrate plant tissues . There is also a positive correlation between the vegetative gp1-like protein gene overexpression and the upregulation of the compound from triterpenoid glycosides class, called (2,3,4-trihydroxy-5-{[11-hydroxy-1-(2-hydroxy-6-methylhept-5-en-2-yl)-3a,3b,6,6,9a-pentamethyl-hexadecahydro-1H-cyclopenta[a]phenanthren-7-yl]oxy}cyclohexyl)methyl acetate (met. ID pos_7652). Terpene glycosides generally belong to the plant secondary metabolites that act as protective factors against fungal pathogens , by inhibiting the fungal (1→3)-β-D-glucan synthase, a key enzyme for rigid fungal cell wall . Another metabolite positively correlated with vegetative gp1-like protein gene expression was (2-methoxyphenyl)methyl 3-(methylsulfanyl)prop-2-enoate (met. ID neg_1070). Specifically, this compound itself has not been described in barley before, but related substances, derivatives of (methylsulfanyl)prop-2-enoate, have been identified as important components of the extract of other plants for which antifungal activity has been demonstrated . The above-described cluster of PEF-upregulated gene expression and metabolite production mutually correlated with coefficient ≥ 0.8 allows for the formulation of a hypothesis that PEF stimulates the biosynthesis of defensive biotic factors as a defense against Fusarium micromycetes, which could theoretically explain the reduction in the incidence of Fusarium species, specifically F. sporotrichioides and F. poae in PEF-treated barley (see ). Another factor upregulated after the PEF treatment is the gene encoding the LEA protein EMB564-like. In plants, these hydrophilic LEA proteins are highly produced in embryos during the late embryogenesis stage and in vegetative organs in response to the plant tissue desiccation and drought stress conditions . As can be seen in E, the expression of the LEA-encoding gene positively correlates with the upregulation of metabolite 6,7-dimethoxy-N-methyl-2-[3-(1H-pyrrol-1-yl)-3-(thiophen-3-yl)propanoyl]-1,2,3,4-tetrahydroisoquinoline-3-carboxamide (met ID pos_4675), the isoquinoline alkaloid derivative. Despite the fact that this metabolite has not been described in any previous barley-associated study, its positive correlation with LEA protein production makes good sense in the context of the results of the study performed on callus plants , where the metabolites upregulated in response to drought stress were identified as intermediates of isoquinoline alkaloid biosynthetic pathways . Given that drought stress cannot be taken into account during barley germination during malting (the usual relative humidity is between 90% and 100%), our data-driven correlation study suggests that the germinating grain accumulates LEA in response to another type of abiotic stress, most likely that caused by PEF treatment. In the recent literature, there are many references cross-connecting the expression of LEA genes and plant hormones, e.g., abscisic acid (ABA), as one of the key players in the adaptive response to drought stress; under the stress conditions, the biosynthesis and accumulation of ABA is enhanced, which triggers LEA gene expression . At the same time, it is known that ABA also triggers key defense mechanisms relevant to Fusarium infection . Despite the fact that ABA was not among the metabolites identified in our study, we can indirectly generalize that the increased production of specific signaling molecules in reaction to PEF-induced abiotic stress triggers not only the accumulation of LEA but also the Fusarium defensive factors. This theory is supported by the PEF-induced upregulation of two metabolites found in our study, in particular 4,12-dihydroxy-6-(hydroxymethyl)-16-methyl-17-[(N’-methylcarbamimidamido)methyl]-2-oxa-25-azapentacyclo [22.3.1.1 3,7 .1 13,17 .0 9,20 ]triaconta-3(30),4,6-trien-18-yn-10-yl acetate (met ID neg_4876), from the group of pentacyclic methylguanidine derivatives, and methyl 15-ethyl-10-[15-ethyl-18-(methoxycarbonyl)-17-methyl-10,17-diazatetracyclo [12.3.1.0 3,11 .0 4,9 ]octadeca-3(11),4,6,8-tetraen-12-yl]-12-hydroxy-17-methyl-10,17-diazatetracyclo [12.3.1.0 3,11 .0 4,9 ]octadeca-3(11),4,6,8-tetraene-18-carboxylate (met ID neg_5133), the derivative of indole alkaloids. As shown in previous studies, some of the polycyclic guanidine derivatives isolated from the natural resources show antifungal activities . Simple polyamines then play a positive role in the response of plants to various stress conditions, including drought , and indole alkaloids have been previously described as substances enhancing tolerance of plants to drought- or Fusarium -induced stress . Neither the methylguanidine derivative nor the indole alkaloid characterized in this study have been previously reported as constituents of barley, and their direct association with LEA has not been demonstrated in the literature before. Nevertheless, the relationships revealed by the data-driven multi-omics allow for the raising of new theories that need to be tested in the future and would never have emerged without the methodology used here. On the other hand, several genes associated with barley’s ability to defend itself against fungal infection were overexpressed also in the PEF untreated (control) variant. One such gene is the gene encoding the ethylene response factor 3-like (ERF3) protein (gene ID LOC123416694), which induces the expression of pathogen-related proteins that help alleviate pathogen infection ; in addition to this role, proteins from the ERF family are involved also in plant growth and fruit maturation . Another significant gene upregulated in untreated control is that encoding putrescine hydroxycinnamoyltransferase 3-like protein (gene ID LOC123405092), which belongs to the hydroxycinnamoyltransferase (HCT) family of enzymes. These enzymes catalyze the transfer of hydroxycinnamoyl groups in the biosynthesis of phenylpropanoids, from which monolignols and lignin, forming a mechanical barrier of the cell walls against infection, are further biosynthesized . Similarly, another overexpression in the control variant was observed for the gene encoding the dirigent protein 21-like (gene ID LOC123451201), which has also been previously described as an important factor in relation to lignin biosynthesis in response to biotic and abiotic stress . These trends may be related to the slightly lower levels of F. culmorum and F. graminearum in ‘control’ germinated barley without PEF intervention and illustrate well the complexity and ambiguity of the events occurring in a living plant organism. The overexpression of all of these control-related genes positively correlates with the upregulation of some lipid species in control samples, in particular with lysophospholipid LysoPE(0:0/22:5(7Z,10Z,13Z,16Z,19Z)), metabolite ID neg_3480, and ceramide N-(1,3-dihydroxy-16-methyloctadec-4-en-2-yl)-2-hydroxyhexadecanamide, metabolite ID neg_4323, (see E). In the current literature, the association of these metabolites with ERF has been described. The recent study by Wang et al. discusses the effects of ethylene and ERF on changes in the lipidome composition of peaches and presents the increased levels of ceramides and phospholipids , which is in agreement with our study. There are also studies in which the lipid profile was directly monitored in the context of resistance to Fusarium disease , and lysophospholipids have been identified as important signaling molecules involved in plant responses to stress . The above discussion represents the interpretation of several correlations that explain well the ‘germinating grain-pathogen’ crosstalk and find support in the existing literature. At the same time, there are still many correlations that have not been elucidated in this study. It is evident that the multi-omics data-driven correlation of barley metabolomics and transcriptomics is very complicated, even though only a relatively small number of statistically significant variables from each omics layer entered the correlation analysis. Although some of the correlations make good sense, others cannot be confirmed on the basis of the available knowledge. The difficulty in interpreting complex biological relationships between expressed genes and metabolites is undoubtedly also related to the fact that correlated variables may not always be the end points of mutual interactions but may be the intermediates of various biosynthetic or signaling pathways that are not yet fully described. Although this study does not provide a comprehensive explanation of all the correlated relationships, it is undoubtedly useful material for any follow-up studies dealing with the issue of plant-pathogen crosstalk, even completely independent of the PEF technology, since all the positively correlated relationships in control-related genes and metabolites actually correspond to normal PEF-unaffected barley. 3.1. Experimental Material Barley of cultivar Bojos, which is most commonly used for the production of Pilsner type of malt, was used as input material for the PEF experiments and follow-up malting. This barley was harvested in 2022, and its natural contamination by four Fusarium species ( F. graminearum , F. culmorum , F. sporotrichioides , and F. poae ) was determined by the below-described RT-PCR method. 3.2. PEF Treatment and Barley Malting One kilogram of barley was pre-soaked for 10 min in 0.05 M phosphate buffer. After pre-soaking, the barley was exposed to PEF under 3.8 kV/cm voltage, 200 A current, 100 bipolar pulses, and 20 µs pulse width . As a control, one kilogram of barley was pre-soaked in 0.05 M phosphate buffer without PEF treatment for the same time period and subjected to malting under the same conditions. The automatic micromalting system (RAVOZ ® , Olomouc, Czech Republic), was used for barley steeping and germination. Both control and PEF-treated barley were malted according to following conditions: steeping for 48 h at 15 °C (8 h steeping; 12 h aeration break; 8 h steeping; 12 h aeration break; 4 h steeping; 4 h dripping), germination for 72 h at 15 °C; humidity 95–98%. After germination, the “green malt” samples from the PEF-treated and control variants were collected in the amount of 20 g, immediately snap-frozen in liquid nitrogen, and then stored at −180 °C until further analysis. The malting experiment was performed in three biological replicates, and from each replicate, the samples were taken in three repetitions. 3.3. Analysis of Fungal DNA The DNA isolation and RT-PCR assay was performed as described by Prusova et al. . In brief, the frozen samples were cryogenically homogenized, and DNA from 200 mg of each sample was isolated in duplicate using the Machery-Nagel (D) Plant II kit, with the follow-up addition of 10U of RNAse and proteinase K (Sigma-Aldrich, St. Louis, MO, USA). RT-PCR reactions were performed in 25 μL volumes consisting of 12.5 μL of TaqMan ® 2× Universal PCR MasterMix (ThermoFisher Scientific, Foster City, CA, USA), 300 nM of each of the forward and reverse primers (Generi Biotech, Hradec Kralove, Czech Republic), 200 nM of the probe (Generi Biotech, Hradec Kralove, Czech Republic), and 100 ng of template DNA and nuclease-free water (Sigma-Aldrich, St. Louis, MO, USA). Real-time primers and probes for the quantification of Fusarium species DNA are provided in . Quantitative RT-PCR was performed in the QuantStudioTM 6 cycler (Thermo Fisher Scientific, Waltham, MA, USA). The DNA concentrations were finally recalculated to dry matter of germinated barley. 3.4. RNA Isolation for Transcriptomics The samples were homogenized by using a cryogenic laboratory mill (IKA A11 basic, Verkon, Prague, Czech Republic). The RNA was extracted from 100 mg of frozen material using the TRIzol reagent (Invitrogen, Carlsbad, CA, USA) and purified with an RNeasy column in the presence of DNase (Qiagen, Hilden, Germany). The quality of the RNA was assessed using agarose gel electrophoresis and the Agilent 4200 TapeStation System (Agilent, Santa Clara, CA, USA). Each sample was represented by three independent replicates. 3.5. NGS Analysis and Data Treatment To ensure the integrity and quality of the RNA samples for subsequent NGS analysis, a secondary check was performed at the company providing the NGS analysis (Seqme, Prague, Czech Republic). This step confirmed that the samples met the required standards. Next, sequencing libraries were prepared for each biological replicate of every sample, targeting the 3′ end of the transcripts. RNA-Seq libraries were prepared using the Lexogen QuantSeq 3′ mRNA-Seq Library Prep Kit FWD, incorporating Unique Molecular Identifiers (UMIs) for error correction and reducing amplification bias, along with dual indexing for accurate sample multiplexing. This approach targeted the 3′ ends of polyadenylated RNA, simplifying the transcriptome sequencing and enabling precise gene expression profiling. Library quality control (QC) was performed using the Agilent Bioanalyzer 2100 (Agilent, Santa Clara, CA, USA), Invitrogen Collibri Library Quantification Kit (Thermo Fisher Scientific Baltics, Vilnius, Lithuania), and Qubit 1X dsDNA High-Sensitivity Assay Kit (Thermo Fisher Scientific, Waltham, MA, USA) to assess size distribution, concentration, and integrity, ensuring high-quality libraries suitable for accurate sequencing with minimal technical variability. Subsequently, individual libraries were normalized to compensate for possible differences in RNA concentration and pooled for further analysis. The NGS analysis was performed using the Illumina NovaSeq platform with the following parameters: single-end sequencing, 50 base pairs per read, and a sequencing depth of 100 million reads. During the sequencing process, raw fastq files were generated for each sample. These files underwent data preprocessing to ensure quality and remove unwanted sequences. The initial quality assessment was conducted using the FastQC (v. 0.11.9) and multiQC (v. 1.12) programs. Trim Galore (v. 0.6.7) was then employed to eliminate low-quality reads and adapter sequences. The resulting data were again evaluated using FastQC and multiQC. After preprocessing, the reads were mapped to the respective genome reference (GCA_904849725.1) using Hisat2 (v. 2.2.1). To analyze the differential expression, the mapped reads were utilized. The count of mapped reads was obtained using featureCounts (v. 2.0.3) and stored in a count matrix. The normalization of read counts and identification of significantly expressed genes were performed using DESeq2 (v. 1.26.0). For further analysis, zeroes in the count matrix were replaced using “R” and routines within missMDA libraryR, which does not disturb the original data structure. 3.6. Isolation of Metabolites All frozen samples were homogenized using a cryogenic laboratory mill (IKA A11 basic, IKA Works GmbH & Co. KG, Staufen im Breisgau, Germany). One gram of dry matter was weighed in to the 50 mL PTFE cuvette, and 10 mL of methanol was added. The cuvette with the homogenized sample was then placed on a laboratory shaker (HS 260 basic, IKA Works GmbH & Co. KG, Staufen im Breisgau, Germany) for 30 min (shaking at 240 RPM). The extract was then centrifuged (Rotina 35 R, Hettich Zentrifugen, Tuttlingen, Germany) for 2 min and 13,528 RCF (relative centrifugal force) and microfiltered through 0.2 µm spin filters (Ciro, Deerfield Beach, FL, USA). Finally, an aliquot of approx. 1 mL was transferred to a glass HPLC vial for further analysis by UHPLC-HRMS/MS. To eliminate potential system drift in the analytical system, the pooled extract (quality control sample (QC)) was prepared by mixing 100 µL of extract from each sample and analyzed. To exclude background signals from the laboratory processing of samples, a “processing blank” sample was prepared together with the analyzed samples. 3.7. Metabolomics Analysis The UHPLC-HRMS/MS analysis was performed according to the study by Stranska et al. , with slight modifications. Briefly, for the separation of metabolites, the UHPLC system (Dionex UltiMate 3000 RS UHPLC system, Thermo Fisher Scientific, Waltham, MA, USA) used a reverse phase column of Acquity UPLC ® BEHC18 (100 mm × 2.1 mm; 1.7 m; Waters, Milford, MA, USA). The injection volume was 2 µL, the autosampler temperature was 10 °C, and the column temperature was 60 °C. The mobile phase consisted of (A) 5 mM ammonium in a mixture of milli-Q water:methanol (95:5, v / v ) with 0.1% formic acid ( v / v ) and (B) 5 mM ammonium in a mixture of isopropanol:methanol:milli-Q water (65:30:5, v / v / v ) with 0.1% formic acid ( v / v ). Methanol, isopropanol, ammonium formate, and formic acid (all of LC-MS grade) were obtained from Merck (Darmstadt, Germany). Deionized water (18 MΩ, 25 °C) was prepared using a Milli-Q ® system (Millipore, Bedford, MA, USA). The SCIEX TripleTOF ® 6600 quadruple time-of-flight mass spectrometer (SCIEX, Concord, ON, Canada) was used for the detection in negative (ESI-) and positive (ESI+) ionization mode. To record the MS1 and MS/MS data, the TOF MS method (full scan) and the Information Dependent Acquisition (IDA) method were used. The working range was 100–1200 m / z for MS1 and 50–1000 m / z for MS/MS, and data were acquired between 0.5 and 19 min. The MS/MS spectra were collected for the eight most intensive ions of the MS spectra throughout the chromatographic run. The collision energy was 35 ± 15 V, and the QC sample was analyzed every ten samples during the sequence. Instrument control and data acquisition were operated with Analyst 1.7.1 TF software (SCIEX, Concord, ON, Canada), and qualitative analysis was performed using SCIEX OS software (v. 1.5.0.23389, SCIEX, Concord, ON, Canada). For data mining and processing, the freely available MS-DIAL (v. 4.80)–MS-CleanR—MS-FINDER (v. 3.52) software platform was used according to the workflow published by Stranska et al. . In the first step, the UHPLC-HRMS/MS data were processed and deconvoluted by MS-DIAL (v. 4.80, 2021, RIKEN, Tokyo, Japan). For peak picking, the MS1 and MS/MS tolerances were set to 0.03 and 0.1 Da in the centroid mode for both the datasets acquired in the ESI− and ESI+ mode. For peak alignment, the QC reference file was used with an RT tolerance of 0.05 min and mass tolerance of 0.015 Da. The minimum peak height for peak detection was set to amplitude 9000 (referring to 70% under the observed baseline for a blank injection). Zero values were replaced with 1/10 of the minimum peak height for all samples during the export of aligned results from MS-DIAL. In the second step, aligned results were cleaned with MS-CleanR (MetaToul-AgromiX Platform, Bordeaux, France). For data cleaning, all filters, including blank filters, a ghost peaks filter, incorrect mass, relative standard deviation, and relative mass defect were activated, with a minimum blank ratio set to 0.8, a maximum relative standard deviation (RSD) set to 30%, and a relative mass defect (RMD) ranging from 50 to 3000. The maximum RT difference and mass difference tolerance values were set to 0.025 min and 0.005 Da for Pearson correlation and ESI+/− data merging. In this step, feature clustering based on Pearson correlation and MS-DIAL peak character estimation algorithm (MS-DIAL-PCE) was implemented, and one peak (the most intense) was kept in each cluster. In the third step, the filtered features were automatically annotated with MS-FINDER (v. 3.52, 2021, RIKEN, Tokyo, Japan). The MS1 and MS/MS tolerances were set to 5 and 15 ppm, respectively. C, H, O, N, P, and S atoms were included for the formula finder parameter module. As a data source for automatic identification, the databases YMDB, ECMDB, PlantCyc, ChEBI, T3DB, NPA, KNApSAcK, LipidMaps, and PubChem were used. On the basis of the accurate mass and isotope ratio of MS1 ions, elemental compositions of candidate ions were proposed, and possible chemical structures of candidates were computed by comparing the experimental and in silico MS/MS spectra. Finally, MS-CleanR was launched, matched information from MS-FINDER automatic annotation with filtered features, and created a final data matrix with filtered and annotated features for further statistical analysis. In this step, MS-CleanR also matched data from the ESI+ and ESI− ionization mode into one data matrix. 3.8. Variable Filtration at the Single-Omics Level Both data matrices from both single-omics levels were filtered using univariate statistical tools (Volcano plot— t -test, Fold change) according to Behner et al. , to exclude statistically insignificant features and prevent the overfitting of final models. Univariate methods were performed using the freely available web platform MetaboAnalyst (v. 5.0, Xia Lab, McGill, Montreal, QC, Canada), with t -test false discovery rate (FDR) adjusted p -value < 0.01 for metabolomics and <0.05 for transcriptomics and Fold change >2 for both single-omics. The general overview of feature reduction during data filtration for all data matrices is summarized in . The metabolomics data matrix was processed with MS Excel to perform the normalization of the data (total area sum normalization). In the next step, both metabolomics and transcriptomics data were separately uploaded to SIMCA software (v. 17.0, 2021, Sartorius, Göttingen, Germany), where multivariate Principal Component Analysis (PCA) and Orthogonal Partial Least Square Discriminant Analysis (OPLS-DA) were performed. Before the creation of every classification model, Pareto scaling and logarithmic transformation of the data were performed. The quality of the models was determined using the R2Y and Q2 validation parameters calculated by a 7-round internal cross-validation. Candidate biomarker compounds and statistically significant genes were selected based on OPLS-DA S-plots and Variable Importance of the Projection (VIP) plots together with Receiver Operating Characteristics (ROCs). Both filters were applied simultaneously in an orthogonal way, to assure proper statistical filtration and to keep only the strong biomarkers/genes relevant for further data interpretation. For selection, VIP scores greater than one (VIP score > 1) and ROC area under curve (AUC = 1) parameters were set. 3.9. Multi-Omics Data Integration Multi-omics data integration was performed according to the workflow developed and described by Ewald et al. based on the web-based freely available software OmicsAnalyst (v. 2.0, Xia Lab, McGill, Montreal, QC, Canada). Although each individual omics dataset (metabolomics + transcriptomics) was uploaded, the ‘Metadata table’ file, which included information about samples, groups, and factors, was also uploaded to the OmicsAnalyst interface. After uploading, auto scaling was applied to both datasets. For the first view of multi-omics data integration, Multiple Co-Inertia Analysis (MCIA) was performed in Dimensionality Reduction module. MCIA represents a robust method of finding related multidimensional components across multiple datasets. In the next step, Data Integration Analysis for Biomarker discovery using Latent cOmponents (DIABLO) was performed. This method is a generalized and supervised version of PLS (multiblock PLS-DA) that seeks to find related multidimensional components that maximally separate sample labels. In the next step, the correlation network based on the DIABLO feature selection method and the classical Spearman correlation analysis similarity matrix method was performed. For building a correlation network, only between-omics connections were considered. The corr. threshold for between-omics was set to >0.8. Barley of cultivar Bojos, which is most commonly used for the production of Pilsner type of malt, was used as input material for the PEF experiments and follow-up malting. This barley was harvested in 2022, and its natural contamination by four Fusarium species ( F. graminearum , F. culmorum , F. sporotrichioides , and F. poae ) was determined by the below-described RT-PCR method. One kilogram of barley was pre-soaked for 10 min in 0.05 M phosphate buffer. After pre-soaking, the barley was exposed to PEF under 3.8 kV/cm voltage, 200 A current, 100 bipolar pulses, and 20 µs pulse width . As a control, one kilogram of barley was pre-soaked in 0.05 M phosphate buffer without PEF treatment for the same time period and subjected to malting under the same conditions. The automatic micromalting system (RAVOZ ® , Olomouc, Czech Republic), was used for barley steeping and germination. Both control and PEF-treated barley were malted according to following conditions: steeping for 48 h at 15 °C (8 h steeping; 12 h aeration break; 8 h steeping; 12 h aeration break; 4 h steeping; 4 h dripping), germination for 72 h at 15 °C; humidity 95–98%. After germination, the “green malt” samples from the PEF-treated and control variants were collected in the amount of 20 g, immediately snap-frozen in liquid nitrogen, and then stored at −180 °C until further analysis. The malting experiment was performed in three biological replicates, and from each replicate, the samples were taken in three repetitions. The DNA isolation and RT-PCR assay was performed as described by Prusova et al. . In brief, the frozen samples were cryogenically homogenized, and DNA from 200 mg of each sample was isolated in duplicate using the Machery-Nagel (D) Plant II kit, with the follow-up addition of 10U of RNAse and proteinase K (Sigma-Aldrich, St. Louis, MO, USA). RT-PCR reactions were performed in 25 μL volumes consisting of 12.5 μL of TaqMan ® 2× Universal PCR MasterMix (ThermoFisher Scientific, Foster City, CA, USA), 300 nM of each of the forward and reverse primers (Generi Biotech, Hradec Kralove, Czech Republic), 200 nM of the probe (Generi Biotech, Hradec Kralove, Czech Republic), and 100 ng of template DNA and nuclease-free water (Sigma-Aldrich, St. Louis, MO, USA). Real-time primers and probes for the quantification of Fusarium species DNA are provided in . Quantitative RT-PCR was performed in the QuantStudioTM 6 cycler (Thermo Fisher Scientific, Waltham, MA, USA). The DNA concentrations were finally recalculated to dry matter of germinated barley. The samples were homogenized by using a cryogenic laboratory mill (IKA A11 basic, Verkon, Prague, Czech Republic). The RNA was extracted from 100 mg of frozen material using the TRIzol reagent (Invitrogen, Carlsbad, CA, USA) and purified with an RNeasy column in the presence of DNase (Qiagen, Hilden, Germany). The quality of the RNA was assessed using agarose gel electrophoresis and the Agilent 4200 TapeStation System (Agilent, Santa Clara, CA, USA). Each sample was represented by three independent replicates. To ensure the integrity and quality of the RNA samples for subsequent NGS analysis, a secondary check was performed at the company providing the NGS analysis (Seqme, Prague, Czech Republic). This step confirmed that the samples met the required standards. Next, sequencing libraries were prepared for each biological replicate of every sample, targeting the 3′ end of the transcripts. RNA-Seq libraries were prepared using the Lexogen QuantSeq 3′ mRNA-Seq Library Prep Kit FWD, incorporating Unique Molecular Identifiers (UMIs) for error correction and reducing amplification bias, along with dual indexing for accurate sample multiplexing. This approach targeted the 3′ ends of polyadenylated RNA, simplifying the transcriptome sequencing and enabling precise gene expression profiling. Library quality control (QC) was performed using the Agilent Bioanalyzer 2100 (Agilent, Santa Clara, CA, USA), Invitrogen Collibri Library Quantification Kit (Thermo Fisher Scientific Baltics, Vilnius, Lithuania), and Qubit 1X dsDNA High-Sensitivity Assay Kit (Thermo Fisher Scientific, Waltham, MA, USA) to assess size distribution, concentration, and integrity, ensuring high-quality libraries suitable for accurate sequencing with minimal technical variability. Subsequently, individual libraries were normalized to compensate for possible differences in RNA concentration and pooled for further analysis. The NGS analysis was performed using the Illumina NovaSeq platform with the following parameters: single-end sequencing, 50 base pairs per read, and a sequencing depth of 100 million reads. During the sequencing process, raw fastq files were generated for each sample. These files underwent data preprocessing to ensure quality and remove unwanted sequences. The initial quality assessment was conducted using the FastQC (v. 0.11.9) and multiQC (v. 1.12) programs. Trim Galore (v. 0.6.7) was then employed to eliminate low-quality reads and adapter sequences. The resulting data were again evaluated using FastQC and multiQC. After preprocessing, the reads were mapped to the respective genome reference (GCA_904849725.1) using Hisat2 (v. 2.2.1). To analyze the differential expression, the mapped reads were utilized. The count of mapped reads was obtained using featureCounts (v. 2.0.3) and stored in a count matrix. The normalization of read counts and identification of significantly expressed genes were performed using DESeq2 (v. 1.26.0). For further analysis, zeroes in the count matrix were replaced using “R” and routines within missMDA libraryR, which does not disturb the original data structure. All frozen samples were homogenized using a cryogenic laboratory mill (IKA A11 basic, IKA Works GmbH & Co. KG, Staufen im Breisgau, Germany). One gram of dry matter was weighed in to the 50 mL PTFE cuvette, and 10 mL of methanol was added. The cuvette with the homogenized sample was then placed on a laboratory shaker (HS 260 basic, IKA Works GmbH & Co. KG, Staufen im Breisgau, Germany) for 30 min (shaking at 240 RPM). The extract was then centrifuged (Rotina 35 R, Hettich Zentrifugen, Tuttlingen, Germany) for 2 min and 13,528 RCF (relative centrifugal force) and microfiltered through 0.2 µm spin filters (Ciro, Deerfield Beach, FL, USA). Finally, an aliquot of approx. 1 mL was transferred to a glass HPLC vial for further analysis by UHPLC-HRMS/MS. To eliminate potential system drift in the analytical system, the pooled extract (quality control sample (QC)) was prepared by mixing 100 µL of extract from each sample and analyzed. To exclude background signals from the laboratory processing of samples, a “processing blank” sample was prepared together with the analyzed samples. The UHPLC-HRMS/MS analysis was performed according to the study by Stranska et al. , with slight modifications. Briefly, for the separation of metabolites, the UHPLC system (Dionex UltiMate 3000 RS UHPLC system, Thermo Fisher Scientific, Waltham, MA, USA) used a reverse phase column of Acquity UPLC ® BEHC18 (100 mm × 2.1 mm; 1.7 m; Waters, Milford, MA, USA). The injection volume was 2 µL, the autosampler temperature was 10 °C, and the column temperature was 60 °C. The mobile phase consisted of (A) 5 mM ammonium in a mixture of milli-Q water:methanol (95:5, v / v ) with 0.1% formic acid ( v / v ) and (B) 5 mM ammonium in a mixture of isopropanol:methanol:milli-Q water (65:30:5, v / v / v ) with 0.1% formic acid ( v / v ). Methanol, isopropanol, ammonium formate, and formic acid (all of LC-MS grade) were obtained from Merck (Darmstadt, Germany). Deionized water (18 MΩ, 25 °C) was prepared using a Milli-Q ® system (Millipore, Bedford, MA, USA). The SCIEX TripleTOF ® 6600 quadruple time-of-flight mass spectrometer (SCIEX, Concord, ON, Canada) was used for the detection in negative (ESI-) and positive (ESI+) ionization mode. To record the MS1 and MS/MS data, the TOF MS method (full scan) and the Information Dependent Acquisition (IDA) method were used. The working range was 100–1200 m / z for MS1 and 50–1000 m / z for MS/MS, and data were acquired between 0.5 and 19 min. The MS/MS spectra were collected for the eight most intensive ions of the MS spectra throughout the chromatographic run. The collision energy was 35 ± 15 V, and the QC sample was analyzed every ten samples during the sequence. Instrument control and data acquisition were operated with Analyst 1.7.1 TF software (SCIEX, Concord, ON, Canada), and qualitative analysis was performed using SCIEX OS software (v. 1.5.0.23389, SCIEX, Concord, ON, Canada). For data mining and processing, the freely available MS-DIAL (v. 4.80)–MS-CleanR—MS-FINDER (v. 3.52) software platform was used according to the workflow published by Stranska et al. . In the first step, the UHPLC-HRMS/MS data were processed and deconvoluted by MS-DIAL (v. 4.80, 2021, RIKEN, Tokyo, Japan). For peak picking, the MS1 and MS/MS tolerances were set to 0.03 and 0.1 Da in the centroid mode for both the datasets acquired in the ESI− and ESI+ mode. For peak alignment, the QC reference file was used with an RT tolerance of 0.05 min and mass tolerance of 0.015 Da. The minimum peak height for peak detection was set to amplitude 9000 (referring to 70% under the observed baseline for a blank injection). Zero values were replaced with 1/10 of the minimum peak height for all samples during the export of aligned results from MS-DIAL. In the second step, aligned results were cleaned with MS-CleanR (MetaToul-AgromiX Platform, Bordeaux, France). For data cleaning, all filters, including blank filters, a ghost peaks filter, incorrect mass, relative standard deviation, and relative mass defect were activated, with a minimum blank ratio set to 0.8, a maximum relative standard deviation (RSD) set to 30%, and a relative mass defect (RMD) ranging from 50 to 3000. The maximum RT difference and mass difference tolerance values were set to 0.025 min and 0.005 Da for Pearson correlation and ESI+/− data merging. In this step, feature clustering based on Pearson correlation and MS-DIAL peak character estimation algorithm (MS-DIAL-PCE) was implemented, and one peak (the most intense) was kept in each cluster. In the third step, the filtered features were automatically annotated with MS-FINDER (v. 3.52, 2021, RIKEN, Tokyo, Japan). The MS1 and MS/MS tolerances were set to 5 and 15 ppm, respectively. C, H, O, N, P, and S atoms were included for the formula finder parameter module. As a data source for automatic identification, the databases YMDB, ECMDB, PlantCyc, ChEBI, T3DB, NPA, KNApSAcK, LipidMaps, and PubChem were used. On the basis of the accurate mass and isotope ratio of MS1 ions, elemental compositions of candidate ions were proposed, and possible chemical structures of candidates were computed by comparing the experimental and in silico MS/MS spectra. Finally, MS-CleanR was launched, matched information from MS-FINDER automatic annotation with filtered features, and created a final data matrix with filtered and annotated features for further statistical analysis. In this step, MS-CleanR also matched data from the ESI+ and ESI− ionization mode into one data matrix. Both data matrices from both single-omics levels were filtered using univariate statistical tools (Volcano plot— t -test, Fold change) according to Behner et al. , to exclude statistically insignificant features and prevent the overfitting of final models. Univariate methods were performed using the freely available web platform MetaboAnalyst (v. 5.0, Xia Lab, McGill, Montreal, QC, Canada), with t -test false discovery rate (FDR) adjusted p -value < 0.01 for metabolomics and <0.05 for transcriptomics and Fold change >2 for both single-omics. The general overview of feature reduction during data filtration for all data matrices is summarized in . The metabolomics data matrix was processed with MS Excel to perform the normalization of the data (total area sum normalization). In the next step, both metabolomics and transcriptomics data were separately uploaded to SIMCA software (v. 17.0, 2021, Sartorius, Göttingen, Germany), where multivariate Principal Component Analysis (PCA) and Orthogonal Partial Least Square Discriminant Analysis (OPLS-DA) were performed. Before the creation of every classification model, Pareto scaling and logarithmic transformation of the data were performed. The quality of the models was determined using the R2Y and Q2 validation parameters calculated by a 7-round internal cross-validation. Candidate biomarker compounds and statistically significant genes were selected based on OPLS-DA S-plots and Variable Importance of the Projection (VIP) plots together with Receiver Operating Characteristics (ROCs). Both filters were applied simultaneously in an orthogonal way, to assure proper statistical filtration and to keep only the strong biomarkers/genes relevant for further data interpretation. For selection, VIP scores greater than one (VIP score > 1) and ROC area under curve (AUC = 1) parameters were set. Multi-omics data integration was performed according to the workflow developed and described by Ewald et al. based on the web-based freely available software OmicsAnalyst (v. 2.0, Xia Lab, McGill, Montreal, QC, Canada). Although each individual omics dataset (metabolomics + transcriptomics) was uploaded, the ‘Metadata table’ file, which included information about samples, groups, and factors, was also uploaded to the OmicsAnalyst interface. After uploading, auto scaling was applied to both datasets. For the first view of multi-omics data integration, Multiple Co-Inertia Analysis (MCIA) was performed in Dimensionality Reduction module. MCIA represents a robust method of finding related multidimensional components across multiple datasets. In the next step, Data Integration Analysis for Biomarker discovery using Latent cOmponents (DIABLO) was performed. This method is a generalized and supervised version of PLS (multiblock PLS-DA) that seeks to find related multidimensional components that maximally separate sample labels. In the next step, the correlation network based on the DIABLO feature selection method and the classical Spearman correlation analysis similarity matrix method was performed. For building a correlation network, only between-omics connections were considered. The corr. threshold for between-omics was set to >0.8. This data-driven multi-omics study identified a number of significant factors (metabolites and expressed genes), the pattern of which was changed in response to the PEF treatment. The most important findings of this study are summarized as follows: In the PEF-treated samples, the upregulation of the gene encoding the vegetative gp1-like protein, previously described as barley protective factor against Fusarium disease and abiotic stress, has been found. At the same time, the upregulation of metabolites from the class of flavonoids, (methylsulfanyl)prop-2-enoates, triterpenoid glycosides, and indole alkaloids, previously described as important factors in the Fusarium pathogen defense, has been demonstrated. The PEF-induced upregulation of these genes/metabolites is well correlated with the apparent and statistically significant decrease in F. sporotrichioides and F. poae in germinated barley after the PEF treatment. On the other hand, the overexpression of some Fusarium defense factors was observed also in the control samples. These include the genes encoding ethylene response factor 3-like, putrescine hydroxycinnamoyltransferase 3-like, and dirigent protein 21-like. This trend may be related to the slight increase in the occurrence of F. culmorum and F. graminearum in germinated barley after the PEF intervention. It should be noted that when working with natural material, which naturally contains the representation of several Fusarium species, a clear interpretation of transcriptomic and metabolomic data in relation to the decrease/increase of fungal pathogens is very difficult. To confirm or refute the hypotheses raised on the basis of these pilot correlations and to correctly interpret the biochemical context, follow-up studies that include the variability of the input material (e.g., different barley varieties with different resistance/susceptibility to fungal diseases or different input levels of Fusarium pathogens, ideally single species) need to be performed. Nevertheless, the current study undoubtedly demonstrates the practical application of the recently published protocol of data-driven multi-omics and has provided the first important results on which future follow-up research can be built. |
Descripción de un programa de intercambio educacional internacional para residentes de medicina familiar en México | 8c944198-c7ed-413c-aab5-c299327c2250 | 9309402 | Family Medicine[mh] | Se han seguido los protocolos de confidencialidad de los datos, así como el derecho a la confidencialidad y consentimiento informado.
Este trabajo no ha recibido ningún tipo de financiación.
Los autores declaran no tener ningún conflicto de intereses.
|
Proteomic Profiling of Extracellular Vesicles in Inflammatory Bowel Diseases | 0ffe5134-1b97-44c3-8781-f18c65a4a74a | 11765028 | Biochemistry[mh] | Inflammatory bowel diseases (IBDs)—including Crohn’s disease (CD) and ulcerative colitis (UC)—are chronic inflammatory diseases of the gastrointestinal tract with a high burden for both patients and society . There is no curative treatment for IBDs, primarily due to their complex and largely unknown etiology, involving interactions between environmental factors, genetics, diet, the intestinal microbiota, and the immune system . Current assessment methods, primarily based on endoscopy, are considered to be the gold standard for evaluating mucosal activity; however, they are invasive and costly, thus limiting their routine application . This has prompted an increasing interest in non-invasive biomarkers, which may facilitate patient monitoring and improve quality of life . Extracellular vesicles (EVs) are membrane-bound particles secreted by cells, which play a crucial role in intercellular communication and can encapsulate various biomolecules, including proteins that contribute to disease pathogenesis . These diverse contents endow EVs with biomarker potential for disease diagnosis, prognosis, and monitoring . Despite the gastrointestinal tract being a primary source of human EVs, their specific roles in IBD pathogenesis and correlation with disease types remain poorly characterized. Preliminary studies suggest that EVs modulate immune responses through proteins involved in macrophage activation . EVs containing Annexin A1 activate wound repair pathways in mice, and their levels are elevated in the sera of active IBD patients compared to healthy controls (HCs) . This suggests that they could be useful in disease monitoring and as potential therapeutic targets. Additionally, EVs derived from inflamed intestinal sites in IBD patients exhibit distinct protein profiles; EVs from patients with severe inflammation contain increased levels of inflammatory cytokines, while those from patients in endoscopic remission have cytokine levels akin to those from HCs . Notably, these active IBD-derived EVs exert pro-inflammatory effects on colonic epithelial cells in vitro. The present study analyzed the proteome composition of serum EVs from IBD (CD and UC) patients, with and without intestinal inflammation, and HCs. The aim was to perform a comprehensive proteomic characterization of EVs to discover potential new biomarkers and identify molecular pathways involved in IBD pathogenesis. The focus on serum EVs aimed to overcome the limitations of traditional serum proteomic analyses, which often hinder the identification of low-abundance proteins that could play crucial roles in IBD pathogenesis. 2.1. Characterization of Serum EVs from IBD Patients and HCs Serum EVs were characterized by NTA. The vesicles showed round morphology by TEM and expression of the typical EV protein markers CD81, CD63, and CD9, as analyzed by Western blotting ( and ). The size and concentration of EVs from CD or UC patients (active and quiescent) and HCs are shown in . No significant differences were observed in EV concentration or size across groups (CD vs. UC vs. HCs) ( A), between active CD (aCD), quiescent CD (qCD), and HCs ( B), or between active UC (aUC), quiescent UC (qUC), and HCs ( C). 2.2. Proteomic Characterization of Serum EVs from Active and Quiescent UC and CD Patients, and HC A total of 324 proteins were identified by at least two unique peptides with an FDR < 1% detected in serum EVs (fold-change ratio > 1.5, p -value < 0.05). Of these, 60 proteins showed differential abundance between CD and HCs, 34 between UC and HCs, and 21 between CD and UC . A Venn diagram revealed that 33 proteins were exclusively found in the comparison between CD patients and HCs, 10 proteins between UC patients and HCs, and 11 proteins between CD and UC . Regarding disease activity, the abundance of 58 and 31 proteins was altered in aCD and aUC patients compared to HCs, respectively ( A, ). In contrast, the abundance of 50 proteins was found to be altered in qCD patients and 34 in qUC patients compared to HCs ( B, ). Notably, only 4 proteins showed differential abundance between aCD and qCD patients, while the abundance of 19 proteins was altered between aUC and qUC ( C, ). Therefore, these results confirm that the protein content from the EVs is related to disease activity and, indeed, is more pronounced in UC than in CD. 2.3. Novel Serum EV Biomarkers for UC and CD, and Their Association with Disease Activity In UC patients, a total of 14 proteins were identified with potential diagnostic value ( p -value < 0.05 and ACC ≥ 0.65) either individually or in combination with other proteins. Among them, the abundance of 10 proteins was altered between UC and HCs, and 4 showed classification potential only in combination with other proteins. A summary of these biomarker candidates is presented in . In aUC patients, 26 classifiers were identified as potential biomarkers. Of these, 21 were single-protein features, while 5 were multi-protein features. Interestingly, 11 proteins did not show differential expression between aUC and HCs but still exhibited strong classification potential. lists only the proteins that were found to have differential abundance in the proteomic analysis. Regarding the comparison between qUC patients and HCs, the abundance of 42 proteins was found to be altered alone and/or in combination with another molecule. Of these, the abundance of five proteins was altered between the qUC and HC cohorts. Nine molecules presented classification power only in combination with another molecule, thirty-three molecules presented classification power on their own, and seven molecules presented classification power both individually and in combination with another molecule. The proteins that demonstrated good discriminatory power and were also found to be differential in the discovery phase are shown in . In relation to CD, 50 out of the 324 evaluated proteins showed differential abundance compared to HCs. Of these, 22 proteins with potential classification power were detected. Out of those 22 proteins, 16 were found to have differential abundance between the CD and HC cohorts. Four molecules possessed classification power only in combination with another molecule, eighteen molecules presented classification power on their own, and five molecules evidenced classification power both individually and in combination with another molecule. shows the results only for those proteins that were found to be differential in the proteomic analysis. The comparison between serum EV samples from aCD patients and those from HCs provided a total of 37 proteins with potential classification power. Out of those 37 proteins, 18 were found to have differential abundance between the aCD and HC cohorts. Three proteins possessed classification power only in combination with another molecule, thirty-four molecules presented classification power on their own, and six molecules presented classification power both individually and in combination with another molecule. shows only the proteins that were found to be differential in the proteomic analysis. In the comparison between qCD patients and HCs, 32 proteins were detected with potential classification power individually and/or in combination with another molecule. Out of those 32 proteins, the abundance of 15 was altered between the aCD and HC cohorts. Four of them had classification power only in combination with another molecule, twenty-eight molecules presented classification power on their own, and twelve molecules presented classification power both individually and in combination with another molecule. Those proteins that demonstrated good discriminatory power and were also found to be differential in the discovery phase are shown in . Based on these findings, two subsets of biomarker candidates were selected as candidates for further validation in an independent cohort in the future. The first subset comprised the proteins that exhibited the best accuracy in distinguishing between the different conditions, representing the most robust discriminatory markers . The second subset included proteins with good accuracy, albeit with slightly lower ACC values compared to the first group, thus indicating a potential relevance with less discriminative power . 2.4. Functional Enrichment Analysis of Differential-Abundance Serum EV Proteins Associated with the Disease Activity To gain a deeper understanding of the proteins potentially associated with disease activity in IBD, and to comprehensively explore the biological significance of differential-abundance proteins in serum EVs from patients with CD and UC in both remission and active phases of the disease, a GO functional enrichment analysis was performed. The GO enrichment analysis bubble plot and the GO enrichment analysis bar plot displayed the top three notably enriched GO terms. The results showed that most of the proteins with great differential expression in serum EVs between aCD patients and HCs were involved in the following biological processes: protein localization to CENP-A-containing chromatin, megakaryocyte differentiation, and detection of molecules of bacterial origin. For the molecular function category, differential proteins were mainly enriched in the complement component C1q complex, hemoglobin binding, and haptoglobin binding; and for the cellular component category, these proteins were enriched in the CENP-A-containing nucleosome, endocytic vesicle lumen, and haptoglobin–hemoglobin complex ( A). The biological processes category of the GO analysis of qCD patients compared to HCs showed that differential-abundance proteins were significantly enriched in the terms positive regulation of apoptotic cell clearance, detection of molecules of bacterial origin, and SNARE complex disassembly ( B). For GO cellular component analysis, the top three significantly enriched terms were cell–substrate junction, membrane attack complex, and fascia adherens, and the significantly enriched molecular function terms included complement component C1q complex, ATP-dependent protein disaggregase activity, and endopeptidase inhibitor activity. Furthermore, GO functional enrichment analysis was conducted on the serum EV proteins with differential abundance between aUC and qUC patients. The main biological processes found in this analysis included negative regulation of hydrogen peroxide catabolic process, regulation of epidermis development, and opsonization ( C). The main cellular component terms included membrane attack complex, endocytic vesicle lumen, and haptoglobin–hemoglobin complex, while the main molecular function terms included choline binding, hemoglobin binding, and antioxidant activity. The comparison between qCD patients and HCs revealed biological processes related to the regulation of opsonization, regulation of epidermis development, and positive regulation of apoptotic cell clearance ( D). For the cellular component category, the differential-abundance proteins were mainly enriched in membrane attack complex, serine-type endopeptidase complex, and high-density lipoprotein particle. The main molecular function terms included the complement component C1q complex, carbohydrate derivative binding, and complement component C3b binding. Finally, in the comparison of UC patients (active vs. quiescent), the GO functional enrichment analysis showed that negative regulation of hydrogen peroxide’s catabolic process, regulation of apoptotic clearance, and opsonization were the main enriched biological processes ( E). For the cellular component category, the differential-abundance proteins were mainly enriched in the terms endocytic vesicle lumen, haptoglobin–hemoglobin complex, and high-density lipoprotein particle, while in the molecular function category these proteins were enriched in the complement component C1q complex, hemoglobin binding, and endopeptidase inhibitor activity. The functional analysis of the comparison between aCD and qCD could not be performed due to the limited differences between groups (the abundance of only four proteins was altered). Serum EVs were characterized by NTA. The vesicles showed round morphology by TEM and expression of the typical EV protein markers CD81, CD63, and CD9, as analyzed by Western blotting ( and ). The size and concentration of EVs from CD or UC patients (active and quiescent) and HCs are shown in . No significant differences were observed in EV concentration or size across groups (CD vs. UC vs. HCs) ( A), between active CD (aCD), quiescent CD (qCD), and HCs ( B), or between active UC (aUC), quiescent UC (qUC), and HCs ( C). A total of 324 proteins were identified by at least two unique peptides with an FDR < 1% detected in serum EVs (fold-change ratio > 1.5, p -value < 0.05). Of these, 60 proteins showed differential abundance between CD and HCs, 34 between UC and HCs, and 21 between CD and UC . A Venn diagram revealed that 33 proteins were exclusively found in the comparison between CD patients and HCs, 10 proteins between UC patients and HCs, and 11 proteins between CD and UC . Regarding disease activity, the abundance of 58 and 31 proteins was altered in aCD and aUC patients compared to HCs, respectively ( A, ). In contrast, the abundance of 50 proteins was found to be altered in qCD patients and 34 in qUC patients compared to HCs ( B, ). Notably, only 4 proteins showed differential abundance between aCD and qCD patients, while the abundance of 19 proteins was altered between aUC and qUC ( C, ). Therefore, these results confirm that the protein content from the EVs is related to disease activity and, indeed, is more pronounced in UC than in CD. In UC patients, a total of 14 proteins were identified with potential diagnostic value ( p -value < 0.05 and ACC ≥ 0.65) either individually or in combination with other proteins. Among them, the abundance of 10 proteins was altered between UC and HCs, and 4 showed classification potential only in combination with other proteins. A summary of these biomarker candidates is presented in . In aUC patients, 26 classifiers were identified as potential biomarkers. Of these, 21 were single-protein features, while 5 were multi-protein features. Interestingly, 11 proteins did not show differential expression between aUC and HCs but still exhibited strong classification potential. lists only the proteins that were found to have differential abundance in the proteomic analysis. Regarding the comparison between qUC patients and HCs, the abundance of 42 proteins was found to be altered alone and/or in combination with another molecule. Of these, the abundance of five proteins was altered between the qUC and HC cohorts. Nine molecules presented classification power only in combination with another molecule, thirty-three molecules presented classification power on their own, and seven molecules presented classification power both individually and in combination with another molecule. The proteins that demonstrated good discriminatory power and were also found to be differential in the discovery phase are shown in . In relation to CD, 50 out of the 324 evaluated proteins showed differential abundance compared to HCs. Of these, 22 proteins with potential classification power were detected. Out of those 22 proteins, 16 were found to have differential abundance between the CD and HC cohorts. Four molecules possessed classification power only in combination with another molecule, eighteen molecules presented classification power on their own, and five molecules evidenced classification power both individually and in combination with another molecule. shows the results only for those proteins that were found to be differential in the proteomic analysis. The comparison between serum EV samples from aCD patients and those from HCs provided a total of 37 proteins with potential classification power. Out of those 37 proteins, 18 were found to have differential abundance between the aCD and HC cohorts. Three proteins possessed classification power only in combination with another molecule, thirty-four molecules presented classification power on their own, and six molecules presented classification power both individually and in combination with another molecule. shows only the proteins that were found to be differential in the proteomic analysis. In the comparison between qCD patients and HCs, 32 proteins were detected with potential classification power individually and/or in combination with another molecule. Out of those 32 proteins, the abundance of 15 was altered between the aCD and HC cohorts. Four of them had classification power only in combination with another molecule, twenty-eight molecules presented classification power on their own, and twelve molecules presented classification power both individually and in combination with another molecule. Those proteins that demonstrated good discriminatory power and were also found to be differential in the discovery phase are shown in . Based on these findings, two subsets of biomarker candidates were selected as candidates for further validation in an independent cohort in the future. The first subset comprised the proteins that exhibited the best accuracy in distinguishing between the different conditions, representing the most robust discriminatory markers . The second subset included proteins with good accuracy, albeit with slightly lower ACC values compared to the first group, thus indicating a potential relevance with less discriminative power . To gain a deeper understanding of the proteins potentially associated with disease activity in IBD, and to comprehensively explore the biological significance of differential-abundance proteins in serum EVs from patients with CD and UC in both remission and active phases of the disease, a GO functional enrichment analysis was performed. The GO enrichment analysis bubble plot and the GO enrichment analysis bar plot displayed the top three notably enriched GO terms. The results showed that most of the proteins with great differential expression in serum EVs between aCD patients and HCs were involved in the following biological processes: protein localization to CENP-A-containing chromatin, megakaryocyte differentiation, and detection of molecules of bacterial origin. For the molecular function category, differential proteins were mainly enriched in the complement component C1q complex, hemoglobin binding, and haptoglobin binding; and for the cellular component category, these proteins were enriched in the CENP-A-containing nucleosome, endocytic vesicle lumen, and haptoglobin–hemoglobin complex ( A). The biological processes category of the GO analysis of qCD patients compared to HCs showed that differential-abundance proteins were significantly enriched in the terms positive regulation of apoptotic cell clearance, detection of molecules of bacterial origin, and SNARE complex disassembly ( B). For GO cellular component analysis, the top three significantly enriched terms were cell–substrate junction, membrane attack complex, and fascia adherens, and the significantly enriched molecular function terms included complement component C1q complex, ATP-dependent protein disaggregase activity, and endopeptidase inhibitor activity. Furthermore, GO functional enrichment analysis was conducted on the serum EV proteins with differential abundance between aUC and qUC patients. The main biological processes found in this analysis included negative regulation of hydrogen peroxide catabolic process, regulation of epidermis development, and opsonization ( C). The main cellular component terms included membrane attack complex, endocytic vesicle lumen, and haptoglobin–hemoglobin complex, while the main molecular function terms included choline binding, hemoglobin binding, and antioxidant activity. The comparison between qCD patients and HCs revealed biological processes related to the regulation of opsonization, regulation of epidermis development, and positive regulation of apoptotic cell clearance ( D). For the cellular component category, the differential-abundance proteins were mainly enriched in membrane attack complex, serine-type endopeptidase complex, and high-density lipoprotein particle. The main molecular function terms included the complement component C1q complex, carbohydrate derivative binding, and complement component C3b binding. Finally, in the comparison of UC patients (active vs. quiescent), the GO functional enrichment analysis showed that negative regulation of hydrogen peroxide’s catabolic process, regulation of apoptotic clearance, and opsonization were the main enriched biological processes ( E). For the cellular component category, the differential-abundance proteins were mainly enriched in the terms endocytic vesicle lumen, haptoglobin–hemoglobin complex, and high-density lipoprotein particle, while in the molecular function category these proteins were enriched in the complement component C1q complex, hemoglobin binding, and endopeptidase inhibitor activity. The functional analysis of the comparison between aCD and qCD could not be performed due to the limited differences between groups (the abundance of only four proteins was altered). This study provides insights into the profile of circulating EVs in IBD patients, suggesting their potential to reflect disease condition and distinguish between active and quiescent sates. Hence, these results have a great impact, not just in unravelling the basis underlying IBD’s pathogenesis, but in identifying proteins with potential utility as novel biomarkers to aid in the diagnosis and/or monitoring of IBD in the absence of an invasive colonoscopy. Despite advancements in the clinical management of IBD patients, there is still a lack of useful non-invasive biomarkers that can reliably facilitate the diagnosis and monitoring of disease progression and treatment response. EVs have emerged as key mediators of intercellular communication, carrying biomolecules that influence inflammation and immune responses in the gastrointestinal tract. Evidence suggests that the protein composition of EVs can reflect the inflammatory state of the intestine . In this context, previous studies have pointed to the potential of EVs as non-invasive biomarkers for IBD. One study identified the proteasome subunit alpha type 7 in salivary EVs as a promising candidate to differentiate IBD patients from HCs . However, that study did not establish a correlation between biomarker expression and disease activity, nor could it effectively compare CD and UC due to its limited sample size. Another study analyzed serum exosomes in IBD patients compared to HCs and in a mouse model of acute colitis . Pregnancy zone protein, known for its immunosuppressive properties, was significantly increased in serum exosomes both in IBD and in mouse colitis. However, like the previous studies, this research did not distinguish between CD and UC patients, and the small sample size limited the findings. Collectively, these studies highlight the premise of EVs as biomarkers for IBD while emphasizing the need for larger, more comprehensive investigations to clarify their clinical utility. This study is unique as it is the first to isolate serum EVs from IBD patients and HCs specifically distinguishing, among IBD patients, between CD and UC and between patients with and without intestinal inflammation. By analyzing EVs’ protein profiles, we identified potential biomarkers associated with IBD presence and disease activity. Our findings are consistent with those of previous studies that have demonstrated that EVs carry biomolecules reflective of inflammatory processes in the gut. Notably, we observed significant differences in the proteomic composition of EVs between active disease in CD and UC compared to HCs, which points to the potential value of EVs as non-invasive biomarkers for disease activity. Our findings revealed no significant differences in EV size or concentration across the different groups, including comparisons between the active and quiescent phases of the disease. These data suggest that the total number of EVs does not vary significantly between IBD patients and HCs. Therefore, functional changes in EV composition, rather than their quantity, may better reflect disease activity. There is a strong association between disease activity and the differential abundance of specific proteins, underscoring distinct molecular alterations in aCD and aUC compared to HCs. These differences suggest that the molecular alterations accompanying disease exacerbations differ between CD and UC. These findings are relevant, as they suggest that targeted therapeutic strategies could be designed based on these specific proteomic signatures, potentially leading to more personalized treatment approaches. Additionally, we identified two distinct subsets of protein biomarkers capable of distinguishing CD and UC from HCs, which deserve further validation in an independent cohort. The first subset showed the highest discriminative power, emerging as the most reliable set of markers. However, the second subset demonstrated moderate accuracy and exhibited slightly lower discriminative capability. Despite this, these proteins may still hold significant potential for disease characterization. Functional enrichment analysis of the differential-abundance proteins in serum EVs has provided valuable insights into the biological processes associated with disease activity in IBD patients. Differential serum EV proteins in aCD were enriched in biological processes such as protein localization to CENP-A-containing chromatin and megakaryocyte differentiation. These processes, which are fundamental for genetic maintenance and the formation of cells involved in hemostasis , have not been previously associated with IBD. Notably, the molecular functions of these proteins are enriched in components like the complement component C1q complex, which plays a pivotal role in immune regulation, and in binding activities associated with hemoglobin and haptoglobin, indicating systemic effects of inflammation. In aUC, differential serum EV proteins were involved in processes such as the negative regulation of hydrogen peroxide catabolic processes, epidermis development, and opsonization, suggesting altered oxidative stress responses and immune modulation during active disease . Previous studies have shown that hydrogen peroxide levels are increased in the non-inflamed colonic epithelium of UC patients, evidencing the role of hydrogen peroxide in both the pathogenesis and relapse of this debilitating form of IBD . Additionally, genetically engineered mice that are unable to neutralize colonic hydrogen peroxide [glutathione (GSH) peroxidase-knockout mice] develop colitis analogous to human UC . This indicates that hydroxide peroxide generated in colonic epithelial cells can diffuse extracellularly and initiate colonic inflammation. At the molecular level, significant functions such as choline binding, hemoglobin binding, and antioxidant activity suggest potential dysregulation of oxidative pathways that may be directly associated with disease activity in aUC. Although the findings described herein pointing to candidate biomarkers for IBD are promising, their preliminary nature is the main limitation of our study. Future studies in larger and independent cohorts of patients will be necessary to validate these biomarkers and evaluate their predictive ability for disease activity. Moreover, further functional studies thoroughly elucidating the biological role of the differential-abundance serum EV proteins in the pathogenesis of IBD will be critical for advancing our understanding of these diseases. This study has several strengths. It is the first study to comprehensively characterize the proteomic profile of serum-derived EVs in CD and UC, in both active and quiescent disease states, and HCs, with the aim of identifying proteins associated with disease activity. Furthermore, the inclusion of a large, well-defined cohort of 150 participants (30 aCD, 30 qCD, 30 aUC, 30 qUC, and 30 HCs) provided a robust dataset for analysis, significantly enhancing this study’s statistical power and serving as a starting point for future validation studies. 4.1. Study Design A total of 150 participants were included in the study: 30 quiescent CD patients (qCD), 30 active CD patients (aCD), 30 quiescent UC patients (qUC), 30 active UC patients (aUC), and 30 HCs. The demographic and clinical characteristics of the study population are shown in . Serum samples from subjects who met the inclusion criteria were provided by the Gastrointestinal Biologic Samples Collection of Dr. Javier P. Gisbert (Reg. C.0003482). The Ethics Committee of Hospital Universitario de La Princesa approved the study protocol. 4.2. Subject Recruitment IBD patients (CD or UC) with active or inactive endoscopic disease, undergoing a colonoscopy indicated by medical criteria, were considered. The CD patients had luminal disease (CD patients with perianal disease were not included in the study). Only patients with complete ileocolonoscopy were considered. Patients were classified into active or quiescent IBD according to endoscopic findings. HCs included subjects undergoing colonoscopy for colorectal cancer surveillance, changes in bowel habit, or rectal bleeding. Only individuals with a macroscopically and histologically normal intestine and no evidence of disease were selected. The exclusion criteria for both groups included pregnancy, active infection, neoplasia, or any chronic condition that could affect the results. 4.3. Data Collection The variables included in the database were IBD type (location and behavior), age at IBD diagnosis, time of disease evolution, smoking habit, surgical interventions due to IBD, IBD treatment, and clinical and endoscopic disease activity. 4.4. Endoscopic IBD Activity Experienced gastroenterologists performed the ileocolonoscopies and graded the findings according to the Simple Endoscopic Score index for CD (SES-CD) and the Mayo endoscopic sub-score for UC. An SES-CD score between 0 and 2 was considered to represent inactive CD. Regarding UC activity, Mayo endoscopic sub-scores of 0 and 1 were considered to represent inactive UC. 4.5. Serum EV Isolation Serum samples (1 mL aliquots) were thawed at room temperature and processed using a series of centrifugations at 4 °C. First, serum was centrifuged to remove small cell debris, followed by ultracentrifugation at 100,000 × g for 75 min to pellet the EVs. The EV pellet was washed and subjected to a second ultracentrifugation under the same conditions. The final EV pellet was resuspended in 20 µL of phosphate-buffered saline (PBS) for immediate use or stored at −80 °C for further experiments. 4.6. Transmission Electron Microscopy (TEM) For the characterization of EVs, PBS-resuspended EV isolates were negatively stained and evaluated by TEM. EV samples were directly adsorbed onto a glow-discharged (60 seg low discharging using a PELCO easy-glow device) carbon-coated copper grid (300 mesh). Afterwards, the grids were fixed with 2% paraformaldehyde (PFA) in 0.2 M PBS (pH 7.4) for 20 min and washed with distilled water. Then, contrast staining was performed by incubating the grids with 4% uranyl acetate at 4 °C for 15 min. TEM images were obtained by using a TECNAI G2 20 C-TWIN high-resolution transmission electron microscope (FEI, Donostia-San Sebastian, Spain), at an acceleration voltage of 200 kV. 4.7. Distribution and Concentration of EVs The size distribution and concentration of the EVs were measured by nanoparticle tracking analysis (NTA) using a NanoSight LM10 system (Malvern, UK). Post-acquisition settings for NTA were standardized for all samples. Each video was analyzed to determine the mode vesicle size and EV concentration. 4.8. Western Blot Analysis Protein expression in the EV samples was analyzed by immunoblotting. Positive (CD63 and CD8) and negative (GRP78) EV markers were evaluated. 4.9. Mass Spectrometry Analysis and Protein Identification The samples were processed with an extraction buffer (7 M urea, 2 M thiourea, 4% CHAPS, and 100 mM DTT) and incubated for 30 min with agitation. Next, Filter-Aided Sample Preparation (FASP) was performed for the digestion of proteins. After digestion, peptides were recovered from the filter units and subjected to ethyl acetate. After careful removal of the last upper ethyl acetate layer, the samples were speed-vacuumed in an RVC2 25 speedvac concentrator (Christ). The samples were further desalted using stage-tip C18 microcolumns (Zip-tip, Millipore, Burlington, MA, USA) and resuspended in 0.1% FA prior to mass spectrometry analysis. Peptide separation was conducted in a nanoACQUITY UPLC system (Waters, Milford, MA, USA) coupled with an LTQ Orbitrap XL (Thermo Electron, Waltham, MA, USA) and/or Synapt G2 Si (Waters) mass spectrometer. A linear gradient of 3–50% acetonitrile over 120 min was used for peptide elution. Peptides were identified using Mascot v2.1 (Matrix Science) through Proteome Discoverer 1.4 (Thermo Electron), with carbamidomethylation of cysteines as a fixed modification and oxidation of methionines as a variable modification. Searches were conducted against the Uniprot/Swissprot database, with a decoy search to estimate the false discovery rate (FDR). Only peptides with an FDR < 1% were selected. For label-free differential protein expression analysis, Progenesis LC-MS (Nonlinear Dynamics) was used. Raw files were imported from both Orbitrap and Synapt runs, with one run designated as the reference for aligning precursor masses in all other samples. The raw abundances of each feature were automatically normalized against the reference run and logarithmized. Samples were grouped according to the comparisons being performed. A peak list containing the information of these significantly different features was generated and exported to the Mascot search engine (Matrix Science Ltd., Boston, MA, USA) for the identification of peptides. The list of identified peptides was imported in Progenesis LC-MS, and the previously quantified features were matched to the corresponding peptides. Only proteins with at least two quantified non-conflicting peptides were selected. Proteins with a p -value < 0.05 and ratio > 1.5 were considered to be significantly deregulated. 4.10. Functional Analysis Gene Ontology (GO) enrichment analysis was carried out using the Database for Annotation, Visualization, and Integrated Discovery (DAVID) online tool ( http://david.abcc.ncifcrf.gov/summary.jsp , accessed on 25 December 2024). DAVID is a GO term annotation and enrichment analysis tool used to highlight the most relevant GO terms associated with a given gene list. Fisher’s exact test was used to determine whether the proportion of genes considered to belong to certain GO terms or categories differed significantly between the dataset and the background. Biological process, molecular function, and cellular component categories were assessed, and only GO terms enriched with an FDR < 5% were considered for comparison and discussion. 4.11. Statistical Analysis Quantitative variables are presented as means and standard deviations. Categorical variables are presented as numbers of events and percentages. The Kolmogorov–Smirnov test was used to determine the normality of data distribution for quantitative variables. Differences in this type of variable between groups were analyzed by Student’s t test or the Wilcoxon rank-sum test according to whether the variables were normally distributed or not. For categorical variables, differences between groups were assessed using the chi-squared test. Differences were considered significant when p < 0.05. For biomarker discovery, the dataset with all of the variables from the individual patients was analyzed by Anaxomics Biotech ( http://www.anaxomics.com , accessed on 25 December 2024) using a data mining approach, employing the algorithm Balanced_Accuracy . Cross-validated balanced accuracy (ACC) was used as a classifier optimization measure together with the cross-validated p -value. In order to prioritize the generalization capability of the conclusion, a K-fold validation analysis was performed, yielding cross-validated quality measures (such as accuracy) for each biomarker . Functional and pathway enrichment analysis was conducted using DAVID to perform GO enrichment analysis using SRplot . A total of 150 participants were included in the study: 30 quiescent CD patients (qCD), 30 active CD patients (aCD), 30 quiescent UC patients (qUC), 30 active UC patients (aUC), and 30 HCs. The demographic and clinical characteristics of the study population are shown in . Serum samples from subjects who met the inclusion criteria were provided by the Gastrointestinal Biologic Samples Collection of Dr. Javier P. Gisbert (Reg. C.0003482). The Ethics Committee of Hospital Universitario de La Princesa approved the study protocol. IBD patients (CD or UC) with active or inactive endoscopic disease, undergoing a colonoscopy indicated by medical criteria, were considered. The CD patients had luminal disease (CD patients with perianal disease were not included in the study). Only patients with complete ileocolonoscopy were considered. Patients were classified into active or quiescent IBD according to endoscopic findings. HCs included subjects undergoing colonoscopy for colorectal cancer surveillance, changes in bowel habit, or rectal bleeding. Only individuals with a macroscopically and histologically normal intestine and no evidence of disease were selected. The exclusion criteria for both groups included pregnancy, active infection, neoplasia, or any chronic condition that could affect the results. The variables included in the database were IBD type (location and behavior), age at IBD diagnosis, time of disease evolution, smoking habit, surgical interventions due to IBD, IBD treatment, and clinical and endoscopic disease activity. Experienced gastroenterologists performed the ileocolonoscopies and graded the findings according to the Simple Endoscopic Score index for CD (SES-CD) and the Mayo endoscopic sub-score for UC. An SES-CD score between 0 and 2 was considered to represent inactive CD. Regarding UC activity, Mayo endoscopic sub-scores of 0 and 1 were considered to represent inactive UC. Serum samples (1 mL aliquots) were thawed at room temperature and processed using a series of centrifugations at 4 °C. First, serum was centrifuged to remove small cell debris, followed by ultracentrifugation at 100,000 × g for 75 min to pellet the EVs. The EV pellet was washed and subjected to a second ultracentrifugation under the same conditions. The final EV pellet was resuspended in 20 µL of phosphate-buffered saline (PBS) for immediate use or stored at −80 °C for further experiments. For the characterization of EVs, PBS-resuspended EV isolates were negatively stained and evaluated by TEM. EV samples were directly adsorbed onto a glow-discharged (60 seg low discharging using a PELCO easy-glow device) carbon-coated copper grid (300 mesh). Afterwards, the grids were fixed with 2% paraformaldehyde (PFA) in 0.2 M PBS (pH 7.4) for 20 min and washed with distilled water. Then, contrast staining was performed by incubating the grids with 4% uranyl acetate at 4 °C for 15 min. TEM images were obtained by using a TECNAI G2 20 C-TWIN high-resolution transmission electron microscope (FEI, Donostia-San Sebastian, Spain), at an acceleration voltage of 200 kV. The size distribution and concentration of the EVs were measured by nanoparticle tracking analysis (NTA) using a NanoSight LM10 system (Malvern, UK). Post-acquisition settings for NTA were standardized for all samples. Each video was analyzed to determine the mode vesicle size and EV concentration. Protein expression in the EV samples was analyzed by immunoblotting. Positive (CD63 and CD8) and negative (GRP78) EV markers were evaluated. The samples were processed with an extraction buffer (7 M urea, 2 M thiourea, 4% CHAPS, and 100 mM DTT) and incubated for 30 min with agitation. Next, Filter-Aided Sample Preparation (FASP) was performed for the digestion of proteins. After digestion, peptides were recovered from the filter units and subjected to ethyl acetate. After careful removal of the last upper ethyl acetate layer, the samples were speed-vacuumed in an RVC2 25 speedvac concentrator (Christ). The samples were further desalted using stage-tip C18 microcolumns (Zip-tip, Millipore, Burlington, MA, USA) and resuspended in 0.1% FA prior to mass spectrometry analysis. Peptide separation was conducted in a nanoACQUITY UPLC system (Waters, Milford, MA, USA) coupled with an LTQ Orbitrap XL (Thermo Electron, Waltham, MA, USA) and/or Synapt G2 Si (Waters) mass spectrometer. A linear gradient of 3–50% acetonitrile over 120 min was used for peptide elution. Peptides were identified using Mascot v2.1 (Matrix Science) through Proteome Discoverer 1.4 (Thermo Electron), with carbamidomethylation of cysteines as a fixed modification and oxidation of methionines as a variable modification. Searches were conducted against the Uniprot/Swissprot database, with a decoy search to estimate the false discovery rate (FDR). Only peptides with an FDR < 1% were selected. For label-free differential protein expression analysis, Progenesis LC-MS (Nonlinear Dynamics) was used. Raw files were imported from both Orbitrap and Synapt runs, with one run designated as the reference for aligning precursor masses in all other samples. The raw abundances of each feature were automatically normalized against the reference run and logarithmized. Samples were grouped according to the comparisons being performed. A peak list containing the information of these significantly different features was generated and exported to the Mascot search engine (Matrix Science Ltd., Boston, MA, USA) for the identification of peptides. The list of identified peptides was imported in Progenesis LC-MS, and the previously quantified features were matched to the corresponding peptides. Only proteins with at least two quantified non-conflicting peptides were selected. Proteins with a p -value < 0.05 and ratio > 1.5 were considered to be significantly deregulated. Gene Ontology (GO) enrichment analysis was carried out using the Database for Annotation, Visualization, and Integrated Discovery (DAVID) online tool ( http://david.abcc.ncifcrf.gov/summary.jsp , accessed on 25 December 2024). DAVID is a GO term annotation and enrichment analysis tool used to highlight the most relevant GO terms associated with a given gene list. Fisher’s exact test was used to determine whether the proportion of genes considered to belong to certain GO terms or categories differed significantly between the dataset and the background. Biological process, molecular function, and cellular component categories were assessed, and only GO terms enriched with an FDR < 5% were considered for comparison and discussion. Quantitative variables are presented as means and standard deviations. Categorical variables are presented as numbers of events and percentages. The Kolmogorov–Smirnov test was used to determine the normality of data distribution for quantitative variables. Differences in this type of variable between groups were analyzed by Student’s t test or the Wilcoxon rank-sum test according to whether the variables were normally distributed or not. For categorical variables, differences between groups were assessed using the chi-squared test. Differences were considered significant when p < 0.05. For biomarker discovery, the dataset with all of the variables from the individual patients was analyzed by Anaxomics Biotech ( http://www.anaxomics.com , accessed on 25 December 2024) using a data mining approach, employing the algorithm Balanced_Accuracy . Cross-validated balanced accuracy (ACC) was used as a classifier optimization measure together with the cross-validated p -value. In order to prioritize the generalization capability of the conclusion, a K-fold validation analysis was performed, yielding cross-validated quality measures (such as accuracy) for each biomarker . Functional and pathway enrichment analysis was conducted using DAVID to perform GO enrichment analysis using SRplot . In conclusion, this research highlights the potential role of EVs in IBD’s pathogenesis. The proteomic characterization of serum EVs offers valuable insights into the underlying pathogenic mechanisms and provides information to identify specific biomarkers of disease activity. These findings represent a starting point for studying the role of EVs in IBD; after validation of these findings in future studies, the incorporation of these biomarkers into clinical practice could significantly improve the monitoring and management of IBD, paving the way for more personalized treatment strategies. |
Mineral substrate quality determines the initial soil microbial development in front of the Nordenskiöldbreen, Svalbard | 9c6078c3-1ea8-425d-b5d9-b15afcec71b4 | 10689212 | Microbiology[mh] | Retreating ice fronts, induced by global climate change, have increasingly exposed landscapes in polar and alpine regions, leading to microbial and plant colonization and soil formation (Hodkinson et al. , Bajerski and Wagner , Bradley et al. ). Microbes are the pioneer colonizers of these deglaciated barren surfaces (Schütte et al. , Bradley et al. ). Their distributions are determined by nutrient availability, and at the same time, they drive nutrient and mineral cycling during the initial phases of soil stabilization and plant establishment (Hodkinson et al. , Tscherko et al. , Borin et al. ). As such, chronosequence approaches have provided a useful tool to investigate the environmental factors that may affect the microbial community during ecosystem development (Bernasconi et al. , Zumsteg et al. , Knelman et al. , Garrido-Benavent et al. ). Plant colonization and plant–microbe interactions have been identified as the predominant factors that shape microbial community composition and nutrient accumulation along alpine and low latitude chronosequences (Tscherko et al. , Zumsteg et al. , Brown and Jumpponen ). In comparison, polar forefields mainly comprise cryptograms, lichens and bryophytes (Garrido-Benavent et al. ), and may remain mostly devoid of plants even a century after glacier retreat (Bradley et al. ). Even without vascular plants, shifts of microbial community occur in polar regions over short time intervals (Nemergut et al. , Garrido-Benavent et al. , Vimercati et al. ), though the possible abiotic driving factors and their complex interaction with carbon sources are still being debated (Tscherko et al. , Knelman et al. ). Together with plants, soil age has also been suggested to be a key factor affecting the microbial life and edaphic properties (Noll and Wellinger , Bradley et al. ), though environmental disturbances may overprint the effect of time (Kim et al. ). The selective pressure of the environmental factors (local geographic and climatic conditions) and the physico-chemical properties of the substrate (e.g. soil texture and porosity, nutrient status, pH, and mineralogical properties) (Lazzaro et al. ) together with accessibility to pioneer sites (Jumpponen et al. ) affect substrate availability and thereby microbial distribution in the glacier forefield (Zhou et al. ). In general, the initial phases of succession select for microbes that are able to fix carbon and nitrogen or otherwise gain the nutrients from weathering (Wojcik et al. ), together with heterotrophs thriving on available organic nutrient sources (Bardgett et al. , Schulz et al. ). The proportion and decomposition of old, microbial necromass gradually increases, leading to the recycling of accumulated nutrients (Bradley et al. ). Recently, among all the possible drivers, low nutrient availability was suggested to be the main constraining factor for microbial community development, which may rival other edaphic properties and harsh climatic conditions of glacier forefields (Knelman et al. , Schmidt et al. , Darcy et al. ). That is, a lack of available nutrients could impose stronger constraints on microbial community development in early succession than the absence of carbon substrates (i.e. energy source) (Castle et al. ). The main factors that determine nutrient availability of newly exposed oligotrophic polar soils remain largely unexplored but are presumably related to soil bedrock type (Barton et al. , Lazzaro et al. , Tytgat et al. , Wojcik et al. ) and/or soil organic matter sources (Fierer et al. ). Bedrock type not only determines physical properties of the landscape, such as texture, porosity (Lazzaro et al. ), and the rate of nutrient release due to weathering (Tytgat et al. ), but also provides minerals that are known to select for specialized bacteria (Carson et al. , Meola et al. ). In this way, the bedrock type controls microbial habitat and release of nutrients, and thereby significantly influences microbial activity in the initial, plant-free phases of soil development (Carson et al. ). In particular, phosphorus availability has been suggested to limit succession in cold and arid climates with low weathering rates (Darcy et al. ). The interplay of life strategies employed by microbial colonizers is depicted by the distinct successional trajectories that have been described for bacteria and fungi (Rime et al. , Jiang et al. , Garrido-Benavent et al. , Vimercati et al. ). This may be caused by ecological differences, such as substrate utilization or stress tolerance of fungi and bacteria (Brunner et al. , Hannula et al. ), which are reflected in the highly specific requirements (Rime et al. ) and more random distribution (Brown and Jumpponen ) of fungi in early succession compared to bacteria. Accordingly, taxonomically diverse bacterial communities evolving on unrelated glacier forefields with different geochemistries have been shown to converge with increasing age of succession (Castle et al. ), while fungi do not exhibit consistent patterns (Brown and Jumpponen ). The present study combines fungal and bacterial successional pathways together with substrate geochemistry to improve our understanding of glacial forefield colonization processes and the implications for future glacier retreat. While most geographical and soil forming conditions of the two forefield chronosequences in front of Nordenskiöldbreen (Svalbard, Norway) are similar, differences in mineral substrate composition offer the possibility to test the effects of contrasting bedrock type and glacial landscape. We, therefore, sampled along the contrasting 80-year successional High Arctic glacial forefield gradients, with the assumption that the controls imposed by physico-chemical properties of the mineral substrate on bacterial and fungal successional trajectories are mitigated with soil age. Our hypotheses are: There is a significant influence of mineral substrate and different quality of organic matter on the microbial assembly in the first stages of the soil development. The influence of the mineral substrate and organic matter on the microbial assembly is mitigated with soil age and the patterns of response are different for bacteria and fungi.
Study sites and sample collection Nordenskiöldbreen (central Spitsbergen, 78.67° N, 16.78° E) is a polythermal valley glacier and largest outlet glacier of Lomonosovfonna Ice Cap in Billefjorden area (Ewertowski et al. , Allaart et al. ). The middle part of glacier tongue is a tidewater terminus and both lateral margins are terrestrial. Nordenskiöldbreen has been retreating since the Little Ice Age (LIA) maximum at the end of the 19th century (Allaart et al. ). Between 1896 and 2013, the northern terrestrial terminated margin retreated about 1.4 km, while the southern terrestrial terminated margin retreated about 3.5 km (Ewertowski et al. ). The newly exposed glacier forefields rises from 2 to 35 m a.s.l. The southern and northern forefields differ in bedrock geology and glacial landscape. The northern silicate forefield is made up of metamorphic rocks (primarily mica schist, calc-pelitic schists, and marble). The relief is dominated by ice-moulded bedrock, which is covered by a thin and discontinuous glacigenic and glaciofluvial deposits comprising monotonous clastic material (mainly gneiss and other metamorphites) (Dallmann et al. , Allaart et al. ). The southern carbonate–silicate forefield is flat and made up of sedimentary carbonates and fewer mica schists. The bedrock is almost completely covered by thick glacigenic deposits of variable and exotic clastic compositions (granitoids, mafic magmatite, and diverse metamorphites; Dallmann et al. ). The annual mean surface ground temperature of the Billefjorden tundra, referred to as a reference site, is −5.2°C, ranging between −19.5°C for April and 10.8°C for July, and the annual precipitation is typically less than 200 mm (Førland et al. , Láska et al. ). Vegetation cover at both forefields is initially absent or sparse (0%–20% vegetation cover, data not shown), consisting primarily of soil crusts and/or single pioneer plants (e.g. Saxifraga oppositifolia, Braya purpurascens, Salix polaris , and Dryas octopetala) on wetter and wind shielded spots. Tundra reference sites are covered predominantly by Carex spp. grasses and woody shrubs ( S. polaris and D. octopetala) . Soils were collected in July 2015 and 2016 from both northern silicate (N1, N2, N3, and NR sites) and southern carbonate–silicate (S1, S2, S3, and SR sites) forefields with a chronosequence approach (Crocker and Major , Nemergut et al. , Bernasconi et al. ). A total of four zones with increasing age were sampled from the glacier front (I. 0–25, II. 26–54, and III. 55–79 years old; Fig. ) to the tundra ‘reference site’ behind the glacier front moraine from the end of the LIA (IV. 10 000 years). Zones were determined in ArcMap 10.6.1 (ESRI ) based on aerial photographs from the Norwegian Polar Institute from the years 1936, 1961, 1990, and 2009 (Norsk Polar Institute ). Proglacial systems are characterized by high spatial variability and redeposition (Hodkinson et al. , Bernasconi et al. ), therefore, the sampling sites were chosen carefully with an effort to find stable surfaces. The sites were first randomly chosen in ArcMap, checked for potential disturbances by glacier streams on the old aerial photos and then again examined by a geologist in the field with the special focus on the visible glacier movement. Mineral surface soil samples (top 5 cm) were collected from unvegetated areas or after the removalof soil crusts when present. At every site, three independent soil samples were pooled from four subsamples, each one meter apart (all together 66 mixed samples, 33 for each forefield; Fig. ). To capture approximate volumetric soil moisture, we also collected 10 cm × 10 cm × 5 cm cubes into plastic bags. Bulk soil was homogenized and sieved with a 2-mm sieve and immediately analyzed for available forms of nitrogen (N-NO 3 − , N-NO 2 − , and N-NH 4 + —the sum is further presented as mineral nitrogen, Nmin) and for available phosphorus (P-PO 4 3− ). Water soil extracts in the ratio 1:5 were prepared to analyze DOC and kept frozen until measured. Part of the soil was immediately frozen for DNA analyses, enzyme activities, and biomarker assessment, and part was air-dried for further physical and chemical analyses. The rest was kept at 4°C until processing for microbial biomass within 1 month. The bulk density samples were dried for 24 h at 105°C in the oven, sieved for 2 mm fraction, and weighted immediately after sampling and drying. Soil biogeochemistry and microbial quantity Dissolved organic carbon (DOC) was extracted with cold water (sample: water ratio of 1:5) by shaking at 20°C for 1 h and subsequently analyzed after transport for total organic carbon on a LiquiTOC II (Elementar, Germany). Water extractable forms of inorganic nitrogen (N-NO 2 − , N-NO 3 − , and N-NH 4 + ; sample: water ratio 1:5) were determined immediately after sampling on the field station using photometric nitrite, nitrate, and ammonium test kits (Spectroquant Test, Merck, Darmstadt, Germany) and a field portable spectrophotometer (Hach Lange DR 2800, Germany). Their concentrations were summed and referred to as mineral nitrogen (Nmin). Water extractable phosphorus (P-PO 4 3− ) was determined as soluble reactive phosphorus according to Murphy and Riley ( ) using a field portable spectrophotometer (Hach Lange DR 2800). Total content of elements in soil matrix (Si, Ca, Al, K, Fe, Mg, P, and S) was determined by X-ray fluorescence (XRF) (Delta Premium XPD 6000, Olympus Innov-X, USA) under standard lab conditions (Tejnecký et al. ). Labile compounds (Ca 2+ , K + , Mg 2+ , and SO 4 2− ) were extracted by cold water (sample: water ratio 1:10) and pH of solution (active pH) was measured potentiometrically (pH meter inoLab pH Level 1 WTW, Germany). The concentration of extracted cations was determined by means of ICP-OES (iCAP 7000, Thermo Scientific, USA) and content of SO 4 2− was determined by means of ion chromatography (ICS 1600, Dionex, USA). The amount of 10 g of the 2-mm sieved sample was air-dried and treated with 30% hydrogen peroxide and the soil suspension was allowed to react under heated conditions with subsequent additions of 30% hydrogen peroxide, until all soil organic matter was removed. The soil suspension was then measured by laser diffraction method (LDM) using a Fritsch Analysette 22 MicroTrec plus in the range of 0.1–2000 µm (ISO 11277 ). As the amount of clay particles was overestimated by the Fraunhofer theory, the pycnometer bottle method was added (Di Stefano et al. ). The fraction < 63 µm was operationally defined as the sum of fine silt and clay, which is typically associated with the majority of the soil organic matter (Hemkemeyer et al. ) and considered to be to be important for soil stabilization through interactions between fine particles and organic carbon (Bird et al. ). TOC and TN contents of the soil were measured using an elemental analyzer (vario MICRO cube, Elementar Analysensysteme GmbH). The detection limits were TOC (1 mg g −1 ) and TN (0.1 mg g −1 ) and it was not possible to determine TN content in samples younger than 54 years. Soil samples contained a significant amount of carbonates, which were removed prior to analysis via acid fumigation with HCl vapours (Harris et al. ). Analyses of TOC and δ 13 C TOC contents of dried soil material were conducted with an NC Elemental analyzer (ThermoQuest, Bremen, Germany) connected to an isotope ratio mass spectrometer (IR-MS Delta X Plus, Finnigan, Bremen, Germany). Carbon, nitrogen, and phosphorus in microbial biomass (C mic , N mic , and P mic ) were estimated using the chloroform-fumigation extraction method (Brookes et al. , , Vance et al. ). The extraction of fumigated soil followed the protocol for available nutrients described above. Nutrient concentrations in microbial biomass were calculated as the difference between concentrations in fresh soil extract and fumigated soil extract. Results were corrected for incomplete extraction of all C mic (Kec = 0.45), all N mic (Ken = 0.54), and all P mic (Kep = 0.4) (Brookes et al. , , Vance et al. ). Tetraether lipids were extracted from six forefield soils (20 g wet weight, one pooled sample for each forefield site) after the protocol of Hopmans et al. ( ). The analysis was conducted solely as a complementary measure, without laboratory repetition. Therefore, caution was exercised when evaluating the data. Lipid extracts to assess the biomarkers were dried and stored frozen until analysis by liquid chromatography mass spectrometry (Becker et al. ). Tetraether lipids were quantified relative to a C46 tetraether injection standard and the relative response of a caldarchaeol standard. The compounds measured include isoprenoidal glycerol dibiphytanyl glycerol tetraethers having zero to three rings (iGDGT0-3; m/z 1302, 1300, and 1298), crenarchaeol (m/z 1292, biomarker for marine archea), and branched tetraethers produced by soil bacteria having varying degrees of rings and methylation (m/z 1018, 1020, 1022, 1032, 1034, 1036, 1046, 1048, and 1050). iGDGT having four rings (iGDGT4) was below the detection limit. The BIT index (branched and isoprenoid tetraethers index), which is a proxy for relative contribution of terrestrial and marine organic matter input to sediments, was calculated as a ratio of the sum of all measured branched tetraethers versus crenarcheol (Schouten et al. ). Soil enzymes analysis, DNA amplicon sequencing, and qPCR Extracellular enzymatic activities were measured for five soil enzymes responsible for organic carbon, nitrogen, and phosphorus utilization with standard fluorometric technique (Marx et al. ) modified by Bárta et al. ( ). The activity of extracellular enzymes was determined in water extracts. Briefly, 0.5 g of soil was homogenized in 50 ml of distilled water via an ultrasonication. The soil suspensions (200 µl) were then transferred to a 96-well microplate. Then, 50 µl of substrates labeled with 4-methylumbelifferone (MUB) or 7-amino-4-methylcoumarin (AMC) were added. Standard curves were measured with MUB/AMC in soil slurries for every sample separately. Concentrations of MUB/AMC for standard curves were: 1, 5, and 10 µM. Microplates were incubated at room temperature for 2 h and fluorescence was measured every 30 min. All fluorescence measurements were performed on the microplate reader INFINITE F200 (Tecan, Germany) using the excitation wavelength of 365 nm and emission wavelength of 450 nm. Enzyme activities were expressed as nmol of liberated MUF/AMC per hour per gram of dry soil. The stoichiometric analysis of enzymes was done according to Hill et al. ( ) and is based on the assumption that microbes release extracellular enzymes in order to gain needed C, N, and P nutrients from complex organic substrates. The analysis compares ratios of C:N and N:P processing enzymes and the results may imply which nutrients are mainly limiting for the microbial growth and prosperity (Schmidt et al. ). Carbon processing enzymes (E C ) comprise β-glucosidase (GLU) and cellobiohydrolase (CELL) activity, nitrogen processing enzymes (E N ) are the sum of leucin aminopeptidase (ALA) and chitinase (CHIT) activity and phosphorus (E P ) is represented by phosphatase (PHO) acitivity. The amount of 0.25 g was used for DNA extraction using PowerSoil DNA isolation kit (MO BIO Laboratories, USA). The DNA was quantified fluorimetrically using Qubit (Thermofisher Scientific, UK) following the standard protocol and kit for dsDNA quantification. The aliquots of DNA extracts were sent to SEQme lab (Prague, Czech Republic) for the preparation of a library and sequencing using MiSeq platform. The Earth Microbiome Project (EMP) protocol was used for library preparation with modified universal primers 515FB/806RB (Caporaso et al. ) and ITS1F/ITS2 (White et al. ) for prokaryotic 16S rRNA and fungal ITS1 amplicons, respectively. The coverage of prokaryotic primer pair 515FB/806RB was additionally tested in silico using ARB Silva database release 128. The primer pair 515FB/806RB covers almost uniformly all major bacterial and archaeal phyla. Bacterial 16S rRNA raw pair-end reads (150 bp) were joined using ea-utils to obtain reads of ~250 bp length. Quality filtering of reads was applied as previously described (Caporaso et al. ). After quality filtering the sequences were trimmed to 250 bp. We obtained 720 625 bacterial and 1746 017 fungal sequences after quality trimming and filtering. Before picking the operational taxonomic units (OTU), the fungal ITS1 region was extracted from reads using ITSx algorithm (Bengtsson-Palme et al. ). Both 16S and ITS1 amplicons were trimmed to equal lengths in order to avoid spurious OTU clusters (Edgar ). Taxonomy was assigned to each read by accepting Silva119 taxonomy string of the best matching Silva119 sequence. Fungal reads were clustered to OTUs using open-reference OTU picking protocol (sequence similarity 98.5%) using UNITE ver. 8.0 database (Koljalg et al. ). Blast algorithm (e-value ≤ 0.001) was used for taxonomic assignment. FAPROTAX (Louca et al. ) and FUNguild (Nguyen et al. ) algorithm was then used for bacterial functional annotation and the fungal lifestyle assignments, respectively. Raw sequences were deposited at ENA under the project n. PRJEB51542. Quantification of bacterial and fungal SSU rRNA genes was performed using the FastStart SybrGREEN Roche® Supermix and Step One system (Life Technologies, USA). Each reaction mixture (20 µl) contained 2 µl DNA template (∼ 1–2 ng DNA), 1 µl each primer (0.5 pmol µl −1 each, final concentration), 6 µl dH2O, 10 µl FastStart SybrGREEN Roche® Supermix (Roche, France), and 1 µl BSA (Fermentas, 20 mg µl −1 ). Initial denaturation (3 min, 95°C) was followed by 30 cycles of 30 s at 95°C, 30 s at 62°C, 15 s at 72°C, and completed by fluorescence data acquisition at 80°C used for target quantification. Product specificity was confirmed by melting point analysis (52–95°C with a plate read every 0.5°C) and amplicon size was verified with agarose gel electrophoresis. Bacterial standards consisted of a dilution series (ranging from 10 1 to 10 9 gene copies µl −1 ) of a known amount of purified plasmid where PCR product using the SSU gene-specific primers 341F/534R (Muyzer et al. ) was inserted. R 2 values for the standard curves were > 0.99. Slope values were −> 3.37 giving an estimated amplification efficiency of > 93%. The qPCR conditions for fungal quantification were as follows: initial denaturation (10 min, 95°C) followed by 40 cycles of 1 min at 95°C, 1 min at 56°C, 1 min at 72°C, and completed by fluorescence data acquisition at 72°C used for target quantification. Fungal DNA standards consisted of a dilution series (ranging from 10 1 to 10 7 gene copies µl −1 ) of a known amount of purified plasmid where PCR product using the SSU gene-specific primers nu-SSU-0817–5′ and nu-SSU1196-3′ (Borneman and Hartin ) was inserted. R 2 values for the fungal standard curves were > 0.99. The slope was between −3.34 and −3.53 giving estimated amplification efficiency between 95% and 93%, respectively. Detection limits for the assays (i.e. lowest standard concentration that is significantly different from the nontemplate controls) were less than 100 gene copies for each of the genes per assay. Samples, standards and nontemplate controls were run in duplicates. To deal with potential inhibition during PCR the enhancers BSA and DMSO were added to the PCR mixture. Also, several dilutions (10x, 20x, 50x, 100x, and 1000x) for each sample were tested to see the dilution effect on Ct values. Alpha diversity metrics the Chao1 richness (measurement of OTUs expected in samples given all the bacterial/fungal species that were identified in the samples), were calculated after rarefying all samples to the same sequencing depth of 3200 and 2800 sequences for bacterial and fungal community, respectively (Chao ). Statistical analyses The effect of the northern versus southern forefield on soil parameters and on parameters changes in time were tested using a generalized linear mixed-effect model with Gamma distribution and logarithmic link function and a likelihood-ratio test, which was run in R version 4.0.2 (R Core Team ) in R package stats . The zone identity for each forefield was used as a random effect (i.e. eight groups). The significant differences in time were evaluated separately for each forefield using Tukey’s HSD post hoc tests in the R package multcomp to examine pairwise differences between the zones (different soil ages). Spearman correlation coefficients were determined in R package ggpubr to assess how strongly chosen parameters were related to each other. The effect of forefield on the total content of chosen soil elements determined by XRF and on labile compounds was visualized using principle component analyses in CANOCO for Windows 5.0 (Ter Braak and Šmilauer ) with zone (age) as a covariate. In all analyses from CANOCO, only the adjusted explained variation is presented in the text. In CANONO, variation partitioning was also applied to quantify the unique effects of TOC, bedrock (total content of elements in soil matrix determined by XRF—Si, Ca, Al, K, Fe, Mg, P, and S) and available nutrients (DOC, P-PO 4 3− , Nmin, Ca 2+ , Mg 2+ , and K + ) on the microbial quantity (C mic , bacteria, and fungi) and quality (CNP processing enzymes), separately during 79 and 10 000 years of soil succession. Subsequently, a forward selection procedure was also performed to identify the single parameters that best explain the microbial quantity and quality and only P -values adjusted with Holms correction were considered. Both analyses were run with zone (age) as covariate. To estimate the beta-diversity in soil microbial communities and in their functional analyses, nonmetric multidimensional scaling (NMDS) ordinations were generated using CANOCO on the basis of Bray–Curtis dissimilarities of square-root transformed relative abundances of OTUs.
Nordenskiöldbreen (central Spitsbergen, 78.67° N, 16.78° E) is a polythermal valley glacier and largest outlet glacier of Lomonosovfonna Ice Cap in Billefjorden area (Ewertowski et al. , Allaart et al. ). The middle part of glacier tongue is a tidewater terminus and both lateral margins are terrestrial. Nordenskiöldbreen has been retreating since the Little Ice Age (LIA) maximum at the end of the 19th century (Allaart et al. ). Between 1896 and 2013, the northern terrestrial terminated margin retreated about 1.4 km, while the southern terrestrial terminated margin retreated about 3.5 km (Ewertowski et al. ). The newly exposed glacier forefields rises from 2 to 35 m a.s.l. The southern and northern forefields differ in bedrock geology and glacial landscape. The northern silicate forefield is made up of metamorphic rocks (primarily mica schist, calc-pelitic schists, and marble). The relief is dominated by ice-moulded bedrock, which is covered by a thin and discontinuous glacigenic and glaciofluvial deposits comprising monotonous clastic material (mainly gneiss and other metamorphites) (Dallmann et al. , Allaart et al. ). The southern carbonate–silicate forefield is flat and made up of sedimentary carbonates and fewer mica schists. The bedrock is almost completely covered by thick glacigenic deposits of variable and exotic clastic compositions (granitoids, mafic magmatite, and diverse metamorphites; Dallmann et al. ). The annual mean surface ground temperature of the Billefjorden tundra, referred to as a reference site, is −5.2°C, ranging between −19.5°C for April and 10.8°C for July, and the annual precipitation is typically less than 200 mm (Førland et al. , Láska et al. ). Vegetation cover at both forefields is initially absent or sparse (0%–20% vegetation cover, data not shown), consisting primarily of soil crusts and/or single pioneer plants (e.g. Saxifraga oppositifolia, Braya purpurascens, Salix polaris , and Dryas octopetala) on wetter and wind shielded spots. Tundra reference sites are covered predominantly by Carex spp. grasses and woody shrubs ( S. polaris and D. octopetala) . Soils were collected in July 2015 and 2016 from both northern silicate (N1, N2, N3, and NR sites) and southern carbonate–silicate (S1, S2, S3, and SR sites) forefields with a chronosequence approach (Crocker and Major , Nemergut et al. , Bernasconi et al. ). A total of four zones with increasing age were sampled from the glacier front (I. 0–25, II. 26–54, and III. 55–79 years old; Fig. ) to the tundra ‘reference site’ behind the glacier front moraine from the end of the LIA (IV. 10 000 years). Zones were determined in ArcMap 10.6.1 (ESRI ) based on aerial photographs from the Norwegian Polar Institute from the years 1936, 1961, 1990, and 2009 (Norsk Polar Institute ). Proglacial systems are characterized by high spatial variability and redeposition (Hodkinson et al. , Bernasconi et al. ), therefore, the sampling sites were chosen carefully with an effort to find stable surfaces. The sites were first randomly chosen in ArcMap, checked for potential disturbances by glacier streams on the old aerial photos and then again examined by a geologist in the field with the special focus on the visible glacier movement. Mineral surface soil samples (top 5 cm) were collected from unvegetated areas or after the removalof soil crusts when present. At every site, three independent soil samples were pooled from four subsamples, each one meter apart (all together 66 mixed samples, 33 for each forefield; Fig. ). To capture approximate volumetric soil moisture, we also collected 10 cm × 10 cm × 5 cm cubes into plastic bags. Bulk soil was homogenized and sieved with a 2-mm sieve and immediately analyzed for available forms of nitrogen (N-NO 3 − , N-NO 2 − , and N-NH 4 + —the sum is further presented as mineral nitrogen, Nmin) and for available phosphorus (P-PO 4 3− ). Water soil extracts in the ratio 1:5 were prepared to analyze DOC and kept frozen until measured. Part of the soil was immediately frozen for DNA analyses, enzyme activities, and biomarker assessment, and part was air-dried for further physical and chemical analyses. The rest was kept at 4°C until processing for microbial biomass within 1 month. The bulk density samples were dried for 24 h at 105°C in the oven, sieved for 2 mm fraction, and weighted immediately after sampling and drying.
Dissolved organic carbon (DOC) was extracted with cold water (sample: water ratio of 1:5) by shaking at 20°C for 1 h and subsequently analyzed after transport for total organic carbon on a LiquiTOC II (Elementar, Germany). Water extractable forms of inorganic nitrogen (N-NO 2 − , N-NO 3 − , and N-NH 4 + ; sample: water ratio 1:5) were determined immediately after sampling on the field station using photometric nitrite, nitrate, and ammonium test kits (Spectroquant Test, Merck, Darmstadt, Germany) and a field portable spectrophotometer (Hach Lange DR 2800, Germany). Their concentrations were summed and referred to as mineral nitrogen (Nmin). Water extractable phosphorus (P-PO 4 3− ) was determined as soluble reactive phosphorus according to Murphy and Riley ( ) using a field portable spectrophotometer (Hach Lange DR 2800). Total content of elements in soil matrix (Si, Ca, Al, K, Fe, Mg, P, and S) was determined by X-ray fluorescence (XRF) (Delta Premium XPD 6000, Olympus Innov-X, USA) under standard lab conditions (Tejnecký et al. ). Labile compounds (Ca 2+ , K + , Mg 2+ , and SO 4 2− ) were extracted by cold water (sample: water ratio 1:10) and pH of solution (active pH) was measured potentiometrically (pH meter inoLab pH Level 1 WTW, Germany). The concentration of extracted cations was determined by means of ICP-OES (iCAP 7000, Thermo Scientific, USA) and content of SO 4 2− was determined by means of ion chromatography (ICS 1600, Dionex, USA). The amount of 10 g of the 2-mm sieved sample was air-dried and treated with 30% hydrogen peroxide and the soil suspension was allowed to react under heated conditions with subsequent additions of 30% hydrogen peroxide, until all soil organic matter was removed. The soil suspension was then measured by laser diffraction method (LDM) using a Fritsch Analysette 22 MicroTrec plus in the range of 0.1–2000 µm (ISO 11277 ). As the amount of clay particles was overestimated by the Fraunhofer theory, the pycnometer bottle method was added (Di Stefano et al. ). The fraction < 63 µm was operationally defined as the sum of fine silt and clay, which is typically associated with the majority of the soil organic matter (Hemkemeyer et al. ) and considered to be to be important for soil stabilization through interactions between fine particles and organic carbon (Bird et al. ). TOC and TN contents of the soil were measured using an elemental analyzer (vario MICRO cube, Elementar Analysensysteme GmbH). The detection limits were TOC (1 mg g −1 ) and TN (0.1 mg g −1 ) and it was not possible to determine TN content in samples younger than 54 years. Soil samples contained a significant amount of carbonates, which were removed prior to analysis via acid fumigation with HCl vapours (Harris et al. ). Analyses of TOC and δ 13 C TOC contents of dried soil material were conducted with an NC Elemental analyzer (ThermoQuest, Bremen, Germany) connected to an isotope ratio mass spectrometer (IR-MS Delta X Plus, Finnigan, Bremen, Germany). Carbon, nitrogen, and phosphorus in microbial biomass (C mic , N mic , and P mic ) were estimated using the chloroform-fumigation extraction method (Brookes et al. , , Vance et al. ). The extraction of fumigated soil followed the protocol for available nutrients described above. Nutrient concentrations in microbial biomass were calculated as the difference between concentrations in fresh soil extract and fumigated soil extract. Results were corrected for incomplete extraction of all C mic (Kec = 0.45), all N mic (Ken = 0.54), and all P mic (Kep = 0.4) (Brookes et al. , , Vance et al. ). Tetraether lipids were extracted from six forefield soils (20 g wet weight, one pooled sample for each forefield site) after the protocol of Hopmans et al. ( ). The analysis was conducted solely as a complementary measure, without laboratory repetition. Therefore, caution was exercised when evaluating the data. Lipid extracts to assess the biomarkers were dried and stored frozen until analysis by liquid chromatography mass spectrometry (Becker et al. ). Tetraether lipids were quantified relative to a C46 tetraether injection standard and the relative response of a caldarchaeol standard. The compounds measured include isoprenoidal glycerol dibiphytanyl glycerol tetraethers having zero to three rings (iGDGT0-3; m/z 1302, 1300, and 1298), crenarchaeol (m/z 1292, biomarker for marine archea), and branched tetraethers produced by soil bacteria having varying degrees of rings and methylation (m/z 1018, 1020, 1022, 1032, 1034, 1036, 1046, 1048, and 1050). iGDGT having four rings (iGDGT4) was below the detection limit. The BIT index (branched and isoprenoid tetraethers index), which is a proxy for relative contribution of terrestrial and marine organic matter input to sediments, was calculated as a ratio of the sum of all measured branched tetraethers versus crenarcheol (Schouten et al. ).
Extracellular enzymatic activities were measured for five soil enzymes responsible for organic carbon, nitrogen, and phosphorus utilization with standard fluorometric technique (Marx et al. ) modified by Bárta et al. ( ). The activity of extracellular enzymes was determined in water extracts. Briefly, 0.5 g of soil was homogenized in 50 ml of distilled water via an ultrasonication. The soil suspensions (200 µl) were then transferred to a 96-well microplate. Then, 50 µl of substrates labeled with 4-methylumbelifferone (MUB) or 7-amino-4-methylcoumarin (AMC) were added. Standard curves were measured with MUB/AMC in soil slurries for every sample separately. Concentrations of MUB/AMC for standard curves were: 1, 5, and 10 µM. Microplates were incubated at room temperature for 2 h and fluorescence was measured every 30 min. All fluorescence measurements were performed on the microplate reader INFINITE F200 (Tecan, Germany) using the excitation wavelength of 365 nm and emission wavelength of 450 nm. Enzyme activities were expressed as nmol of liberated MUF/AMC per hour per gram of dry soil. The stoichiometric analysis of enzymes was done according to Hill et al. ( ) and is based on the assumption that microbes release extracellular enzymes in order to gain needed C, N, and P nutrients from complex organic substrates. The analysis compares ratios of C:N and N:P processing enzymes and the results may imply which nutrients are mainly limiting for the microbial growth and prosperity (Schmidt et al. ). Carbon processing enzymes (E C ) comprise β-glucosidase (GLU) and cellobiohydrolase (CELL) activity, nitrogen processing enzymes (E N ) are the sum of leucin aminopeptidase (ALA) and chitinase (CHIT) activity and phosphorus (E P ) is represented by phosphatase (PHO) acitivity. The amount of 0.25 g was used for DNA extraction using PowerSoil DNA isolation kit (MO BIO Laboratories, USA). The DNA was quantified fluorimetrically using Qubit (Thermofisher Scientific, UK) following the standard protocol and kit for dsDNA quantification. The aliquots of DNA extracts were sent to SEQme lab (Prague, Czech Republic) for the preparation of a library and sequencing using MiSeq platform. The Earth Microbiome Project (EMP) protocol was used for library preparation with modified universal primers 515FB/806RB (Caporaso et al. ) and ITS1F/ITS2 (White et al. ) for prokaryotic 16S rRNA and fungal ITS1 amplicons, respectively. The coverage of prokaryotic primer pair 515FB/806RB was additionally tested in silico using ARB Silva database release 128. The primer pair 515FB/806RB covers almost uniformly all major bacterial and archaeal phyla. Bacterial 16S rRNA raw pair-end reads (150 bp) were joined using ea-utils to obtain reads of ~250 bp length. Quality filtering of reads was applied as previously described (Caporaso et al. ). After quality filtering the sequences were trimmed to 250 bp. We obtained 720 625 bacterial and 1746 017 fungal sequences after quality trimming and filtering. Before picking the operational taxonomic units (OTU), the fungal ITS1 region was extracted from reads using ITSx algorithm (Bengtsson-Palme et al. ). Both 16S and ITS1 amplicons were trimmed to equal lengths in order to avoid spurious OTU clusters (Edgar ). Taxonomy was assigned to each read by accepting Silva119 taxonomy string of the best matching Silva119 sequence. Fungal reads were clustered to OTUs using open-reference OTU picking protocol (sequence similarity 98.5%) using UNITE ver. 8.0 database (Koljalg et al. ). Blast algorithm (e-value ≤ 0.001) was used for taxonomic assignment. FAPROTAX (Louca et al. ) and FUNguild (Nguyen et al. ) algorithm was then used for bacterial functional annotation and the fungal lifestyle assignments, respectively. Raw sequences were deposited at ENA under the project n. PRJEB51542. Quantification of bacterial and fungal SSU rRNA genes was performed using the FastStart SybrGREEN Roche® Supermix and Step One system (Life Technologies, USA). Each reaction mixture (20 µl) contained 2 µl DNA template (∼ 1–2 ng DNA), 1 µl each primer (0.5 pmol µl −1 each, final concentration), 6 µl dH2O, 10 µl FastStart SybrGREEN Roche® Supermix (Roche, France), and 1 µl BSA (Fermentas, 20 mg µl −1 ). Initial denaturation (3 min, 95°C) was followed by 30 cycles of 30 s at 95°C, 30 s at 62°C, 15 s at 72°C, and completed by fluorescence data acquisition at 80°C used for target quantification. Product specificity was confirmed by melting point analysis (52–95°C with a plate read every 0.5°C) and amplicon size was verified with agarose gel electrophoresis. Bacterial standards consisted of a dilution series (ranging from 10 1 to 10 9 gene copies µl −1 ) of a known amount of purified plasmid where PCR product using the SSU gene-specific primers 341F/534R (Muyzer et al. ) was inserted. R 2 values for the standard curves were > 0.99. Slope values were −> 3.37 giving an estimated amplification efficiency of > 93%. The qPCR conditions for fungal quantification were as follows: initial denaturation (10 min, 95°C) followed by 40 cycles of 1 min at 95°C, 1 min at 56°C, 1 min at 72°C, and completed by fluorescence data acquisition at 72°C used for target quantification. Fungal DNA standards consisted of a dilution series (ranging from 10 1 to 10 7 gene copies µl −1 ) of a known amount of purified plasmid where PCR product using the SSU gene-specific primers nu-SSU-0817–5′ and nu-SSU1196-3′ (Borneman and Hartin ) was inserted. R 2 values for the fungal standard curves were > 0.99. The slope was between −3.34 and −3.53 giving estimated amplification efficiency between 95% and 93%, respectively. Detection limits for the assays (i.e. lowest standard concentration that is significantly different from the nontemplate controls) were less than 100 gene copies for each of the genes per assay. Samples, standards and nontemplate controls were run in duplicates. To deal with potential inhibition during PCR the enhancers BSA and DMSO were added to the PCR mixture. Also, several dilutions (10x, 20x, 50x, 100x, and 1000x) for each sample were tested to see the dilution effect on Ct values. Alpha diversity metrics the Chao1 richness (measurement of OTUs expected in samples given all the bacterial/fungal species that were identified in the samples), were calculated after rarefying all samples to the same sequencing depth of 3200 and 2800 sequences for bacterial and fungal community, respectively (Chao ).
The effect of the northern versus southern forefield on soil parameters and on parameters changes in time were tested using a generalized linear mixed-effect model with Gamma distribution and logarithmic link function and a likelihood-ratio test, which was run in R version 4.0.2 (R Core Team ) in R package stats . The zone identity for each forefield was used as a random effect (i.e. eight groups). The significant differences in time were evaluated separately for each forefield using Tukey’s HSD post hoc tests in the R package multcomp to examine pairwise differences between the zones (different soil ages). Spearman correlation coefficients were determined in R package ggpubr to assess how strongly chosen parameters were related to each other. The effect of forefield on the total content of chosen soil elements determined by XRF and on labile compounds was visualized using principle component analyses in CANOCO for Windows 5.0 (Ter Braak and Šmilauer ) with zone (age) as a covariate. In all analyses from CANOCO, only the adjusted explained variation is presented in the text. In CANONO, variation partitioning was also applied to quantify the unique effects of TOC, bedrock (total content of elements in soil matrix determined by XRF—Si, Ca, Al, K, Fe, Mg, P, and S) and available nutrients (DOC, P-PO 4 3− , Nmin, Ca 2+ , Mg 2+ , and K + ) on the microbial quantity (C mic , bacteria, and fungi) and quality (CNP processing enzymes), separately during 79 and 10 000 years of soil succession. Subsequently, a forward selection procedure was also performed to identify the single parameters that best explain the microbial quantity and quality and only P -values adjusted with Holms correction were considered. Both analyses were run with zone (age) as covariate. To estimate the beta-diversity in soil microbial communities and in their functional analyses, nonmetric multidimensional scaling (NMDS) ordinations were generated using CANOCO on the basis of Bray–Curtis dissimilarities of square-root transformed relative abundances of OTUs.
Soil properties XRF analysis of main soil elements has identified Si to be a prevailing element in the soil matrix of both forefields (from 13.6% to 22.5%), followed by calcium (from 3.5% to 8.3%) and/or aluminium (from 2.6% to 6.0%) ( , ). The significant difference in the content of most soil elements between northern and southern forefield corresponds to the overall lithological distinction described in detail within the methods (Fig. ; , ). This geochemical distinction is further evident in the higher amount of water extractable cations and in higher content of fine soil particles < 63 μm of the more easily weatherable sedimentary rocks, which are present in the glacial till of the southern forefield (Fig. ; , ). The higher proportion of fine particles < 63 µm in the soil matrix correlates with a higher soil moisture observed in the whole southern forefield ( r = 0.7, P < .05; , ). Besides the geochemical distinction, forefields did not significantly differ in pH, TOC, TN, or water extractable C and P, but they differed in water extractable N across all sites ( , ). All samples were alkaline with pH between 8.28 and 9.81. TOC levels showed very distinct trends in time within each forefield. TOC gradually increased with age on the northern forefield, from low values at the N1 site (0.11%) to high values at the NR site (2.78%) ( , ). In contrast, on the southern forefield, TOC was higher at the S1 site (0.16%) than at the N1 site (0.11%), decreased at the S2 site (0.09%), and grew in the later stages (S3 0.19%, SR 3.25%) (Fig. ; , ). TN was below the detection limit in sites younger than 54 years and then increased on both forefields ( , ). All water extractable nutrients generally increased with time, only SO 4 2− had the opposite trend. Both forefields contained a proportion of sulphate minerals, which released especially high amounts of SO 4 2− on the S1 site together with water extractable Ca, Mg, and K ( , ). The stable C isotope composition of TOC (δ 13 C TOC ) varied from −20.8‰ to −24.3‰ ( , ). TOC samples collected on the southern forefield were typically more depleted in 13 C than on the northern forefield, and N1 site exhibited the highest δ 13 C TOC values (−20.8‰). The interpretation of the biomarker data is constrained by the limited replication, as only one pooled sample was used for each forefield site. However, despite this limitation, we have included the data in order to propose a broader range of explanations for the various successional developments observed. The BIT index on the northern forefield steadily increased with soil age (0.39–0.59; , ), suggesting increasing inputs of terrestrial (brGDGT) relative to pelagic/glacial (crenarchaeol) OM over time (Hopmans et al. , Brady and Daniel ). In contrast, the BIT index was consistently high on the southern forefield (> 0.69; , ) suggesting relatively higher inputs of OM from terrestrial sources. The organic matter on the southern forefield generally comprised higher content of mono- and disaccharides, fatty acids, and sitosterol and brassicasterol ( , ). Microbial quantity and stoichiometry Distinct temporal trends in TOC dynamics within each forefield were reflected in (i) microbial abundance expressed as C mic , (ii) the quantity of bacteria and fungi expressed as SSU gene copies per gram of dry soil, and (iii) bacteria species richness (Fig. ). These microbial parameters differed the most between the youngest sites (i.e. N1 and S1). Values were the lowest at N1 sites, whereas southern S1 sites were characterized by significantly higher microbial biomass and bacterial and fungal abundance (Fig. ). A significant drop of microbial biomass was observed at southern S2 sites, thus the drop in microbial quantity was significant only for fungi, not for bacteria (Fig. ). The trends and values in the older sites (> 54 years) were similar in both forefields (Fig. ). Bacterial species richness was equivalent or increased with time on both forefields (Chao index: northern forefield from 890 to 1396, respectively, southern forefield from 1306 to 1648, respectively) (Fig. ), and was consistently higher in the southern forefield. However, fungal species richness in the youngest plots showed a completely different trend between the two forefields (Fig. ). In the southern foreland, the Chao index was significantly lower in the youngest plots S1 (679) and S2 (616) compared to the oldest plots S3 (480) and SR (513). The fungi in the northern foreland showed a completely opposite trend (Fig. ). Forefields also differed in C mic normalized to TOC, with extremely low values at the N1 sites (9.6 ± 11.0), compared to the significantly higher values at the S1 sites (143 ± 64.8) (Fig. ). However, similarly to TOC, most of all microbial parameters converged at N3 and S3 sites (after 79 years) and the forefields develop similarly in later stages. Fungi to bacteria ratio was very low in the youngest sites of both forefields (N1, N2, S1, and S2; from 7.24 × 10 −5 to 4.00 × 10 −4 ), indicating the general prevalence of bacterial over fungal communities. In later stages, especially in S3 sites (1.54 × 10 −3 ), an expansion of fungi was observed (Figs and ). Similarly to C mic , also N mic and P mic were considerably higher at the S1 site (21.7 µgN g dw − 1 and 5.5 µgP g dw −1 , respectively) compared to the N1 site (0.95 µgN g dw − 1 and 1.5 µgP g dw −1 , respectively) and did not change significantly at the S2 and S3 sites except of a decrease in N mic at S2 sites, while they gradually increased on the northern forefield over time (Fig. and ; , ). Enzyme activity and stoichiometry The sum of enzyme activity increased along the northern forefield, while there was no clear pattern on the southern forefield (Table ). Despite remarkable differences in the sum of enzyme activity per gram of soil among N1 (31 nmol h −1 g −1 ) and S1 sites (359 nmol h −1 g −1 ), the values were similar when normalized to microbial biomass (2.82 nmol h −1 g −1 and 2.86 nmol h −1 g −1 , respectively). At these youngest sites, phosphatase (PHO) was the main processing enzyme measured (84.3% and 59.5%, respectively), but its proportion decreased in time on behalf of N processing enzymes (Table ). Carbon acquiring enzymes (GLU and CELL) had very low activities across both chronosequences until the reference sites. Stoichiometric analysis of CNP processing enzymes (Hill et al. ) revealed a clear distinction of the youngest deglaciated N1 and S1 sites in terms of potential nutrient limitation (Fig. ). While the northern forefield appeared to be C and P colimited, the southern forefield was mainly P limited. Later, both forefields showed a similar trend of gradual shift from P limitation (S1 and N2 sites) to N limitation (S2, S3, and N3 sites) and finally to almost no nutrient limitation (S3, SR, and NR sites). Effect of chemical parameters on microbial quantity and enzyme activity Both microbial quantity (microbial biomass, abundance of bacteria and fungi determined by qPCR) and enzyme activities were strongly positively correlated with soil age (C mic r = 0.72, bacteria r = 0.68, fungi r = 0.82, and enzymes r = 0.66, all P < .001) and with TOC (C mic r = 0.86, bacteria r = 0.75, fungi r = 0.81, and enzymes r = 0.63 respectively, P < .001). The exception was the soil development within 79 years in the S1–S3 sites, where the correlations of microbial parameters with age were not significant and the soil development was clearly driven by different parameters. Variation partitioning analyses of both forefields revealed that, together with time and TOC, bedrock type and available nutrients also played an important role. Specifically, bedrock (16.5%) and available nutrients (10.5%) explained together more variability than TOC (18.9%) during the first 79 years of soil succession (Fig. ). The forward selection of single environmental parameters revealed TOC to be the only explanatory variable of the microbial quantity (47.7%, pseudo-F = 58.3, P = .022) and microbial activity (53.6%, pseudo-F = 73.8, P = .022) across all sites. Interestingly, only microbial activity of the southern forefield was significantly influenced also by P-PO 4 3− (7.4%, pseudo-F = 6.4, P = .03). This available phosphorus seemed to be especially important for the abundance of fungi, which showed a close correlation (r = 0.72, P < .001). Microbial community composition and functioning After quality trimming and filtering, we obtained 737 010 (7046 OTUs) and 1521 308 (2774 OTUs) high quality sequences for bacterial and fungal communities, respectively. We were able to taxonomically affiliate 3932 bacterial OTUs (56%) and 602 fungal OTUs (22%) up to the genus level. In the fungal community, we found 535 OTUs (19%) that were taxonomically assigned only to the domain level. These fungal species might represent novel, unknown arctic species that are still not represented in recent fungal databases. Bacteria and fungi were rarefied to 3212 and 2866 sequences per sample, respectively. NMDS analysis showed that the bacterial communities were different between the early and late stages (10 kyr) and were also different between the two forefields (Fig. ). In contrast, functional diversity was only different between the early stages and the late (10 kyr) stage but was not as markedly different between the forefields. Thus, our data showed that (i) although the bacterial communities of the early-stage forefields were distinct, they exhibited a higher degree of functional overlap and (ii) that the two forefields became more similar in both species composition and function after about 10 kyr. This trend of community convergence was consistent with the data from chemical analyses of the soils of the two forefields ( , ). Functional analyses further showed critical differences in trends of specific fungal lifestyles, mainly the proportions of lichenized fungi between the northern and southern forefield ( , ). Lichenized fungi in the youngest southern S1 sites represented 55% of the community, on average, and steadily decreased with time. In contrast, they represented only 12% in the youngest northern N1 sites and gradually increased from N1 to N3 sites. The high proportion of OTU of lichenized fungi at the youngest southern S1 sites was accompanied by relatively higher proportions of cyanobacteria and diazotrophs ( , ). These functional guilds decreased with time on the southern forefield while an opposite trend was seen on the northern forefield. This can relate to decreasing trends of C and N flux to the soil in the southern forefield, presumably via lower CO 2 and N 2 fixation performed by cyanobacteria and diazotrophs.
XRF analysis of main soil elements has identified Si to be a prevailing element in the soil matrix of both forefields (from 13.6% to 22.5%), followed by calcium (from 3.5% to 8.3%) and/or aluminium (from 2.6% to 6.0%) ( , ). The significant difference in the content of most soil elements between northern and southern forefield corresponds to the overall lithological distinction described in detail within the methods (Fig. ; , ). This geochemical distinction is further evident in the higher amount of water extractable cations and in higher content of fine soil particles < 63 μm of the more easily weatherable sedimentary rocks, which are present in the glacial till of the southern forefield (Fig. ; , ). The higher proportion of fine particles < 63 µm in the soil matrix correlates with a higher soil moisture observed in the whole southern forefield ( r = 0.7, P < .05; , ). Besides the geochemical distinction, forefields did not significantly differ in pH, TOC, TN, or water extractable C and P, but they differed in water extractable N across all sites ( , ). All samples were alkaline with pH between 8.28 and 9.81. TOC levels showed very distinct trends in time within each forefield. TOC gradually increased with age on the northern forefield, from low values at the N1 site (0.11%) to high values at the NR site (2.78%) ( , ). In contrast, on the southern forefield, TOC was higher at the S1 site (0.16%) than at the N1 site (0.11%), decreased at the S2 site (0.09%), and grew in the later stages (S3 0.19%, SR 3.25%) (Fig. ; , ). TN was below the detection limit in sites younger than 54 years and then increased on both forefields ( , ). All water extractable nutrients generally increased with time, only SO 4 2− had the opposite trend. Both forefields contained a proportion of sulphate minerals, which released especially high amounts of SO 4 2− on the S1 site together with water extractable Ca, Mg, and K ( , ). The stable C isotope composition of TOC (δ 13 C TOC ) varied from −20.8‰ to −24.3‰ ( , ). TOC samples collected on the southern forefield were typically more depleted in 13 C than on the northern forefield, and N1 site exhibited the highest δ 13 C TOC values (−20.8‰). The interpretation of the biomarker data is constrained by the limited replication, as only one pooled sample was used for each forefield site. However, despite this limitation, we have included the data in order to propose a broader range of explanations for the various successional developments observed. The BIT index on the northern forefield steadily increased with soil age (0.39–0.59; , ), suggesting increasing inputs of terrestrial (brGDGT) relative to pelagic/glacial (crenarchaeol) OM over time (Hopmans et al. , Brady and Daniel ). In contrast, the BIT index was consistently high on the southern forefield (> 0.69; , ) suggesting relatively higher inputs of OM from terrestrial sources. The organic matter on the southern forefield generally comprised higher content of mono- and disaccharides, fatty acids, and sitosterol and brassicasterol ( , ).
Distinct temporal trends in TOC dynamics within each forefield were reflected in (i) microbial abundance expressed as C mic , (ii) the quantity of bacteria and fungi expressed as SSU gene copies per gram of dry soil, and (iii) bacteria species richness (Fig. ). These microbial parameters differed the most between the youngest sites (i.e. N1 and S1). Values were the lowest at N1 sites, whereas southern S1 sites were characterized by significantly higher microbial biomass and bacterial and fungal abundance (Fig. ). A significant drop of microbial biomass was observed at southern S2 sites, thus the drop in microbial quantity was significant only for fungi, not for bacteria (Fig. ). The trends and values in the older sites (> 54 years) were similar in both forefields (Fig. ). Bacterial species richness was equivalent or increased with time on both forefields (Chao index: northern forefield from 890 to 1396, respectively, southern forefield from 1306 to 1648, respectively) (Fig. ), and was consistently higher in the southern forefield. However, fungal species richness in the youngest plots showed a completely different trend between the two forefields (Fig. ). In the southern foreland, the Chao index was significantly lower in the youngest plots S1 (679) and S2 (616) compared to the oldest plots S3 (480) and SR (513). The fungi in the northern foreland showed a completely opposite trend (Fig. ). Forefields also differed in C mic normalized to TOC, with extremely low values at the N1 sites (9.6 ± 11.0), compared to the significantly higher values at the S1 sites (143 ± 64.8) (Fig. ). However, similarly to TOC, most of all microbial parameters converged at N3 and S3 sites (after 79 years) and the forefields develop similarly in later stages. Fungi to bacteria ratio was very low in the youngest sites of both forefields (N1, N2, S1, and S2; from 7.24 × 10 −5 to 4.00 × 10 −4 ), indicating the general prevalence of bacterial over fungal communities. In later stages, especially in S3 sites (1.54 × 10 −3 ), an expansion of fungi was observed (Figs and ). Similarly to C mic , also N mic and P mic were considerably higher at the S1 site (21.7 µgN g dw − 1 and 5.5 µgP g dw −1 , respectively) compared to the N1 site (0.95 µgN g dw − 1 and 1.5 µgP g dw −1 , respectively) and did not change significantly at the S2 and S3 sites except of a decrease in N mic at S2 sites, while they gradually increased on the northern forefield over time (Fig. and ; , ).
The sum of enzyme activity increased along the northern forefield, while there was no clear pattern on the southern forefield (Table ). Despite remarkable differences in the sum of enzyme activity per gram of soil among N1 (31 nmol h −1 g −1 ) and S1 sites (359 nmol h −1 g −1 ), the values were similar when normalized to microbial biomass (2.82 nmol h −1 g −1 and 2.86 nmol h −1 g −1 , respectively). At these youngest sites, phosphatase (PHO) was the main processing enzyme measured (84.3% and 59.5%, respectively), but its proportion decreased in time on behalf of N processing enzymes (Table ). Carbon acquiring enzymes (GLU and CELL) had very low activities across both chronosequences until the reference sites. Stoichiometric analysis of CNP processing enzymes (Hill et al. ) revealed a clear distinction of the youngest deglaciated N1 and S1 sites in terms of potential nutrient limitation (Fig. ). While the northern forefield appeared to be C and P colimited, the southern forefield was mainly P limited. Later, both forefields showed a similar trend of gradual shift from P limitation (S1 and N2 sites) to N limitation (S2, S3, and N3 sites) and finally to almost no nutrient limitation (S3, SR, and NR sites).
Both microbial quantity (microbial biomass, abundance of bacteria and fungi determined by qPCR) and enzyme activities were strongly positively correlated with soil age (C mic r = 0.72, bacteria r = 0.68, fungi r = 0.82, and enzymes r = 0.66, all P < .001) and with TOC (C mic r = 0.86, bacteria r = 0.75, fungi r = 0.81, and enzymes r = 0.63 respectively, P < .001). The exception was the soil development within 79 years in the S1–S3 sites, where the correlations of microbial parameters with age were not significant and the soil development was clearly driven by different parameters. Variation partitioning analyses of both forefields revealed that, together with time and TOC, bedrock type and available nutrients also played an important role. Specifically, bedrock (16.5%) and available nutrients (10.5%) explained together more variability than TOC (18.9%) during the first 79 years of soil succession (Fig. ). The forward selection of single environmental parameters revealed TOC to be the only explanatory variable of the microbial quantity (47.7%, pseudo-F = 58.3, P = .022) and microbial activity (53.6%, pseudo-F = 73.8, P = .022) across all sites. Interestingly, only microbial activity of the southern forefield was significantly influenced also by P-PO 4 3− (7.4%, pseudo-F = 6.4, P = .03). This available phosphorus seemed to be especially important for the abundance of fungi, which showed a close correlation (r = 0.72, P < .001).
After quality trimming and filtering, we obtained 737 010 (7046 OTUs) and 1521 308 (2774 OTUs) high quality sequences for bacterial and fungal communities, respectively. We were able to taxonomically affiliate 3932 bacterial OTUs (56%) and 602 fungal OTUs (22%) up to the genus level. In the fungal community, we found 535 OTUs (19%) that were taxonomically assigned only to the domain level. These fungal species might represent novel, unknown arctic species that are still not represented in recent fungal databases. Bacteria and fungi were rarefied to 3212 and 2866 sequences per sample, respectively. NMDS analysis showed that the bacterial communities were different between the early and late stages (10 kyr) and were also different between the two forefields (Fig. ). In contrast, functional diversity was only different between the early stages and the late (10 kyr) stage but was not as markedly different between the forefields. Thus, our data showed that (i) although the bacterial communities of the early-stage forefields were distinct, they exhibited a higher degree of functional overlap and (ii) that the two forefields became more similar in both species composition and function after about 10 kyr. This trend of community convergence was consistent with the data from chemical analyses of the soils of the two forefields ( , ). Functional analyses further showed critical differences in trends of specific fungal lifestyles, mainly the proportions of lichenized fungi between the northern and southern forefield ( , ). Lichenized fungi in the youngest southern S1 sites represented 55% of the community, on average, and steadily decreased with time. In contrast, they represented only 12% in the youngest northern N1 sites and gradually increased from N1 to N3 sites. The high proportion of OTU of lichenized fungi at the youngest southern S1 sites was accompanied by relatively higher proportions of cyanobacteria and diazotrophs ( , ). These functional guilds decreased with time on the southern forefield while an opposite trend was seen on the northern forefield. This can relate to decreasing trends of C and N flux to the soil in the southern forefield, presumably via lower CO 2 and N 2 fixation performed by cyanobacteria and diazotrophs.
Global warming has accelerated the retreat of glaciers in polar ecosystems and the exposure of new terrain that may be subject to succession by many species, including microbial species. Studies in Alaska have shown that limited nutrient availability on poor geological substrates after the glacier retreat may be more important for the early microbial succession than climatic conditions, such as temperature and precipitation (Schmidt et al. ). These studies have also highlighted that the influence of local climate is difficult to separate from that of nutrient availability (Nemergut et al. ). Both questions remain unanswered because the available data were mostly compared between distant glacier forefields, where the influence of climatic conditions could not be appropriately filtered out (Lazzaro et al. , Bajerski and Wagner , Alfaro et al. ). Our study provides, for the first time, a detailed insight into the biogeochemical and microbial succession in two closely spaced forefields of the retreating Nordenskiöldbreen (Svalbard, Norway) with similar climatic conditions but different bedrock geology (Fig. ). Given the close proximity of the two forefields, this study offers a unique possibility to test the hypothesis that the substrate quality is critically important for microbial succession during the first decades after deglaciation (Schmidt et al. ) and may considerably influence the development of microbial communities and biogeochemistry in later successional stages of soil evolution. Effect of mineral substrate on nutrient availability Our study confirmed that different substrate parameters and nutrient availability resulted in different soil succession of the microbial community in nearby forefields. The most pronounced differences were in mineral substrate composition (Fig. ). While the mineral substrate of the northern forefield was dominated by glacially formed rocks and thin glacial deposits with silicate rocks, the southern forefield was completely covered by thick glacial deposits with high proportion of carbonates. Carbonate sediments typically weather more rapidly and can increase the content of fine soil particles < 63 µm, leading to higher soil moisture ( , ) and higher nutrient availability for microbes (Fig. ). The higher content of simple saccharides and available fatty acids ( , ) and higher content of water-extractable Ca, Mg, and K (Fig. ; , ) in the southern forefield indicated higher quality OM that can be more rapidly utilized by microorganisms, leading to higher microbial biomass in the southern forefield, especially at the youngest sites. Our data, therefore, show that substrate chemistry and nutrient availability are more important for microbial activity and early soil development (Figs and ) than the commonly reported TOC (Fig. ) (Tytgat et al. , Kim et al. ). However, the above-mentioned differences between the forefields do not provide us with any information on whether C, N, or P are available in sufficient quantities for microorganisms and whether they are limiting for their growth and development. To answer this question, we used an enzymatic model that provides information on the utilization of C, N, and P in the context of their limitation for microorganisms and their enzymatic activity (Fig. ) (Hill et al. ). Enzymes of heterotrophic bacteria and fungi involved in C, N, and P processing showed distinct trends, especially in early successional stages on both forefields. We found a shift from P to N limitation in the southern forefield, whereas the northern forefield was initially more colimited by C–P and shifted to greater N limitation in later stages, which is consistent with a previous study (Bueno de Mesquita et al. ). Therefore, we hypothesize that the lower complexity and higher quality of C-rich organic material on the southern forefield required less investment by microorganisms in C enzymes to obtain energy and carbon, which allowed microbes to grow faster in the early stages but resulted in limitation of N and P. This can also be observed in different microbial biomass and abundance of bacteria and fungi in these forefields. The later shift to N limitation on both forefields corroborates other observations and a general paradigm of higher N requirements with more developed soils, among others, in association with increasing vegetation cover (Vitousek et al. , Bueno de Mesquita et al. ). The different character of the organic material was further supported by an increasing BIT index and decreasing δ 13 C with age in the northern forefield, indicating gradual accumulation of soil-derived organic material, as could be expected along a classical chronosequence. These findings are consistent with reports of taxa mainly from supra- and subglacial habitats in front of the Damma Glacier, Switzerland (Rime et al. ) or a gradual transition from glacial to edaphic taxa in front of the Fourcade Glacier, maritime Antarctica (Gyeong et al. ). We are aware that the inclusion of biomarker data with low replication (one pooled sample for each forefield site) needs to be justified. Based on our results, we would like to suggest that biomarker distributions may help to unravel microbial assembly dynamics that are still poorly understood and may not be fully captured by measuring soil physico-chemical parameters alone. Different succession of the microbiome in the northern and southern forefield In Arctic extreme environments, the successional pattern deviates from the classical model of directional change and species replacement (Matthews , Svoboda and Henry ). Under excessive climatic stress, competition is reduced and directional, nonsubstituting succession is more common, with native species remaining and new species being added during succession (Svoboda and Henry , Jones and Henry ). Previous studies from Svalbard glacier forefields suggest that colonization of both plants and root-bound fungi follow this model of directional, nonsubstituting succession (Hodkinson et al. , Davey et al. ). Our data show that the bacterial and fungal microbiomes of the northern and southern forefields evolved differently in biomass, composition, and activity during the early stages of deglaciation. This different development lasted roughly until the first 79 years after deglaciation. Thereafter, convergence occurred and the differences between the forefields almost disappeared (Figs – and ). Total microbial biomass was significantly higher on the youngest S1 sites than on the N1 sites (Fig. ). Compared to the northern, the southern microbiome experienced a significant decline in both abundance and diversity over a relatively short period of time, probably related to the decline of fungi (Fig. ) and probably also related to a significant rebuilding of the fungal microbiome (Fig. ; , ). The Chao index showed very low fungal richness of the youngest southern sites (Fig. ), indicating a dominance of few fungal species. Thus, the later increase in species richness may be a result of community rearrangement. Fungal communities developed differently from the beginning on both forefields, and differences were also reflected in the relative abundance of specific functional groups of fungi. Lichen-forming fungi are often found in early successional stages after glacial retreat (Syers and Iskandar , Zumsteg et al. , Garrido-Benavent et al. ), and they also strongly and relatively rapidly dominated the microbiome of the southern forefield, which correlates with higher abundance of bacterial phototrophs ( , ) and with higher levels of sitosterol and brassicasterol ( , ), biomarkers indicating algal and/or plant phytomass (Volkman , Rontani et al. ). However, the lichen-forming fungi abundance in the fungal community declined significantly over time on southern forefield and they were gradually replaced by saprotrophs and partially ericoid species in the oldest sites after deglaciation (Selosse et al. ). Thus, the fungal microbiome of the southern forefield experienced more pronounced fluctuations in both abundance and diversity, and a significant rebuilding of the fungal microbiome was observed, indicating a direct evolution with partial replacement of fungal species compared to the northern forefield with a rather nonsubstituting succession, more common for the High Arctic (Svoboda and Henry , Jones and Henry ). In contrast to the fungal microbiome, bacterial abundance and richness increased gradually with soil development on both forefields, consistent with other studies (Schütte et al. , Jiang et al. , Gyeong et al. ). However, we found specific differences in some major functional groups that might be important for initial succession in soil. The microbiome of the northern forefield had a higher proportion of bacteria degrading more complex and less available compounds in the early stages of degradation. This was mainly reflected by a higher proportion of cellulolytic and chitinolytic bacteria ( , ) and probably related to lower nutrient availability and higher C mining (Fig. ; , ). In contrast, the young southern forefield contained a higher proportion of phototrophic and nitrogen-fixing bacteria ( , ), also supported by lower δ 13 C TOC values ( , ). These two key functional guilds in the southern forefield mediate the flux of atmospheric C and N during early soil succession to microbial community. When they have sufficient P for energy production (Darcy and Schmidt , Darcy et al. ), they ‘pump’ available C and N into the system, allowing other heterotrophic microbes to develop (Chapin et al. ). Indeed, more P mic was measured on the southern forefield (Fig. ), suggesting that higher availability of P promoted the development of heterotrophic bacteria and fungi and increased the overall microbial biomass in the early stages of deglaciation compared to the northern forefield. Our findings are consistent with other studies that have found that P is essential for accelerating phototrophic growth (Darcy and Schmidt , Bueno de Mesquita et al. ) and for microbial succession in general (Knelman et al. ). Thus, the two forefields evolved quite differently in terms of the number, composition, and functional potential of the fungal and bacterial microbiome in the early stages up to 79 years after glacial retreat. Different trends in convergence and redundancy of bacteria and fungi over time Similarly to other studies (Rime et al. , Alfaro et al. , Jiang et al. , Garrido-Benavent et al. , Vimercati et al. ), the trajectories of bacteria and fungi showed different patterns during succession (Fig. ). As discussed above, the bacterial and fungal microbiomes were clearly affected by the geochemistry and nutrient availability in the youngest sites (Fig. ). Only bacteria showed a gradual convergence with time, whereas the lack of a clear pattern of fungal assembly is consistent with previous studies (Brown and Jumpponen , Castle et al. ). Fungal succession is thought to be controlled by stochastic processes (Jiang et al. , Gyeong et al. , Lin et al. , Vimercati et al. ), which may be caused by their different dispersal possibilities or their close relationships with vegetation (Schmidt et al. ), together with high geomorphological and geological heterogeneity of the glacial forefields (Ozerskaya et al. , Wojcik et al. , Gyeong et al. ). Compared to the microbial composition, functional groups of bacteria and fungi were shaped much less by the geochemistry and exhibited partial functional redundancy. In summary, in the northern forefield, most geochemical and microbial parameters gradually increased with time and may be considered as the classical chronosequence reported by most previous studies (Crocker and Major , Chapin et al. , Bernasconi et al. ). In contrast, in the southern forefield, nutrients and microbial biomass were already high at the youngest sites, leading to significant later fluctuations in bacterial and fungal abundance and diversity that last for approximately the first 79 years after deglaciation. We suggest that after deglaciation in the southern forefield, the early microbial communities thrived on the higher availability of nutrients, most likely provided by the mineral substrate and microbial guilds involved in CO 2 and N 2 fixation (mainly lichenized fungi and oxygenic phototrophs) and, therefore, evolved much faster. Subsequently, nutrient depletion occurred rapidly, driving the habitat to nutrient limitation, and the microbial biomass again declined and rearranged in composition and in functional potential (e.g. lichen-forming fungi were replaced by saprotrophic fungi) as partially observed in another study (Schmidt et al. ). Our findings support the hypothesis that different bedrock may accelerate early soil succession after deglaciation, providing limiting nutrients during microorganism assembly, thereby promoting more rapid soil stabilization and providing higher quality organic matter. Our study also showed distinct patterns of bacterial and fungal trajectories during succession. The chemical parameters and microbiomes of both forefields converged gradually with time, while the fungi did not show a clear pattern.
Our study confirmed that different substrate parameters and nutrient availability resulted in different soil succession of the microbial community in nearby forefields. The most pronounced differences were in mineral substrate composition (Fig. ). While the mineral substrate of the northern forefield was dominated by glacially formed rocks and thin glacial deposits with silicate rocks, the southern forefield was completely covered by thick glacial deposits with high proportion of carbonates. Carbonate sediments typically weather more rapidly and can increase the content of fine soil particles < 63 µm, leading to higher soil moisture ( , ) and higher nutrient availability for microbes (Fig. ). The higher content of simple saccharides and available fatty acids ( , ) and higher content of water-extractable Ca, Mg, and K (Fig. ; , ) in the southern forefield indicated higher quality OM that can be more rapidly utilized by microorganisms, leading to higher microbial biomass in the southern forefield, especially at the youngest sites. Our data, therefore, show that substrate chemistry and nutrient availability are more important for microbial activity and early soil development (Figs and ) than the commonly reported TOC (Fig. ) (Tytgat et al. , Kim et al. ). However, the above-mentioned differences between the forefields do not provide us with any information on whether C, N, or P are available in sufficient quantities for microorganisms and whether they are limiting for their growth and development. To answer this question, we used an enzymatic model that provides information on the utilization of C, N, and P in the context of their limitation for microorganisms and their enzymatic activity (Fig. ) (Hill et al. ). Enzymes of heterotrophic bacteria and fungi involved in C, N, and P processing showed distinct trends, especially in early successional stages on both forefields. We found a shift from P to N limitation in the southern forefield, whereas the northern forefield was initially more colimited by C–P and shifted to greater N limitation in later stages, which is consistent with a previous study (Bueno de Mesquita et al. ). Therefore, we hypothesize that the lower complexity and higher quality of C-rich organic material on the southern forefield required less investment by microorganisms in C enzymes to obtain energy and carbon, which allowed microbes to grow faster in the early stages but resulted in limitation of N and P. This can also be observed in different microbial biomass and abundance of bacteria and fungi in these forefields. The later shift to N limitation on both forefields corroborates other observations and a general paradigm of higher N requirements with more developed soils, among others, in association with increasing vegetation cover (Vitousek et al. , Bueno de Mesquita et al. ). The different character of the organic material was further supported by an increasing BIT index and decreasing δ 13 C with age in the northern forefield, indicating gradual accumulation of soil-derived organic material, as could be expected along a classical chronosequence. These findings are consistent with reports of taxa mainly from supra- and subglacial habitats in front of the Damma Glacier, Switzerland (Rime et al. ) or a gradual transition from glacial to edaphic taxa in front of the Fourcade Glacier, maritime Antarctica (Gyeong et al. ). We are aware that the inclusion of biomarker data with low replication (one pooled sample for each forefield site) needs to be justified. Based on our results, we would like to suggest that biomarker distributions may help to unravel microbial assembly dynamics that are still poorly understood and may not be fully captured by measuring soil physico-chemical parameters alone.
In Arctic extreme environments, the successional pattern deviates from the classical model of directional change and species replacement (Matthews , Svoboda and Henry ). Under excessive climatic stress, competition is reduced and directional, nonsubstituting succession is more common, with native species remaining and new species being added during succession (Svoboda and Henry , Jones and Henry ). Previous studies from Svalbard glacier forefields suggest that colonization of both plants and root-bound fungi follow this model of directional, nonsubstituting succession (Hodkinson et al. , Davey et al. ). Our data show that the bacterial and fungal microbiomes of the northern and southern forefields evolved differently in biomass, composition, and activity during the early stages of deglaciation. This different development lasted roughly until the first 79 years after deglaciation. Thereafter, convergence occurred and the differences between the forefields almost disappeared (Figs – and ). Total microbial biomass was significantly higher on the youngest S1 sites than on the N1 sites (Fig. ). Compared to the northern, the southern microbiome experienced a significant decline in both abundance and diversity over a relatively short period of time, probably related to the decline of fungi (Fig. ) and probably also related to a significant rebuilding of the fungal microbiome (Fig. ; , ). The Chao index showed very low fungal richness of the youngest southern sites (Fig. ), indicating a dominance of few fungal species. Thus, the later increase in species richness may be a result of community rearrangement. Fungal communities developed differently from the beginning on both forefields, and differences were also reflected in the relative abundance of specific functional groups of fungi. Lichen-forming fungi are often found in early successional stages after glacial retreat (Syers and Iskandar , Zumsteg et al. , Garrido-Benavent et al. ), and they also strongly and relatively rapidly dominated the microbiome of the southern forefield, which correlates with higher abundance of bacterial phototrophs ( , ) and with higher levels of sitosterol and brassicasterol ( , ), biomarkers indicating algal and/or plant phytomass (Volkman , Rontani et al. ). However, the lichen-forming fungi abundance in the fungal community declined significantly over time on southern forefield and they were gradually replaced by saprotrophs and partially ericoid species in the oldest sites after deglaciation (Selosse et al. ). Thus, the fungal microbiome of the southern forefield experienced more pronounced fluctuations in both abundance and diversity, and a significant rebuilding of the fungal microbiome was observed, indicating a direct evolution with partial replacement of fungal species compared to the northern forefield with a rather nonsubstituting succession, more common for the High Arctic (Svoboda and Henry , Jones and Henry ). In contrast to the fungal microbiome, bacterial abundance and richness increased gradually with soil development on both forefields, consistent with other studies (Schütte et al. , Jiang et al. , Gyeong et al. ). However, we found specific differences in some major functional groups that might be important for initial succession in soil. The microbiome of the northern forefield had a higher proportion of bacteria degrading more complex and less available compounds in the early stages of degradation. This was mainly reflected by a higher proportion of cellulolytic and chitinolytic bacteria ( , ) and probably related to lower nutrient availability and higher C mining (Fig. ; , ). In contrast, the young southern forefield contained a higher proportion of phototrophic and nitrogen-fixing bacteria ( , ), also supported by lower δ 13 C TOC values ( , ). These two key functional guilds in the southern forefield mediate the flux of atmospheric C and N during early soil succession to microbial community. When they have sufficient P for energy production (Darcy and Schmidt , Darcy et al. ), they ‘pump’ available C and N into the system, allowing other heterotrophic microbes to develop (Chapin et al. ). Indeed, more P mic was measured on the southern forefield (Fig. ), suggesting that higher availability of P promoted the development of heterotrophic bacteria and fungi and increased the overall microbial biomass in the early stages of deglaciation compared to the northern forefield. Our findings are consistent with other studies that have found that P is essential for accelerating phototrophic growth (Darcy and Schmidt , Bueno de Mesquita et al. ) and for microbial succession in general (Knelman et al. ). Thus, the two forefields evolved quite differently in terms of the number, composition, and functional potential of the fungal and bacterial microbiome in the early stages up to 79 years after glacial retreat.
Similarly to other studies (Rime et al. , Alfaro et al. , Jiang et al. , Garrido-Benavent et al. , Vimercati et al. ), the trajectories of bacteria and fungi showed different patterns during succession (Fig. ). As discussed above, the bacterial and fungal microbiomes were clearly affected by the geochemistry and nutrient availability in the youngest sites (Fig. ). Only bacteria showed a gradual convergence with time, whereas the lack of a clear pattern of fungal assembly is consistent with previous studies (Brown and Jumpponen , Castle et al. ). Fungal succession is thought to be controlled by stochastic processes (Jiang et al. , Gyeong et al. , Lin et al. , Vimercati et al. ), which may be caused by their different dispersal possibilities or their close relationships with vegetation (Schmidt et al. ), together with high geomorphological and geological heterogeneity of the glacial forefields (Ozerskaya et al. , Wojcik et al. , Gyeong et al. ). Compared to the microbial composition, functional groups of bacteria and fungi were shaped much less by the geochemistry and exhibited partial functional redundancy. In summary, in the northern forefield, most geochemical and microbial parameters gradually increased with time and may be considered as the classical chronosequence reported by most previous studies (Crocker and Major , Chapin et al. , Bernasconi et al. ). In contrast, in the southern forefield, nutrients and microbial biomass were already high at the youngest sites, leading to significant later fluctuations in bacterial and fungal abundance and diversity that last for approximately the first 79 years after deglaciation. We suggest that after deglaciation in the southern forefield, the early microbial communities thrived on the higher availability of nutrients, most likely provided by the mineral substrate and microbial guilds involved in CO 2 and N 2 fixation (mainly lichenized fungi and oxygenic phototrophs) and, therefore, evolved much faster. Subsequently, nutrient depletion occurred rapidly, driving the habitat to nutrient limitation, and the microbial biomass again declined and rearranged in composition and in functional potential (e.g. lichen-forming fungi were replaced by saprotrophic fungi) as partially observed in another study (Schmidt et al. ). Our findings support the hypothesis that different bedrock may accelerate early soil succession after deglaciation, providing limiting nutrients during microorganism assembly, thereby promoting more rapid soil stabilization and providing higher quality organic matter. Our study also showed distinct patterns of bacterial and fungal trajectories during succession. The chemical parameters and microbiomes of both forefields converged gradually with time, while the fungi did not show a clear pattern.
fiad104_Supplemental_Files Click here for additional data file.
|
Therapeutic outcomes of thermal ablation versus repeated hepatic resection for recurrent hepatocellular carcinoma by using propensity score analysis: a multicenter real-world study | a552a2ca-797d-427e-bde8-0581512ba391 | 11841235 | Surgical Procedures, Operative[mh] | Hepatocellular carcinoma (HCC) is the most common primary malignant liver tumor, ranking sixth among the most frequently diagnosed cancers globally and the third leading cause of cancer-related mortality in 2022 . Its incidence and mortality rates are continually rising, imposing an increasing burden on public health . Hepatic resection remains an effective therapeutic strategy for early-stage HCC. However, owing to its histopathologic, genetic, and immunophenotypic heterogeneity, the long-term prognosis of patients following curative resection varies significantly. The 5-year recurrence rate reaches levels as high as 70% . Early recurrence and intrahepatic metastasis (within 2 years of resection) contribute to frequent relapses and high mortality rates in HCC . Currently, therapeutic approaches for recurrent HCC (rHCC) largely mirror those used for initial HCC . Curative treatment modalities for early-stage rHCC include repeated hepatic resection (RHR), thermal ablation (TA), and salvage liver transplantation (SLT). While SLT proves to be the most advantageous option, its utilization is limited in regions with a scarcity of liver donors, such as China. Therefore, RHR and TA are the primary first-line treatments for rHCC. The selection of the appropriate treatment is influenced by intra-abdominal adhesions, residual liver function, and the size, location, and number of recurrent tumors. Unfortunately, only 20% of patients with recurrence can benefit from repeated resection due to these factors . TA has advantages over RHR in terms of lower mortality and complication rates. Some studies have suggested that TA is as effective as RHR in achieving long-term survival outcomes for rHCC . However, other studies have indicated that TA may yield inferior outcomes compared to RHR in patients with these characteristics, such as tumor size exceeding 3 cm and alpha-fetoprotein levels above 200 mg/L . Additionally, previous research found that RHR was associated with superior overall survival (OS) compared to radiofrequency ablation (RFA) . Conversely, other studies have indicated that RHR was linked to prolonged recurrence-free survival but did not show a significant difference in OS compared to RFA . Diverse viewpoints exist regarding the comparative effectiveness of RHR and TA in the management of rHCC. In this study, we compared long-term survival outcomes between TA and RHR for treating rHCC within the Milan criteria. Moreover, to our knowledge, there is no comprehensive assessment of selection bias using various propensity score methods. Therefore, we adopted the propensity score matching (PSM) and inverse probability of treatment weighting (IPTW) to reduce potential selection bias. Patients and study design This retrospective multicenter study was approved by the institutional review board of all the participating hospitals. The necessity for written informed consent for publication was waived due to the retrospective nature of the study. Patients who underwent hepatectomy were enrolled between January 2015 and August 2023 from two centers: Center A (The First Affiliated Hospital, Zhejiang University School of Medicine) in southern China, and Center B (Peking University Cancer Hospital) in northern China. The inclusion criteria were: (1) hepatectomy as initial curative treatment; (2) confirmation of HCC diagnosed through biopsy or pathologic histology; (3) rHCC within the Milan criteria: either a solitary tumor with a diameter < 5 cm or 2–3 tumors each with a maximum diameter < 3 cm. The exclusion criteria were: (1) clinical diagnosis of rHCC subsequent to initial resection, followed by other treatments for rHCC prior to RHR or TA; (2) occurrence of HCC recurrence within 1 month of the initial hepatic resection; (3) presence of macrovascular invasion or extrahepatic metastasis before treatments; (4) patients with a previous liver transplant or kidney-liver transplant; (5) severe hepatic and renal dysfunction or other substantial organ disorders; (6) a follow-up duration less than 6 months. Finally, a total of 473 patients with rHCC were enrolled, with 340 in the TA group and 133 in RHR group. Details of patient participation are shown in Fig. . Covariates Baseline covariates were gathered from the electronic health record systems of both institutions. General clinical characteristics included sex, age, liver cirrhosis, etiology, and Child-Turcotte-Pugh grade. Laboratory indicators within 2 weeks included aspartate aminotransferase (AST, cut-off value: 45 U/L), alanine aminotransferase (ALT, cut-off value: 40 U/L), alkaline phosphatase (ALP, cut-off value: 160 U/L), total bilirubin (TBIL, cut-off value: 17.1 μmol/L), platelet (PLT, cut-off value: 100 × 10 9 /L), international normalized ratio (INR, cut-off value: 1.15), alpha-fetoprotein (AFP, cut-off value: 200 ng/mL), systemic immune-inflammation index (SII, cut-off value: 119.72), neutrophil–lymphocyte ratio (NLR, cut-off value: 3.83), lymphocyte-to-monocyte ratio (LMR, cut-off value: 2.55), and prognostic nutritional index (PNI, cut-off value: 40.33). SII was calculated using the standard formula (Platelet count × Neutrophil count / Lymphocyte count). PNI was Albumin + 5 × Lymphocyte count. Imaging variables, including tumor location, tumor size, tumor number, portal hypertension, subcapsular lesions, and perivascular lesions, were collected from the picture archiving and communication system. Treatment procedure and follow-up At each institution, all patients diagnosed with intrahepatic rHCC underwent R0 liver resection, characterized by the complete removal of all macroscopically detectable tumors with histologically tumor-free margins along the parenchymal transection line. Percutaneous ablation was performed by two radiologists with more than 10 years of experience in TA for HCC nodules under real-time ultrasound guidance, as previously reported . The Cool-tip RFA system (Covidien, USA) was utilized for ablation, with an active tip of either 2 or 3 cm. Power settings ranged from 150 to 200 W, with durations of 5 to 30 min per nodule. The objective was to form an ablative margin of at least 0.5 cm in the surrounding normal tissue. Microwave ablation (MWA) was performed using the KY-2000 MWA equipment (Kangyou Medical, China), with power levels between 50 and 150 W for durations of 4 to 20 min per nodule. Postoperatively, patients underwent imaging using dynamic contrast-enhanced ultrasound, multiphase contrast-enhanced CT, or MRI after one month to assess complete ablation. Complete ablation was defined as an absence of contrast enhancement in an area equal to or larger than the ablated tumor one month post-TA. Patients were subsequently monitored every 3 months for the first year, followed by intervals of 3 to 6 months thereafter. In case of tumor recurrence, the treatment approach was determined based on multidisciplinary discussions, including repeated ablation, transcatheter arterial chemoembolization, and systemic therapies like targeted therapy and immunotherapy. The final follow-up dates were July 15, 2023 (Center A) and January 25, 2024 (Centers B). Outcomes The primary study endpoint was recurrence-free survival (RFS), and the secondary endpoints were local tumor progression (LTP) and post-recurrence survival (PRS). LTP was defined as tumor recurrence along the margin of the ablation zone. RFS was defined as the duration between secondary TA or RHR and tumor progression (including LTP, intrahepatic distance recurrence, or extrahepatic metastasis) or until the last follow-up . PRS was defined as the period from the first diagnosis of recurrence to death or last follow-up . Statistical analysis We utilized R software (version 4.2.2; https://www.rproject.org ) for data analysis. To reduce potential confounding bias and obtains effects similar to those of randomized controlled trials, we adopted the PSM and IPTW methods . For PSM, we utilized the "MatchIt" package, and for IPTW, we used the "WeightIt" package. Propensity scores were estimated through multivariable logistic regression, with dependent variables including sex, age, etiology, liver cirrhosis, CTP grade, location, tumor size, tumor number, portal hypertension, ascites, subcapsular lesion, perivascular lesion, AST, ALT, ALP, TBIL, PLT, INR, AFP, SII, NLR, LMR, and PNI. PSM was conducted using a 1:1 nearest neighbor algorithm with a caliper of 0.02. To corroborate our findings, we repeated all statistical analyses following IPTW, considering the same potential confounders as those used for PSM. IPTW was conducted using weighted with inverse propensity scores and inverse of 1 minus the propensity scores. Standardized mean differences (SMD) were computed for all confounders to assess covariate balance between the TA and RHR groups. An SMD of less than 0.300 indicated satisfactory covariate balance between the two groups. For categorical variables, comparisons between the two groups were performed using either the chi-square test or Fisher’s exact test. Survival curves were generated using Kaplan–Meier analysis and compared using the log-rank test. Univariable and multivariable Cox proportional hazards regression analyses were conducted to identify clinical factors associated with RFS and PRS. All clinical variables in univariable analysis were subsequently subjected to stepwise multivariable analysis using Akaike information criterion to determine their inclusion in multivariable analysis . The optimal cutoff values of the SII, NLR, LMR, and PNI were determined by maximizing the Youden index based on receiver operating characteristic (ROC) curves. The results provided P-values, hazard ratios (HR), and 95% confidence intervals (CIs). P < 0.05 was considered statistically significant. This retrospective multicenter study was approved by the institutional review board of all the participating hospitals. The necessity for written informed consent for publication was waived due to the retrospective nature of the study. Patients who underwent hepatectomy were enrolled between January 2015 and August 2023 from two centers: Center A (The First Affiliated Hospital, Zhejiang University School of Medicine) in southern China, and Center B (Peking University Cancer Hospital) in northern China. The inclusion criteria were: (1) hepatectomy as initial curative treatment; (2) confirmation of HCC diagnosed through biopsy or pathologic histology; (3) rHCC within the Milan criteria: either a solitary tumor with a diameter < 5 cm or 2–3 tumors each with a maximum diameter < 3 cm. The exclusion criteria were: (1) clinical diagnosis of rHCC subsequent to initial resection, followed by other treatments for rHCC prior to RHR or TA; (2) occurrence of HCC recurrence within 1 month of the initial hepatic resection; (3) presence of macrovascular invasion or extrahepatic metastasis before treatments; (4) patients with a previous liver transplant or kidney-liver transplant; (5) severe hepatic and renal dysfunction or other substantial organ disorders; (6) a follow-up duration less than 6 months. Finally, a total of 473 patients with rHCC were enrolled, with 340 in the TA group and 133 in RHR group. Details of patient participation are shown in Fig. . Baseline covariates were gathered from the electronic health record systems of both institutions. General clinical characteristics included sex, age, liver cirrhosis, etiology, and Child-Turcotte-Pugh grade. Laboratory indicators within 2 weeks included aspartate aminotransferase (AST, cut-off value: 45 U/L), alanine aminotransferase (ALT, cut-off value: 40 U/L), alkaline phosphatase (ALP, cut-off value: 160 U/L), total bilirubin (TBIL, cut-off value: 17.1 μmol/L), platelet (PLT, cut-off value: 100 × 10 9 /L), international normalized ratio (INR, cut-off value: 1.15), alpha-fetoprotein (AFP, cut-off value: 200 ng/mL), systemic immune-inflammation index (SII, cut-off value: 119.72), neutrophil–lymphocyte ratio (NLR, cut-off value: 3.83), lymphocyte-to-monocyte ratio (LMR, cut-off value: 2.55), and prognostic nutritional index (PNI, cut-off value: 40.33). SII was calculated using the standard formula (Platelet count × Neutrophil count / Lymphocyte count). PNI was Albumin + 5 × Lymphocyte count. Imaging variables, including tumor location, tumor size, tumor number, portal hypertension, subcapsular lesions, and perivascular lesions, were collected from the picture archiving and communication system. At each institution, all patients diagnosed with intrahepatic rHCC underwent R0 liver resection, characterized by the complete removal of all macroscopically detectable tumors with histologically tumor-free margins along the parenchymal transection line. Percutaneous ablation was performed by two radiologists with more than 10 years of experience in TA for HCC nodules under real-time ultrasound guidance, as previously reported . The Cool-tip RFA system (Covidien, USA) was utilized for ablation, with an active tip of either 2 or 3 cm. Power settings ranged from 150 to 200 W, with durations of 5 to 30 min per nodule. The objective was to form an ablative margin of at least 0.5 cm in the surrounding normal tissue. Microwave ablation (MWA) was performed using the KY-2000 MWA equipment (Kangyou Medical, China), with power levels between 50 and 150 W for durations of 4 to 20 min per nodule. Postoperatively, patients underwent imaging using dynamic contrast-enhanced ultrasound, multiphase contrast-enhanced CT, or MRI after one month to assess complete ablation. Complete ablation was defined as an absence of contrast enhancement in an area equal to or larger than the ablated tumor one month post-TA. Patients were subsequently monitored every 3 months for the first year, followed by intervals of 3 to 6 months thereafter. In case of tumor recurrence, the treatment approach was determined based on multidisciplinary discussions, including repeated ablation, transcatheter arterial chemoembolization, and systemic therapies like targeted therapy and immunotherapy. The final follow-up dates were July 15, 2023 (Center A) and January 25, 2024 (Centers B). The primary study endpoint was recurrence-free survival (RFS), and the secondary endpoints were local tumor progression (LTP) and post-recurrence survival (PRS). LTP was defined as tumor recurrence along the margin of the ablation zone. RFS was defined as the duration between secondary TA or RHR and tumor progression (including LTP, intrahepatic distance recurrence, or extrahepatic metastasis) or until the last follow-up . PRS was defined as the period from the first diagnosis of recurrence to death or last follow-up . We utilized R software (version 4.2.2; https://www.rproject.org ) for data analysis. To reduce potential confounding bias and obtains effects similar to those of randomized controlled trials, we adopted the PSM and IPTW methods . For PSM, we utilized the "MatchIt" package, and for IPTW, we used the "WeightIt" package. Propensity scores were estimated through multivariable logistic regression, with dependent variables including sex, age, etiology, liver cirrhosis, CTP grade, location, tumor size, tumor number, portal hypertension, ascites, subcapsular lesion, perivascular lesion, AST, ALT, ALP, TBIL, PLT, INR, AFP, SII, NLR, LMR, and PNI. PSM was conducted using a 1:1 nearest neighbor algorithm with a caliper of 0.02. To corroborate our findings, we repeated all statistical analyses following IPTW, considering the same potential confounders as those used for PSM. IPTW was conducted using weighted with inverse propensity scores and inverse of 1 minus the propensity scores. Standardized mean differences (SMD) were computed for all confounders to assess covariate balance between the TA and RHR groups. An SMD of less than 0.300 indicated satisfactory covariate balance between the two groups. For categorical variables, comparisons between the two groups were performed using either the chi-square test or Fisher’s exact test. Survival curves were generated using Kaplan–Meier analysis and compared using the log-rank test. Univariable and multivariable Cox proportional hazards regression analyses were conducted to identify clinical factors associated with RFS and PRS. All clinical variables in univariable analysis were subsequently subjected to stepwise multivariable analysis using Akaike information criterion to determine their inclusion in multivariable analysis . The optimal cutoff values of the SII, NLR, LMR, and PNI were determined by maximizing the Youden index based on receiver operating characteristic (ROC) curves. The results provided P-values, hazard ratios (HR), and 95% confidence intervals (CIs). P < 0.05 was considered statistically significant. Baseline characteristics In this study, we enrolled 473 patients from two Chinese medical centers, with 340 in the TA group and 133 in the RHR group. Among these, 214 patients were selected for 1:1 PSM, resulting in 107 patients in each group. IPTW generated 340 and 133 patients in the two groups, respectively. The clinical baseline characteristics are presented in Table . In the entire cohort, there were no significant differences in sex, age, etiology, liver cirrhosis, CTP grade, location, tumor number, portal hypertension, ascites, perivascular lesions, ALP, TBIL, PLT, INR, and AFP ( P = 0.084–1.000). However, significant differences were observed in tumor size ( P < 0.001), subcapsular lesions ( P = 0.003), AST ( P = 0.019), ALT ( P = 0.008), SII ( P = 0.006), LMR ( P = 0.025) and PNI ( P < 0.001) between the TA and RHR groups. After PSM and IPTW, the baseline characteristics of the two groups were better balanced, as indicated by all SMD being less than 0.300 (Table ). Analysis of risk factors for RFS and PRS The results of univariable and multivariable Cox proportional hazards regression analyses for RFS and PRS are summarized in Table , S1, and S2. Among all patients, multiple tumors (HR: 2.28; 95% CI: 1.69–3.07; P < 0.001) and PLT > 100 × 10 9 /L (HR: 0.73; 95% CI: 0.54–0.98; P = 0.038) were independent prognostic factors for RFS. For PRS, tumor location in right lobe (HR: 0.55; 95% CI: 0.37–0.83; P = 0.004), 3–5 cm in tumor size (HR: 1.95; 95% CI: 1.21–3.15; P = 0.006), multiple tumors (HR: 1.67; 95% CI: 1.06–2.61; P = 0.026), ascites (HR: 2.18; 95% CI: 1.15–4.14; P = 0.017), AST > 45 U/L (HR: 1.72; 95% CI: 1.02–2.90; P = 0.042), and AFP ≥ 200 ng/mL (HR: 1.69; 95% CI: 1.01–2.84; P = 0.046) were independent prognostic factors. Additionally, both univariable and multivariable analyses revealed that the treatment method (TA vs. RHR) was not an independent prognostic factor for RFS and PRS, regardless of whether PSM or IPTW was applied. Analysis of LTP, RFS, and PRS In the entire rHCC cohort, 47 out of 340 patients (13.8%) in the TA group and 16 out of 133 (12.0%) in the RHR group experienced LTP. The 1-, 3-, and 5-year cumulative LTP rates were 7.4%, 14.7%, and 17.5%, respectively, in the TA group, and 6.5%, 15.4%, and 29.5%, respectively, in the RHR group. LTP rates were not significantly different between the two groups (HR = 0.98, 95% CI = 0.55–1.73, P = 0.940; Fig. a). In terms of RFS, 182 patients (53.5%) in the TA group and 70 patients (52.6%) in the RHR group experienced it. The 1-, 3-, and 5-year RFS rates were 68.8%, 45.4%, and 38.1% in the TA group, compared to 69.8%, 39.8%, and 9.6% in the RHR group. The differences in RFS rates were not statistically significant (HR = 1.21, 95% CI = 0.92–1.60, P = 0.180; Fig. b). Regarding PRS, 89 patients (26.2%) in the TA group and 26 (19.5%) in the RHR group had died. The PRS rates at 1, 3, and 5 years were 92.2%, 76.5%, and 67.1% for the TA group, and 96.8%, 80.7%, and 63.6% for the RHR group. The PRS rates also showed no significant difference between the groups (HR = 0.92, 95% CI = 0.59–1.43, P = 0.700; Fig. c). Analysis of LTP, RFS, and PRS after PSM Kaplan–Meier curves for LTP, RFS, and PRS were evaluated by PSM. After PSM, the cumulative LTP rates were 16.8% (18/107) for the TA group and 12.1% (13/107) for the RHR group. The 1-, 3-, and 5-year cumulative LTP rates were 12.6%, 12.6% and 12.6% in the TA group and 5.9%, 16.4%, 19.2% in the RHR group, respectively (HR = 0.74, 95% CI = 0.36–1.52, P = 0.420; Fig. a). The median RFS rates for the TA (22.7 months) and RHR (22.6 months) groups were closely monitored. The RFS rates were found to be 63.6% (68/107) in the TA group and 56.1% (60/107) in the RHR group. At 1-, 3-, and 5- year, the TA group exhibited RFS rates of 56.5%, 36.9%, and 29.2%, respectively, whereas the RHR group had rates of 68.4%, 37.9%, and 16.1% (HR = 0.93, 95% CI = 0.65–1.32, P = 0.680; Fig. b). PRS rates were evaluated, revealing 28.0% (30/107) in the TA group and 23.4% (25/107) in the RHR group. The PRS rates at 1-, 3-, and 5-year for the TA group were 89.7%, 70.9%, and 67.0%, respectively, in contrast to the RHR group, which had rates of 96.1%, 77.5%, and 65.0% (HR = 0.94, 95% CI = 0.55–1.60, P = 0.810; Fig. c). Analysis of LTP, RFS, and PRS after IPTW To validate findings of PSM, Kaplan–Meier curves for LTP, RFS, and PRS were also evaluated using IPTW. The cumulative LTP rates post-IPTW were 17.5% (59.5/340.0) in the TA group and 16.0% (12.0/133.0) in the RHR group. Specifically, the 1-, 3-, and 5-year LTP rates for the TA group were 7.7%, 19.5%, and 26.7%, respectively, compared to 6.5%, 15.4%, and 29.5% in the RHR group (HR = 0.73, 95% CI = 0.34–1.58, P = 0.940; Fig. d). In terms of RFS, the TA group had a median RFS of 19.2 months, while the RHR group's median RFS was 24.1 months. The RFS rates were observed to be 61.0% (207.5/340.0) in the TA group and 52.6% (70.0/133.0) in the RHR group. The 1-, 3-, and 5-year RFS rates were 59.5%, 35.7%, and 31.3% for the TA group, compared to 69.8%, 39.8%, and 9.6% for the RHR group (HR = 0.89, 95% CI = 0.61–1.29, P = 0.180; Fig. e). Evaluating PRS, the TA group had 31.7% (107.7/340.0) of patients, while the RHR group had 19.5% (26.0/133) who died. The 1-, 3-, and 5-year PRS rates were 83.7%, 68.2%, and 64.8% for the TA group, and 96.8%, 80.7%, and 63.6% for the RHR group (HR = 0.63, 95% CI = 0.35–1.14, P = 0.700; Fig. f). Subgroup analysis Tumor size, tumor number, subcapsular lesions, and perivascular lesions not only affect the difficulty of surgical resection or ablation but also determine the prognosis of patients. Therefore, this subgroup analysis was stratified based on these factors to observe their influence on LTP, RFS, and PRS in both groups (Fig. and S1-2). After PSM, no statistically significant differences in RFS were observed for the subgroups with tumor sizes < 3 cm ( P = 0.940) or ≥ 3 cm ( P = 0.380) and solitary tumor ( P = 0.630) or multiple tumors ( P = 0.110) (Fig. b). Similarly, no statistically significant differences were found for the subgroups after IPTW ( P = 0.320 and P = 0.120; P = 0.890 and P = 0.630; Fig. c). RFS did not show statistically significant differences between the subgroups without and with perivascular lesions after PSM ( P = 0.780 and P = 0.055, respectively) and IPTW ( P = 0.052 and P = 0.130, respectively; Fig. b and c) There were no statistically significant differences in RFS for the subgroups without and with subcapsular lesions after PSM ( P = 0.100 and P = 0.220, respectively; Fig. b). However, a statistically significant difference in RFS was observed for the subcapsular subgroup after IPTW ( P = 0.008; Fig. c). Figures S1 and S2 display the analyses of tumor size, tumor number, subcapsular, and perivascular subgroups for LTP and PRS, revealing no statistically significant differences between the two groups. In this study, we enrolled 473 patients from two Chinese medical centers, with 340 in the TA group and 133 in the RHR group. Among these, 214 patients were selected for 1:1 PSM, resulting in 107 patients in each group. IPTW generated 340 and 133 patients in the two groups, respectively. The clinical baseline characteristics are presented in Table . In the entire cohort, there were no significant differences in sex, age, etiology, liver cirrhosis, CTP grade, location, tumor number, portal hypertension, ascites, perivascular lesions, ALP, TBIL, PLT, INR, and AFP ( P = 0.084–1.000). However, significant differences were observed in tumor size ( P < 0.001), subcapsular lesions ( P = 0.003), AST ( P = 0.019), ALT ( P = 0.008), SII ( P = 0.006), LMR ( P = 0.025) and PNI ( P < 0.001) between the TA and RHR groups. After PSM and IPTW, the baseline characteristics of the two groups were better balanced, as indicated by all SMD being less than 0.300 (Table ). The results of univariable and multivariable Cox proportional hazards regression analyses for RFS and PRS are summarized in Table , S1, and S2. Among all patients, multiple tumors (HR: 2.28; 95% CI: 1.69–3.07; P < 0.001) and PLT > 100 × 10 9 /L (HR: 0.73; 95% CI: 0.54–0.98; P = 0.038) were independent prognostic factors for RFS. For PRS, tumor location in right lobe (HR: 0.55; 95% CI: 0.37–0.83; P = 0.004), 3–5 cm in tumor size (HR: 1.95; 95% CI: 1.21–3.15; P = 0.006), multiple tumors (HR: 1.67; 95% CI: 1.06–2.61; P = 0.026), ascites (HR: 2.18; 95% CI: 1.15–4.14; P = 0.017), AST > 45 U/L (HR: 1.72; 95% CI: 1.02–2.90; P = 0.042), and AFP ≥ 200 ng/mL (HR: 1.69; 95% CI: 1.01–2.84; P = 0.046) were independent prognostic factors. Additionally, both univariable and multivariable analyses revealed that the treatment method (TA vs. RHR) was not an independent prognostic factor for RFS and PRS, regardless of whether PSM or IPTW was applied. In the entire rHCC cohort, 47 out of 340 patients (13.8%) in the TA group and 16 out of 133 (12.0%) in the RHR group experienced LTP. The 1-, 3-, and 5-year cumulative LTP rates were 7.4%, 14.7%, and 17.5%, respectively, in the TA group, and 6.5%, 15.4%, and 29.5%, respectively, in the RHR group. LTP rates were not significantly different between the two groups (HR = 0.98, 95% CI = 0.55–1.73, P = 0.940; Fig. a). In terms of RFS, 182 patients (53.5%) in the TA group and 70 patients (52.6%) in the RHR group experienced it. The 1-, 3-, and 5-year RFS rates were 68.8%, 45.4%, and 38.1% in the TA group, compared to 69.8%, 39.8%, and 9.6% in the RHR group. The differences in RFS rates were not statistically significant (HR = 1.21, 95% CI = 0.92–1.60, P = 0.180; Fig. b). Regarding PRS, 89 patients (26.2%) in the TA group and 26 (19.5%) in the RHR group had died. The PRS rates at 1, 3, and 5 years were 92.2%, 76.5%, and 67.1% for the TA group, and 96.8%, 80.7%, and 63.6% for the RHR group. The PRS rates also showed no significant difference between the groups (HR = 0.92, 95% CI = 0.59–1.43, P = 0.700; Fig. c). Kaplan–Meier curves for LTP, RFS, and PRS were evaluated by PSM. After PSM, the cumulative LTP rates were 16.8% (18/107) for the TA group and 12.1% (13/107) for the RHR group. The 1-, 3-, and 5-year cumulative LTP rates were 12.6%, 12.6% and 12.6% in the TA group and 5.9%, 16.4%, 19.2% in the RHR group, respectively (HR = 0.74, 95% CI = 0.36–1.52, P = 0.420; Fig. a). The median RFS rates for the TA (22.7 months) and RHR (22.6 months) groups were closely monitored. The RFS rates were found to be 63.6% (68/107) in the TA group and 56.1% (60/107) in the RHR group. At 1-, 3-, and 5- year, the TA group exhibited RFS rates of 56.5%, 36.9%, and 29.2%, respectively, whereas the RHR group had rates of 68.4%, 37.9%, and 16.1% (HR = 0.93, 95% CI = 0.65–1.32, P = 0.680; Fig. b). PRS rates were evaluated, revealing 28.0% (30/107) in the TA group and 23.4% (25/107) in the RHR group. The PRS rates at 1-, 3-, and 5-year for the TA group were 89.7%, 70.9%, and 67.0%, respectively, in contrast to the RHR group, which had rates of 96.1%, 77.5%, and 65.0% (HR = 0.94, 95% CI = 0.55–1.60, P = 0.810; Fig. c). To validate findings of PSM, Kaplan–Meier curves for LTP, RFS, and PRS were also evaluated using IPTW. The cumulative LTP rates post-IPTW were 17.5% (59.5/340.0) in the TA group and 16.0% (12.0/133.0) in the RHR group. Specifically, the 1-, 3-, and 5-year LTP rates for the TA group were 7.7%, 19.5%, and 26.7%, respectively, compared to 6.5%, 15.4%, and 29.5% in the RHR group (HR = 0.73, 95% CI = 0.34–1.58, P = 0.940; Fig. d). In terms of RFS, the TA group had a median RFS of 19.2 months, while the RHR group's median RFS was 24.1 months. The RFS rates were observed to be 61.0% (207.5/340.0) in the TA group and 52.6% (70.0/133.0) in the RHR group. The 1-, 3-, and 5-year RFS rates were 59.5%, 35.7%, and 31.3% for the TA group, compared to 69.8%, 39.8%, and 9.6% for the RHR group (HR = 0.89, 95% CI = 0.61–1.29, P = 0.180; Fig. e). Evaluating PRS, the TA group had 31.7% (107.7/340.0) of patients, while the RHR group had 19.5% (26.0/133) who died. The 1-, 3-, and 5-year PRS rates were 83.7%, 68.2%, and 64.8% for the TA group, and 96.8%, 80.7%, and 63.6% for the RHR group (HR = 0.63, 95% CI = 0.35–1.14, P = 0.700; Fig. f). Tumor size, tumor number, subcapsular lesions, and perivascular lesions not only affect the difficulty of surgical resection or ablation but also determine the prognosis of patients. Therefore, this subgroup analysis was stratified based on these factors to observe their influence on LTP, RFS, and PRS in both groups (Fig. and S1-2). After PSM, no statistically significant differences in RFS were observed for the subgroups with tumor sizes < 3 cm ( P = 0.940) or ≥ 3 cm ( P = 0.380) and solitary tumor ( P = 0.630) or multiple tumors ( P = 0.110) (Fig. b). Similarly, no statistically significant differences were found for the subgroups after IPTW ( P = 0.320 and P = 0.120; P = 0.890 and P = 0.630; Fig. c). RFS did not show statistically significant differences between the subgroups without and with perivascular lesions after PSM ( P = 0.780 and P = 0.055, respectively) and IPTW ( P = 0.052 and P = 0.130, respectively; Fig. b and c) There were no statistically significant differences in RFS for the subgroups without and with subcapsular lesions after PSM ( P = 0.100 and P = 0.220, respectively; Fig. b). However, a statistically significant difference in RFS was observed for the subcapsular subgroup after IPTW ( P = 0.008; Fig. c). Figures S1 and S2 display the analyses of tumor size, tumor number, subcapsular, and perivascular subgroups for LTP and PRS, revealing no statistically significant differences between the two groups. RHR may be the optimal first-line treatment for rHCC within the Milan criteria. However, it is only feasible in a small proportion of patients, particularly those with smaller tumor size, solitary tumors, and better residual liver function . Aggressive local treatments may improve the prognosis of patients with intrahepatic rHCC after surgery and should therefore be attempted. In this study, we analyzed long-term survival outcomes of TA and RHR in 473 rHCC patients within the Milan criteria, revealing no statistically significant difference in 1-, 3-, and 5-year LTP, RFS, and PRS through PSM and IPTW analyses. Subgroup analysis also showed no significant differences in LTP, RFS, and PRS between the two treatments among patients with tumors smaller or larger than 3 cm, solitary or multiple, with or without subcapsular, and with or without perivascular involvement after PSM. However, a significant difference in RFS was observed among patients with subcapsular lesions after IPTW. Using both PSM and IPTW, we reduced the SMD for all potential confounding factors to less than 0.300, effectively adjusting for these factors. PSM is regarded as more robust and less biased compared to other propensity score methods . However, a notable disadvantage of PSM is the reduction in sample size, especially in datasets with highly imbalanced groups. To address this limitation, our study also employed IPTW, which helps maintain a larger sample size while still effectively controlling for confounding factors. By using PSM and IPTW, we ensured a comprehensive adjustment for confounding factors, leveraging the strengths of both methods to enhance the reliability of our findings. Therefore, the results of our study, which balanced patient demographics, liver function reserves, and tumor characteristics between the RHR and TA groups, provide important data for clinical decision-making in rHCC after initial hepatic resection. In this study, LTP was observed in 13.3% (63 out of 473) of patients with rHCC: 13.8% (47 out of 340) in the TA group and 12.0% (16 out of 133) in the RHR group. These results align with previously reported LTP rates of 8.4–14.8% . The cumulative LTP rates showed no significant difference between the TA and RHR groups ( P = 0.420–0.940), both before and after applying PSM and IPTW. This lack of significant difference can be attributed to the relatively small tumor size and complete thermal ablation in our study. Additionally, the median RFS was nearly the same between the TA and RHR groups, both before and after PSM. Although there was a tendency toward a longer RFS in RHR group, this difference remained statistically insignificant after applying IPTW. This tendency aligns with the findings of Zhong et al. , where RFS for rHCC was significantly longer in the RHR group compared to the RFA group, with a statistically significant difference. This discrepancy can be explained by the different methods used to reduce confounding factors or the smaller sample size in our study. Similarly, the RHR group had a longer PRS compared to the TA group before and after applying PSM and IPTW, but this difference was not statistically significant. This finding is consistent with previous studies . We analyzed prognostic variables with rHCC and found that long-term RFS and PRS results were associated with ascites, tumor number, size, location, AST, PLT, AFP, and NLR. Liver function damage emerged as a significant indicator of poor prognosis. Ascites and elevated transaminase levels, representing liver damage, suggest high tumor invasiveness and poor liver function, leading to a worse prognosis . A recent meta-analysis confirmed that PLT count is a prognostic marker in HCC, particularly with a PLT count cutoff of < 100 × 10 9 /L . Liu et al. further demonstrated that intra-platelet 5-HT might affect progression and prognosis of HCC by regulating Yes-associated protein expression. Tang et al. identified that tumors located in the left lobe were linked to OS and RFS after hepatectomy for HCC. Additionally, previous studies have reported that systemic inflammation biomarkers can evaluate HCC prognosis, finding that higher NLR levels and lower LMR levels were associated with worse survival outcomes . Therefore, our prognostic analysis is consistent with previous reports and confirms that the underlying biological features of HCC are the main determinants of survival. In this study, we compared two groups of patients with rHCC: those who underwent TA and those who underwent RHR. The subgroup analysis was based on tumor size, tumor number, subcapsular lesions, and perivascular lesions. Of the tumors, 35.1% (166/473) were located subcapsularly, and 16.7% (79/473) were close to large central vessels. Our study demonstrated that both TA and RHR are effective treatments for early-stage rHCC (within the Milan criteria) in terms of LTP, RFS, and PRS, regardless of whether the tumors were larger (3–5 cm), multiple (≤ 3), subcapsular or perivascular. Liu et al. indicated that MWA showed comparable LTP, OS, and disease-free survival to surgical resection for subcapsular HCC meeting the Milan criteria. Lee et al. compared long-term outcomes between surgical resection and RFA as first-line treatments, finding both to be viable options for perivascular small HCC (≤ 3 cm). However, few studies have reported on the long-term prognosis of subcapsular or perivascular rHCC treated with TA and RHR. Our results provide additional evidence in this area. Our study had several limitations. Firstly, we used a multicenter retrospective approach. In the future, we will employ large-sample, prospective randomized controlled trials to validate these results. Secondly, despite performing PSM and IPTW to enhance intergroup comparison, there were still unavoidable and unidentified confounding factors and selection bias. In conclusion, the cumulative LTP, RFS, and PRS were not significantly different between TA and RHR for rHCC patients within the Milan criteria. TA showed comparable therapeutic outcomes to RHR for rHCC within the Milan criteria, and may be a viable curative option for early-stage rHCC patients. Supplementary Material 1. |
Wound closure techniques after wide excision for hidradenitis suppurativa: a systematic review and meta‐analysis | 691dc3e8-7b93-440e-b6ea-9b446e4a5937 | 11931091 | Surgical Procedures, Operative[mh] | HS is a chronic inflammatory skin condition that affects the intertriginous areas bearing apocrine glands. The obstruction and rupture of the folliculopilosebaceous unit (FPSU) generates a consistent, chronic immune response. Apocrine glands cannot excrete their secretion product because of ductal keratinocyte proliferation, so they dilate and rupture. The created anoxic environment stimulates bacterial overgrowth, which maintains the inflammatory process, leading to recurrent nodules, sinus tracts, and severe scars. The pain, chronic drainage, and scars have a significant psychological impact and affect the patients' quality of life. Mechanical stress on the skin, obesity, smoking, androgen excess, and genetic predisposition are associated with disease development. When mild to moderate disease is not efficiently controlled by medical therapy, surgical techniques such as punch debridement, unroofing, deroofing with scissors, electrosurgery, and CO 2 laser excision have been proposed, with variable results. In patients with severe disease, chronic inflammation leads to irreversible tissue damage, resulting in tunnel development and scarring (Hurley stage II/III); systemic treatments provide inadequate relief, while wide surgical excision results in long disease‐free periods. Wide excision is defined as the removal of the entire affected dermis reaching the subcutaneous tissue, with the intent of leaving a border of unaffected tissue. Recently, the consensus of Bui et al. on the surgical procedure's definitions for hidradenitis suppurativa (HS) provided standardized definitions for accurate communication. Based on this consensus, the authors suggested replacing the term “wide excision” with “complete regional excision”. Since we used the term “wide excision” to conduct the literature research in our systematic review and meta‐analysis, we maintain the same term in the present paper. Several options are available for closing the wounds (Table ), but currently, there is no international consensus regarding the most effective one. We conducted a systematic review and meta‐analysis of the existing literature to evaluate the most effective wound closure technique after wide excision for HS. We compared the wound closure methods in terms of recurrence, outcome, and the patient's quality of life. The objective is to determine which method offers the best functionality outcome and lowest recurrence rates. Although lesion recurrence is a primary outcome measure, several studies emphasize the difficulty of providing a standardized definition.
Systematic literature search A literature search was conducted according to the Preferred Reporting Items for Systematic Reviews and Meta‐Analysis Protocols (PRISMA 2020) statement. PubMed (MEDLINE) and Embase databases (Scopus) were systematically searched for original studies concerning the methods of wound closure in wide excision for HS published from 1965 to 2024. Cochrane Register, https://www.clinicaltrials.gov/ , and the first 10 pages of Google Academic were also reviewed. The literature was last consulted on February 23, 2024. The reference lists of identified articles were also consulted for further data. The search strategy included the following combination of terms and their synonyms: “hidradenitis suppurativa”, “wide excision” and methods of wound closure (“primary closure”, “secondary closure”, “split skin graft”, “flap”, “secondary intention healing”, “reconstruction”). Where available, MeSH terms were included (File ). The literature search was restricted to articles written in English; here, the automation tools from PubMed and Scopus were used. Original studies with available full texts were included. Case reports, reviews, and papers concerning closure techniques for different types of surgical intervention other than wide excision were excluded from the analysis. Data collection From each eligible paper, data for the following features were extracted: author, country of origin, year of study, study design, number of patients by sex, Hurley stage, the average duration of the lesion before surgery, number of body sites treated with surgery, type of surgical closure, recurrence rate, postoperative complications, Dermatology Life Quality Index (DLQI), pain visual analog scale (VAS) and follow‐up period. Not all included articles contained information for each of the items mentioned above. Quality assessment The Newcastle‐Ottawa Scale (NOS) was used as a quality assessment form for cohort studies. Two authors (CCI and CI) independently assigned a maximum of one star for each of the four items of the selection category (representativeness of the exposed cohort, i.e., ≥20 patients); selection of the non‐exposed cohort; ascertainment of exposure, that is, surgical records; a demonstration that outcome of interest was not present at study initiation, and of the three items of the Outcome category (assessment of outcome‐record linkage; long enough follow‐up for outcomes to occur, i.e., mean >24 months; adequacy of follow‐up of cohorts, i.e., complete follow‐up/follow‐up rate >70%), while the Comparability category (comparability of cohorts based on the design or analysis) was scored with a maximum of two stars. If the total score ranged between 0 and 2, the quality of the included study was poor; 3–5 meant fair quality, and 6–9 good/high quality. Outcome measures The primary outcome was the recurrence rate (number of events divided by the number of interventions). Secondary outcomes included the evaluation of postoperative complications (number of events divided by the number of interventions), specifically infection, wound dehiscence, bleeding, and movement impairments. Data analysis Two data‐analysis approaches have been used. For each primary and secondary outcome, we used multi‐intervention network meta‐analysis, which integrates data from both direct and indirect comparisons between wound closure techniques across all primary (recurrence rate) and secondary outcomes (infection, wound dehiscence, bleeding, and movement impairment rates), including the large number of studies that did not directly report on each specific comparison. We used the CINeMA web application (Confidence in Network Meta‐Analysis), whose methodology provides the contribution matrix reporting how much information each study contributes to the results from network meta‐analysis. Estimates are presented using odds ratios and their 95% confidence intervals. The primary outcome has been comparatively evaluated between wound closure techniques through direct head‐to‐head comparison meta‐analysis using Review Manager (RevMan), version 5.4.1 (The Nordic Cochrane Centre, The Cochrane Collaboration, 2009, Copenhagen, Denmark). Sufficient data (at least three independent studies) were unavailable to compare secondary outcome variables head‐to‐head. A Mantel–Haenszel random effect model was applied to define the odds ratios (ORs) illustrated using forest plots, with an indication of 95% confidence interval (CI) and a cumulative weighted mean effect for all the studies analyzed. The degree of heterogeneity between studies was measured using the inverse variance index ( I 2 ). The I 2 statistic [ I 2 25% was considered low heterogeneity, 50% moderate heterogeneity, and 75% high heterogeneity] and test Z for the overall effect of <0.05 was considered for heterogeneity significance between studies.
A literature search was conducted according to the Preferred Reporting Items for Systematic Reviews and Meta‐Analysis Protocols (PRISMA 2020) statement. PubMed (MEDLINE) and Embase databases (Scopus) were systematically searched for original studies concerning the methods of wound closure in wide excision for HS published from 1965 to 2024. Cochrane Register, https://www.clinicaltrials.gov/ , and the first 10 pages of Google Academic were also reviewed. The literature was last consulted on February 23, 2024. The reference lists of identified articles were also consulted for further data. The search strategy included the following combination of terms and their synonyms: “hidradenitis suppurativa”, “wide excision” and methods of wound closure (“primary closure”, “secondary closure”, “split skin graft”, “flap”, “secondary intention healing”, “reconstruction”). Where available, MeSH terms were included (File ). The literature search was restricted to articles written in English; here, the automation tools from PubMed and Scopus were used. Original studies with available full texts were included. Case reports, reviews, and papers concerning closure techniques for different types of surgical intervention other than wide excision were excluded from the analysis.
From each eligible paper, data for the following features were extracted: author, country of origin, year of study, study design, number of patients by sex, Hurley stage, the average duration of the lesion before surgery, number of body sites treated with surgery, type of surgical closure, recurrence rate, postoperative complications, Dermatology Life Quality Index (DLQI), pain visual analog scale (VAS) and follow‐up period. Not all included articles contained information for each of the items mentioned above.
The Newcastle‐Ottawa Scale (NOS) was used as a quality assessment form for cohort studies. Two authors (CCI and CI) independently assigned a maximum of one star for each of the four items of the selection category (representativeness of the exposed cohort, i.e., ≥20 patients); selection of the non‐exposed cohort; ascertainment of exposure, that is, surgical records; a demonstration that outcome of interest was not present at study initiation, and of the three items of the Outcome category (assessment of outcome‐record linkage; long enough follow‐up for outcomes to occur, i.e., mean >24 months; adequacy of follow‐up of cohorts, i.e., complete follow‐up/follow‐up rate >70%), while the Comparability category (comparability of cohorts based on the design or analysis) was scored with a maximum of two stars. If the total score ranged between 0 and 2, the quality of the included study was poor; 3–5 meant fair quality, and 6–9 good/high quality.
The primary outcome was the recurrence rate (number of events divided by the number of interventions). Secondary outcomes included the evaluation of postoperative complications (number of events divided by the number of interventions), specifically infection, wound dehiscence, bleeding, and movement impairments.
Two data‐analysis approaches have been used. For each primary and secondary outcome, we used multi‐intervention network meta‐analysis, which integrates data from both direct and indirect comparisons between wound closure techniques across all primary (recurrence rate) and secondary outcomes (infection, wound dehiscence, bleeding, and movement impairment rates), including the large number of studies that did not directly report on each specific comparison. We used the CINeMA web application (Confidence in Network Meta‐Analysis), whose methodology provides the contribution matrix reporting how much information each study contributes to the results from network meta‐analysis. Estimates are presented using odds ratios and their 95% confidence intervals. The primary outcome has been comparatively evaluated between wound closure techniques through direct head‐to‐head comparison meta‐analysis using Review Manager (RevMan), version 5.4.1 (The Nordic Cochrane Centre, The Cochrane Collaboration, 2009, Copenhagen, Denmark). Sufficient data (at least three independent studies) were unavailable to compare secondary outcome variables head‐to‐head. A Mantel–Haenszel random effect model was applied to define the odds ratios (ORs) illustrated using forest plots, with an indication of 95% confidence interval (CI) and a cumulative weighted mean effect for all the studies analyzed. The degree of heterogeneity between studies was measured using the inverse variance index ( I 2 ). The I 2 statistic [ I 2 25% was considered low heterogeneity, 50% moderate heterogeneity, and 75% high heterogeneity] and test Z for the overall effect of <0.05 was considered for heterogeneity significance between studies.
Study characteristics A total of 1181 records were identified from a systematic literature search. After removing duplicates and entries in a non‐English language, there were 655 unique articles to screen as per the eligibility assessment (Figure ). Double independent review (CCI and CIM) of the available evidence excluded publications that did not meet the inclusion criteria (single case reports, unrelated topics, systematic reviews). Overall, 121 full‐text articles were included in the analysis. Of these, six were prospective case series, cohort, pilot, or observational studies, while the remaining 115 were retrospective studies (Table ). The Newcastle‐Ottawa Scale, as per quality assessment, scored as poor quality [(score: 0–2)] the 22.3% (27/121), fair quality [(score: 3–5)] the 54.5% (66/121) and good/high quality [(score: 6–9)] the 23.1% (28/121) of studies included in the analysis. Patient baseline characteristics The overall subjects in the primary closure cohort ( N = 2404; female = 1279, male = 701, the sum was not equal to the total because of missing values) had a mean (±SD) age of 34.8 (4.8) years, with a mean (±SD) body mass index of 31.3 (3.4) kg/m 2 . The split/skin graft involved a total of 2700 subjects (female = 1490, male = 1042, among those with available records) with a mean (±SD) age of 36.5 (4.8) years and a mean (±SD) body mass index of 30.6 (3.6) kg/m 2 , while local/distal flaps procedure was applied to a total of 2347 individuals (female = 1154, male = 956, among those with available records) showing a mean (±SD) age of 36.5 (4.8) years and a mean (±SD) body mass index of 29.6 (2.9) kg/m 2 . Likewise, the cohort undergoing secondary intention healing comprised a total of 2191 patients (female = 1240, male = 884 with available records), with a mean (±SD) age of 36.2 (3.6) years and a mean (±SD) body mass index of 28.8 (2.8) kg/m 2 . Defects surgically repaired with local/distal flaps showed a significantly lower size area (cm 2 ) compared to those repaired with other methods of wound closure [101.4 ± 58.8 cm 2 versus 243.4 ± 243.6 cm 2 , P = 0.017]. The average (±SD) size defect in each wound closure cohort was 82.3 (47.0) cm 2 for primary closure, 233.9 (207.8) cm 2 for split/skin graft, 101.4 (58.8) cm 2 for local/distant flap and 84.3 (85.5) cm 2 for secondary intention healing. Meta‐analysis The meta‐analysis aimed to integrate the available evidence and evaluate the recurrence rate (primary outcome) and complications (secondary outcomes) associated with different wound closure techniques after wide excision in HS patients. Each wound closure technique, including primary closure, secondary closure (split/skin graft, local/distant flap), and secondary intention healing, has been comparatively evaluated regarding the primary outcome, that is, the proportion of recurrence. Overall, six possible head‐to‐head comparisons between wound closure techniques (primary vs. flap; primary closure vs. skin graft; primary closure vs. secondary intention healing; flap vs. skin graft; flap vs. primary closure; flap vs. secondary intention healing) have been analyzed if at least three independent studies had available data about recurrence events for at least one head‐to‐head comparison. Also, postoperative secondary outcomes, specifically infection, wound dehiscence, bleeding, and movement impairments, have been evaluated for comparisons between wound closure techniques. Because of the large number of studies that did not directly report on each specific comparison, both traditional head‐to‐head comparisons and a multi‐intervention network meta‐analysis have been performed to integrate primary (recurrence rate) and secondary outcomes (infection, wound dehiscence, bleeding, and movement impairments rates) across all possible comparisons between wound closure techniques. Primary Outcome : Disease recurrence . Network meta‐analysis Table shows the matrix table for the network of real‐life studies comparing the recurrence rate between wound closure techniques. There was evidence that local/distant flap reduced the risk of recurrence by 55% [OR = 0.450; 95% CI = 0.260, 0.780] and split skin graft by 45% [OR = 0.549; 95% CI = 0.334, 0.903], compared to primary closure. Figure shows the geometry of the study network of real‐life studies comparing the recurrence rate between wound closure strategies in wide excision for HS. Wound closure techniques had been compared among themselves. The results of direct comparison meta‐analyses were consistent with these findings. A direct head‐to‐head meta‐analysis confirmed local/distant flap in reducing the risk of recurrence by 59% [0.41 (0.17, 1.01), P = 0.05, I 2 = 49%] compared to primary closure, and secondary intention healing tended to favor a reduced recurrence rate by 26% compared to primary closure [0.74 (0.53, 1.04), P = 0.08, I 2 = 0%]. A low degree of heterogeneity across studies was demonstrated ( I 2 = 0% or 49%, respectively). Figure shows a direct head‐to‐head meta‐analysis comparatively evaluating the recurrence rate associated with different wound closure techniques after wide excision in HS patients. Secondary outcomes: Post‐treatment complications such as dehiscence, infection, bleeding, and movement impairments . Postoperative complications have been evaluated through network meta‐analysis because sufficient data (at least three independent studies) were not available to conduct head‐to‐head comparisons on these variables. Network meta‐analysis: Dehiscence . Data pooled by random effect (Table ) showed a lower likelihood of wound dehiscence complications among primary closure (OR = 0.543, 95% CI = 0.363–0.811) and secondary intention healing (OR = 0.084, 95% CI = 0.010–0.705) compared to local/distal flap. Figure shows the geometry of the network of real‐life studies comparing the dehiscence rate between wound closure strategies in wide excision for HS. Statistically insignificant results have been obtained for comparing the infection, bleeding, and movement impairment rates as postoperative complications between wound closure techniques. For infection (Table ) and bleeding (Table ), the comparison between wound closure techniques did not provide statistically significant results. The number of studies and/or events for movement impairments is too low to provide CINEMA output. Bias and heterogeneity analysis In this meta‐analysis, we assessed the potential bias and heterogeneity among the included studies to ensure the reliability of our findings. Bias was evaluated using the Newcastle‐Ottawa Scale (NOS), with scores indicating that 22.3% (27/121) of the studies were of poor quality, 54.5% (66/121) of fair quality, and 23.1% (28/121) of good/high quality. The variability in study quality highlights the presence of potential bias, particularly in retrospective analyses, which comprise most of the included studies (116/121). To address heterogeneity, we measured the degree of variability between studies using the inverse variance index ( I 2 ). The I 2 statistic values were categorized as low (25%), moderate (50%), and high (75%) heterogeneity. For the primary outcome of recurrence rates, heterogeneity was assessed using network and direct head‐to‐head meta‐analyses. We found low to moderate heterogeneity ( I 2 ranging from 0% to 49%) in direct comparisons such as primary closure versus local/distant flap. However, some comparisons' high degree of heterogeneity underscores the variability in study populations, interventions, and outcome measures. The presence of significant heterogeneity and the inclusion of studies with varying quality levels necessitate cautious interpretation of the results. Further randomized clinical trials with standardized methodologies are essential to provide more robust and generalizable evidence regarding HS's optimal wound closure techniques following wide excision.
A total of 1181 records were identified from a systematic literature search. After removing duplicates and entries in a non‐English language, there were 655 unique articles to screen as per the eligibility assessment (Figure ). Double independent review (CCI and CIM) of the available evidence excluded publications that did not meet the inclusion criteria (single case reports, unrelated topics, systematic reviews). Overall, 121 full‐text articles were included in the analysis. Of these, six were prospective case series, cohort, pilot, or observational studies, while the remaining 115 were retrospective studies (Table ). The Newcastle‐Ottawa Scale, as per quality assessment, scored as poor quality [(score: 0–2)] the 22.3% (27/121), fair quality [(score: 3–5)] the 54.5% (66/121) and good/high quality [(score: 6–9)] the 23.1% (28/121) of studies included in the analysis.
The overall subjects in the primary closure cohort ( N = 2404; female = 1279, male = 701, the sum was not equal to the total because of missing values) had a mean (±SD) age of 34.8 (4.8) years, with a mean (±SD) body mass index of 31.3 (3.4) kg/m 2 . The split/skin graft involved a total of 2700 subjects (female = 1490, male = 1042, among those with available records) with a mean (±SD) age of 36.5 (4.8) years and a mean (±SD) body mass index of 30.6 (3.6) kg/m 2 , while local/distal flaps procedure was applied to a total of 2347 individuals (female = 1154, male = 956, among those with available records) showing a mean (±SD) age of 36.5 (4.8) years and a mean (±SD) body mass index of 29.6 (2.9) kg/m 2 . Likewise, the cohort undergoing secondary intention healing comprised a total of 2191 patients (female = 1240, male = 884 with available records), with a mean (±SD) age of 36.2 (3.6) years and a mean (±SD) body mass index of 28.8 (2.8) kg/m 2 . Defects surgically repaired with local/distal flaps showed a significantly lower size area (cm 2 ) compared to those repaired with other methods of wound closure [101.4 ± 58.8 cm 2 versus 243.4 ± 243.6 cm 2 , P = 0.017]. The average (±SD) size defect in each wound closure cohort was 82.3 (47.0) cm 2 for primary closure, 233.9 (207.8) cm 2 for split/skin graft, 101.4 (58.8) cm 2 for local/distant flap and 84.3 (85.5) cm 2 for secondary intention healing.
The meta‐analysis aimed to integrate the available evidence and evaluate the recurrence rate (primary outcome) and complications (secondary outcomes) associated with different wound closure techniques after wide excision in HS patients. Each wound closure technique, including primary closure, secondary closure (split/skin graft, local/distant flap), and secondary intention healing, has been comparatively evaluated regarding the primary outcome, that is, the proportion of recurrence. Overall, six possible head‐to‐head comparisons between wound closure techniques (primary vs. flap; primary closure vs. skin graft; primary closure vs. secondary intention healing; flap vs. skin graft; flap vs. primary closure; flap vs. secondary intention healing) have been analyzed if at least three independent studies had available data about recurrence events for at least one head‐to‐head comparison. Also, postoperative secondary outcomes, specifically infection, wound dehiscence, bleeding, and movement impairments, have been evaluated for comparisons between wound closure techniques. Because of the large number of studies that did not directly report on each specific comparison, both traditional head‐to‐head comparisons and a multi‐intervention network meta‐analysis have been performed to integrate primary (recurrence rate) and secondary outcomes (infection, wound dehiscence, bleeding, and movement impairments rates) across all possible comparisons between wound closure techniques. Primary Outcome : Disease recurrence .
Table shows the matrix table for the network of real‐life studies comparing the recurrence rate between wound closure techniques. There was evidence that local/distant flap reduced the risk of recurrence by 55% [OR = 0.450; 95% CI = 0.260, 0.780] and split skin graft by 45% [OR = 0.549; 95% CI = 0.334, 0.903], compared to primary closure. Figure shows the geometry of the study network of real‐life studies comparing the recurrence rate between wound closure strategies in wide excision for HS. Wound closure techniques had been compared among themselves. The results of direct comparison meta‐analyses were consistent with these findings. A direct head‐to‐head meta‐analysis confirmed local/distant flap in reducing the risk of recurrence by 59% [0.41 (0.17, 1.01), P = 0.05, I 2 = 49%] compared to primary closure, and secondary intention healing tended to favor a reduced recurrence rate by 26% compared to primary closure [0.74 (0.53, 1.04), P = 0.08, I 2 = 0%]. A low degree of heterogeneity across studies was demonstrated ( I 2 = 0% or 49%, respectively). Figure shows a direct head‐to‐head meta‐analysis comparatively evaluating the recurrence rate associated with different wound closure techniques after wide excision in HS patients. Secondary outcomes: Post‐treatment complications such as dehiscence, infection, bleeding, and movement impairments . Postoperative complications have been evaluated through network meta‐analysis because sufficient data (at least three independent studies) were not available to conduct head‐to‐head comparisons on these variables. Network meta‐analysis: Dehiscence . Data pooled by random effect (Table ) showed a lower likelihood of wound dehiscence complications among primary closure (OR = 0.543, 95% CI = 0.363–0.811) and secondary intention healing (OR = 0.084, 95% CI = 0.010–0.705) compared to local/distal flap. Figure shows the geometry of the network of real‐life studies comparing the dehiscence rate between wound closure strategies in wide excision for HS. Statistically insignificant results have been obtained for comparing the infection, bleeding, and movement impairment rates as postoperative complications between wound closure techniques. For infection (Table ) and bleeding (Table ), the comparison between wound closure techniques did not provide statistically significant results. The number of studies and/or events for movement impairments is too low to provide CINEMA output.
In this meta‐analysis, we assessed the potential bias and heterogeneity among the included studies to ensure the reliability of our findings. Bias was evaluated using the Newcastle‐Ottawa Scale (NOS), with scores indicating that 22.3% (27/121) of the studies were of poor quality, 54.5% (66/121) of fair quality, and 23.1% (28/121) of good/high quality. The variability in study quality highlights the presence of potential bias, particularly in retrospective analyses, which comprise most of the included studies (116/121). To address heterogeneity, we measured the degree of variability between studies using the inverse variance index ( I 2 ). The I 2 statistic values were categorized as low (25%), moderate (50%), and high (75%) heterogeneity. For the primary outcome of recurrence rates, heterogeneity was assessed using network and direct head‐to‐head meta‐analyses. We found low to moderate heterogeneity ( I 2 ranging from 0% to 49%) in direct comparisons such as primary closure versus local/distant flap. However, some comparisons' high degree of heterogeneity underscores the variability in study populations, interventions, and outcome measures. The presence of significant heterogeneity and the inclusion of studies with varying quality levels necessitate cautious interpretation of the results. Further randomized clinical trials with standardized methodologies are essential to provide more robust and generalizable evidence regarding HS's optimal wound closure techniques following wide excision.
This systematic review and meta‐analysis suggest that secondary closure techniques may offer superior outcomes in managing HS compared to primary closure methods. It shows that both secondary closure techniques, local/distant flap and split skin graft techniques, significantly reduce the risk of recurrence by 55% and 45%, respectively, compared to primary direct closure. The long‐standing effect of surgery is a key point in the management of patients with severe HS. When performing wide excision of HS lesions, wound closure may be achieved by direct sutures, skin grafts, flaps, or secondary intention healing. Antibiotics, short courses of corticosteroids, or biologic agents can be used prior to intervention. , It is recommended that anti‐inflammatory therapy be maintained throughout the surgical procedure, as it facilitates the differentiation between healthy and inflamed tissue, thereby accelerating the healing process. While simpler and faster than the secondary closure technique, the primary closure technique showed higher recurrence rates, likely because of its limited capacity to manage the extensive tissue defects and residual inflammation characteristic of HS. In contrast, local/distant flaps and split skin grafts provide more robust coverage and integration with surrounding tissues, potentially explaining their lower recurrence rates. However, the complexity and invasiveness of flap procedures may pose higher risks of certain complications, including wound dehiscence, tissue necrosis, and hemorrhage, that must be considered. Similarly, while split skin grafts provide good functional and aesthetic outcomes, they also carry risks of donor site morbidity and healing issues. Secondary intention healing, although not significantly reducing recurrence rates compared to primary closure, offers the advantage of avoiding donor site morbidity and maintaining a reasonable range of motion. However, the prolonged healing period and the necessity for meticulous wound care limit its practicality, especially in larger defects. This meta‐analysis provides evidence that secondary closure techniques, such as local and distant flaps and split‐thickness skin grafts, significantly reduce recurrence rates following wide excision for HS compared to primary closure. Although these advanced techniques are associated with greater complexity and a higher risk of complications, their superior efficacy in preventing recurrence supports their consideration in clinical practice. Despite being less effective in recurrence reduction, secondary intention healing remains a viable option for patients where other techniques are contraindicated or impractical. We observed no significant differences in infection and bleeding rates among the different closure techniques, indicating that these complications are relatively consistent across methods. These results are in line with previous studies. Ovadja et al. performed a systematic review and meta‐analysis of 79 studies regarding reconstructive strategies after wide excision, finding that primary closure had the highest recurrence rate (22%), followed by secondary intention healing (11%) and that recurrence was lowest for skin graft and fasciocutaneous flaps (2%). A review by Riddle et al. demonstrated that wide excision combined with flap‐based reconstruction is correlated with a reduced recurrence rate of postsurgical HS. However, this advantage must be weighed against the potential for increased morbidity associated with extensive surgical procedures. Bouazzi et al. reviewed 41 wide excision‐relevant articles for a total of 2844 patients, reporting the highest recurrence rate after primary closure (39.9%), followed by split skin graft (9.3%), reconstruction by flaps (6.1%), and secondary healing (5.4%). A further review of the recurrence of HS following surgical management synthesized data from 22 studies. The review revealed the following average recurrence rates for defects repair after wide excision: 15% for primary closure, 8% for flap use, and 6.0% for grafting, although highlighting poor quality of the evidence and potential issues with reporting accuracy. Our meta‐analysis demonstrated low to moderate heterogeneity in direct comparisons between closure techniques across studies ( I 2 ranging from 0% to 49%). However, evaluating the potential bias using the Newcastle‐Ottawa Scale (NOS), we found high variability in study quality related to the retrospective nature of most of the included studies (116/121), which involves variability in the study populations, interventions, and outcome measures. Multicentric studies on large patient cohorts are needed to establish standardized methodologies to be used as guidelines for wound closure techniques in HS. Our study has some limitations. Most of the included studies were retrospective, which may introduce bias and affect the reliability of the results. The heterogeneity in study designs, patient populations, and follow‐up periods across studies may also impact our findings. Notably, this study is the first of its kind to include such a comprehensive number of articles, making it a significant contribution to the field. Further high‐quality prospective studies are needed to validate these results and establish standardized guidelines for wound closure techniques in HS.
Among the techniques analyzed, local flaps emerged as the most effective in minimizing recurrence. While primary closure and secondary intention healing are simpler to perform, they are linked with higher recurrence rates. Though somewhat effective, skin grafts still need to catch up to the outcomes achieved with flaps.
Table S1. Network meta‐analysis (NMA) matrix table from the network of real‐life studies comparing infection complications rate between wound closure techniques. Table S2. Network meta‐analysis (NMA) matrix table from the network of real‐life studies comparing bleeding complications rate between wound closure techniques. Table A1. Patients demographic and study characteristics. Figure A1. Network plot comparing recurrence rates between wound closure strategies. Edge width refers to the number of studies and the node size of the sample size. Links represent head‐to‐head comparisons between linked nodes, with the thickness proportional to the number of studies on each specific comparison. Figure A2. Direct head‐to‐head meta‐analysis comparing recurrence rates between wound closure techniques throughout six possible head‐to‐head comparisons. (a) Primary closure versus local/distant flap. (b) Primary closure versus skin graft. (c) Primary closure versus secondary intention healing. (d) Skin graft versus local/distant flap. (e) Skin graft versus secondary intention healing. (f) Local/distant flap versus secondary intention healing. Figure A3. Network plot from the network meta‐analysis of real‐life studies comparing wound dehiscence rates between wound closure strategies in wide excisions for HS. Edge width refers to the number of studies and the node size of the sample size.
|
A Combination of Conventional Transconjunctival Lower Blepharoplasty and Percutaneous Nanofat Grafting in a Young Chinese Population With Eyelid Bags and Tear Trough Deformities | 0f8c45c3-2228-4e68-b449-eaea720a451d | 11755396 | Ophthalmologic Surgical Procedures[mh] | Introduction Patients seeking lower blepharoplasty often attribute their problem to a baggy appearance of the lower eyelids. With the development of the facial rejuvenation surgery, a better understanding of the anatomical structures of the periorbital region, including the lower eyelid, has been achieved . Lower eyelid bags are a kind of complex and comprehensive problem caused by several anatomical changes . Apart from orbital fat herniation , the tear trough depression [ , , ] is frequently present in the medial third part of the infraorbital rim. Other anatomical abnormalities such as dark circles , skin laxity and fine wrinkles are also associated with eyelid bags. All these anatomical elements, both congenital and acquired with age, lead to a tired and aged appearance. In order to restore a youthful contour, the lower blepharoplasty has evolved from classic fat excision to current fat preservation . Fat repositioning and fat grafting are two major categories of surgical procedures [ , , , , ] widely used to eliminate the protruding baggy appearance and to add tissue volume to improve the hollow infraorbital area, including the tear trough. Although both procedures can provide satisfactory cosmetic results, each has its own pros and cons. Compared with fat repositioning, autologous fat grafting following orbital fat removal in lower blepharoplasty has some merits [ , , , , , ] including convenient operation, minimal invasiveness and quick recovery, but it may still carry risks such as uneven skin contour in periorbital and perioral areas. Many plastic surgeons have developed and streamlined individualized surgical methods to address the specific anatomical problems of eyelids confirmed in their population [ , , , , , , , ]. Due to the anatomical characteristics of the lower eyelids in our young population, a combined application of two surgeries, transconjunctival orbital fat excision followed by autologous orbital fat transplantation, was adopted in our study. During fat grafting, the harvested orbital fat is processed into nanofat [ , , ], which is more suitable for injection into the periorbital area. Our study aimed to validate the efficacy of this simple and minimally invasive combination of transconjunctival lower blepharoplasty and percutaneous nanofat injection in our young Chinese population.
Patients and Methods The retrospective study of consecutive young patients for the treatment of lower eyelid bags was conducted between February 2020 and June 2024 at the Affiliated Hangzhou First People's Hospital, Westlake University School of Medicine. A total of 183 patients who had lower eyelid bags without skin laxity or with slight skin laxity were recruited for our study. All patients wanted to reduce their tired appearance and all were under the age of 40. Lower eyelids with excessive skin redundancy, a history of surgery or trauma to the infraorbital region within 1 year, the injection of filler within the previous 6 months, and other systemic diseases such as coagulopathy were the exclusion criteria. All the patients signed an informed consent and underwent a combined procedure including transconjunctival septal fat resection and transcutaneous resected fat grafting in our hospital. This study was approved by the Ethics Committee of our hospital and adhered to the tenets of the Declaration of Helsinki. 2.1 Patients' Information Collection Patient demographics including age, gender, race was collected and standard preoperative photographs were taken. The operative time and injection volume of autologous resected septal fat were recorded immediately after surgery. The surgical history of a lower blepharoplasty (more than 1 year prior) was also obtained. All patients were followed up for at least 6 months, follow‐up period was calculated, postoperative photographs were taken and complications related to both septal fat removal and autologous fat transplantation were documented. Patient satisfaction was subjectively rated by themselves on a 4‐point grade (worse being 1; low satisfaction or not satisfied being 2; satisfied being 3; very satisfied being 4) according to the therapeutic outcome at the end of follow‐up. All the patients came to our clinic to address the problem of the lower eyelids, with the main complaint being the appearance of fatigue and aging. However, in addition to orbital fat herniation, tear trough deformity and dark circles were also common in most of the young Chinese patients. In order to objectively evaluate the improvement in aesthetic appearance before and after surgery, our study adopted five major components that contribute to the anatomy of the baggy eyelids in the Chinese population, referring to the scoring system of Goldberg et al. , Sadick et al. , and Cheng et al. . Thus, orbital fat prolapse, tear trough depression, dark circles, orbicularis prominence, and loss of skin elasticity were analyzed in detail by blindly comparing the pre‐ and post‐operative photographs by the two independent observers. In addition, both sides of the lower eyelids should be assessed respectively, as the two sides of lower eyelid bags are usually asymmetrical in the same person. Each indicator of the five elements was graded into five scales, recording no involvement, mild, moderate, marked or severe. Therefore, the scale for each category was scored before and after surgery for each eyelid of all patients in this study. The number and percentage of eyelids with grades from no involvement to severe were then calculated. 2.2 Preoperative Design Lower blepharoplasty is routinely performed with the patient in the supine position, which hides the three bulging septal fat pads. Therefore, the preoperative evaluation and planning of tranconjunctival lower blepharoplasty in our study was conducted with the patient in an upright position. In the sitting position, the patients were asked to look straight ahead and look upwards, then the location of the orbital fat was briefly marked for precise removal of redundant medial, central, and lateral fat pads. The zone of the tear trough depression was depicted in the sunken area of the anterior maxilla below the medial infraorbital rim and was the recipient area for autologous septal fat injection (Figure ). 2.3 A Combined Technique of Transconjunctival Blepharoplasty and Nanofat Grafting 2.3.1 Transconjunctival Resection of Orbital Fat Pads All of the surgeries in our study were carried out under local anesthesia, with the lower eyelid being retracted caudally by the assistant (Figure ). 2% lidocaine with 1:100000 epinephrine was injected into the subconjunctival layer, and then into each of the medial, central, and lateral orbital fat pads under the orbicularis oculi muscle and orbital septum. After 5 min of compression on the anesthetized area, the conjunctiva and capsulopalpebral fascia were both incised at the site of five millimeters inferior to the lower eyelid margin using a No. 11 blade (Figure ). The eyelid margin of the split conjunctiva was pulled back in a caudal direction by the assistant using a lid retractor. Blunt dissection (Figure ) was performed towards the infraorbital rim below the orbicularis oculi muscle using ophthalmic scissors and mosquito clamps to secure the orbital septum intact. After completion of the retro‐orbicularis oculi muscle dissection, the three orbital fat compartments were exposed by incising the orbital septum with a small window. The three fat pads were herniated with gentle pressure on the globe, and then the central, medial and lateral orbital fat pads were resected in sequence (Figure ). The medial fat pad was carefully excised to avoid damaging the inferior oblique muscle (Figure ). The predicted amount of lower eyelid fat was removed conservatively with the patient in the supine position, and then the amount of fat removed was confirmed to be appropriate for the appearance of the infraorbital region when the patient was in the sitting position. All of the resected fat pads were immediately placed on the sterile saline‐soaked gauze until further processing (Figure ). The incised conjunctiva and capsulopalpebral fascia were restored to their original position and left in place without any sutures (Figure ). 2.3.2 Transcutaneous Resected Fat Injection for Tear Trough Deformity The resected orbital fat pads were washed with normal saline and chopped up with scissors (Figure ). Then the minced fat was transferred into two 1 mL syringes connected to each other by a Luer‐Lock connector with an internal diameter of 1.2 mm (Figure ). The fat was mechanically transferred at least 20 times between two syringes using a Luer‐Lock connector (Figure ). The yellow‐white fat emulsion obtained by shifting was nanofat. Nanofat [ , , ], rich in adipose‐derived stem cells (ASCs), was injected to smooth the eyelid‐cheek junction and improve skin texture, resulting in a more even and natural contour of the infraorbital area. During this emulsification process of the adipose tissue, the remnants of the connective tissue blocking the Luer‐Lock connector were removed (Figure ). After completion of the transconjunctival removal of the orbital fat pads and processing of the fat emulsion, the patient adjusted his posture to a sitting position (Figure ). The percutaneous entry point for nanofat injection was marked at the intersection of the tear trough extension line and a line from the angulus oris to the lateral canthus (Figure ). This insertion point was then punctured with a 22 G sharp needle (as shown in Figure ) after local anesthesia with lidocaine. With the patient sitting upright, the nanofat injection was administered using a 23 G blunt injection cannula with a single side hole (as depicted in Figure ). During the procedure, the surgeon's left index finger was used to feel the injection layer of the needle tip and to sense the injection volume of autologous fat to avoid lump formation (Figure ). The nanofat graft was injected into different anatomical layers in multiple directions in the hollow area of the tear trough. In our study, the fat grafting was performed in three anatomical layers in a fan shape. The deep layer was located above the maxillary periosteum and approximately 25% of the total fat volume was transplanted into this layer. The middle layer was the space between the orbicularis oculi muscle and the periosteum, and about 50% of the total volume was injected into this layer. The superficial layer was a space between the skin dermis and the orbicularis oculi muscle, and a further 25% of the total fat was placed in this layer. The total volume of nanofat graft for both sides of the lower eyelids in one person was recorded. Once the fat injection was completed, digital pressure was slightly applied to shape the injection area until the desired aesthetic result was achieved when the patient looked forward and upward (Figure ). 2.4 Postoperative Management A cold compress was applied for 20 min immediately after operation and used intermittently for the first 3 days to reduce postoperative bleeding and swelling. If necessary, eye drops were used to relieve postoperative discomfort. Oral antibiotics were not routinely administered after surgery. Patients were instructed not to put pressure on the surgical site for 1 week after surgery. Patients were instructed to minimize facial expressions, including smiling, and not to apply pressure to the surgical site for 1 week after surgery. 2.5 Statistical Analysis Statistical analysis was performed with SPSS version 20 (IBM Corp, USA). Descriptions of quantitative variables were presented as mean with standard deviation, while qualitative variables were expressed as number and percentage. Five elements of each eye including orbital fat prolapse, tear trough depression, dark circles, orbicularis prominence and loss of skin elasticity were evaluated. And the scale of each indicator of the eyes before and after operation was documented. The number and percentage of eyelids with scales being no involvement, mild, moderate, marked or severe were calculated, and then the chi‐square test was analyzed for the qualitative variables to compare the differences of each index before and after surgery, The level of p < 0.05 was regarded as statistically significant.
Patients' Information Collection Patient demographics including age, gender, race was collected and standard preoperative photographs were taken. The operative time and injection volume of autologous resected septal fat were recorded immediately after surgery. The surgical history of a lower blepharoplasty (more than 1 year prior) was also obtained. All patients were followed up for at least 6 months, follow‐up period was calculated, postoperative photographs were taken and complications related to both septal fat removal and autologous fat transplantation were documented. Patient satisfaction was subjectively rated by themselves on a 4‐point grade (worse being 1; low satisfaction or not satisfied being 2; satisfied being 3; very satisfied being 4) according to the therapeutic outcome at the end of follow‐up. All the patients came to our clinic to address the problem of the lower eyelids, with the main complaint being the appearance of fatigue and aging. However, in addition to orbital fat herniation, tear trough deformity and dark circles were also common in most of the young Chinese patients. In order to objectively evaluate the improvement in aesthetic appearance before and after surgery, our study adopted five major components that contribute to the anatomy of the baggy eyelids in the Chinese population, referring to the scoring system of Goldberg et al. , Sadick et al. , and Cheng et al. . Thus, orbital fat prolapse, tear trough depression, dark circles, orbicularis prominence, and loss of skin elasticity were analyzed in detail by blindly comparing the pre‐ and post‐operative photographs by the two independent observers. In addition, both sides of the lower eyelids should be assessed respectively, as the two sides of lower eyelid bags are usually asymmetrical in the same person. Each indicator of the five elements was graded into five scales, recording no involvement, mild, moderate, marked or severe. Therefore, the scale for each category was scored before and after surgery for each eyelid of all patients in this study. The number and percentage of eyelids with grades from no involvement to severe were then calculated.
Preoperative Design Lower blepharoplasty is routinely performed with the patient in the supine position, which hides the three bulging septal fat pads. Therefore, the preoperative evaluation and planning of tranconjunctival lower blepharoplasty in our study was conducted with the patient in an upright position. In the sitting position, the patients were asked to look straight ahead and look upwards, then the location of the orbital fat was briefly marked for precise removal of redundant medial, central, and lateral fat pads. The zone of the tear trough depression was depicted in the sunken area of the anterior maxilla below the medial infraorbital rim and was the recipient area for autologous septal fat injection (Figure ).
A Combined Technique of Transconjunctival Blepharoplasty and Nanofat Grafting 2.3.1 Transconjunctival Resection of Orbital Fat Pads All of the surgeries in our study were carried out under local anesthesia, with the lower eyelid being retracted caudally by the assistant (Figure ). 2% lidocaine with 1:100000 epinephrine was injected into the subconjunctival layer, and then into each of the medial, central, and lateral orbital fat pads under the orbicularis oculi muscle and orbital septum. After 5 min of compression on the anesthetized area, the conjunctiva and capsulopalpebral fascia were both incised at the site of five millimeters inferior to the lower eyelid margin using a No. 11 blade (Figure ). The eyelid margin of the split conjunctiva was pulled back in a caudal direction by the assistant using a lid retractor. Blunt dissection (Figure ) was performed towards the infraorbital rim below the orbicularis oculi muscle using ophthalmic scissors and mosquito clamps to secure the orbital septum intact. After completion of the retro‐orbicularis oculi muscle dissection, the three orbital fat compartments were exposed by incising the orbital septum with a small window. The three fat pads were herniated with gentle pressure on the globe, and then the central, medial and lateral orbital fat pads were resected in sequence (Figure ). The medial fat pad was carefully excised to avoid damaging the inferior oblique muscle (Figure ). The predicted amount of lower eyelid fat was removed conservatively with the patient in the supine position, and then the amount of fat removed was confirmed to be appropriate for the appearance of the infraorbital region when the patient was in the sitting position. All of the resected fat pads were immediately placed on the sterile saline‐soaked gauze until further processing (Figure ). The incised conjunctiva and capsulopalpebral fascia were restored to their original position and left in place without any sutures (Figure ). 2.3.2 Transcutaneous Resected Fat Injection for Tear Trough Deformity The resected orbital fat pads were washed with normal saline and chopped up with scissors (Figure ). Then the minced fat was transferred into two 1 mL syringes connected to each other by a Luer‐Lock connector with an internal diameter of 1.2 mm (Figure ). The fat was mechanically transferred at least 20 times between two syringes using a Luer‐Lock connector (Figure ). The yellow‐white fat emulsion obtained by shifting was nanofat. Nanofat [ , , ], rich in adipose‐derived stem cells (ASCs), was injected to smooth the eyelid‐cheek junction and improve skin texture, resulting in a more even and natural contour of the infraorbital area. During this emulsification process of the adipose tissue, the remnants of the connective tissue blocking the Luer‐Lock connector were removed (Figure ). After completion of the transconjunctival removal of the orbital fat pads and processing of the fat emulsion, the patient adjusted his posture to a sitting position (Figure ). The percutaneous entry point for nanofat injection was marked at the intersection of the tear trough extension line and a line from the angulus oris to the lateral canthus (Figure ). This insertion point was then punctured with a 22 G sharp needle (as shown in Figure ) after local anesthesia with lidocaine. With the patient sitting upright, the nanofat injection was administered using a 23 G blunt injection cannula with a single side hole (as depicted in Figure ). During the procedure, the surgeon's left index finger was used to feel the injection layer of the needle tip and to sense the injection volume of autologous fat to avoid lump formation (Figure ). The nanofat graft was injected into different anatomical layers in multiple directions in the hollow area of the tear trough. In our study, the fat grafting was performed in three anatomical layers in a fan shape. The deep layer was located above the maxillary periosteum and approximately 25% of the total fat volume was transplanted into this layer. The middle layer was the space between the orbicularis oculi muscle and the periosteum, and about 50% of the total volume was injected into this layer. The superficial layer was a space between the skin dermis and the orbicularis oculi muscle, and a further 25% of the total fat was placed in this layer. The total volume of nanofat graft for both sides of the lower eyelids in one person was recorded. Once the fat injection was completed, digital pressure was slightly applied to shape the injection area until the desired aesthetic result was achieved when the patient looked forward and upward (Figure ).
Transconjunctival Resection of Orbital Fat Pads All of the surgeries in our study were carried out under local anesthesia, with the lower eyelid being retracted caudally by the assistant (Figure ). 2% lidocaine with 1:100000 epinephrine was injected into the subconjunctival layer, and then into each of the medial, central, and lateral orbital fat pads under the orbicularis oculi muscle and orbital septum. After 5 min of compression on the anesthetized area, the conjunctiva and capsulopalpebral fascia were both incised at the site of five millimeters inferior to the lower eyelid margin using a No. 11 blade (Figure ). The eyelid margin of the split conjunctiva was pulled back in a caudal direction by the assistant using a lid retractor. Blunt dissection (Figure ) was performed towards the infraorbital rim below the orbicularis oculi muscle using ophthalmic scissors and mosquito clamps to secure the orbital septum intact. After completion of the retro‐orbicularis oculi muscle dissection, the three orbital fat compartments were exposed by incising the orbital septum with a small window. The three fat pads were herniated with gentle pressure on the globe, and then the central, medial and lateral orbital fat pads were resected in sequence (Figure ). The medial fat pad was carefully excised to avoid damaging the inferior oblique muscle (Figure ). The predicted amount of lower eyelid fat was removed conservatively with the patient in the supine position, and then the amount of fat removed was confirmed to be appropriate for the appearance of the infraorbital region when the patient was in the sitting position. All of the resected fat pads were immediately placed on the sterile saline‐soaked gauze until further processing (Figure ). The incised conjunctiva and capsulopalpebral fascia were restored to their original position and left in place without any sutures (Figure ).
Transcutaneous Resected Fat Injection for Tear Trough Deformity The resected orbital fat pads were washed with normal saline and chopped up with scissors (Figure ). Then the minced fat was transferred into two 1 mL syringes connected to each other by a Luer‐Lock connector with an internal diameter of 1.2 mm (Figure ). The fat was mechanically transferred at least 20 times between two syringes using a Luer‐Lock connector (Figure ). The yellow‐white fat emulsion obtained by shifting was nanofat. Nanofat [ , , ], rich in adipose‐derived stem cells (ASCs), was injected to smooth the eyelid‐cheek junction and improve skin texture, resulting in a more even and natural contour of the infraorbital area. During this emulsification process of the adipose tissue, the remnants of the connective tissue blocking the Luer‐Lock connector were removed (Figure ). After completion of the transconjunctival removal of the orbital fat pads and processing of the fat emulsion, the patient adjusted his posture to a sitting position (Figure ). The percutaneous entry point for nanofat injection was marked at the intersection of the tear trough extension line and a line from the angulus oris to the lateral canthus (Figure ). This insertion point was then punctured with a 22 G sharp needle (as shown in Figure ) after local anesthesia with lidocaine. With the patient sitting upright, the nanofat injection was administered using a 23 G blunt injection cannula with a single side hole (as depicted in Figure ). During the procedure, the surgeon's left index finger was used to feel the injection layer of the needle tip and to sense the injection volume of autologous fat to avoid lump formation (Figure ). The nanofat graft was injected into different anatomical layers in multiple directions in the hollow area of the tear trough. In our study, the fat grafting was performed in three anatomical layers in a fan shape. The deep layer was located above the maxillary periosteum and approximately 25% of the total fat volume was transplanted into this layer. The middle layer was the space between the orbicularis oculi muscle and the periosteum, and about 50% of the total volume was injected into this layer. The superficial layer was a space between the skin dermis and the orbicularis oculi muscle, and a further 25% of the total fat was placed in this layer. The total volume of nanofat graft for both sides of the lower eyelids in one person was recorded. Once the fat injection was completed, digital pressure was slightly applied to shape the injection area until the desired aesthetic result was achieved when the patient looked forward and upward (Figure ).
Postoperative Management A cold compress was applied for 20 min immediately after operation and used intermittently for the first 3 days to reduce postoperative bleeding and swelling. If necessary, eye drops were used to relieve postoperative discomfort. Oral antibiotics were not routinely administered after surgery. Patients were instructed not to put pressure on the surgical site for 1 week after surgery. Patients were instructed to minimize facial expressions, including smiling, and not to apply pressure to the surgical site for 1 week after surgery.
Statistical Analysis Statistical analysis was performed with SPSS version 20 (IBM Corp, USA). Descriptions of quantitative variables were presented as mean with standard deviation, while qualitative variables were expressed as number and percentage. Five elements of each eye including orbital fat prolapse, tear trough depression, dark circles, orbicularis prominence and loss of skin elasticity were evaluated. And the scale of each indicator of the eyes before and after operation was documented. The number and percentage of eyelids with scales being no involvement, mild, moderate, marked or severe were calculated, and then the chi‐square test was analyzed for the qualitative variables to compare the differences of each index before and after surgery, The level of p < 0.05 was regarded as statistically significant.
Results From February 2020 to June 2024, a total of 183 consecutive patients with 366 lower eyelid bags were enrolled in our study. All the patients received this simplified application of conventional transconjunctival blepharoplasty combined with resected nanofat grafting at the Affiliated Hangzhou First People's Hospital, Westlake University School of Medicine, and all the surgeries were performed by the attending surgeon, Chuanbo Liu. 3.1 General Information of the Study Population The mean age of the 183 patients was 26.9 years old, and 173 (95%) were female. 180 (98%) of the patients were of Han nationality and only three (2%) were from other ethnic minorities. Of the 183 patients, four (2%) had a previous transconjunctival lower blepharoplasty several years ago and underwent their second surgery for recurrent lower eyelid bags. All of the 183 patients were followed for at least 6 months after surgery based on the protocol, and the mean duration of follow‐up was 29.8 weeks. The average surgery time for the entire procedure including transconjunctival septal fat removal and percutaneous incisional fat injection was 36.1 min. The mean volume of total nanofat injected for bilateral lower eyelids in each patient was 1.1 mL. These baseline demographic characteristics and surgery‐related data of our study population were listed in Table . 3.2 Assessment of Surgical Outcomes by Pre‐ and Post‐Operative Comparison The preoperative and postoperative images of the patients were shown in Figure . Firstly, the subjective evaluation of the surgical outcomes was made in each patient based on patient satisfaction. Of the 183 patients, 175 (96.6%) were satisfied with the results of their surgery, with 75 (41.0%) being “very satisfied” and 100 (54.6%) “satisfied”. Only eight patients (4.4%) rated the outcomes of their operation as “low satisfaction or not satisfied”. Secondly, the objective assessment of the operative results was conducted in each lower eyelid according to the scores of orbital fat prolapse, tear trough depression, dark circles, orbicularis prominence and loss of skin elasticity before and after surgery respectively. The number and percentage of lower eyelids with scales of no involvement, mild, moderate, marked or severe were recorded, and the differences of each index before and after operation were analyzed by chi‐square test as shown in Tables , , , , below. These results of our study showed that the appearance of orbital fat prolapse ( p < 0.01, Table ), tear trough depression ( p < 0.01, Table ) and dark circles ( p < 0.01, Table ) were all improved after surgery. Postoperative orbicularis prominence was also more evident ( p < 0.01, Table ). However, there was no significant difference in loss of skin elasticity after operation ( p > 0.05, Table ), although someone felt more wrinkles in the lower lids. 3.3 Postoperative Complications and Management Postoperative complications occurred in both transconjunctival fat removal and transcutaneous fat grafting procedures and the detailed information was given in Table . None of the patients in our study developed serious complications associated with transconjunctival blepharoplasty including infection, hematoma, retrobulbar hemorrhage, lid retraction, ectropion, and scleral show. No severe lipoinjection‐related complications, such as fat embolism, were reported during the follow‐up period. Almost all the patients experienced mild to moderate bruising, edema or swelling of the periorbital area, which resolved spontaneously after one to 2 weeks without further treatment. Five patients had transient bloodshot eyes, three patients experienced chemosis, and one patient showed dry eye symptoms. All of these minor complaints mentioned above were transient and resolved after one to 3 weeks with the application of ophthalmic drops or ointment. In addition, one patient had a palpable but nonvisible submuscular nodule that regressed spontaneously within 2 months. Ten patients developed more fine wrinkles on their lower eyelids after surgery, but this didn't affect their overall satisfaction. Five patients had undercorrection of the tear trough depression and two patients had infraorbital hollowness, while three patients complained of recurrent lower eyelid bags due to residual orbital fat prolapse. Surgical repair was performed by either additional fat transplantation or transconjunctival septal fat resection after at least 6 months.
General Information of the Study Population The mean age of the 183 patients was 26.9 years old, and 173 (95%) were female. 180 (98%) of the patients were of Han nationality and only three (2%) were from other ethnic minorities. Of the 183 patients, four (2%) had a previous transconjunctival lower blepharoplasty several years ago and underwent their second surgery for recurrent lower eyelid bags. All of the 183 patients were followed for at least 6 months after surgery based on the protocol, and the mean duration of follow‐up was 29.8 weeks. The average surgery time for the entire procedure including transconjunctival septal fat removal and percutaneous incisional fat injection was 36.1 min. The mean volume of total nanofat injected for bilateral lower eyelids in each patient was 1.1 mL. These baseline demographic characteristics and surgery‐related data of our study population were listed in Table .
Assessment of Surgical Outcomes by Pre‐ and Post‐Operative Comparison The preoperative and postoperative images of the patients were shown in Figure . Firstly, the subjective evaluation of the surgical outcomes was made in each patient based on patient satisfaction. Of the 183 patients, 175 (96.6%) were satisfied with the results of their surgery, with 75 (41.0%) being “very satisfied” and 100 (54.6%) “satisfied”. Only eight patients (4.4%) rated the outcomes of their operation as “low satisfaction or not satisfied”. Secondly, the objective assessment of the operative results was conducted in each lower eyelid according to the scores of orbital fat prolapse, tear trough depression, dark circles, orbicularis prominence and loss of skin elasticity before and after surgery respectively. The number and percentage of lower eyelids with scales of no involvement, mild, moderate, marked or severe were recorded, and the differences of each index before and after operation were analyzed by chi‐square test as shown in Tables , , , , below. These results of our study showed that the appearance of orbital fat prolapse ( p < 0.01, Table ), tear trough depression ( p < 0.01, Table ) and dark circles ( p < 0.01, Table ) were all improved after surgery. Postoperative orbicularis prominence was also more evident ( p < 0.01, Table ). However, there was no significant difference in loss of skin elasticity after operation ( p > 0.05, Table ), although someone felt more wrinkles in the lower lids.
Postoperative Complications and Management Postoperative complications occurred in both transconjunctival fat removal and transcutaneous fat grafting procedures and the detailed information was given in Table . None of the patients in our study developed serious complications associated with transconjunctival blepharoplasty including infection, hematoma, retrobulbar hemorrhage, lid retraction, ectropion, and scleral show. No severe lipoinjection‐related complications, such as fat embolism, were reported during the follow‐up period. Almost all the patients experienced mild to moderate bruising, edema or swelling of the periorbital area, which resolved spontaneously after one to 2 weeks without further treatment. Five patients had transient bloodshot eyes, three patients experienced chemosis, and one patient showed dry eye symptoms. All of these minor complaints mentioned above were transient and resolved after one to 3 weeks with the application of ophthalmic drops or ointment. In addition, one patient had a palpable but nonvisible submuscular nodule that regressed spontaneously within 2 months. Ten patients developed more fine wrinkles on their lower eyelids after surgery, but this didn't affect their overall satisfaction. Five patients had undercorrection of the tear trough depression and two patients had infraorbital hollowness, while three patients complained of recurrent lower eyelid bags due to residual orbital fat prolapse. Surgical repair was performed by either additional fat transplantation or transconjunctival septal fat resection after at least 6 months.
Discussion Removal of herniated orbital fat alone during a conventional blepharoplasty can only improve the protruding appearance of the lower eyelids. Tear trough deformities and dark circles associated with eyelid bags are not corrected because this technique doesn't improve the anatomical volume loss of the tear trough depression and skin texture . As a result, the current mainstream of lower blepharoplasty has been refined into techniques of fat preservation with conservative resection of the orbital fat . Many techniques [ , , , , , , , , , , , , ] of lower blepharoplasty, categorized as fat repositioning or fat grafting, not only eliminate the prominent baggy appearance of the lower eyelid, but also increase the volume of the tear trough depression in the midface. Each of these methods has its own merits and demerits respectively. If the appropriate procedure is chosen based on the characteristics of the patient's lower eyelid, a satisfactory surgical result can be achieved . The fat transposition technique during transcutaneous or transconjunctival lower blepahroplasty was performed by the tear trough ligament release and septal fat reset. Many plastic surgeons [ , , ] have modified their individual fat repositioning techniques by releasing the arcus marginalis and then advancing and fixing the orbital fat pedicles either in the subperiosteal, supraosteal, submuscular or even supramuscular plane. Although widely and effectively used in lower blepahroplasty, fat redistribution still has some drawbacks , including the possible disruption of the middle lamella caused by extensive dissection. In addition, the learning curve for surgeons is steep due to the relative complexity of the technique. For these above reasons, some experts attempted to combine traditional lower blepharoplasty in conjunction with a minimally invasive fat grafting technique [ , , , , , , ]. In this study, we provided a combined application of conventional transconjunctival lower blepharoplasty and transcutaneous resected orbital fat injection to address the concerns of lower eyelid bags in the young Chinese patients, which proved to be simple, effective and safe. The main benefit of our combined application of traditional lower blepharoplasty and nanofat grafting was simplification, and the average total operative time was 36.1 min. This procedure met the desire of young patients for a minimally invasive operation with rapid recovery. Transconjunctival orbital fat removal and transcutaneous resected fat grafting were two separate but sequential steps in our technique. Both procedures were simple and less invasive, so the combined therapy was more acceptable to patients and the learning curve for surgeons was relatively short. The first step was to expose and incise the three orbital fat pads of each eyelid after completing the retro‐obicularis oculi muscle dissection. In this procedure, we didn't carry out an extensive dissection up to the orbital rim or a complete release of the tear trough ligament. All three fat compartments under the orbital septum were removed through one small incision. The second step was the process of fat emulsification and injection. The procedure was convenient because all the autologous fat used for grafting was harvested from the resected orbital fat only, with no further liposuction from other donor sites [ , , , , , ]. After cutting the resected septal fat into pieces with scissors, the chopped fat was transferred into a 1 mL syringe. The fat emulsion was finally produced by mechanical transfer between two syringes using a Lock‐Leur connector. The single entry point for percutaneous fat injection was simply incised with an 22 G sharp needle with less injury, and then fat was injected in different layers of the tear trough groove in a fan shape using a 23 G blunt cannula. In our study, following transconjunctival orbital fat removal, the resected fat was processed into nanofat and injected. The application of nanofat was one of the features of our combined surgery, as other research used various forms of autologous fat grafting, including free excised 2‐3 mm or even larger orbital fat transplantation , fractionated fat or microfat injection , or stromal vascular fraction gel injection . In our study, patients ranged in age from 18 to 40 years, with an average age of 26.9 years, and 95% were female. There was no severe skin laxity or obvious wrinkles in the eyelids, and orbital fat herniation, tear trough deformity and dark circles were common in our young Chinese patients. Tear trough deformities in the young are usually associated with congenital maxillary retrusion in addition to the cutaneous, muscular and ligamentous structures of the infraorbital region while dark circles are usually associated with the thin and transparent skin of the lower eyelids. Nanofat is rich in adipose‐derived stem cells (ASCs) with a good capacity for proliferation and differentiation [ , , ], and many studies [ , , ] have confirmed the effect of nanofat on skin rejuvenation. By combining with traditional lower blepharoplasty, percutaneous nanofat injection was confirmed to be effective in our study, with overall subjective patient satisfaction as high as 96.6%. Tear trough deformities were improved ( p < 0.01, Table ), while dark circles were significantly alleviated ( p < 0.01, Table ) after a few months of nanofat filling in our young population with lower eyelid bags. These favorable results may be related to the measurements we took during the nanofat grafting in our study. Nanofat was a yellow‐white fat emulsion harvested after mechanical mincing and transfer of the resected septal fat. Nanofat has lost its normal adipose tissue structure and the main concern with nanofat grafting is the unpredictable rate of fat retention . Therefore, the volumetric capacity of the nanofat is considered to be limited . However, in our study, although the exact figure was not calculated, the fat retention rate might still be acceptable. To improve the effect of the operation, nanofat injection was distributed in multiple anatomical layers and in small amounts increased the blood supply to the transplanted fat . In addition, the tear trough ligament was released to some extent during the process of multilayer injection using a blunt cannula . Thus, the increased volume of fat in the subperiosteal, submuscular and subcutaneous layers provided the supporting effect . Furthermore, the lower eyelid was less involved in facial expression movements and patients were instructed to minimize facial expressions, especially smiling, for 1 week after surgery. These three factors, taken together, may have favored the volume retention rate of the nanofat graft . However, further research is needed to clarify the complex causes of this speculation. Because a small amount of nanofat was injected into the superficial subcutaneous layer, dark circles were also reduced, similar to the results of other studies . By communicating with patients during follow‐up visits, another underlying cause of high patient satisfaction may exist. The main purpose of fat grafting after orbital fat removal was to mold a youthful contour with a smooth lid‐cheek junction rather than simply to augment the volume. Hence, the flat appearance of the lower eyelids without the convex‐concave deformity (protruding fat and tear trough groove) increased patient satisfaction, even though the infraorbital zone was slightly hollow. Based on the anatomical characteristics of the eyelid bags in our study population, a five‐factor evaluation, modified from the scoring system of Goldberg et al. , Sadick et al. , and Cheng et al. , was adopted to more objectively and comprehensively assess our surgical outcomes. Apart from orbital fat prolapse, tear trough depression and dark circles, the objective assessment of orbicularis prominence and skin elasticity before and after surgery was also important and had an impact on patient satisfaction. The orbicularis prominence, also known as pretarsal roll, is considered a significant symbol of youth in the Asian population . After removal of orbital fat, a smoother lower eyelid probably made the orbicularis muscle more prominent. Our study confirmed that the prominence of the orbicularis was enhanced after surgery ( p < 0.01, Table ), which is consistent with another similar study . Although 10 patients developed more fine wrinkles in their lower eyelids after surgery, there was still no significant difference in the loss of skin elasticity after operation ( p > 0.05, Table ). However, this didn't affect their overall satisfaction. On the one hand, this was related to the choice of surgical technique, and patients were informed of the advantages and disadvantages of transconjunctival and transcutaneous approaches to lower blepharoplasty respectively. It was therefore understandable that the limitations of transconjunctival blepharoplasty could lead to increased wrinkling after surgery . On the other hand, nanofat injection could improve skin texture through the rejuvenating effect of ASCs [ , , ]. This could compensate for the increased skin relaxation caused by the transconjunctival approach. This study confirmed that our combination application of lower blepharoplasty and nanofat grafting is safe with few complications. No major complications such as infections or vascular thrombosis occurred in our population. In our study, the incidence of minor complications was low as shown in Table , and only one patient developed a liponodule. It was not visible but palpable and resolved with any therapy. In order to prevent and reduce the complications associated with fat embolism , a very tiny volume dose of fat was delivered with each syringe push. What is more, the injection area was gently massaged during the fat grafting process. Besides, in order to place the autologous fat in the precise location to correct the tear trough deformity, the nanofat injection was performed with the patient in an upright position . Due to the different position and volume of the tear trough in the supine and sitting positions, this maneuver was superior to lipoinjection through the same incision . There are still some limitations in our study. Firstly, it was a retrospective study and there were some inherent flaws. The study protocol was confirmed post‐operatively, so pre‐ and post‐operative photographs were used to evaluate the aesthetic outcome of lower eyelid bags. It also failed to take photos of the patient making facial expressions, such as laughing, which were the best way to assess the safety of superficial injection . In addition, some of the photos were taken with an iPhone against different backgrounds, and the lighting was not consistent before and after surgery, which could affect the assessment of surgical outcomes. Moreover, dark circles were evaluated and graded in five levels, from none to severe, and were not classified as hyperpigmented, vascular or mixed dark circles according to the photographs. Therefore, prospective research with longer follow‐up may provide more persuasive evidence. Secondly, it was a non‐controlled study and we did not directly compare the outcomes and benefits of traditional transconjunctival lower blepharoplasty with fat repositioning versus fat grafting. Instead, Table presents a comparative analysis of surgical techniques from various studies, encompassing surgical outcomes, patient satisfaction, and complications. Each method has its inherent advantages and disadvantages. Consequently, a case–control study should be conducted to offer a more definitive perspective on the comparative efficacy between fat repositioning and fat grafting. Thirdly, the retention rate of the grafted nanofat was not quantified, although patient satisfaction was high and the improvements in tear trough deformities were satisfactory. In further studies, the application of three‐dimensional assessment can more precisely measure the infraorbital volume before and after lower blepharoplasty and then assist to objectively evaluate the fat retention rate.
Conclusion Our study offered a combination of conventional transconjunctival lower blepharoplasty and percutaneous excised orbital fat grafting to treat the lower eyelid bags and tear trough deformities based on the anatomical characteristics of our young Chinese population. This combined technique was simple and minimally invasive, and favorable results were achieved with high patient satisfaction, rapid recovery and low complication rate. Therefore, this combined application of traditional blepharoplasty and nanofat grafting may be a promising alternative surgical candidate for the young Chinese with lower eyelid bags to yield a satisfactory treatment outcome.
C.L., J.L., and X.S. designed the research study; C.L., Z.W., and L.T. performed the research; Z.Z. and J.L. analyzed the data; C.L. wrote the paper; J.L. and X.S. reviewed and edited the paper. All authors have read and approved the final manuscript.
This retrospective study was approved by the Ethics Committee of Affiliated Hangzhou First People's Hospital, Westlake University School of Medicine (ZN‐20240401‐0109‐01) and adhered to the tenets of the Declaration of Helsinki.
The authors declare no conflicts of interest.
|
Nomogram to predict periprosthetic joint infection after total hip arthroplasty using laboratory tests | 8935e8a4-e34e-4bc9-b7f8-a8e770ee361a | 11923320 | Surgical Procedures, Operative[mh] | Periprosthetic joint infection (PJI) is one of the primary reasons for revisions after total knee arthroplasty (TKA) and total hip arthroplasty (THA) and is associated with high morbidity rates and substantial economic burdens . Diagnosing PJI, particularly of the hip, is challenging due to the lack of a single definitive test . The diagnostic process requires an integrated evaluation of clinical symptoms, patient history, auxiliary examination results, and intraoperative findings . The complexity of the hip anatomy often impedes early detection of PJI, thereby delaying interventions and potentially exacerbating patient outcomes. Recent studies highlight the potential of serological markers such as white blood cell count (WBC), erythrocyte sedimentation rate (ESR), and C-reactive protein (CRP) in diagnosing PJI . The diagnostic criteria for PJI are based on updated criteria of the Musculoskeletal Infection Society (MSIS) ; the criteria for diagnosing PJI require meeting at least one primary diagnostic criterion or four minor criteria, with serological results playing a significant role. However, diagnosis remains comprehensive, involving both serological markers and clinical signs such as sinus tracts and pus. For patients with a high clinical suspicion of PJI, performing hip joint aspiration and culture and synovial fluid analysis is generally recommended. Still, joint puncture is an invasive operation that may cause iatrogenic infection and other risks . Variability in aspiration techniques and laboratory analyses can affect results. Therefore, quickly screening potential PJI patients through simple hematological indicators is crucial. There is a gap in research on the combined use of various hematological indicators for evaluating PJI risk post-THA. Developing a visual scoring system based on serum tests could enable rapid assessment of potential PJI in patients with THA, prompting further examinations and improving prognosis. This study aims to identify and validate PJI risk factors using the MSIS criteria and determine optimal thresholds. We intend to create a nomogram for quick screening of potential PJI after THA, facilitating early diagnosis and management, which could enhance patient care and reduce healthcare costs.
This retrospective study, approved by the Ethics Committee of the First Affiliated Hospital of Wenzhou Medical University, analyzed patients who underwent THA between January 2015 and December 2020. Patients were followed up with at 3-, 6-, and 12-months postsurgery to assess wound status and limb function recovery. A comprehensive hematological examination was typically recommended at 3 months after discharge and recorded in the system. The study was based on this independent database. Inclusion criteria were (1) an age ≥ 18 years, (2) complete pre- and postoperative medical records, and (3) continuous follow-up for at least 1 year. Exclusion criteria were (1) potential infectious diseases (respiratory, urinary, gastrointestinal, or skin/soft tissue) based on case records, admission examinations, physical examinations, and laboratory tests; (2) conditions that could affect serological indicators, including malignancies, severe liver dysfunction, coagulation disorders, and systemic inflammatory diseases (“Diseases including rheumatoid arthritis, systemic lupus erythematosus, psoriatic arthritis, ankylosing spondylitis, Sjogren’s syndrome, inflammatory bowel disease, systemic sclerosis, and autoimmune liver disease.”); (3) a diagnosis of aseptic loosening for noninfectious reasons (periprosthetic fractures, implant wear, instability, misalignment, or unexplained pain); and (4) history of joint revision arthroplasty. A total of 589 patient records post-THA were included in this study. These records, encompassing hospitalization, readmission, rehabilitation, and outpatient services, were thoroughly reviewed to determine outcomes. Patients’ baseline information, such as age, sex, body mass index (BMI), side of surgery (left/right), smoking and drinking history, and comorbidities, was obtained from their initial hospital admission electronic medical records following THA. Some patients had their initial THA surgery elsewhere, and data for these cases were obtained from readmission and outpatient records. Collected data included laboratory tests such as complete blood count, biochemistry, coagulation function, and serum markers. Laboratory data for patients with PJI were collected as part of routine presurgical examinations. Blood samples for fibrinogen, D-dimer, and other markers were taken on the morning after admission and then promptly sent to the medical laboratory center for testing. The laboratory data for patients No-PJI patients were primarily obtained from outpatient follow-up examinations. Through the database, we collected detailed hematological tests for patients, including liver function, kidney function, coagulation function, D-dimer, ESR, and CRP. Statistical analysis Continuous variables were expressed as mean ± standard deviation (SD) or median (range) and compared using an independent Student t -test or Mann–Whitney test. Qualitative data are described in numbers and percentages and compared using chi-squared or Fisher exact tests where appropriate. P < 0.05 variables were substituted into multivariate logistics regression to identify independent risk factors, and LR (Likelihood Ratio) was used for variable selection. When the Youden index (sensitivity + specificity − 1) was the largest, the sensitivity and specificity of each variable for the diagnosis of PJI were found through ROC curve analysis, and the cutoff value was determined. A predictive model was established, and a nomogram was created to calculate the total score of each risk factor as a risk score, providing a visual system for predicting PJI. The Hosmer–Lemeshow (H–L) goodness-of-fit test and calibration curve were used to evaluate the model’s fit. The area under the ROC and the optimal cutoff values were calculated to assess the model’s discriminative ability. Decision curve analysis (DCA) was used to evaluate the clinical validity and net benefit of the nomogram. Statistical analysis was performed using SPSS (version 18.0; SPSS Company, Chicago, IL, USA) and R software version 3.22, and ROC curves were constructed by Med-Calc 20.0. Statistical significance was set as P < 0.05.
Continuous variables were expressed as mean ± standard deviation (SD) or median (range) and compared using an independent Student t -test or Mann–Whitney test. Qualitative data are described in numbers and percentages and compared using chi-squared or Fisher exact tests where appropriate. P < 0.05 variables were substituted into multivariate logistics regression to identify independent risk factors, and LR (Likelihood Ratio) was used for variable selection. When the Youden index (sensitivity + specificity − 1) was the largest, the sensitivity and specificity of each variable for the diagnosis of PJI were found through ROC curve analysis, and the cutoff value was determined. A predictive model was established, and a nomogram was created to calculate the total score of each risk factor as a risk score, providing a visual system for predicting PJI. The Hosmer–Lemeshow (H–L) goodness-of-fit test and calibration curve were used to evaluate the model’s fit. The area under the ROC and the optimal cutoff values were calculated to assess the model’s discriminative ability. Decision curve analysis (DCA) was used to evaluate the clinical validity and net benefit of the nomogram. Statistical analysis was performed using SPSS (version 18.0; SPSS Company, Chicago, IL, USA) and R software version 3.22, and ROC curves were constructed by Med-Calc 20.0. Statistical significance was set as P < 0.05.
A total of 589 patients post-THA were included, of which 87 were PJI patients. The demographic characteristics and laboratory and clinical data of the PJI and No-PJI groups are summarized in Table . In the univariate analysis, potential risk factors for PJI after THA included age, BMI, CRP, ESR, WBC, platelet count (PC), D-dimer, and diabetes ( P < 0.05). After entering each potential risk variable from Table into a univariate logistics analysis, those with P < 0.05 were selected as candidate predictive variables to be included in the multivariate logistics. Ultimately, five independent risk factors were established: CRP (odds ratio [OR], 5.03; 95% confidence interval [CI] 2.74–9.12, P < 0.001), ESR (OR, 3.65; 95% CI 1.93–6.87, P < 0.001), PC (OR, 2.35; 95% CI 1.17–4.73, P = 0.017), D-dimer (OR, 3.55; 95% CI 1.60–7.88, P = 0.002), and polymorphonuclear neutrophils (PMN; OR, 2.01; 95% CI 1.03–3.93, P = 0.041) (Table ). ROC curve analysis evaluated the five independent risk factors’ area under curve (AUC) values, sensitivity, specificity, and optimal cutoff values (Table ; Fig. ). The ROC curve shows that ESR has the highest AUC (AUC, 0.816; 95% CI 0.782–0.847; P = 0.02), with a sensitivity of 78.16%, specificity of 69.32%, and the best cutoff point of > 20 mm/hr; CRP has the highest sensitivity (90.8%), an AUC of 0.805, and the best cutoff point of > 5.46 mg/L; and PC has the highest specificity (86.45%), an AUC of 0.534, and the best cutoff point of > 298 × 10 9 /L; A nomogram was constructed on the basis of the five independent risk factors mentioned above (Fig. ). In the nomogram, each independent risk factor is assigned a total score or a weighted total score (superscript), and the probability of PJI occurrence after post-THA is calculated according to the total score (subscript). The predictive performance of the model was assessed using the ROC curve (Fig. ). The AUC of the nomogram was 0.832 (95% CI 0.798–0.861), with an optimal cutoff value of ≤ 0.883. At this cutoff, the Youden index was 0.5252, the sensitivity was 77.01%, and the specificity was 75.51%, indicating good discrimination by the nomogram. The Hosmer–Lemeshow test and calibration curve were used to evaluate the calibration of the model. The P -value of the Hosmer–Lemeshow test was 0.225, indicating good calibration. The calibration curve closely approximated the ideal 45° line, demonstrating a good agreement between the model’s predictions and observations (Fig. ). Decision curve analysis (DCA) was used to evaluate the clinical utility of the nomogram. The DCA curve indicated that, when the threshold probability of PJI in patients with THA ranged from 0 to 0.8, the net benefit of using the nomogram was significantly higher than that of the “no intervention” and “all intervention” strategies (Fig. ), suggesting that the model is clinically useful.
Our study attempts to fill this gap by determining and establishing the optimal thresholds for risk factors associated with PJI after hip arthroplasty and proposing a diagnostic nomogram for preliminary assessment. The nomogram includes CRP, ESR, PC, D-dimer, and PMN as predictive indicators for PJI following THA. Previous studies have established that serum ESR and CRP are susceptible tests for diagnosing PJI [ , – ] and are practical screening tools. Multivariate logistic regression analysis and ROC curves have shown that CRP has the highest sensitivity, with the best cutoff point being 5.46 mg/L, close to the upper standard limit. One study reviewed a 1.2% reoperation rate within the first 6 weeks post-THA, with some cases presenting with PJI. It was found that serum CRP serves as an excellent screening test. However, the optimal threshold for diagnosing PJI in the acute postoperative phase is higher than the traditional values used for chronic PJI diagnosis . Another study set the optimal cutoff value for diagnosing PJI after hip arthroplasty using CRP at 10 mg/L, with a sensitivity of 85.1% and a specificity of 67.6% . These thresholds are significantly higher than those in our study. We believe there is a significant difference in CRP threshold values based on postoperative time thresholds . The above studies only defined the postoperative time thresholds of 6 weeks and 4 weeks, whereas CRP levels gradually return to normal as the postoperative period extends . According to the definitions by MSIS, PJI is classified into two types, acute and chronic, on the basis of a 90-day postoperative period, but this classification is currently controversial. This binary classification does not reflect the infection continuum and may not correspond to the infection’s pathophysiology or the pathogens’ behavior. Factors unique to each patient, such as immune status and comorbidities, can significantly influence the timing and presentation of PJI . In this study, we excluded patients with systemic inflammatory diseases, because these can cause inflammatory reactions, which may lead to elevated laboratory indicators, thereby interfering with the accurate diagnosis of PJI. Most PJIs are thought to be caused by low-virulence microorganisms, which may trigger a weak or absent CRP response . These microorganisms preferentially adhere to the implant’s surface, forming a biofilm and evading the immune system, thus reducing the inflammatory response [ – ]. CRP, with its high sensitivity, is susceptible to influence from other infections and helps monitor the treatment of acute conditions rather than diagnosing PJI . To minimize false positives, it may be necessary to consider changes in CRP levels in PJI patients in the context of their overall health status. The longer half-life of ESR suggests it is more suitable for assessing chronic infections. Our study indicates that ESR has potentially higher diagnostic value, with the highest area under the curve, and warrants special attention in patients post-THA. Our findings are similar to those of Stephen P. Maier et al. , where ESR may remain elevated in chronic infections (ongoing active infection), while CRP is normal. Compared with No-PJI patients, PJI patients had higher PMN levels within the normal range, with a statistically significant difference. This is consistent with a study analyzing 1856 revision surgeries, which found the optimal threshold for percentage of PMN for sensitivity and specificity to be 69% . In another study, when a cutoff of 90% was used for PMN, the summarized estimates of sensitivity, specificity, and LR for septic arthritis PMN were 60%, 78%, and 2.7 , respectively. However, our study’s best cutoff value for PMN only showed low sensitivity and moderate specificity. In our PJI cohort, PMN levels did not indicate systemic inflammation but gradually increased, suggesting PMN evaluation might be necessary, but its diagnostic cutoff remains contentious. Systemic or local infections can alter coagulation and fibrinolytic activity, providing a new potential avenue for PJI diagnosis. During infection and inflammation, activated platelets can inhibit pathogen growth by activating immune cells and promoting their clearance . Additionally, related research has confirmed that platelets become significant coordinators of inflammation and innate and adaptive immune responses through interactions with monocytes, neutrophils, lymphocytes, and endothelial cells . Therefore, platelet count has emerged as a vital candidate indicator for PJI diagnosis in our study. Although the sensitivity of platelet count is not as high as CRP or ESR, it has shown the highest specificity (86.45%), indicating its potential contributory value in PJI diagnosis. In previous studies, serum D-dimer has been considered a promising biomarker for PJI, showing higher sensitivity and specificity than ESR and CRP. Still, our study did not confirm this finding. Although D-dimer is a risk factor, its sensitivity and specificity did not exceed those of CRP or ESR. In the research by Shahi et al. , racial differences in D-dimer levels between predominantly white (European American) and African American populations and our Asian cohort may account for this discrepancy. Therefore, the role of D-dimer as a diagnostic biomarker seems limited and should not replace established serum parameters. Our study, through calibration analysis, has proven that traditional risk factor screening and a robust predictive nomogram are superior to univariate models. The DCA has verified the significant net benefit of the model, indicating its practical utility. According to the total score derived from Youden’s index, the probability of PJI is 0.128, with the highest sensitivity (82.76%) and specificity (70.12%). The nomogram suggests that performing joint fluid aspiration or culture is recommended for patients with a score greater than 200. For those with a score lower than 200, close monitoring and follow-up are advised, paying attention to the dynamic changes in serum markers and nomogram scores. However, the study has several limitations. First, this is a retrospective study, and selection bias is inevitable. Secondly, due to the limited number of PJI patients, this preliminary prediction model requires multi-center external validation with a larger sample size. Third, as with most studies, we could not include all confounding factors or completely rule out the question of survivor bias. PJI often presents with systemic inflammatory diseases. By excluding these patients, our study sample may differ from the actual clinical population, thereby limiting the generalizability of our findings. This exclusion criterion means that our results may not be fully applicable to PJI patients with underlying autoimmune conditions. Future research should consider including patients with systemic inflammatory diseases to evaluate the predictive value of laboratory markers across a broader and more representative population. In summary, the results of this study suggest that CRP, ESR, D-dimer, PMN, and PC may be independent risk factors for PJI after THA and have potential roles in the diagnosis of PJI. Furthermore, this study introduces a predictive model with good accuracy, which may assist clinicians in predicting the risk of PJI after THA, thereby facilitating timely medical interventions for both clinicians and patients. The effectiveness of the nomogram-based prediction model in early detection of PJI and improvement of patient outcomes warrants further investigation.
|
Reformvorhaben „Gelegenheits-Onkochirurgie“ | e1c3d40a-dc55-4896-92ee-079c0a5dbcb8 | 11933218 | Internal Medicine[mh] | Nach wie vor wird Krebs als zweithäufigste Todesursache in Deutschland in der Todesursachenstatistik des Statistischen Bundesamtes geführt . Die Zahl der Neuerkrankungen steigt zudem stetig an . Dass eine Leistungssteuerung und Spezialisierungskonzepte in der Medizin Potenziale bergen, die die Versorgung und die Überlebenschancen von Patientinnen und Patienten positiv beeinflussen, scheint bereits gesundheitspolitische Aufmerksamkeit zu genießen: So dient beispielsweise der gesetzliche Auftrag an den Gemeinsamen Bundesausschuss (G-BA), elektive stationäre Leistungen zu identifizieren und zu katalogisieren, bei denen ein evidenter Zusammenhang zwischen medizinischer Behandlungshäufigkeit und Ergebnisqualität besteht (Mindestmengen; ). Für die Krebsmedizin im Speziellen zeigte nicht zuletzt das durch den Innovationsfonds beim G‑BA geförderte Projekt „WiZen“ (Wirksamkeit der Versorgung in onkologischen Zentren), wie sich „Zentreneffekte“ auf die Überlebenschancen Tumorekrankter auswirken: Basierend auf bundesweiten GKV-Abrechnungsdaten sowie Daten aus vier klinischen Krebsregistern konnte demonstriert werden, dass die Erstbehandlung elf unterschiedlicher Krebserkrankungen in zertifizierten Zentren im Vergleich zu nichtzertifizierten Einrichtungen durchweg mit einer niedrigeren Mortalität assoziiert ist. Die relativen Überlebensvorteile lagen bei den verschiedenen Entitäten zwischen 3 und 23 % . Darüber hinaus konnten positive Effekte innerklinischer Qualitätsindikatoren im Zusammenhang mit der Behandlung in einem nach den Kriterien der Deutschen Krebsgesellschaft zertifizierten Zentrum gezeigt werden, wie bspw. ein verbessertes Komplikationsmanagement und eine sinkende Krankenhaussterblichkeit . Um die Konzentrationsbestrebungen auf dem Gebiet der onkochirurgischen Versorgung zu fördern, sieht der Gesetzentwurf des KHVVG die Einführung des neuen § 40 Krankenhausfinanzierungsgesetz (KHG) vor. Die Regelung normiert Instruktionen an das Bundesinstitut für Arzneimittel und Medizinprodukte (BfArM) und das Institut für das Entgeltsystem im Krankenhaus (InEK), um die Spezialisierung und Bündelung bei der Erbringung chirurgischer Leistungen im Zusammenhang mit einer onkologischen Diagnose (onkochirurgische Leistungen) zu unterstützen. Ausweislich der Begründung im Gesetzentwurf des KHVVG soll ab dem Jahr 2027 ein wesentlicher Anteil der Finanzierung für onkochirurgische Leistungen in einem onkochirurgischen Indikationsbereich für Krankenhäuser mit geringen Fallzahlen, die zusammen 15 % der Fälle des einschlägigen Indikationsbereichs erbringen, entfallen, wodurch eine Leistungsverlagerung in der Versorgungslandschaft zu erwarten ist. Eine plakative Darstellung dieser Logik ist in der Abb. dargestellt. Der Gesetzgeber statuiert im Gesetzentwurf die Intension, durch die Bündelung von Ressourcen und Expertise in ausgewählten Gesundheitseinrichtungen die Qualität und Effizienz der onkochirurgischen Versorgung weiter zu verbessern. Die Effekte müssen leistungserbringerseitig antizipiert und in entsprechende medizinstrategische Überlegungen zur regionalen Gesundheitsversorgung einfließen.
Das Gesetzesvorhaben soll den Anreiz setzen, Gelegenheitsversorgung auf dem Gebiet der onkochirurgischen Versorgung zu vermeiden, um die Qualität dieser Behandlungen durch Zentralisierung zu verbessern. Die Hauptziele des Reformvorhabens lassen sich in diesem Kontext auf drei wesentliche Punkte zusammenfassen: Förderung einer sachgerechten Konzentration von Versorgungsstrukturen auf dem Gebiet der Onkochirurgie, Sicherstellung einer bedarfsgerechten onkochirurgischen Versorgung, Gewährleistung einer qualitativ hochwertigen Patientenversorgung durch gezielte Ressourcenallokation und Prozessoptimierung mit dem Potenzial ökonomischer Positiveffekte.
Das Reformvorhaben umfasst nach derzeitigem Stand drei wesentliche Komponenten, darunter die Definition des einschlägigen Leistungsgerüsts, die Identifikation der betroffenen Standorte sowie das Abrechnungsverbot für bestimmte Entgelte. Prozessdarstellung Definition des Leistungsgerüsts Das BfArM und das InEK sind verantwortlich, onkochirurgische Leistungen mithilfe von ICD-Codes und Operationen- und Prozedurenschlüsseln (OPS) zu definieren und diese onkochirurgischen Indikationsbereichen zuzuordnen. Diese Codes werden entsprechend onkochirurgischer Indikationsbereiche kategorisiert, um die Identifizierung und Erfassung onkochirurgischer Leistungen zu erleichtern. Auf der Basis der Klassifikationsversionen des Jahres 2023 wird das BfArM beauftragt, sämtliche onkochirurgischen Leistungen anhand von ICD- und OPS-Codes zu definieren und diese Aufstellung erstmals zum 28.02.2025 an das InEK zu übermitteln. In einem zweiten Schritt soll sodann das InEK Indikationsbereiche für alle onkochirurgischen Leistungen festlegen. In diesem Kontext ist erwähnenswert, dass eine Stellungnahme bzw. Beteiligung der Deutschen Krebsgesellschaft in beiden Schritten vorgesehen ist, wodurch eine Orientierung an den in der Medizin praxisrelevanten onkologischen Bereichen sichergestellt werden kann. Darüber hinaus erhält das InEK den Auftrag, die ICD- und OPS-Codes aus der Aufstellung des BfArM den Leistungsgruppen und definierten onkochirurgischen Indikationsbereichen zuzuordnen. Eine barrierefreie Veröffentlichung dieses nach Leistungsgruppen und onkochirurgischen Indikationsbereichen differenzierten Verzeichnisses ist erstmalig zum 30.04.2025 auf der Internetseite des InEK zu erwarten. Abb. veranschaulicht den Prozess grafisch. Ab dem Jahr 2026 soll schließlich eine jährliche Überarbeitung der Datengrundlage (hier: Klassifikationen) und Veröffentlichung der relevanten Liste (im Folgenden: Liste nach § 40 Abs. 1 S. 4 KHG) zum 31.12. eines Jahres durch das InEK erfolgen. Prozessdarstellung Standortidentifikation Das InEK erstellt unter Hinzuziehung der nach Leistungsgruppen und Indikationsbereichen differenzierten Liste nach § 40 Abs. 1 S. 4 KHG (s. Leistungsgerüst) und den Leistungsdaten nach § 21 Krankenhausentgeltgesetz (KHEntgG) eine Auswertung zur Identifikation aller Krankenhausstandorte, die in 2023 onkochirurgische Leistungen erbracht haben. Kongruent wird auch bei den sog. § 21-er Daten auf das Jahr 2023 Bezug genommen. Sind die Krankenhausstandorte, welche im Jahr 2023 Leistungen der Liste nach § 40 Abs. 1 S. 4 KHG erbracht haben, identifiziert, werden diese je onkochirurgischem Indikationsbereich aufsteigend nach Fallzahl sortiert. Aus dieser Gesamtheit werden diejenigen Krankenhausstandorte, die die wenigsten Fälle mit onkochirurgischen Leistungen und kumuliert 15 % dieser Fälle aller Krankenhausstandorte in einem onkochirurgischen Indikationsbereich aufweisen, bis zum 31.05.2025 durch das InEK barrierefrei auf dessen Internetseite veröffentlicht (im Folgenden: Liste nach § 40 Abs. 2 S. 2 KHG). Nach dem Wortlaut dieser Regelung ist/sind damit auch der/die Krankenhausstandort/e zu veröffentlichen, mit dessen/deren Fällen die 15 %-Perzentile eines relevanten Indikationsbereichs erreicht bzw. überschritten wird/werden. Durch diese Veröffentlichung der Liste nach § 40 Abs. 2 S. 2 KHG erhalten Träger Kenntnis über das für sie geltende Abrechnungsverbot nach § 8 Abs. 4 S. 6 KHEntgG (neue Fassung). Der Prozess der Standortidentifikation wird in Abb. dargestellt. Die Evaluation eines „Konzentrationseffektes“ soll erstmals zum 30.06.2032 erfolgen. Darüber hinaus sollen o. g. Informationen perspektivisch im Transparenzverzeichnis nach § 135d SGB V verarbeitet werden, um damit Patientinnen und Patienten bei der Suche nach onkochirurgischen Behandlungsangeboten zu unterstützen. Aus Patientensicht werden nur solche Krankenhausstandorte angezeigt, bei denen aufgrund der Häufigkeit der Eingriffe von einer besonderen Expertise der Einrichtung und höheren Behandlungsqualität auszugehen ist. Finanzielle Restriktionen (Abrechnungsverbot) Eine neue Bestimmung in § 8 Abs. 4 des Krankenhausentgeltgesetzes (KHEntgG) soll ein Abrechnungsverbot für bestimmte Leistungen einführen, die in Zusammenhang mit der Erbringung onkochirurgischer Leistungen stehen. Ab dem Jahr 2027 ist vorgesehen, Krankenhausstandorten für Fälle, bei denen Leistungen in einem onkochirurgischen Indikationsbereich der Liste nach § 40 Abs. 1 S. 4 KHG (Leistungsgerüst) erbracht werden und der Krankenhausstandort auf der Liste nach § 40 Abs. 2 S. 2 KHG (Standortidentifikation) benannt ist, die Abrechnungsmöglichkeit der Entgelte nach § 7 Abs. 1 S. 1 Nr. 1–6 sowie Nr. 8 KHEntgG faktisch zu entleeren. Bei Einbezug aller im Gesetzentwurf des KHVVG dargestellten Änderungsvorschlägen sollen nicht mehr vergütet werden: Fallpauschalen und damit die zukünftigen rDRG (§ 7 Abs. 1 S. 1 Nr. 1 KHEntgG), bundeseinheitlich bepreiste Zusatzentgelte (§ 7 Abs. 1 S. 1 Nr. 2 KHEntgG), unbepreiste Zusatzentgelte (§ 7 Abs. 1 S. 1 Nr. 3 KHEntgG), Zu- und Abschläge (§ 7 Abs. 1 S. 1 Nr. 4 KHEntgG), Entgelte für besondere Einrichtungen (§ 7 Abs. 1 S. 1 Nr. 5 KHEntgG), Entgelte für neue Untersuchungs- und Behandlungsmethoden (§ 7 Abs. 1 S. 1 Nr. 6 KHEntgG) sowie krankenhausindividuelle Tagesentgelte (neu: § 7 Abs. 1 S. 1 Nr. 8 KHEntgG). Lediglich die tagesbezogenen Pflegeentgelte zur Abzahlung des Pflegebudgets (§ 7 Abs. 1 S. 1 Nr. 6a KHEntgG), die Vorhaltevergütung zur Abzahlung des Vorhaltebudgets (neu: 7 Abs. § 6b KHEntgG) sowie der Pflegezuschlag (§ 7 Abs. 1 S. 1 Nr. 7 KHEntgG) sollen für Krankenhäuser abrechenbar bleiben. Auf diesem Wege soll die Leistungskonzentration für onkochirurgische Behandlungsangebote und die damit einhergehende Spezialisierung gefördert werden. Gleichzeitig werden etwaige finanzielle Anreize in erheblichem Maße ausgeräumt, was annahmegemäß zu einer Vermeidung onkochirurgischer „Gelegenheitsversorgung“ in der Krankenhauslandschaft führen wird.
Das BfArM und das InEK sind verantwortlich, onkochirurgische Leistungen mithilfe von ICD-Codes und Operationen- und Prozedurenschlüsseln (OPS) zu definieren und diese onkochirurgischen Indikationsbereichen zuzuordnen. Diese Codes werden entsprechend onkochirurgischer Indikationsbereiche kategorisiert, um die Identifizierung und Erfassung onkochirurgischer Leistungen zu erleichtern. Auf der Basis der Klassifikationsversionen des Jahres 2023 wird das BfArM beauftragt, sämtliche onkochirurgischen Leistungen anhand von ICD- und OPS-Codes zu definieren und diese Aufstellung erstmals zum 28.02.2025 an das InEK zu übermitteln. In einem zweiten Schritt soll sodann das InEK Indikationsbereiche für alle onkochirurgischen Leistungen festlegen. In diesem Kontext ist erwähnenswert, dass eine Stellungnahme bzw. Beteiligung der Deutschen Krebsgesellschaft in beiden Schritten vorgesehen ist, wodurch eine Orientierung an den in der Medizin praxisrelevanten onkologischen Bereichen sichergestellt werden kann. Darüber hinaus erhält das InEK den Auftrag, die ICD- und OPS-Codes aus der Aufstellung des BfArM den Leistungsgruppen und definierten onkochirurgischen Indikationsbereichen zuzuordnen. Eine barrierefreie Veröffentlichung dieses nach Leistungsgruppen und onkochirurgischen Indikationsbereichen differenzierten Verzeichnisses ist erstmalig zum 30.04.2025 auf der Internetseite des InEK zu erwarten. Abb. veranschaulicht den Prozess grafisch. Ab dem Jahr 2026 soll schließlich eine jährliche Überarbeitung der Datengrundlage (hier: Klassifikationen) und Veröffentlichung der relevanten Liste (im Folgenden: Liste nach § 40 Abs. 1 S. 4 KHG) zum 31.12. eines Jahres durch das InEK erfolgen.
Das InEK erstellt unter Hinzuziehung der nach Leistungsgruppen und Indikationsbereichen differenzierten Liste nach § 40 Abs. 1 S. 4 KHG (s. Leistungsgerüst) und den Leistungsdaten nach § 21 Krankenhausentgeltgesetz (KHEntgG) eine Auswertung zur Identifikation aller Krankenhausstandorte, die in 2023 onkochirurgische Leistungen erbracht haben. Kongruent wird auch bei den sog. § 21-er Daten auf das Jahr 2023 Bezug genommen. Sind die Krankenhausstandorte, welche im Jahr 2023 Leistungen der Liste nach § 40 Abs. 1 S. 4 KHG erbracht haben, identifiziert, werden diese je onkochirurgischem Indikationsbereich aufsteigend nach Fallzahl sortiert. Aus dieser Gesamtheit werden diejenigen Krankenhausstandorte, die die wenigsten Fälle mit onkochirurgischen Leistungen und kumuliert 15 % dieser Fälle aller Krankenhausstandorte in einem onkochirurgischen Indikationsbereich aufweisen, bis zum 31.05.2025 durch das InEK barrierefrei auf dessen Internetseite veröffentlicht (im Folgenden: Liste nach § 40 Abs. 2 S. 2 KHG). Nach dem Wortlaut dieser Regelung ist/sind damit auch der/die Krankenhausstandort/e zu veröffentlichen, mit dessen/deren Fällen die 15 %-Perzentile eines relevanten Indikationsbereichs erreicht bzw. überschritten wird/werden. Durch diese Veröffentlichung der Liste nach § 40 Abs. 2 S. 2 KHG erhalten Träger Kenntnis über das für sie geltende Abrechnungsverbot nach § 8 Abs. 4 S. 6 KHEntgG (neue Fassung). Der Prozess der Standortidentifikation wird in Abb. dargestellt. Die Evaluation eines „Konzentrationseffektes“ soll erstmals zum 30.06.2032 erfolgen. Darüber hinaus sollen o. g. Informationen perspektivisch im Transparenzverzeichnis nach § 135d SGB V verarbeitet werden, um damit Patientinnen und Patienten bei der Suche nach onkochirurgischen Behandlungsangeboten zu unterstützen. Aus Patientensicht werden nur solche Krankenhausstandorte angezeigt, bei denen aufgrund der Häufigkeit der Eingriffe von einer besonderen Expertise der Einrichtung und höheren Behandlungsqualität auszugehen ist.
Eine neue Bestimmung in § 8 Abs. 4 des Krankenhausentgeltgesetzes (KHEntgG) soll ein Abrechnungsverbot für bestimmte Leistungen einführen, die in Zusammenhang mit der Erbringung onkochirurgischer Leistungen stehen. Ab dem Jahr 2027 ist vorgesehen, Krankenhausstandorten für Fälle, bei denen Leistungen in einem onkochirurgischen Indikationsbereich der Liste nach § 40 Abs. 1 S. 4 KHG (Leistungsgerüst) erbracht werden und der Krankenhausstandort auf der Liste nach § 40 Abs. 2 S. 2 KHG (Standortidentifikation) benannt ist, die Abrechnungsmöglichkeit der Entgelte nach § 7 Abs. 1 S. 1 Nr. 1–6 sowie Nr. 8 KHEntgG faktisch zu entleeren. Bei Einbezug aller im Gesetzentwurf des KHVVG dargestellten Änderungsvorschlägen sollen nicht mehr vergütet werden: Fallpauschalen und damit die zukünftigen rDRG (§ 7 Abs. 1 S. 1 Nr. 1 KHEntgG), bundeseinheitlich bepreiste Zusatzentgelte (§ 7 Abs. 1 S. 1 Nr. 2 KHEntgG), unbepreiste Zusatzentgelte (§ 7 Abs. 1 S. 1 Nr. 3 KHEntgG), Zu- und Abschläge (§ 7 Abs. 1 S. 1 Nr. 4 KHEntgG), Entgelte für besondere Einrichtungen (§ 7 Abs. 1 S. 1 Nr. 5 KHEntgG), Entgelte für neue Untersuchungs- und Behandlungsmethoden (§ 7 Abs. 1 S. 1 Nr. 6 KHEntgG) sowie krankenhausindividuelle Tagesentgelte (neu: § 7 Abs. 1 S. 1 Nr. 8 KHEntgG). Lediglich die tagesbezogenen Pflegeentgelte zur Abzahlung des Pflegebudgets (§ 7 Abs. 1 S. 1 Nr. 6a KHEntgG), die Vorhaltevergütung zur Abzahlung des Vorhaltebudgets (neu: 7 Abs. § 6b KHEntgG) sowie der Pflegezuschlag (§ 7 Abs. 1 S. 1 Nr. 7 KHEntgG) sollen für Krankenhäuser abrechenbar bleiben. Auf diesem Wege soll die Leistungskonzentration für onkochirurgische Behandlungsangebote und die damit einhergehende Spezialisierung gefördert werden. Gleichzeitig werden etwaige finanzielle Anreize in erheblichem Maße ausgeräumt, was annahmegemäß zu einer Vermeidung onkochirurgischer „Gelegenheitsversorgung“ in der Krankenhauslandschaft führen wird.
Die Reformierung onkochirurgischer Leistungen im Rahmen des KHVVG stellt grundsätzlich, basierend auf den Erkenntnissen der Mindestmengen- und Zentrumseffekte (s. oben), einen proaktiven Schritt zur Verbesserung der Qualität und Effizienz der onkologischen Versorgung dar. Durch die Förderung von Spezialisierung und Konzentration zielt die Reform darauf ab, den sich wandelnden Bedürfnissen der Patientinnen und Patienten gerecht zu werden und die Ressourcennutzung im Gesundheitssystem zu optimieren. Eine kontinuierliche Einbindung der Stakeholder und Evaluation wird entscheidend sein, um die erfolgreiche Umsetzung und langfristige Wirkung der Reforminitiative sicherzustellen. Allerdings sind unserer Auffassung nach zum jetzigen Zeitpunkt noch eine Reihe an Punkten ungeklärt, die im Kontext des Reformvorhabens jedoch einer Klärung bedürfen: So bleibt beispielsweise der Umgang mit Notfällen bzw. Zufallsbefunden im aktuellen Gesetzentwurf unberücksichtigt und eine Evaluation der Maßnahmen ist erstmals zum 30.06.2032 vorgesehen – vor dem Hintergrund der dynamischen Entwicklungen um die „Krankenhausreform“ ein aus unserer Sicht sehr langer Zeitraum. Weiterhin stellt sich uns die Frage, ob die definierte Größe von 15 % arbiträr festgelegt oder auf einer wissenschaftlichen Herleitung beruht. Sollte ersteres zutreffen, muss diese Zahl aus unserer Sicht nach einer Übergangszeit kritisch auf Effekte evaluiert und bestenfalls auch mit der Deutschen Krebsgesellschaft oder anderen relevanten Fachgesellschaften abgestimmt werden. Unter wissenschaftlichen Gesichtspunkten sollten die Maßnahmen eng im Sinne der tatsächlichen Auswirkungen auf die Behandlungs- und Prozessqualität begleitet, erfasst und perspektivisch näher hinsichtlich der unterstellten Effizienz- und Qualitätssteigerungseffekte in der stationären Versorgung onkologischer Patientinnen und Patienten quantifiziert werden. Dies insbesondere vor dem Hintergrund, dass es sich nach unserer Interpretation um eine einmalige „Bereinigung“ von Versorgungsstrukturen handelt (s. erstmalige Maßnahmenevaluation im Jahr 2032). Darüber hinaus möchten wir darauf hinweisen, dass zum gegenwärtigen Zeitpunkt keine Regelung zu notwendigen investiven Mitteln in die Infrastruktur solcher Krankenhäuser mitgedacht wird, die perspektivisch mit den Leistungsverlagerungen aus benachbarten Krankenhäusern konfrontiert sein werden. Die Berücksichtigung eines potenziellen „Zentralisierungseffektes“ und folgelogischer monetärer Konsequenzen auf die verbleibenden Leistungserbringer im System findet auch keine explizite Erwähnung in den Fördertatbeständen des neu einzurichtenden Transformationsfonds nach § 12b KHG. Zudem scheint der gesetzgeberische Gedanke, die Leistungskonzentration lediglich an eine mengenmäßige Komponente zu knüpfen, nicht weit genug zu gehen. So könnte darüber nachgedacht werden, die Initiative flankierende Qualitätsanforderungen an die potenziell im System verbleibenden Leistungserbringer zu stellen und gesetzlich zu verankern (z. B. analog Zentrumszertifizierung nach Deutscher Krebsgesellschaft). Auch sollte der Harmonisierungsgrad mit den Zentrumsregelungen des G‑BA näher beleuchtet werden, um etwaige Redundanzen zu identifizieren. Mit dem Regelwerk liegen bereits seit dem Jahr 2020 bundeseinheitliche Kriterien vor, welche als Anforderungen an die Ausweisung und Finanzierung von G‑BA-Zentren formuliert sind (z. B. Mindestfallzahlen, Forschungstätigkeiten, Kooperationen, Art und Anzahl von Fachabteilungen). Als positiv hervorheben möchten wir die Absicht, die Deutsche Krebsgesellschaft in die Leistungsidentifikation einzubeziehen. Hierdurch ist notwendiger wissenschaftlicher Sachverstand frühzeitig im Prozess repräsentiert. Nach unserer Auffassung könnte die Einflussnahme der Deutschen Krebsgesellschaft durch ihre Stellungnahmen detaillierter im Gesetzestext spezifiziert werden. Strategische Handlungsfelder aus Sicht des Universitätsklinikums Freiburg (UKF ) Analyse des Versorgungsbedarfs in der Region: Unerlässlich ist unseres Erachtens eine frühestmögliche, z. B. datenbankgestützte Begutachtung der krankenhauseigenen Versorgungsregion hin auf die Systemrelevanz onkologischer Leistungsangebote, auch unter Einbezug des Angebots benachbarter Krankenhäuser. Hieraus lassen sich erste Erkenntnisse hinsichtlich der Notwendigkeit der Bildung von Leistungsclustern abschätzen. Problematisch sind dabei die Grenzen der zur Verfügung stehenden Datengrundlagen (z. B. Qualitätsberichte) bis zur in Aussicht gestellten Veröffentlichung durch das InEK im Jahr 2025. Regionale intersektorale Vernetzung mit stationären Leistungserbringern: Von zentralem medizinstrategischem Interesse ist aus unserer Perspektive die Intensivierung der Vernetzung mit Partnerkrankenhäusern im Sinne abgestimmter patientenorientierter Schwerpunktsetzungen bis hin zum Austausch von Leistungsgruppen in der Versorgungslandschaft. Krankenhäuser sind angehalten, intelligente Wege zu finden, um auf innovative und medizinisch sinnvolle Art Hand in Hand zum Wohle der onkologischen Patientinnen und Patienten zu handeln. Zum Beispiel könnten vermehrt „Chefarzt-Modelle“ (Führung mehrerer Fachabteilungen in Personalunion) etabliert oder regionale Absprachen zu Leistungsaustauschen getroffen werden (z. B. onkochirurgische Eingriffe gesamthaft an ein zertifiziertes Zentrum, dafür die chirurgische Grund- und Regelversorgung an ein kompetentes, qualitätsgesichertes Partnerkrankenhaus verlagern). Aus- und Weiterbildungskonzepte können in diesem Kontext mitgedacht werden. Digitale Infrastruktur ausweiten: Der konsequente Ausbau zur Entwicklung eines Gesundheitssystems, das die Chancen der Digitalisierung für eine neue, leistungsfähigere Form der Gesundheitsversorgung zum Wohle von Patientinnen und Patienten nutzt und die Potenziale der Gesundheitswirtschaft hebt, ist aus unserer Sicht ein Schlüsselfaktor für die erfolgreiche Verwirklichung des Reformvorhabens. Flankierende Maßnahmen: Teilweise können die vorgeschlagenen Maßnahmen als Eingriff in das ärztliche Standesrecht interpretiert werden. Unser Appell an dieser Stelle ist daher, auf Bundesebene die entsprechenden berufs- und standesrechtlichen Vorkehrungen zu treffen, um die Zielverwirklichung des Reformvorhabens nicht zu konterkarieren. Dies gilt auch für ungenutzte Chancen im Hinblick auf die Zurverfügungstellung notwendiger finanzieller Mittel zur Ertüchtigung einer bedarfsgerechten Krankenhausinfrastruktur. Darüber hinaus halten wir eine durchgehende wissenschaftliche Begleitung sowie Evaluation der geplanten Maßnahmen für den Erfolg des Reformvorhabens und dessen Weiterentwicklung für unentbehrlich.
) Analyse des Versorgungsbedarfs in der Region: Unerlässlich ist unseres Erachtens eine frühestmögliche, z. B. datenbankgestützte Begutachtung der krankenhauseigenen Versorgungsregion hin auf die Systemrelevanz onkologischer Leistungsangebote, auch unter Einbezug des Angebots benachbarter Krankenhäuser. Hieraus lassen sich erste Erkenntnisse hinsichtlich der Notwendigkeit der Bildung von Leistungsclustern abschätzen. Problematisch sind dabei die Grenzen der zur Verfügung stehenden Datengrundlagen (z. B. Qualitätsberichte) bis zur in Aussicht gestellten Veröffentlichung durch das InEK im Jahr 2025. Regionale intersektorale Vernetzung mit stationären Leistungserbringern: Von zentralem medizinstrategischem Interesse ist aus unserer Perspektive die Intensivierung der Vernetzung mit Partnerkrankenhäusern im Sinne abgestimmter patientenorientierter Schwerpunktsetzungen bis hin zum Austausch von Leistungsgruppen in der Versorgungslandschaft. Krankenhäuser sind angehalten, intelligente Wege zu finden, um auf innovative und medizinisch sinnvolle Art Hand in Hand zum Wohle der onkologischen Patientinnen und Patienten zu handeln. Zum Beispiel könnten vermehrt „Chefarzt-Modelle“ (Führung mehrerer Fachabteilungen in Personalunion) etabliert oder regionale Absprachen zu Leistungsaustauschen getroffen werden (z. B. onkochirurgische Eingriffe gesamthaft an ein zertifiziertes Zentrum, dafür die chirurgische Grund- und Regelversorgung an ein kompetentes, qualitätsgesichertes Partnerkrankenhaus verlagern). Aus- und Weiterbildungskonzepte können in diesem Kontext mitgedacht werden. Digitale Infrastruktur ausweiten: Der konsequente Ausbau zur Entwicklung eines Gesundheitssystems, das die Chancen der Digitalisierung für eine neue, leistungsfähigere Form der Gesundheitsversorgung zum Wohle von Patientinnen und Patienten nutzt und die Potenziale der Gesundheitswirtschaft hebt, ist aus unserer Sicht ein Schlüsselfaktor für die erfolgreiche Verwirklichung des Reformvorhabens. Flankierende Maßnahmen: Teilweise können die vorgeschlagenen Maßnahmen als Eingriff in das ärztliche Standesrecht interpretiert werden. Unser Appell an dieser Stelle ist daher, auf Bundesebene die entsprechenden berufs- und standesrechtlichen Vorkehrungen zu treffen, um die Zielverwirklichung des Reformvorhabens nicht zu konterkarieren. Dies gilt auch für ungenutzte Chancen im Hinblick auf die Zurverfügungstellung notwendiger finanzieller Mittel zur Ertüchtigung einer bedarfsgerechten Krankenhausinfrastruktur. Darüber hinaus halten wir eine durchgehende wissenschaftliche Begleitung sowie Evaluation der geplanten Maßnahmen für den Erfolg des Reformvorhabens und dessen Weiterentwicklung für unentbehrlich.
Medizinstrategisch bleiben vor dem Hintergrund des noch laufenden politischen Prozesses noch mannigfaltige Unsicherheiten. Unserer Meinung nach sollten Krankenhäuser jedoch bereits jetzt proaktiv das Gespräch mit benachbarten Krankenhäusern suchen und Kooperationsgespräche zur Gestaltung der regionalen Gesundheitsversorgung führen, um die lokale Allokation onkologischer Patientinnen und Patienten im Sinne einer bestmöglichen Behandlung an einem Zentrum optimal abzubilden, den Verlust von Patientinnen und Patienten an den betroffenen Standorten abzufedern sowie auf der anderen Seite den Patientenaufwuchs bei den Einrichtungen, die weiterhin onkochirurgisch versorgen werden, vorzubereiten. Die oben skizzierten Handlungsfelder sind von globaler Natur und können daher, unabhängig von der Versorgungsstufe eines Krankenhauses, Impulse für die leistungserbringerseitige Auseinandersetzung mit dem Reformvorhaben „Gelegenheits-Onkochirurgie“ setzen. Mutmaßlich werden aber insbesondere universitäre Standorte und Maximalversorger mit einem zusätzlichen Aufkommen onkologischer Patienten rechnen müssen, welche über die aktuell vorhandenen – personellen sowie infrastrukturellen – Versorgungskapazitäten hinaus behandelt werden müssen. Einschränkend muss gesagt werden, dass die relevanten Codes frühestens im ersten Quartal, realistisch auch erst im zweiten Quartal 2025 identifiziert und öffentlichkeitswirksam zugänglich werden; entsprechende Berechnungsmodelle, insbesondere unter regionalen Gesichtspunkten, können erst nach Bekanntwerden der einschlägigen Kodierungen erstellt werden. Eine belastbare detailgenaue Analyse ist damit derzeit nicht möglich. Diese muss, wie oben bereits erwähnt, neben medizinisch-prozessualen Gesichtspunkten und der neu zu organisierenden Ressourcenallokation an den betroffenen Standorten auch zwingend die wissenschaftliche Aufarbeitung des Reformvorhabens zu Qualitäts- und Strukturaspekten und somit seines Gelingens berücksichtigen.
|
Establishment of reference intervals for hematological parameters of adult population in the western region of Saudi Arabia | 98a641e7-a7c1-479a-8324-fe076650fc09 | 9907849 | Internal Medicine[mh] | Clinical interpretations and medical decisions primarily rely on laboratory test results provided by laboratory reports and reference intervals (RIs) . Reference interval defined as a central 95% range of reference values (RVs) from well-defined healthy individuals has been considered to represent the results expected for healthy individuals . It is considered as a golden tool in guiding clinicians to diagnose and monitor diseases precisely, track patients’ responses to treatment, and predict any potential risk factor. RIs often vary with gender, age, geographic area, ethnicity, and other factors leading to variations in RIs among clinical laboratories . Therefore, it is imperative to establish population-specific RIs for major laboratory analytes in each laboratory. With this background, the Committee on Reference Intervals and Decision Limits (C-RIDL) of the International Federation of Clinical Chemistry and Laboratory Medicine (IFCC) has initiated an international multicenter study aiming to find appropriate RIs for each laboratory by implementing population-specific RIs . Following this initiative, we aimed to establish Saudi-related RIs for basic hematological analytes as a part of a nationwide multicenter study. A complete blood count (CBC) and other hematology parameters are widely measured in clinical laboratories. Such parameters have clinical values for assessing and diagnosing certain blood conditions and tracking treatment response. Hematological tests and their corresponding RIs are often influenced by age, sex, geographical regions, genetic predisposition, environmental factors, and dietary habits . Additionally, Arab countries, particularly Saudi Arabia, are well known for high prevalence of consanguineous marriages compared with other countries . This may have a potential impact on peripheral blood composition . Several studies have been attempted to establish hematological RIs for healthy individuals in many countries, including the Arab gulf . There are significant racial and gender differences in almost all studies in calculated RIs for complete blood count parameters demonstrating the necessity to establish population-specific RIs In Saudi Arabia, a recent multicenter study was conducted in 2019 to derive typical RIs for five hematological analytes. In this study, nearly 1127 healthy volunteers from three Saudi regions were recruited. They found that the calculated RIs for WBC, hemoglobin, platelet, MCV, and neutrophils were vastly lower than those currently used in our clinical laboratories . Three other studies were also carried out in Saudi Arabia, attempting to establish hematological RIs . However, these studies were performed without rigorous attempt to exclude individuals with latent diseases, especially iron deficiency states, and without application of up-to-date statistical methods for data analysis and determination of RIs. Since we have just established the RIs for clinical biochemistry and immunoassay analytes as a part of the global multicenter project coordinated by IFCC/C-RIDL, this study aimed to derive a Saudi-based RIs for hematological parameters again in accordance with the harmonized protocol .
Subject recruitment Following the harmonized IFCC/C-RIDL protocol for recruitment, apparently healthy Saudi citizens were enrolled in this study irrespective of ethnic backgrounds. However, the preference was given to native residents of the Arabian Peninsula. The current study was conducted in the Makkah province located in the western coast of Saudi Arabia. This region includes two large cities, namely Makkah and Jeddah. The study was announced for recruitment by using different tools of communication including direct contact, email distribution, social media, posters or by official letters. The healthy subjects were recruited from different professions and governmental sectors including health, education, military, retirement office, etc. Ethical approval was obtained from Research Ethics Committee, King Abdullah International Medical Research Center (KAIMRC), King Abdulaziz Medical City, Jeddah, Saudi Arabia (Study number RCJ0212-209). A total of 409 Saudi adults were recruited aged between 18 and 65 years (51.1% males, 48.9% females). After signing a written informed consent for participation, each volunteer was asked to fill out a questionnaire about lifestyle, ABO blood type, menstrual status, their medical history including recent infection or allergic disorders, family history, whether they took any medication or supplements, and other factors. Height and weight were measured using standardized instruments and techniques, and body mass index (BMI) was computed for each subject. The questionnaire was derived from C-RIDL protocol with some modification to be more suited for Saudi culture. After applying the inclusion and exclusion criteria provided in the C-RIDL protocol, six subjects were excluded because of pathological conditions such as abnormal Hb variants, sickle cell anemia, and leukemia. All abnormalities were confirmed by using more specific tests for the associated conditions such as microscopic examination, sickle cell test, and hemoglobin electrophoresis. Those subjects were informed about their pathological status and referred for treatment. Subjects were not screened for malaria, intestinal parasites, human immunodeficiency virus, and other infectious diseases as their incidences are low in Saudi Arabia, Blood collection and handling The procedure for blood drawing was performed according to the established protocol. The included healthy participants were requested to avoid robust physical activity for three days prior to the blood sampling and to abstain from excessive eating/drinking the night before. Venipuncture was set between 7 and 10 AM following fasting for at least 10 hours. Blood samples were collected in the phlebotomy area of pathology and laboratory medicine at King Abdulaziz Medical City, Jeddah. The participants were living in Jeddah (⁓81%), Makkah (⁓14%), and other cities (⁓5%) of the western region, Saudi Arabia. Jeddah is located on the sea level while Makkah’s altitude is about 200m above the sea level. Participants were asked to remain seated for 30 minutes to avoid any variations in the results owing to postural changes . During the waiting time, blood pressure, height, and weight were taken. Thereafter, 15–20 mL of blood were drawn in six vacutainers. Two EDTA vacutainer tubes were collected for the determination of complete blood count (CBC), glycated hemoglobin (HbA1c), hemoglobin variants and erythrocyte sedimentation rate (ESR). Four plain tubes were collected for chemistry and immunoassay parameters. The measurement of HbA1c was performed for the purpose of future research and to help in detecting abnormal hemoglobin variants. The two EDTA samples were gently inverted 3 to 4 times to make sure EDTA anticoagulant was mixed homogenously with the blood and to prevent blood clotting. After EDTA samples collection they were transported by porters to the main laboratory (hematology section) within 30 minutes for analysis. For non-CBC measurement, the four collected plain tubes were centrifuged, separated and stored at −80°C until analysis as described in our previously published RI studies for chemistry and immunoassay parameters. Measurements Whole blood samples were examined for CBC using a Cell-Dyn Sapphire Hematology Analyzer (Abbott Diagnostic, USA). Hematological parameters included in the CBC were white blood cell count (WBC), neutrophil absolute count (Neu), lymphocyte absolute count (Lym), monocyte absolute count (Mon), eosinophil absolute count (Eos), basophil absolute count (Bas), neutrophil percentage (Neu%), lymphocyte percentage (Lym%), monocyte percentage (Mon%), basophil percentage (Bas%), eosinophil percentage (Eos%), red blood cell count (RBC), hemoglobin (Hb), hematocrit (Hct), mean corpuscular volume (MCV), mean corpuscular hemoglobin (MCH), mean corpuscular hemoglobin concentration (MCHC), red cell distribution width (RDW), platelet (PLT), mean platelet volume (MPV), platelet distribution width (PDW), and plateletcrit (PCT). TEST 1 (Alifax, Padova, Italy) was used to measure ESR. All measurements were conducted at King Abdulaziz Medical City Laboratory (Jeddah, Saudi Arabia), utilizing the manufacturer’s protocol and reagents, including calibrators and quality controls. For uncovering subjects with latent iron deficiency anemia, iron (Fe), unsaturated iron binding capacity (UIBC), transferrin (TF), and ferritin were measured in the aliquot of sera as described in our first two reports by the auto-analyzer of Architect 16000c for the first three and by 2000i for ferritin. Quality control Quality control (QC) assessments were routinely performed in the laboratories. All laboratory measurements were carried out in accordance with the C-RIDL standardized protocol and standard operation procedure 2 (SOP 2). The department of pathology and laboratory medicine at King Abdulaziz Medical City is accredited by the College of American Pathologists (CAP) laboratory accreditation. The laboratory applies rigorous internal and external quality control regulations. The results of the controls (Bio-Rad Liquichek TM ) and calibrators were within acceptable limits on every day of the analysis of the research samples. Statistical analyses Partitioning criteria The partitioning method was adapted from Ichihara K, 2008. Briefly, subgrouping of reference values (RVs) by sex and age was based on inter-subgroup variations expressed as a standard deviation (SD) ratio or SDR. The magnitude of between-sex SD (SDsex), between-age SD (SDage), and net-between individual SD (SDindiv) were computed by the two-level nested ANOVA, and SDR for each factor was calculated as a ratio of each SD to the SDindiv: i.e. SDRsex and SDRage. The threshold level of SDR to consider the partitioning of RVs by a given factor was set to 0.40 . In applying the ANOVA, the RVs were divided into four age groups: 18−29, 30−39, 40−49 and 50−65 years. When RVs are highly skewed, like RDW, Eos, Bas, and ESR, their RVs were first logarithmically transformed, and then the SD in the transformed scale was reverse-transformed as described elsewhere . We adopted a secondary criterion termed bias ratio (BR) at LL (or BR LL ) and at UL (or BR UL ) to determine the need for partitioning RVs since we occasionally meet situations where the SDR does not represent a true between-subgroup variation at the lower or upper bounds (LL, UL) of the RI . This is defined by the following formula illustrated for sex partitioning: B R L L = | L L M − L L F | ( U L M F − L L M F ) / 3.92 , B R U L = | U L M − U L F | ( U L M F − L L M F ) / 3.92 where subscript M, F, and MF represent male, female, and male + female, respectively. The denominator reflects the SD comprising the RI, which corresponds to between-individual SD, while the numerator represents the real between-sex bias in LL or UL. BR was also calculated for assessing the effect of the LAVE procedure (see below) as a bias (difference) observed at the LL and UL with/without LAVE. As a threshold of |BR|, 0.375 was used for assessing the effect of LAVE procedure in analogy to the convention of “allowable analytical bias” at a minimum level. Since the threshold 0.375 was shown to be overly sensitive in determining the need for sex or age partitioning, we used 0.57, which corresponds to the SDR threshold of 0.40. It was based on the relationship of bias and SD for two values x 1 and x 2 : (SD of x 1 , x 2 = 2 |x 1 −x 2 |) . Hence, the BR threshold of 0.375 was multiplied by 2 and set to 0.57 so that it is comparable to the SDR threshold of 0.4: i.e., the denominator of BR and SDR is common or 1/4 th of the RI. Derivation of RI The parametric technique, based on Gaussian transformation of RVs using the two-parameter Box-Cox formula, was utilized to calculate RIs . The bootstrap method was used to calculate the confidence intervals (CIs) of the lower and upper limits (LL and UL): i.e. after the secondary exclusion steps (see below), the final dataset was randomly resampled, allowing replacement until the data size was the same as the source dataset. RIs were then calculated using the resampled dataset. This resampling and recalculation of RIs was repeated 50 times, and CIs for LL and UL were predicted from the repeatedly calculated LLs and ULs of the RIs. In the process of calculating the RI, we sought to apply LAVE method to reduce the influence of latent anemia in determining erythrocyte-related analytes, and to reduce possible influence of latent inflammation in determining leukocyte and platelet-related parameters. The reference tests for use in the procedure were chosen based on Spearman’s correlation coefficient among the parameters . In the initial calculation of the RIs, no exclusion was made, but in the subsequent iterative calculation, subjects with abnormal results among the reference tests (other than the one under calculation) were excluded to obtain refined RIs. The number of iterations was fixed to six times.
Following the harmonized IFCC/C-RIDL protocol for recruitment, apparently healthy Saudi citizens were enrolled in this study irrespective of ethnic backgrounds. However, the preference was given to native residents of the Arabian Peninsula. The current study was conducted in the Makkah province located in the western coast of Saudi Arabia. This region includes two large cities, namely Makkah and Jeddah. The study was announced for recruitment by using different tools of communication including direct contact, email distribution, social media, posters or by official letters. The healthy subjects were recruited from different professions and governmental sectors including health, education, military, retirement office, etc. Ethical approval was obtained from Research Ethics Committee, King Abdullah International Medical Research Center (KAIMRC), King Abdulaziz Medical City, Jeddah, Saudi Arabia (Study number RCJ0212-209). A total of 409 Saudi adults were recruited aged between 18 and 65 years (51.1% males, 48.9% females). After signing a written informed consent for participation, each volunteer was asked to fill out a questionnaire about lifestyle, ABO blood type, menstrual status, their medical history including recent infection or allergic disorders, family history, whether they took any medication or supplements, and other factors. Height and weight were measured using standardized instruments and techniques, and body mass index (BMI) was computed for each subject. The questionnaire was derived from C-RIDL protocol with some modification to be more suited for Saudi culture. After applying the inclusion and exclusion criteria provided in the C-RIDL protocol, six subjects were excluded because of pathological conditions such as abnormal Hb variants, sickle cell anemia, and leukemia. All abnormalities were confirmed by using more specific tests for the associated conditions such as microscopic examination, sickle cell test, and hemoglobin electrophoresis. Those subjects were informed about their pathological status and referred for treatment. Subjects were not screened for malaria, intestinal parasites, human immunodeficiency virus, and other infectious diseases as their incidences are low in Saudi Arabia,
The procedure for blood drawing was performed according to the established protocol. The included healthy participants were requested to avoid robust physical activity for three days prior to the blood sampling and to abstain from excessive eating/drinking the night before. Venipuncture was set between 7 and 10 AM following fasting for at least 10 hours. Blood samples were collected in the phlebotomy area of pathology and laboratory medicine at King Abdulaziz Medical City, Jeddah. The participants were living in Jeddah (⁓81%), Makkah (⁓14%), and other cities (⁓5%) of the western region, Saudi Arabia. Jeddah is located on the sea level while Makkah’s altitude is about 200m above the sea level. Participants were asked to remain seated for 30 minutes to avoid any variations in the results owing to postural changes . During the waiting time, blood pressure, height, and weight were taken. Thereafter, 15–20 mL of blood were drawn in six vacutainers. Two EDTA vacutainer tubes were collected for the determination of complete blood count (CBC), glycated hemoglobin (HbA1c), hemoglobin variants and erythrocyte sedimentation rate (ESR). Four plain tubes were collected for chemistry and immunoassay parameters. The measurement of HbA1c was performed for the purpose of future research and to help in detecting abnormal hemoglobin variants. The two EDTA samples were gently inverted 3 to 4 times to make sure EDTA anticoagulant was mixed homogenously with the blood and to prevent blood clotting. After EDTA samples collection they were transported by porters to the main laboratory (hematology section) within 30 minutes for analysis. For non-CBC measurement, the four collected plain tubes were centrifuged, separated and stored at −80°C until analysis as described in our previously published RI studies for chemistry and immunoassay parameters.
Whole blood samples were examined for CBC using a Cell-Dyn Sapphire Hematology Analyzer (Abbott Diagnostic, USA). Hematological parameters included in the CBC were white blood cell count (WBC), neutrophil absolute count (Neu), lymphocyte absolute count (Lym), monocyte absolute count (Mon), eosinophil absolute count (Eos), basophil absolute count (Bas), neutrophil percentage (Neu%), lymphocyte percentage (Lym%), monocyte percentage (Mon%), basophil percentage (Bas%), eosinophil percentage (Eos%), red blood cell count (RBC), hemoglobin (Hb), hematocrit (Hct), mean corpuscular volume (MCV), mean corpuscular hemoglobin (MCH), mean corpuscular hemoglobin concentration (MCHC), red cell distribution width (RDW), platelet (PLT), mean platelet volume (MPV), platelet distribution width (PDW), and plateletcrit (PCT). TEST 1 (Alifax, Padova, Italy) was used to measure ESR. All measurements were conducted at King Abdulaziz Medical City Laboratory (Jeddah, Saudi Arabia), utilizing the manufacturer’s protocol and reagents, including calibrators and quality controls. For uncovering subjects with latent iron deficiency anemia, iron (Fe), unsaturated iron binding capacity (UIBC), transferrin (TF), and ferritin were measured in the aliquot of sera as described in our first two reports by the auto-analyzer of Architect 16000c for the first three and by 2000i for ferritin.
Quality control (QC) assessments were routinely performed in the laboratories. All laboratory measurements were carried out in accordance with the C-RIDL standardized protocol and standard operation procedure 2 (SOP 2). The department of pathology and laboratory medicine at King Abdulaziz Medical City is accredited by the College of American Pathologists (CAP) laboratory accreditation. The laboratory applies rigorous internal and external quality control regulations. The results of the controls (Bio-Rad Liquichek TM ) and calibrators were within acceptable limits on every day of the analysis of the research samples.
Partitioning criteria The partitioning method was adapted from Ichihara K, 2008. Briefly, subgrouping of reference values (RVs) by sex and age was based on inter-subgroup variations expressed as a standard deviation (SD) ratio or SDR. The magnitude of between-sex SD (SDsex), between-age SD (SDage), and net-between individual SD (SDindiv) were computed by the two-level nested ANOVA, and SDR for each factor was calculated as a ratio of each SD to the SDindiv: i.e. SDRsex and SDRage. The threshold level of SDR to consider the partitioning of RVs by a given factor was set to 0.40 . In applying the ANOVA, the RVs were divided into four age groups: 18−29, 30−39, 40−49 and 50−65 years. When RVs are highly skewed, like RDW, Eos, Bas, and ESR, their RVs were first logarithmically transformed, and then the SD in the transformed scale was reverse-transformed as described elsewhere . We adopted a secondary criterion termed bias ratio (BR) at LL (or BR LL ) and at UL (or BR UL ) to determine the need for partitioning RVs since we occasionally meet situations where the SDR does not represent a true between-subgroup variation at the lower or upper bounds (LL, UL) of the RI . This is defined by the following formula illustrated for sex partitioning: B R L L = | L L M − L L F | ( U L M F − L L M F ) / 3.92 , B R U L = | U L M − U L F | ( U L M F − L L M F ) / 3.92 where subscript M, F, and MF represent male, female, and male + female, respectively. The denominator reflects the SD comprising the RI, which corresponds to between-individual SD, while the numerator represents the real between-sex bias in LL or UL. BR was also calculated for assessing the effect of the LAVE procedure (see below) as a bias (difference) observed at the LL and UL with/without LAVE. As a threshold of |BR|, 0.375 was used for assessing the effect of LAVE procedure in analogy to the convention of “allowable analytical bias” at a minimum level. Since the threshold 0.375 was shown to be overly sensitive in determining the need for sex or age partitioning, we used 0.57, which corresponds to the SDR threshold of 0.40. It was based on the relationship of bias and SD for two values x 1 and x 2 : (SD of x 1 , x 2 = 2 |x 1 −x 2 |) . Hence, the BR threshold of 0.375 was multiplied by 2 and set to 0.57 so that it is comparable to the SDR threshold of 0.4: i.e., the denominator of BR and SDR is common or 1/4 th of the RI. Derivation of RI The parametric technique, based on Gaussian transformation of RVs using the two-parameter Box-Cox formula, was utilized to calculate RIs . The bootstrap method was used to calculate the confidence intervals (CIs) of the lower and upper limits (LL and UL): i.e. after the secondary exclusion steps (see below), the final dataset was randomly resampled, allowing replacement until the data size was the same as the source dataset. RIs were then calculated using the resampled dataset. This resampling and recalculation of RIs was repeated 50 times, and CIs for LL and UL were predicted from the repeatedly calculated LLs and ULs of the RIs. In the process of calculating the RI, we sought to apply LAVE method to reduce the influence of latent anemia in determining erythrocyte-related analytes, and to reduce possible influence of latent inflammation in determining leukocyte and platelet-related parameters. The reference tests for use in the procedure were chosen based on Spearman’s correlation coefficient among the parameters . In the initial calculation of the RIs, no exclusion was made, but in the subsequent iterative calculation, subjects with abnormal results among the reference tests (other than the one under calculation) were excluded to obtain refined RIs. The number of iterations was fixed to six times.
The partitioning method was adapted from Ichihara K, 2008. Briefly, subgrouping of reference values (RVs) by sex and age was based on inter-subgroup variations expressed as a standard deviation (SD) ratio or SDR. The magnitude of between-sex SD (SDsex), between-age SD (SDage), and net-between individual SD (SDindiv) were computed by the two-level nested ANOVA, and SDR for each factor was calculated as a ratio of each SD to the SDindiv: i.e. SDRsex and SDRage. The threshold level of SDR to consider the partitioning of RVs by a given factor was set to 0.40 . In applying the ANOVA, the RVs were divided into four age groups: 18−29, 30−39, 40−49 and 50−65 years. When RVs are highly skewed, like RDW, Eos, Bas, and ESR, their RVs were first logarithmically transformed, and then the SD in the transformed scale was reverse-transformed as described elsewhere . We adopted a secondary criterion termed bias ratio (BR) at LL (or BR LL ) and at UL (or BR UL ) to determine the need for partitioning RVs since we occasionally meet situations where the SDR does not represent a true between-subgroup variation at the lower or upper bounds (LL, UL) of the RI . This is defined by the following formula illustrated for sex partitioning: B R L L = | L L M − L L F | ( U L M F − L L M F ) / 3.92 , B R U L = | U L M − U L F | ( U L M F − L L M F ) / 3.92 where subscript M, F, and MF represent male, female, and male + female, respectively. The denominator reflects the SD comprising the RI, which corresponds to between-individual SD, while the numerator represents the real between-sex bias in LL or UL. BR was also calculated for assessing the effect of the LAVE procedure (see below) as a bias (difference) observed at the LL and UL with/without LAVE. As a threshold of |BR|, 0.375 was used for assessing the effect of LAVE procedure in analogy to the convention of “allowable analytical bias” at a minimum level. Since the threshold 0.375 was shown to be overly sensitive in determining the need for sex or age partitioning, we used 0.57, which corresponds to the SDR threshold of 0.40. It was based on the relationship of bias and SD for two values x 1 and x 2 : (SD of x 1 , x 2 = 2 |x 1 −x 2 |) . Hence, the BR threshold of 0.375 was multiplied by 2 and set to 0.57 so that it is comparable to the SDR threshold of 0.4: i.e., the denominator of BR and SDR is common or 1/4 th of the RI.
The parametric technique, based on Gaussian transformation of RVs using the two-parameter Box-Cox formula, was utilized to calculate RIs . The bootstrap method was used to calculate the confidence intervals (CIs) of the lower and upper limits (LL and UL): i.e. after the secondary exclusion steps (see below), the final dataset was randomly resampled, allowing replacement until the data size was the same as the source dataset. RIs were then calculated using the resampled dataset. This resampling and recalculation of RIs was repeated 50 times, and CIs for LL and UL were predicted from the repeatedly calculated LLs and ULs of the RIs. In the process of calculating the RI, we sought to apply LAVE method to reduce the influence of latent anemia in determining erythrocyte-related analytes, and to reduce possible influence of latent inflammation in determining leukocyte and platelet-related parameters. The reference tests for use in the procedure were chosen based on Spearman’s correlation coefficient among the parameters . In the initial calculation of the RIs, no exclusion was made, but in the subsequent iterative calculation, subjects with abnormal results among the reference tests (other than the one under calculation) were excluded to obtain refined RIs. The number of iterations was fixed to six times.
Recruited subjects After excluding 6 subjects, a total of 403 recruited subjects participated in the study. The number of male and female participants were 206 and 197, respectively. The mean age ± SD was 39.3±11.6 years for males and 37.4±13.1 years for females with the BMI of 28.7±5.7 kg/m 2 and 27.8±6.3 kg/m 2 , respectively. The proportions of participants with allergic conditions (rhinitis, atopic dermatitis, or asthma) were 4.4% (males) and 7.1% (females). Current smokers were 58 (29.1%) and 12 (6.4%) of males and females, respectively. Source of variation and correlation among analytes Multiple regression analysis (MRA) was independently performed for each gender by defining RVs of each analyte as an objective variable and a fixed set of explanatory variables (source of variations): age, BMI, smoking (in four levels), and allergy (binary) as shown in . As a practical level of association for the partial regression coefficients (r p ), a value of ≥ 0.20 was considered as a significant effect size. An age-related reduction of RVs was observed for RBC, Hb, and Hct solely in males, while the r p value of Ht increased with age in females (r p = 0.208) and decreased in males (r p = −0.200). At the same time, the r p value of ESR was increased with age in males (r p = 0.328) but not in females (r p = −0.008). The r p value of ESR and Neu were increased with BMI in females (r p = 0.371; r p = 0.308) more than males (rp = 0.272; r p = 0.114) respectively . shows that there was no association of smoking (Smk) with WBC in both males (rp = -0.018) and females (rp = 0.053). Allergy had a slight correlation with Eos level in males (r p = 0.246) more than females (r p = 0.049). shows the magnitude of increased Eos in subjects with allergy compared to those without allergy. In our previous reports about Saudi Ris, we found that by performing the MRA in the same way, the iron markers (Fe, TRF, UIBC, and ferritin) in each gender were not significantly associated with age, BMI, and smoking . Using a cut-off of 0.4 as a guide, SDRs have shown that sex was a significant source of variation for RBC, Hb, Hct, PCT and ESR, while PLT and Eos counts had high SDRs but still <0.4 . Therefore, partition of RVs by sex was decided for RBC, Hb, Ht, ESR, and PCT . Meanwhile, our previous reports about that the level of SDRsex were significantly high (≥ 0.4) for Fe, TRF, and UIBC , and ferritin . For all hematology parameters, age and BMI did not appear to be significant sources of variation based on SDR values . Hence the age was not considered for partitioning and BMI was not treated as a clue for secondary exclusion of RVs for any analyte. The association study among erythrocyte parameters and iron- markers (iron, Fe; unsaturated iron-binding capacity, UIBC; transferrin, TRF; and ferritin, Ferr) showed generally very high levels of correlations by using the Spearman’s correlation coefficient, especially in females . This result was interpreted as a rationale for applying LAVE method by selecting reference tests from among those analytes. On the other hand, there was no appreciable association of total protein (TP), albumin (Alb), transferrin (TRF), and C-reactive protein (CRP) with leukocytes and platelet related parameters, and thus LAVE method was not applied . Derivation of reference intervals The RIs were derived by using parametric method, with/without application of the LAVE procedure. For selecting reference tests for use in the LAVE procedure, RIs were calculated in two parts, one for erythrocyte-related parameters and the other for leukocyte and platelet related (or non-erythrocyte) parameters as shown in and Tables, respectively. For the former, there were significant associations between iron-markers and erythrocyte parameters. The utility of LAVE was expected for reducing the influence of latent anemia. As reference tests for use in LAVE procedure, seven analytes (Fe, UIBC, transferrin, ferritin, Hb, MCV, and RDW) were empirically chosen to attain an optimal effect for the secondary exclusion. In contrast, no remarkable cross-correlations between inflammatory markers (TP, Alb, transferrin, and CRP) and non-erythrocyte parameters were observed. Hence, the application of LAVE was withheld for them. We considered the requirement for partitioning RVs by gender based on SDRsex and/or BRs, and the utility of the LAVE method was assessed based on BRs at LL and/or UL. The list of RIs calculated in multiple ways are fully listed in for erythrocyte and iron markers and in for leukocyte and platelet parameters. It is of note that RIs derived by use of nonparametric method were omitted from the tables. This was because the RIs by nonparametric method often gave: 1) a wider interval, 2) a broader 90%CI at RI limits, 3) raised UL when the RV distribution showed a prominent tailing to a higher-side (Ferr, Eos%, Bas%, etc), 4) lowered LL when the distribution showed more scatter to a lower-side (Hb, Ht, MCV, MCH, MCHC) as shown in . Based on SDRsex and BRs, partitioning by sex was adopted for RBC, Hb, Ht, RDW, and ESR among erythrocyte parameters , and for PLT, MPV, Eos, Eos%, and PCT among non-erythrocyte parameters . In , the comprehensive list of RIs determined by parametric method with/without LAVE are listed for all erythrocyte parameters and iron markers. The effect of the LAVE method was judged from bias ratio at LL or UL (BR LL , BR UL ) using the threshold of 0.375. Hence, the LAVE method was applied in calculating RIs for RBC, Hb, Ht, MCV, MCH, MCHC, RDW, and ESR. Sex and LAVE effects on RIs By applying the LAVE method, the RI of Hb in males was changed from 13.0 to 13.6 g/dL for the LL and from 17.8 to 18.0 g/dL for the UL. In females, the LL was changed from 9.7 to 11.0 g/dL and the UL was changed from 15.1 to 15.2 g/dL. Similarly, appreciable changes in RI limits were also observed for MCV, RDW, and ESR as illustrated in . The RIs for RBC, Hb, Ht, Eos, and Eos% were shifted to a lower side in female than male. The RIs for ESR, PLT, MPV, and PCT were shifted to a higher side in female than male as shown in . The final list of RIs selected from and Tables are listed in . Saudi RIs in comparison to other countries shows our obtained Saudi RIs in comparison to other countries i.e. Turkey , Kenya and Ghana involved in the in the IFCC global project. It seems that countries involved in this multicenter study have the highest upper limits for RBC, Hb and Hct in both males and females. This is because of LAVE effect in excluding subjects with latent anemia in comparison with other countries. Other CBC parameters including RDW, MCV, MCH and MCHC are comparable to all other countries as shown in .
After excluding 6 subjects, a total of 403 recruited subjects participated in the study. The number of male and female participants were 206 and 197, respectively. The mean age ± SD was 39.3±11.6 years for males and 37.4±13.1 years for females with the BMI of 28.7±5.7 kg/m 2 and 27.8±6.3 kg/m 2 , respectively. The proportions of participants with allergic conditions (rhinitis, atopic dermatitis, or asthma) were 4.4% (males) and 7.1% (females). Current smokers were 58 (29.1%) and 12 (6.4%) of males and females, respectively.
Multiple regression analysis (MRA) was independently performed for each gender by defining RVs of each analyte as an objective variable and a fixed set of explanatory variables (source of variations): age, BMI, smoking (in four levels), and allergy (binary) as shown in . As a practical level of association for the partial regression coefficients (r p ), a value of ≥ 0.20 was considered as a significant effect size. An age-related reduction of RVs was observed for RBC, Hb, and Hct solely in males, while the r p value of Ht increased with age in females (r p = 0.208) and decreased in males (r p = −0.200). At the same time, the r p value of ESR was increased with age in males (r p = 0.328) but not in females (r p = −0.008). The r p value of ESR and Neu were increased with BMI in females (r p = 0.371; r p = 0.308) more than males (rp = 0.272; r p = 0.114) respectively . shows that there was no association of smoking (Smk) with WBC in both males (rp = -0.018) and females (rp = 0.053). Allergy had a slight correlation with Eos level in males (r p = 0.246) more than females (r p = 0.049). shows the magnitude of increased Eos in subjects with allergy compared to those without allergy. In our previous reports about Saudi Ris, we found that by performing the MRA in the same way, the iron markers (Fe, TRF, UIBC, and ferritin) in each gender were not significantly associated with age, BMI, and smoking . Using a cut-off of 0.4 as a guide, SDRs have shown that sex was a significant source of variation for RBC, Hb, Hct, PCT and ESR, while PLT and Eos counts had high SDRs but still <0.4 . Therefore, partition of RVs by sex was decided for RBC, Hb, Ht, ESR, and PCT . Meanwhile, our previous reports about that the level of SDRsex were significantly high (≥ 0.4) for Fe, TRF, and UIBC , and ferritin . For all hematology parameters, age and BMI did not appear to be significant sources of variation based on SDR values . Hence the age was not considered for partitioning and BMI was not treated as a clue for secondary exclusion of RVs for any analyte. The association study among erythrocyte parameters and iron- markers (iron, Fe; unsaturated iron-binding capacity, UIBC; transferrin, TRF; and ferritin, Ferr) showed generally very high levels of correlations by using the Spearman’s correlation coefficient, especially in females . This result was interpreted as a rationale for applying LAVE method by selecting reference tests from among those analytes. On the other hand, there was no appreciable association of total protein (TP), albumin (Alb), transferrin (TRF), and C-reactive protein (CRP) with leukocytes and platelet related parameters, and thus LAVE method was not applied .
The RIs were derived by using parametric method, with/without application of the LAVE procedure. For selecting reference tests for use in the LAVE procedure, RIs were calculated in two parts, one for erythrocyte-related parameters and the other for leukocyte and platelet related (or non-erythrocyte) parameters as shown in and Tables, respectively. For the former, there were significant associations between iron-markers and erythrocyte parameters. The utility of LAVE was expected for reducing the influence of latent anemia. As reference tests for use in LAVE procedure, seven analytes (Fe, UIBC, transferrin, ferritin, Hb, MCV, and RDW) were empirically chosen to attain an optimal effect for the secondary exclusion. In contrast, no remarkable cross-correlations between inflammatory markers (TP, Alb, transferrin, and CRP) and non-erythrocyte parameters were observed. Hence, the application of LAVE was withheld for them. We considered the requirement for partitioning RVs by gender based on SDRsex and/or BRs, and the utility of the LAVE method was assessed based on BRs at LL and/or UL. The list of RIs calculated in multiple ways are fully listed in for erythrocyte and iron markers and in for leukocyte and platelet parameters. It is of note that RIs derived by use of nonparametric method were omitted from the tables. This was because the RIs by nonparametric method often gave: 1) a wider interval, 2) a broader 90%CI at RI limits, 3) raised UL when the RV distribution showed a prominent tailing to a higher-side (Ferr, Eos%, Bas%, etc), 4) lowered LL when the distribution showed more scatter to a lower-side (Hb, Ht, MCV, MCH, MCHC) as shown in . Based on SDRsex and BRs, partitioning by sex was adopted for RBC, Hb, Ht, RDW, and ESR among erythrocyte parameters , and for PLT, MPV, Eos, Eos%, and PCT among non-erythrocyte parameters . In , the comprehensive list of RIs determined by parametric method with/without LAVE are listed for all erythrocyte parameters and iron markers. The effect of the LAVE method was judged from bias ratio at LL or UL (BR LL , BR UL ) using the threshold of 0.375. Hence, the LAVE method was applied in calculating RIs for RBC, Hb, Ht, MCV, MCH, MCHC, RDW, and ESR.
By applying the LAVE method, the RI of Hb in males was changed from 13.0 to 13.6 g/dL for the LL and from 17.8 to 18.0 g/dL for the UL. In females, the LL was changed from 9.7 to 11.0 g/dL and the UL was changed from 15.1 to 15.2 g/dL. Similarly, appreciable changes in RI limits were also observed for MCV, RDW, and ESR as illustrated in . The RIs for RBC, Hb, Ht, Eos, and Eos% were shifted to a lower side in female than male. The RIs for ESR, PLT, MPV, and PCT were shifted to a higher side in female than male as shown in . The final list of RIs selected from and Tables are listed in .
shows our obtained Saudi RIs in comparison to other countries i.e. Turkey , Kenya and Ghana involved in the in the IFCC global project. It seems that countries involved in this multicenter study have the highest upper limits for RBC, Hb and Hct in both males and females. This is because of LAVE effect in excluding subjects with latent anemia in comparison with other countries. Other CBC parameters including RDW, MCV, MCH and MCHC are comparable to all other countries as shown in .
For proper derivation of hematology RIs, it is important to analyze factors which can affect test results such as sex, age, ethnicity, BMI, or smoking. They can be used for partitioning RIs, or for secondary exclusion. We used SDR as a primary guide for judging the need of partitioning by sex and age. By setting its threshold at 0.4 as in other studies, SDR for between-age differences (SDRage) was below the threshold in all the hematological parameters we examined, while SDR for between-sex differences (SDRsex) exceeded the level in reference values only for RBC, Hb, Ht, and ESR. In addition to this, the criterion of high |BR| (>0.57), partition of RVs by sex was found to be necessary for RDW, PLT, MPV, Eos, Eos%, and PCT. Our results are consistent with the previous results in Turkey , Ghana , Kenya , and Japan , which used the same statistical scheme. In addition to erythrocyte parameters, between-sex difference was noted for ESR (SDRsex = 0.96). This indicates that it is important to establish separate RIs for males and females and as shown in , the ESR value increased with BMI (SDRmbi = 0.302) in both males and females. This finding is consistent with the finding from a previous study . In addition to this, we found that the SDRsex values were below 0.4 for PLT (0.321) and Eos (0.315), but when between-sex bias ratio at the upper limit (BR UL ) was calculated, they exceeded the threshold of 0.57. This indicates that it is not appropriate to rely on the SDRsex only, but BR also must be considered to make the final decision for partitioning. Our results showed that the RIs for CBC and ESR were not appreciably age-related (SDRage<0.4) although partial correlation coefficient by MRA results revealed a weak level of age-related changes for ESR and some of analytes of CBC. Therefore, we did not partition RIs by age in analytes for Saudis. In this study, we found that Saudi median values for WBC and Neu counts were higher than some African counties e.g. Ghana, Kenya, Uganda, Zimbabwe, and Ethiopia , but at the same time they were comparable to those in Morocco Turkey and Mayo-clinic laboratory in the United States . The reason for low WBC and Neu counts in Africans compared to Saudis could be due to African-derived null variant of the Duffy antigen receptor for chemokines (DARC-null genotype) as reported in previous studies . Further details about benign ethnic neutropenia in Africans were summarized by Atallah-Yunes et.al, 2019 . The UL of Eos count is less than some African countries such as Kenya, Uganda and Zembabwe but still higher than those in Malaysia , Spain and Australia . Eos is known to be elevated with parasitic infection. Therefore, it is probable that the prevalence of parasitic infection and allergic individuals in Saudi population is less common than in African countries, as the total percentage of allergic subjects among Saudi males and females participating in the study were 5.7%. The increased upper limit of Hb in Kenyan males and females compared to other IFCC countries involved in the global studies can be attributed to the high altitude of Nairobi where the study was conducted. The high altitude can induce erythropoietic process and this consequently will increase the hemoglobin concentration . The extended RIs obtained for the Emiratis were due to the genotypic defined criteria for thalassemia in the recruited subjects . The RI for Fe which was recalculated in this study was consistent with that in our previous report , which was calculated from Fe RVs of three cities without reference to CBC test results: i.e., CBC was not tested in two other cities. This comparison of two sets of RIs indicates that RI for Fe did not change much with/without application of the LAVE method considering CBC test results. The findings from this study agree with a previous study findings as both show that Saudi females have low iron stores and the deficiency of iron is known to be a common cause of thrombocytosis . This kind of association between platelet and Fe levels may explain the high median platelet level in this study for adult Saudi females compared to other countries but at the same time it is compatible to levels for Kenyan females as shown in . The scientific mechanism behind this association is not clear, but it is suggested that a decrease in iron can induce thrombocytosis by changing the proliferation and differentiation of pluripotent erythroid and megakaryocyte precursors in bone marrow . Although a recent study showed that thrombocytosis in Saudi females is higher than males (6.34% vs. 1.08%) , but still no available study investigated the incidence of thrombocytosis in Saudi females in comparison to other countries. Our study shows that the median value of MPV in Saudi females (8.16 fL) is lower than those in Turkey (8.9fL) and Ghana (10.7 fL). Therefore, we believe that the deficiency of iron in Saudi females and its association with increased platelet count needs further investigation. In comparison with other RIs studies, we obtained higher UL of Hb (18.0 g/dL) in males and PLT in females (443 ×10 6 /μL) compared to other Arab countries (Oman, Iraq, Qatar, Morocco, Sudan) for Hb and PLT . Moreover, shows that our UL of Hb in males and PLT in females are mostly higher than those of four previously published studies on Saudi population (17.9 g/dL, 303×10 9 /μL; 17.2 g/dL, 334×10 9 /μL ; 17.5g/dL, 337×10 6 /μL ; 18.0g/dL, 367×10 6 /μL ) respectively. The reason for the higher-side shift of RVs for hemoglobin and PLT compared in this study could be due to our use of improved statistical approaches including: 1) the parametric method which is robust to extreme values prevalent in skewed distributions unlike nonparametric method , and, 2) the LAVE procedure which is effective in reducing the influence of highly prevalent subclinical conditions such as anemia with and without iron deficiency. The effect of parametric and LAVE methods on RBC parameters are shown in . By the application of both approaches, the 90% CI of the RI limits became narrower but at the same time are more reliable and efficient specially with the exclusion of recruited subjects with latent diseases . In summary, the Saudi healthy subjects recruited in this study were well-defined, as guided in the IFCC/CRIDL harmonized protocol. The application of parametric and LAVE methods helped to improve the reliability and precision of calculated RIs and to reduce the influence of subjects with latent diseases. The limitation of this study was that the healthy subjects were recruited only in the western region of Saudi Arabia. Although some of the recruited subjects were born in other regions including eastern, central, northern, and southern regions, collecting samples from subjects based on other regions would have been more favorable for the study. Nevertheless, it is of note that we experienced no regional difference in RIs of commonly measured chemistry and immunoassay analytes among subjects recruited in three major regions: western, central, and eastern of Saudi Arabia. .
This is the only study in the Arab countries to establish RIs for the hematological parameters using the internationally harmonized protocol with participation of a well-defined healthy Saudi population. Compared to the conventional studies, we believe that the RIs determined in this study are more reliable by using the following up-to-date statistical methods: 1) the parametric method that is less influenced by peripheral extreme values than the nonparametric method, 2) LAVE method that effectively reduced the influence of latent anemia in high prevalence, and 3) combined use of SDR and BR that helped make objective decision for the need of partitioning RVs by sex and age.
S1 Table Spearman’s correlation coefficients between erythrocyte parameters. Iron related markers (Fe, UIBC, TRF, Ferr) showed variable degrees of closed associations with erthryocyte related parameters. Therefore, reference tests for use in LAVE procedure were set to: Fe, UIBC, TRF, Ferr, Hb, MCV, RDW, and ESR. (XLSX) Click here for additional data file. S2 Table Spearman correlation coefficient between leukocyte parameters and inflammatory makers. Inflammatory markers TP, Alb, TRF, and CRP showed no remarkeable association with leukocyte parameters and PLT, whereas there were close associations within-leukocyte parameters. This situation is not appropriate to apply LAVE methods for leukocyte parameters because self-trimming of peripheral values may occur among them. (XLSX) Click here for additional data file. S3 Table List of RIs partitioned by sex with/without LAVE method for erhyrocyte-related parameters. LL, lower limit; UL,upper limit; SDRsex, standard deviation ratio between sex; BR-LL, bias ratio at LL; BR-UL, bias ratio at UL; CI, confidence interval; LAVE, latent abnormal values exclusion method. The thresholds for SDR and BR were set to 0.4 and 0.57, respectively. LAVE allowing no abnormal results among Hb, MCV, RDW, Fe, UIBC, TF and Ferr. (XLSX) Click here for additional data file. S4 Table List of RIs partitioned by sex for leukocyte and platelet-related parameters. LL, lower limit; UL,upper limit; SDRsex, standard deviation ratio between sex; BR-LL, bias ratio at LL; BR-UL, bias ratio at UL; CI, confidence interval. The thresholds for SDR and BR were set to 0.4 and 0.57, respectively. (XLSX) Click here for additional data file. S1 Fig BMI-related change in ESR. (TIF) Click here for additional data file. S2 Fig Association of allergy with Eos and Eos%. (TIF) Click here for additional data file. S3 Fig Hematology RIs of Saudi compared to other Saudi studies. Saudi 1, our study; Saudi 2 ref. , Saudi 3 ref. , Saudi 4 ref. , Saudi 5 Ref . (TIF) Click here for additional data file. S1 Data (XLSX) Click here for additional data file.
|
Assessing impact of a community-based screening campaign to address social determinants of cervical cancer | 7e982e24-d3a0-4733-9aab-65db7406c1c1 | 11878794 | Biopsy[mh] | Prevention and screening represent cornerstones of the public health strategy to combat cervical cancer. However, traditional outreach has often been inadequate to increase screening among underserved populations, including the Hispanic community. Hispanic women experience one of the highest cervical cancer rates across racial and ethnic groups and are more often diagnosed at advanced stages, perpetuating cancer disparities. , The social determinants of this disparity are multifactorial, including insurance status, limited English proficiency, lack of health education, and sociocultural beliefs. , As a result, Hispanic women are screened at lower rates, with the latest estimates from the National Cancer Institute (NCI) showing that 67.9% were up-to-date on cervical cancer screening guidelines in 2021 compared with 72.4% of women overall during the same year. These data represent a decline from the 2018 rate of 80.8% among Hispanic women, as US cervical cancer screening rates diminished during the COVID-19 pandemic. , Catch-up efforts have not been as effective in Hispanic populations compared with non-Hispanic White populations, highlighting the need to rethink screening outreach in these communities. In the east area of Los Angeles, where 95% of residents identify as Hispanic or Latino, women experience multiple social barriers to care such as lower education levels (12.6% of women having a bachelor’s degree or higher in 2022 vs 36.8% nationally), limited English proficiency (26.6% of adults vs 5.2% nationally), and higher uninsured rates (13.2% of women vs 7.0% nationally). , Recognizing the impact of social determinants of health, targeted cancer screening efforts are a key priority in this population. Increasingly, community–academic partnerships and culturally tailored media have emerged as promising strategies to increase cancer screening in underserved communities. , Thus, this study assesses a culturally tailored cervical cancer screening campaign implemented through academic–community–government partnership in eastern neighborhoods of Los Angeles.
The Es Tiempo outdoor media campaign From 2015 to 2018, intercept surveys evaluated the recall, interpretation, and impact of Es Tiempo, a public health campaign conducted in predominantly Hispanic neighborhoods in eastern areas of Los Angeles (Boyle Heights, Lincoln Heights, and El Sereno). Impact was operationalized as participants reported intention to call the campaign’s 800 number or schedule a Pap test after seeing the campaign. Es Tiempo was enacted through multidisciplinary partnership of the University of South Carolina Norris Comprehensive Cancer Center, Los Angeles General Medical Center (a large safety net hospital), Clínica Monseñor Oscar Romero (a Federally Qualified Health Center), and the Los Angeles Office of Women’s Health at the Los Angeles County Department of Public Health (LADPH). The campaign design has been detailed previously. Relevant to the present study, the outdoor media campaign targeted Hispanic women eligible for cervical cancer screening and was guided by English and Spanish focus groups, community advisory board meetings, and visual culture studies. This work revealed themes of family values and the culturally significant jacaranda tree, which produces abundant purple blossoms each spring in Los Angeles. Imagery of the blooming jacaranda tree was used as a visual cue to pursue screening (see ). A second campaign image featuring a Hispanic mother hugging her children was used to convey the importance of family values related to preventive screening (see ). Materials were displayed on bus benches, billboards, and light post banners with messaging in Spanish and English (“It’s important. It’s easy. It’s time.” “You can prevent cervical cancer.”). All materials included a message to call the campaign’s 800 number, through which LADPH staff could answer questions about screening and schedule Pap test appointments. The protocol for this study was approved by the University of South Carolina institutional review board and adhered to the Standards for Reporting Qualitative Research (SRQR) ( ). Conceptual model and researcher characteristics This study employed a community-based participatory research conceptual model guided by principles of community empowerment, equitable partnership, and cultural humility (see ). Under this model, collaboration among academic, government, and community stakeholders was emphasized at all phases of the study. The theory of change involved logical progression from campaign co-creation, which incorporated stakeholder input into design and implementation, to outcomes of increased cervical cancer awareness and screening behavior. In accordance with SRQR guidelines and in recognition that researcher characteristics influence the study process, we report here characteristics of the investigators. Relevant to the study population and survey methodology, researchers were bicultural and bilingual in Spanish, with lived experience in the geographic area of focus. Interviewers were recruited from the local community to ensure cultural and linguistic congruence with study participants. All interviewers received training on participant recruitment and consent procedures, survey administration, cultural humility, and communication of scientific information in English and Spanish. Training was further augmented through role-play of challenging field scenarios interviewers may encounter during survey administration. Inclusion and survey design Included participants were Hispanic women between ages 21 and 65 years who spoke English or Spanish. Given the study’s focus on screening uptake, groups ineligible for cervical cancer screening under current US Preventive Services Taskforce (USPSTF) guidelines (biological males, and females younger than 21 years or older than 65 years) were excluded. Participants reported demographic characteristics of age, primary language spoken at home (English, Spanish, both English and Spanish), and insurance status (insured and uninsured). They additionally reported previous cervical cancer screening behavior, detailed in . Campaign recall was assessed using close-ended questioning corroborated by open-ended follow-up questions. Campaign interpretation and most liked campaign aspects were assessed using open-ended questions, Screening intention (operationalized as intention to call the campaign’s 800 number or schedule a Pap test) was evaluated using Likert-type scales. Survey questions are included verbatim in . Data collection Intercept surveys were conducted in person from 2015 to 2018 at community locations including parks, bus stops, laundromats, and clinics. This sampling strategy was used to better capture information from local residents, including those who may not participate in either online surveys or organized activities such as health fairs. Written consent was obtained in English or Spanish, and participants received a $5 gift card for their time. Each standardized survey was completed on paper in participants’ preferred language, with assistance from trained bilingual interviewers. Data entry was monitored daily by 2 coders who entered survey information and checked for quality and accuracy. Data were further audited by independent reviewers, who randomly sampled 25% of the data collected to verify correct data entry. De-identified data were then analyzed using Google Sheets, with tabulation of close-ended Likert-type question responses and thematic analysis using keywords identified in both Spanish and English. Statistical analysis Thematic analysis for open-ended responses was conducted using an iterative coding process to identify emerging themes. Saturation of themes was determined when no new themes emerged from successive coding rounds. Likert-type responses for screening intention were dichotomized as “likely or extremely likely” vs “neutral, unlikely, or extremely unlikely.” This approach was chosen for clearer interpretation and improved model stability, given small sample sizes in certain response categories. Dichotomized responses were then analyzed using χ 2 goodness-of-fit tests. Additionally, logistic regression models were employed to examine differences in dichotomized screening intention by participant characteristics. These models were fitted for age, primary language spoken at home, insurance status, human papillomavirus (HPV) vaccination status, and adherence to USPSTF cervical cancer screening guidelines. Missing data were excluded at item level using available case analysis, with sample size for each question excluding nonrespondents. Subgroup analyses were performed to ascertain interpretation among those who recalled seeing the campaign prior to survey and screening intention among those with no Pap test in the last 3 years.
From 2015 to 2018, intercept surveys evaluated the recall, interpretation, and impact of Es Tiempo, a public health campaign conducted in predominantly Hispanic neighborhoods in eastern areas of Los Angeles (Boyle Heights, Lincoln Heights, and El Sereno). Impact was operationalized as participants reported intention to call the campaign’s 800 number or schedule a Pap test after seeing the campaign. Es Tiempo was enacted through multidisciplinary partnership of the University of South Carolina Norris Comprehensive Cancer Center, Los Angeles General Medical Center (a large safety net hospital), Clínica Monseñor Oscar Romero (a Federally Qualified Health Center), and the Los Angeles Office of Women’s Health at the Los Angeles County Department of Public Health (LADPH). The campaign design has been detailed previously. Relevant to the present study, the outdoor media campaign targeted Hispanic women eligible for cervical cancer screening and was guided by English and Spanish focus groups, community advisory board meetings, and visual culture studies. This work revealed themes of family values and the culturally significant jacaranda tree, which produces abundant purple blossoms each spring in Los Angeles. Imagery of the blooming jacaranda tree was used as a visual cue to pursue screening (see ). A second campaign image featuring a Hispanic mother hugging her children was used to convey the importance of family values related to preventive screening (see ). Materials were displayed on bus benches, billboards, and light post banners with messaging in Spanish and English (“It’s important. It’s easy. It’s time.” “You can prevent cervical cancer.”). All materials included a message to call the campaign’s 800 number, through which LADPH staff could answer questions about screening and schedule Pap test appointments. The protocol for this study was approved by the University of South Carolina institutional review board and adhered to the Standards for Reporting Qualitative Research (SRQR) ( ).
This study employed a community-based participatory research conceptual model guided by principles of community empowerment, equitable partnership, and cultural humility (see ). Under this model, collaboration among academic, government, and community stakeholders was emphasized at all phases of the study. The theory of change involved logical progression from campaign co-creation, which incorporated stakeholder input into design and implementation, to outcomes of increased cervical cancer awareness and screening behavior. In accordance with SRQR guidelines and in recognition that researcher characteristics influence the study process, we report here characteristics of the investigators. Relevant to the study population and survey methodology, researchers were bicultural and bilingual in Spanish, with lived experience in the geographic area of focus. Interviewers were recruited from the local community to ensure cultural and linguistic congruence with study participants. All interviewers received training on participant recruitment and consent procedures, survey administration, cultural humility, and communication of scientific information in English and Spanish. Training was further augmented through role-play of challenging field scenarios interviewers may encounter during survey administration.
Included participants were Hispanic women between ages 21 and 65 years who spoke English or Spanish. Given the study’s focus on screening uptake, groups ineligible for cervical cancer screening under current US Preventive Services Taskforce (USPSTF) guidelines (biological males, and females younger than 21 years or older than 65 years) were excluded. Participants reported demographic characteristics of age, primary language spoken at home (English, Spanish, both English and Spanish), and insurance status (insured and uninsured). They additionally reported previous cervical cancer screening behavior, detailed in . Campaign recall was assessed using close-ended questioning corroborated by open-ended follow-up questions. Campaign interpretation and most liked campaign aspects were assessed using open-ended questions, Screening intention (operationalized as intention to call the campaign’s 800 number or schedule a Pap test) was evaluated using Likert-type scales. Survey questions are included verbatim in .
Intercept surveys were conducted in person from 2015 to 2018 at community locations including parks, bus stops, laundromats, and clinics. This sampling strategy was used to better capture information from local residents, including those who may not participate in either online surveys or organized activities such as health fairs. Written consent was obtained in English or Spanish, and participants received a $5 gift card for their time. Each standardized survey was completed on paper in participants’ preferred language, with assistance from trained bilingual interviewers. Data entry was monitored daily by 2 coders who entered survey information and checked for quality and accuracy. Data were further audited by independent reviewers, who randomly sampled 25% of the data collected to verify correct data entry. De-identified data were then analyzed using Google Sheets, with tabulation of close-ended Likert-type question responses and thematic analysis using keywords identified in both Spanish and English.
Thematic analysis for open-ended responses was conducted using an iterative coding process to identify emerging themes. Saturation of themes was determined when no new themes emerged from successive coding rounds. Likert-type responses for screening intention were dichotomized as “likely or extremely likely” vs “neutral, unlikely, or extremely unlikely.” This approach was chosen for clearer interpretation and improved model stability, given small sample sizes in certain response categories. Dichotomized responses were then analyzed using χ 2 goodness-of-fit tests. Additionally, logistic regression models were employed to examine differences in dichotomized screening intention by participant characteristics. These models were fitted for age, primary language spoken at home, insurance status, human papillomavirus (HPV) vaccination status, and adherence to USPSTF cervical cancer screening guidelines. Missing data were excluded at item level using available case analysis, with sample size for each question excluding nonrespondents. Subgroup analyses were performed to ascertain interpretation among those who recalled seeing the campaign prior to survey and screening intention among those with no Pap test in the last 3 years.
The study included 673 participants, with a mean age of 45.4 years ( ). The majority of participants primarily spoke Spanish at home (85.9%), and 26.1% were uninsured. In the past year, 57.7% of participants reported asking their health provider about getting a Pap test, and 22.4% reported asking their provider about cervical cancer. In the last 3 years, 87.2% of participants had undergone a Pap test. Among all participants, 25.1% recalled seeing the campaign prior to survey administration ( ). Among those who had seen the campaign prior, the main message was interpreted as prevention and early detection of cancer by 64.5% of participants, followed by health and regular checkups (14.2%) and protecting family and loved ones (7.7%). When asked what they liked most about the campaign, participants most commonly cited the emphasis on family (37.1%), followed by the campaign’s message and information provided (29.7%) and the jacaranda tree and/or nature imagery (22.6%). Exemplar responses can be seen in . After seeing the campaign, 83.7% of participants were likely or extremely likely to call the 800 number provided. Additionally, 89.5% were likely or extremely likely to schedule a Pap test. In subgroup analysis among participants who had not had a Pap test in the last 3 years, 78.9% and 83.5% were likely or extremely likely to call the 800 number or schedule a Pap test, respectively. In , Pap test intention did not differ by age, primary language spoken at home, insurance status, HPV vaccination status, or Pap test in the last 3 years. However, participants who were older, had received the HPV vaccine, and primarily spoke Spanish at home were more likely to call the 800 number.
Compared with national statistics over the same period, a greater proportion of our sample was uninsured (25.2% vs 8.3% of women nationally). Additionally, compared with 2018 American Community Survey estimates for East Los Angeles, our sample was older (sample, 29.1% aged 55-65 years, vs American Community Survey, 10.7% aged 55-64 years), and a greater proportion primarily spoke Spanish at home (88.7% vs 82.6%). Despite this, the proportion of participants who reported a Pap test in the last 3 years (87.2%) was higher than NCI estimates among Hispanic women over the same time period (2018: 80.8%), which may reflect our inclusion of clinics as survey locations or action taken in response to the campaign prior to survey date. The results reveal the impact of community–academic partnerships and culturally tailored outreach on social determinants of cervical cancer. Specifically, this campaign improved screening awareness and integrated culturally significant values , to increase screening intention in majority Hispanic neighborhoods. The 25.1% recall rate among survey participants indicates greater recall than other health campaigns in the United States, with a previous large sample survey by Goodman et al., which assessed government healthy eating campaigns using similar survey methods, finding US campaign awareness of only 13.0% (n = 4612). Our campaign recall is even more unusual when considering our study setting and target population, as Goodman and colleagues also found that females and those with low education levels were less likely to recall having seen a campaign (campaign awareness in both groups: 10.5%). With our campaign recall being approximately double that of other US health campaigns, these findings underscore the potential to improve outreach efficacy through community-based participatory research. Among participants who recalled seeing the campaign, the majority identified its message to be health oriented. Additionally, 7.7% of participants interpreted the main message to be centered around family and protecting loved ones, with family values also shown to be the campaign’s most liked element. These findings are in line with our group’s previous work on cervical cancer screening, which found a narrative film featuring a Hispanic family to significantly increase supportive attitudes around Pap tests among Hispanic women compared with a nonnarrative film, which used more traditional public health campaign approaches ( P = .05). Recognition of these values may help combat sociocultural drivers of cancer disparities. In numerous studies on cervical cancer perceptions, , , participants reported feeling that cervical cancer carried with it stigma around sexual activity, with a study among immigrant Latinas finding that a major barrier to Pap tests was fear that pursuing screening would label them as sexually promiscuous. The same study showed that family support was an important facilitator for screening, corroborating findings from the present work. Thus, a culturally tailored approach may present a strategy to improve social determinants of cervical cancer screening by replacing negative cultural associations around sexual activity with positive associations of protecting family and loved ones. Participants exhibited high intention to pursue screening after campaign exposure. The observed differences in intention to call the 800 number—with participants who were older, HPV-vaccinated, and speaking Spanish as the primary language more likely to call—may indicate increased efficacy among the intended target audience for Es Tiempo. For Pap test scheduling intent, rates were similar among women who reported not having a Pap test in the last 3 years, a clinically important subgroup given current USPSTF guidelines. Based on the screening intention rates observed in this subgroup, culturally tailored outreach appears to be similarly effective in general, but additional facilitators may be necessary to drive specific actions such as calling the 800 number. Besides addressing cultural stigma, facilitators identified in prior research to increase Pap tests among Hispanic women include increasing insurance rates, reducing language barriers, and addressing fears regarding the pelvic exam and prospect of finding cancer. These facilitators are consistent with our study demographics, as our Los Angeles–based participants were disproportionately primarily Spanish-speaking and uninsured. To mitigate insurance barriers in this population, California’s 2024 Medicaid expansion to all low-income adults, regardless of immigration status, represents an important policy change. It must be matched with community-based outreach to effectively close gaps in screening. High campaign awareness and screening intention in our sample may reflect the intended effects of our multidisciplinary partnership. Led by our Office of Community Outreach and Education (COE), our NCI-designated comprehensive cancer center has fostered bidirectional relationships with community stakeholders. Recognizing that many cancer centers have established COE offices and all communities may benefit from them, we highlight key factors that have enabled our program to better address social determinants of health. Internally, our COE office has strong representation of East Los Angeles natives among our staff, including in leadership roles. Further, to operationalize the principles outlined in our community-based participatory research conceptual model, we continually adapt programming to meet community needs. For example, in a recent effort that built on the culturally tailored design and multidisciplinary partnerships of the present work, Stay Connected Los Angeles collaborated with local artists to increase COVID-19 vaccination and mitigation behaviors in the eastern part of Los Angeles. Such adaptation has been highlighted by other chronic disease–focused community–academic partnerships as a key factor for maintaining their partnerships throughout the pandemic. Especially in the historical context of research involving ethnic minorities, responsiveness to community needs is necessary to foster and maintain trust. , , Previous research has further highlighted how meeting the distinct needs of each community strengthens the quality of community partnerships. Our center’s role in the present work included contributing expertise and resources for funding acquisition, study design, and implementation. Likewise, the LADPH contributed expertise in large-scale care coordination and integration with public services. LADPH staffed the campaign’s 800 number, answering callers’ questions about cervical cancer and connecting them to Pap test appointments. Such follow-through was only possible thanks to the department’s expertise navigating the public health-care system. We posit that leveraging stakeholders’ unique strengths through academic–community–government partnerships can enhance the impact of public health campaigns, with present findings supporting its efficacy for enhanced recall and screening intention. Though this study was conducted annually over 4 years in diverse community settings, its intercept survey design limits generalizability. Women with limited mobility or time constraints or those unavailable during daytime hours may be underrepresented. However, in-person survey administration and bilingual interviewers may have captured additional participants who would otherwise not participate. The study assessed participants’ self-reported screening intention, which may not always reflect actual behavior. Meta-analysis on concordance between self-reported behavioral intention and behavioral change has shown that medium-to-large changes in intention lead to small-to-medium changes in behavior. Additionally, querying specific intentions, as done here, has been shown to improve predictive value. Further, previous work has found that in a sample of Hispanic women in East Los Angeles (n = 1428), completion of Pap tests was 46% among those who received Es Tiempo campaign materials vs 33% among those who did not ( P < .01). Missing data can bias findings, with previous research highlighting that bias is more likely to be present when there is more than 10% missing data. The proportion of missing responses for our study endpoints, as seen in , was 1.1%-3.5%. As data were mostly complete, missing responses were handled using a standard approach of available case analysis, with missing values excluded at item level. Finally, the cross-sectional nature of the survey limits our ability to infer causality between campaign exposure and screening intentions. Future research should include pre- and postcampaign assessment to strengthen the evidence for campaign impact and incorporation of electronic medical records to evaluate changes in screening rates. In a sample of Hispanic women who were disproportionately uninsured and primarily Spanish speaking, we observed campaign recall rates double that of other US health campaigns and high intention among participants to engage in cancer screening after viewing the campaign. Intention was high even among those who had not undergone screening in the last 3 years, highlighting the potential of community-based interventions. Future studies should adapt these approaches to different populations and comprehensive packages of preventive services. Our findings underscore several important strategies to reduce cervical cancer disparities: (1) identifying positive cultural values with screening to decrease stigma, (2) combining culturally tailored outreach with interventions that target other known screening barriers, (3) facilitating long-term relationships that are responsive to community needs, and (4) leveraging the unique strengths of academic–community–government partnerships. In conclusion, this study provides evidence that culturally tailored outreach and partnerships can improve social determinants of cervical cancer. By deepening stakeholder collaboration, gaps in cervical cancer screening can be narrowed and ultimately eliminated in historically underserved communities.
pkaf006_Supplementary_Data
|
Deciphering clinical significance of BCL11A isoforms and protein expression roles in triple-negative breast cancer subtype | 3951b9b9-9867-4a5b-bd49-498efb25b0de | 10314865 | Anatomy[mh] | Triple negative breast cancer, which accounts for 10–20% of all invasive breast cancer (BC) subtypes, is characterized by the lack of immunohistochemical expression of estrogen receptor (ER), progesterone receptor (PR), and HER2 and/or HER2 gene amplification. TNBC is most prevalent in women aged < 50 years and shows aggressive clinical behavior (i.e., high histological grade, significantly high metastatic rate and it is responsible for about 25% of BC related deaths) (Angius et al. ). Its heterogeneity can be associated with different clinical outcomes. A recent study evaluated the outcome of TNBC patients highlighting that an accurate and reliable histopathologic definition of TNBC subtypes has a significant clinical utility and is an effective tool during the therapeutic decision making process (Sanges et al. ). Using gene expression profiling, the molecular signature of TNBC divided the molecular subclassification into four groups: basal-like 1 and 2, mesenchymal, and luminal androgen receptor (LAR) (Lehmann et al. ). Gene expression profiling, morphological and immunohistochemical analysis of TNBC represent prognostic and therapeutic tools to customize therapy and improve patient outcomes. TNBC molecular biomarkers could predict the prognosis (Cagney et al. ). We demonstrated that modification of miR-135b might improve the outcome of TNBCs with basal-like features (Uva et al. ). The subclassification of patients in our TNBC cohort, based on the high proportion of genetic alterations involving PI3K/AKT pathways, provides evidence that specific genomic abnormalities can select patients who can benefit from targeted therapies (Cossu-Rocca et al. ). BCL11A was initially detected due to an aberrant chromosomal translocation t(2;14)(p13;q32.3) in human B-cell non-Hodgkin’s lymphomas (Nakamura et al. ). BCL11A gene is located on human chromosome 2p13 and is ~ 102 kb in length. BCL11A codes for a protein with an uncommon C2HC zinc finger at the N-terminus and six Krüppel-like C2H2 zinc fingers near the C terminus. Three main mRNA variants were found: BCL11A-XL, BCL11A-L and BCL11A-S, each contains differing numbers of C-terminal C2H2 finger motifs. All 3 isoforms contained the first 3 exons, and only the longest isoform expresses sequences from exons one to four (Satterwhite et al. ). BCL11A-XL protein isoform was expressed in brain and hematopoietic tissues (Liu et al. ). Also BCL11A-XL expressed in a range of tumor-derived cell lines (Pulford et al. ). Functional studies demonstrated that BCL11A-XL was a transcriptional repressor working in association with itself, other BCL11A isoforms, and with BCL6 gene. So BCL11A-XL might play an essential role in tumor development (Liu et al. ; Pulford et al. ). High level expression of BCL11A-S was observed in human Hodgkin’s lymphoma cell line [8]. BCL11A-L isoform was expressed preferentially in derived B -cell malignant cell lines (Satterwhite et al. ). Growing evidence demonstrated that BCL11A also plays an essential role in the pathogenesis of solid tumors, including prostate cancer, lung cancer, laryngeal squamous cell carcinoma and acute leukemia (Kapatai and Murray ; Chetaille et al. ; Boelens et al. ; Agueli et al. ; Jin et al. ; Podgornik et al. ). Khaled et al. determined that BCL11A acts as an oncogene in TNBC, and its overexpression is key for tumor formation and invasion. BCL11A supports the development of normal and malignant mammary epithelial stem/progenitor populations (Khaled et al. ). Furthermore, its silencing re duces tumor initiating cells population in TNBC xenograft model (Zhu et al. ). In the mouse mammary gland, BCL11A is part of a specific subsets of embryonic mammary genes, silenced in adult epithelia and reactivated in mouse and human basal-like breast cancer (Zvelebil et al. ). The aim of the present study was to assess the clinical role of BCL11A in the molecular TNBC subtype.
A retrospective cohort of BC patients diagnosed between 2000 and 2015 was selected. Samples were obtained from the archives of the Department of Histopathology of the Oncology Hospital of Cagliari, Italy. Inclusion criteria were complete review of surgical specimens and medical records and availability of formalin-fixed, paraffin-embedded (FFPE) tumor blocks from surgical specimens. Three experienced pathologists independently reviewed all cases. Histologic subtyping was performed according to current WHO classification (Rakha et al. ). Three µm thick tissue sections of FFPE specimens were cut for hematoxylin and eosin staining, IHC, in situ hybridization (SISH) and genetic analysis. The study protocol was approved by the Azienda Sanitaria Locale Sassari Bioethics Committee (n. 1140/L, 05/21/2013); and followed the Italian law on guidelines for the implementation of retrospective observational studies (G.U. n. 76, 31 March 2008). Only coded data were collected to protect patient confidentiality. Immunohistochemistry ER, PR, HER2 and Ki-67 immunohistochemical expression and/or HER2 gene amplification, as defined by silver enhanced SISH, established the surrogate intrinsic subtypes of BC, based on the St. Gallen Consensus 2013 (Goldhirsch et al. ). AR Clone SP107 (Cell-MarqueTM, Rocklin, CA, USA) was used to determine AR expression. IHC and SISH analysis were performed as previously described (Orrù et al. ). BCL11A clone 14B 5 (dilution 1:100, ab19487, AbCam, Cambridge, USA) was used to determine BCL11A expression. The ab19487 antibody, whose epitope is in core of amino acids 172–434, can identify the BCL11A-XL and BCL11A-L isoforms. BCL11A immunostaining was performed using the Ventana Benchmark XT staining system with an Optiview DAB detection kit. IHC analysis was performed on 87 BC and 12 normal breast tissue (NBT) FFPE block samples. Also, 343 TNBC tissue microarrays (TMAs) were used. Evaluation of immunohistochemical staining ER and PR expression were positive if at least 1% immunostained tumor nuclei were detected in the sample, according to the American Society of Clinical Oncology/College of American Pathologists (ASCO/CAP) recommendations for immunohistochemical testing of hormone receptors in BC (Hammond et al. ), whose criteria have recently been adopted by WHO classification (Rakha et al. ). The Ki67 cut-offs < 14, 15–35% and > 35% were based on results previously obtained (Urru et al. ); AR expression was considered positive if at least 10% immunostained tumor nuclei were detected in the sample (Park et al. ). All IHC expressions were categorized using a semi-quantitative method. Based on IHC approach the following BC surrogate intrinsic subtypes were found: nine Luminal A [ER and PR expression positive, with PR cut point of ≥ 20%, HER2 negative and Ki-67 low (< 14%)]; nine Luminal B [ER expression positive, PR expression negative or low, HER2 expression negative and Ki-67 high (> 14%), or ER expression positive, HER2 protein positive or HER2 gene amplified, any PR and any Ki-67]; eight HER2-enriched [ER and PR expression negative, HER2 protein positive or HER2 gene amplified]; sixty-one TNBC [ER, PR and HER2 expression negative or HER2 gene not amplified]. The ordinal Allred scoring system was used to assess BCL11A immunostaining quantity in tumor cells, based on intensity (0, negative; 1 + , weak; 2 + , moderate; 3 + , strong) and percentage of stained cells (0 = 0%, 1 = < 1%, 2 = 1–10%, 3 = 11–33%, 4 = 34–66% and 5 = > 66%); the combination of intensity + percentage gives an Allred score between 0 and 8. Tumor with Allred score > 2 was defined as positive for BCL11A expression (Khaled et al. ). Acid nucleic extraction Genomic DNA was obtained from neoplastic tissue, and total RNA was obtained from neoplastic and non-neoplastic specimens. Nucleic acids were extracted using the QIAmp DNA Mini Kit and miRNeasy Mini Kit (Qiagen, Hilden, Germany). The quantity and the quality of nucleic acids were assessed using Nanodrop ND1000 (Euro-Clone, Milan, Italy). The RNA quantity was evaluated by Qubit ® RNA BR Assay Kit (ThermoFisher Scientific, Waltham, USA). The RNA integrity was assessed by the RNA Integrity Number (RIN) using the Agilent RNA 6000 Nano Kit on the BioAnalyzer 2100 (Agilent, Santa Clara, USA). Quantitative real time PCR Gene expression profiles of BCL11A were analyzed in all BC molecular intrinsic subtypes. Two µg of total RNA were reverse transcribed to cDNA using the High-Capacity cDNA Reverse Transcription Kit (Applied Biosystem, Foster City, CA, USA). BCL11A encodes three mRNA variants and each isoform of BCL11A has specific expression patterns. Primers for BCL11A (Hs01076078_m1, 60 bp), the isoforms BCL11A-S (Hs01093198_m1), BCL11A-L (Hs01093199-m1), BCL11A-XL (Hs00250581_s1) and 18S rRNA (Hs99999901_S1, 187 bp) human genes were chosen using Assays-on-Demand™-Products (Applied Biosystems). Neoplastic and non-neoplastic tissues were analyzed by quantitative real time PCR (qRT-PCR) using the ABI 7900HT Sequence Detection System (Applied Biosystems) (Cossu-Rocca et al. ). The relative mRNA expression level was analyzed according to the Applied Biosystem User Bulletin N°2. The calculation 2-ΔΔCt (Fold Change, FC) was chosen to represent the level of expression, with a FC > 2 being considered as overexpression. Mutation analysis BCL11A gene mutation analysis was performed on exon 4 encoding five of the six Kruppel-like zinc-finger domains (C2H2) of the BCL11A-XL protein, where several most common missense mutations were identified in patients affected by autism, intelligence disabilities (Cai et al. ), and ovarian cancer (Er et al. ): the exon 4 contains almost all the BCL11A single nucleotide polymorphisms. Amplification of the exon 4 and Sanger sequencing analysis were performed in all BC molecular subtypes analyzed for gene expression profile, using the following sequence primers: BCL11A_ex4_F2:5ʹ-ACCGCATAGACGATGGCAC-3ʹ and BCL11A_ex4_R2:5ʹ-CCCCGAGATCCCTCCGT-3ʹ (De Miglio et al. ). Statistical analysis An ad hoc electronic form was created to collect qualitative and quantitative variables. Qualitative data were summarized with absolute and relative (percentages) frequencies. Chi-squared or Fisher exact tests were used to detect any statistical differences in the comparison of qualitative variables between down and up regulation of BCL11A gene or low and high protein expression. Logistic regression analysis was performed to assess the relationship between BCL11A upregulation or high protein expression and clinicopathological TNBC characteristics. Survival rate differences between down and upregulation or low and high protein expression were detected with Kaplan–Meier analysis. P-value less than 0.05 was considered statistically significant. Stata 17 (StataCorp, TX) statistical software was used for every statistical computation.
ER, PR, HER2 and Ki-67 immunohistochemical expression and/or HER2 gene amplification, as defined by silver enhanced SISH, established the surrogate intrinsic subtypes of BC, based on the St. Gallen Consensus 2013 (Goldhirsch et al. ). AR Clone SP107 (Cell-MarqueTM, Rocklin, CA, USA) was used to determine AR expression. IHC and SISH analysis were performed as previously described (Orrù et al. ). BCL11A clone 14B 5 (dilution 1:100, ab19487, AbCam, Cambridge, USA) was used to determine BCL11A expression. The ab19487 antibody, whose epitope is in core of amino acids 172–434, can identify the BCL11A-XL and BCL11A-L isoforms. BCL11A immunostaining was performed using the Ventana Benchmark XT staining system with an Optiview DAB detection kit. IHC analysis was performed on 87 BC and 12 normal breast tissue (NBT) FFPE block samples. Also, 343 TNBC tissue microarrays (TMAs) were used.
ER and PR expression were positive if at least 1% immunostained tumor nuclei were detected in the sample, according to the American Society of Clinical Oncology/College of American Pathologists (ASCO/CAP) recommendations for immunohistochemical testing of hormone receptors in BC (Hammond et al. ), whose criteria have recently been adopted by WHO classification (Rakha et al. ). The Ki67 cut-offs < 14, 15–35% and > 35% were based on results previously obtained (Urru et al. ); AR expression was considered positive if at least 10% immunostained tumor nuclei were detected in the sample (Park et al. ). All IHC expressions were categorized using a semi-quantitative method. Based on IHC approach the following BC surrogate intrinsic subtypes were found: nine Luminal A [ER and PR expression positive, with PR cut point of ≥ 20%, HER2 negative and Ki-67 low (< 14%)]; nine Luminal B [ER expression positive, PR expression negative or low, HER2 expression negative and Ki-67 high (> 14%), or ER expression positive, HER2 protein positive or HER2 gene amplified, any PR and any Ki-67]; eight HER2-enriched [ER and PR expression negative, HER2 protein positive or HER2 gene amplified]; sixty-one TNBC [ER, PR and HER2 expression negative or HER2 gene not amplified]. The ordinal Allred scoring system was used to assess BCL11A immunostaining quantity in tumor cells, based on intensity (0, negative; 1 + , weak; 2 + , moderate; 3 + , strong) and percentage of stained cells (0 = 0%, 1 = < 1%, 2 = 1–10%, 3 = 11–33%, 4 = 34–66% and 5 = > 66%); the combination of intensity + percentage gives an Allred score between 0 and 8. Tumor with Allred score > 2 was defined as positive for BCL11A expression (Khaled et al. ).
Genomic DNA was obtained from neoplastic tissue, and total RNA was obtained from neoplastic and non-neoplastic specimens. Nucleic acids were extracted using the QIAmp DNA Mini Kit and miRNeasy Mini Kit (Qiagen, Hilden, Germany). The quantity and the quality of nucleic acids were assessed using Nanodrop ND1000 (Euro-Clone, Milan, Italy). The RNA quantity was evaluated by Qubit ® RNA BR Assay Kit (ThermoFisher Scientific, Waltham, USA). The RNA integrity was assessed by the RNA Integrity Number (RIN) using the Agilent RNA 6000 Nano Kit on the BioAnalyzer 2100 (Agilent, Santa Clara, USA).
Gene expression profiles of BCL11A were analyzed in all BC molecular intrinsic subtypes. Two µg of total RNA were reverse transcribed to cDNA using the High-Capacity cDNA Reverse Transcription Kit (Applied Biosystem, Foster City, CA, USA). BCL11A encodes three mRNA variants and each isoform of BCL11A has specific expression patterns. Primers for BCL11A (Hs01076078_m1, 60 bp), the isoforms BCL11A-S (Hs01093198_m1), BCL11A-L (Hs01093199-m1), BCL11A-XL (Hs00250581_s1) and 18S rRNA (Hs99999901_S1, 187 bp) human genes were chosen using Assays-on-Demand™-Products (Applied Biosystems). Neoplastic and non-neoplastic tissues were analyzed by quantitative real time PCR (qRT-PCR) using the ABI 7900HT Sequence Detection System (Applied Biosystems) (Cossu-Rocca et al. ). The relative mRNA expression level was analyzed according to the Applied Biosystem User Bulletin N°2. The calculation 2-ΔΔCt (Fold Change, FC) was chosen to represent the level of expression, with a FC > 2 being considered as overexpression.
BCL11A gene mutation analysis was performed on exon 4 encoding five of the six Kruppel-like zinc-finger domains (C2H2) of the BCL11A-XL protein, where several most common missense mutations were identified in patients affected by autism, intelligence disabilities (Cai et al. ), and ovarian cancer (Er et al. ): the exon 4 contains almost all the BCL11A single nucleotide polymorphisms. Amplification of the exon 4 and Sanger sequencing analysis were performed in all BC molecular subtypes analyzed for gene expression profile, using the following sequence primers: BCL11A_ex4_F2:5ʹ-ACCGCATAGACGATGGCAC-3ʹ and BCL11A_ex4_R2:5ʹ-CCCCGAGATCCCTCCGT-3ʹ (De Miglio et al. ).
An ad hoc electronic form was created to collect qualitative and quantitative variables. Qualitative data were summarized with absolute and relative (percentages) frequencies. Chi-squared or Fisher exact tests were used to detect any statistical differences in the comparison of qualitative variables between down and up regulation of BCL11A gene or low and high protein expression. Logistic regression analysis was performed to assess the relationship between BCL11A upregulation or high protein expression and clinicopathological TNBC characteristics. Survival rate differences between down and upregulation or low and high protein expression were detected with Kaplan–Meier analysis. P-value less than 0.05 was considered statistically significant. Stata 17 (StataCorp, TX) statistical software was used for every statistical computation.
BCL11A expression in molecular intrinsic subtypes of breast cancer Eighty-seven primary BC, comprising all molecular subtypes, were analyzed by gene expression profiling by qRT-PCR. The overall high expression of BCL11A and each of its transcripts (BCL11A-XL, BCL11A-L and BCL11A-S) significantly correlated with TNBC pathology ( P < 0.05) (Fig. A). We found a significant BCL11A overexpression in TNBC compared to Luminal A (P: 0.004) and B (P: 0.002) while a significant BCL11A downregulation was present in Luminal A and B compared to NBT (P: 0.002 and P < 0.001, respectively). No significant differences were shown between HER2-enriched and other molecular intrinsic subtypes and NBT. BCL11A-XL was overexpressed in TNBC vs Luminal A and B (P: 0.012 and P: 0.040, respectively), whereas BCL11A-L and BCL11A-S were overexpressed in TNBC vs Luminal B (P: 0.003 and P: 0.011, respectively) (Fig. B). Focusing on BCL11A protein expression profile we performed IHC on the 87 primary BC molecular subtypes used for gene expression profile. A BCL11A protein overexpression was found in 16 out of 61 (26.2%) TNBCs, in 4 out of 12 (33.3%) NBT, whereas no protein expression was detected in Luminal A (nine cases), Luminal B (nine cases) and HER2-enriched (eight cases) tumors. The tumors immunostained positively showed high mRNA levels compared with those with negative immunostaining. Immunohistochemistry analysis of BCL11A in an independent validation cohort of 343 TNBC samples, confirmed that BCL11A protein expression agreed with the first cohort examined: 79 BCL11A-overexpressing TNBCs out of 343 (23.0%). Figure showed a representative BCL11A protein expression of BC molecular intrinsic subtypes. BCL11A expression profile and association with TNBC clinic-pathological data Table showed the clinic-pathological features of the 61 TNBC patients included in the expression profile analysis. The median (interval quartile range, IQR) age at diagnosis was 57 (31–84) years, with 39 (63.9%) older than 50 years. Forty (65.6%) tumors were ductal, 9 (14.8%) medullary, 4 (6.6%) metaplastic. Tumor staging was pT1 in 24 (42.9%) cases, pT2 in 26 (46.4%), pT3 in 3 cases (5.4%), pT4 in 3 cases (5.4%). Lymph node status was divided into 31 pN0 (53.5%), 16 pN1 (27.6%), 6 pN2 (10.3%) and 5 pN3 (8.6%). Moreover, 24.1% of tumors were stage I, 53.7% stage II, and 22.2% stage III; 4.9% of TNBCs were G1, 13.1% G2, and 82.0% G3. Ki-67 expression was > 20% in 80.3% of TNBCs. Necrosis was present in 35.1%. Tumor infiltrating lymphocyte (TIL) and lymphovascular invasion (LVI) were detected in 52.9 and 25.5%, respectively. AR expression was found in 30.9% cases. A total of 8 patients out of 61 (13.1%) died. The clinicopathological data of the validation cohort is reported in Table S1. TNBCs with BCL11A and BCL11A-L mRNA overexpression were more frequently associated with AR expression < 10% ( P : 0.05). BCL11A-L mRNA overexpression was associated with some histological types such as medullary and metaplastic carcinomas (P: 0.04) (Table ). BCL11A protein expression was associated with ki-67 > 35% (P: 0.004), and with absence of LIV and AR downregulation (P: 0.03 and P: 0.02, respectively) (Table ). BCL11A expression profile and association with TNBC survival Logistic regression analysis revealed that histological type (HR, 0.2; 95% CI 0.1–0.8; P: 0.02) and AR expression (HR, 0.2; 95% CI 0.0–1.0; P: 0.05) are independent prognostic factors for overall survival (OS) in BCL11A-L transcripts overexpressing TNBCs. High protein expression levels of BCL11A (HR, 17.1; 95% CI 4.0–72.2; P < 0.001) are independent prognostic factors for TNBCs overexpressing mRNA BCL11A or its isoforms (Tables ). LIV (HR, 0.52; 95% CI 0.29–0.92; P: 0.03) and AR (HR, 0.37; 95% CI 0.16–0.88; P: 0.02) are independent prognostic factors for TNBCs showing high BCL11A protein expression levels (Table ). Kaplan–Meier curve for OS showed no differences among TNBCs with overexpression of BCL11A transcripts and its isoforms in comparison with those downregulated. We observed the same trend for TNBCs with high protein expression levels, analyzing the entire cohort of tumors included in the study (Fig. ). BCL11A mutational analysis in molecular intrinsic subtypes of breast cancer Sequencing of BCL11A exons 4 did not find any genomic variation in our BC molecular cohort, expect the rs7569946. This synonymous substitution C vs T (Phe699Phe), was detected in all BC molecular subtypes. CC genotype was prevalent in all BC molecular subtypes (60–62.5%). In TNBC subtype, no TT homozygous were present while 40% of them showed CT genotype.
Eighty-seven primary BC, comprising all molecular subtypes, were analyzed by gene expression profiling by qRT-PCR. The overall high expression of BCL11A and each of its transcripts (BCL11A-XL, BCL11A-L and BCL11A-S) significantly correlated with TNBC pathology ( P < 0.05) (Fig. A). We found a significant BCL11A overexpression in TNBC compared to Luminal A (P: 0.004) and B (P: 0.002) while a significant BCL11A downregulation was present in Luminal A and B compared to NBT (P: 0.002 and P < 0.001, respectively). No significant differences were shown between HER2-enriched and other molecular intrinsic subtypes and NBT. BCL11A-XL was overexpressed in TNBC vs Luminal A and B (P: 0.012 and P: 0.040, respectively), whereas BCL11A-L and BCL11A-S were overexpressed in TNBC vs Luminal B (P: 0.003 and P: 0.011, respectively) (Fig. B). Focusing on BCL11A protein expression profile we performed IHC on the 87 primary BC molecular subtypes used for gene expression profile. A BCL11A protein overexpression was found in 16 out of 61 (26.2%) TNBCs, in 4 out of 12 (33.3%) NBT, whereas no protein expression was detected in Luminal A (nine cases), Luminal B (nine cases) and HER2-enriched (eight cases) tumors. The tumors immunostained positively showed high mRNA levels compared with those with negative immunostaining. Immunohistochemistry analysis of BCL11A in an independent validation cohort of 343 TNBC samples, confirmed that BCL11A protein expression agreed with the first cohort examined: 79 BCL11A-overexpressing TNBCs out of 343 (23.0%). Figure showed a representative BCL11A protein expression of BC molecular intrinsic subtypes.
Table showed the clinic-pathological features of the 61 TNBC patients included in the expression profile analysis. The median (interval quartile range, IQR) age at diagnosis was 57 (31–84) years, with 39 (63.9%) older than 50 years. Forty (65.6%) tumors were ductal, 9 (14.8%) medullary, 4 (6.6%) metaplastic. Tumor staging was pT1 in 24 (42.9%) cases, pT2 in 26 (46.4%), pT3 in 3 cases (5.4%), pT4 in 3 cases (5.4%). Lymph node status was divided into 31 pN0 (53.5%), 16 pN1 (27.6%), 6 pN2 (10.3%) and 5 pN3 (8.6%). Moreover, 24.1% of tumors were stage I, 53.7% stage II, and 22.2% stage III; 4.9% of TNBCs were G1, 13.1% G2, and 82.0% G3. Ki-67 expression was > 20% in 80.3% of TNBCs. Necrosis was present in 35.1%. Tumor infiltrating lymphocyte (TIL) and lymphovascular invasion (LVI) were detected in 52.9 and 25.5%, respectively. AR expression was found in 30.9% cases. A total of 8 patients out of 61 (13.1%) died. The clinicopathological data of the validation cohort is reported in Table S1. TNBCs with BCL11A and BCL11A-L mRNA overexpression were more frequently associated with AR expression < 10% ( P : 0.05). BCL11A-L mRNA overexpression was associated with some histological types such as medullary and metaplastic carcinomas (P: 0.04) (Table ). BCL11A protein expression was associated with ki-67 > 35% (P: 0.004), and with absence of LIV and AR downregulation (P: 0.03 and P: 0.02, respectively) (Table ).
Logistic regression analysis revealed that histological type (HR, 0.2; 95% CI 0.1–0.8; P: 0.02) and AR expression (HR, 0.2; 95% CI 0.0–1.0; P: 0.05) are independent prognostic factors for overall survival (OS) in BCL11A-L transcripts overexpressing TNBCs. High protein expression levels of BCL11A (HR, 17.1; 95% CI 4.0–72.2; P < 0.001) are independent prognostic factors for TNBCs overexpressing mRNA BCL11A or its isoforms (Tables ). LIV (HR, 0.52; 95% CI 0.29–0.92; P: 0.03) and AR (HR, 0.37; 95% CI 0.16–0.88; P: 0.02) are independent prognostic factors for TNBCs showing high BCL11A protein expression levels (Table ). Kaplan–Meier curve for OS showed no differences among TNBCs with overexpression of BCL11A transcripts and its isoforms in comparison with those downregulated. We observed the same trend for TNBCs with high protein expression levels, analyzing the entire cohort of tumors included in the study (Fig. ).
Sequencing of BCL11A exons 4 did not find any genomic variation in our BC molecular cohort, expect the rs7569946. This synonymous substitution C vs T (Phe699Phe), was detected in all BC molecular subtypes. CC genotype was prevalent in all BC molecular subtypes (60–62.5%). In TNBC subtype, no TT homozygous were present while 40% of them showed CT genotype.
BCL11A is a proto-oncogene which maps on chromosome 2p16. Alternative splicing generates at least three most common BCL11A transcripts, BCL11A-XL, BCL11A-L and BCL11A-S containing differing numbers of C-terminal C2H2 finger motifs, and showing low expression in normal human tissue, except in fetal liver, hematopoietic tissue and brain (Yin et al. ). The BCL11A-XL mRNA is the prevalent transcript (Satterwhite et al. ). BCL11A acts as a transcription repressor directly binding to its DNA target sequence, 5ʹ-GGCCGG-3ʹ (Avram et al. ) and/or indirectly interacting with and repressing other sequence specific transcription factors, such as COUP-TFs (Avram et al. ). BCL11A is an oncogene of different malignant hematological diseases (Weniger et al. ; Nakamura et al. ). Recently, the pathogenetic role of BCL11A was also highlighted in solid tumors (e.g., lung, prostate, breast cancer, endometrial carcinoma, laryngeal squamous carcinoma) (Zhang et al. , ; Jiang et al. ; Khaled et al. ; Zhou et al. ; Chen et al. ; Wang et al. ). In our study, BCL11A was significantly overexpressed in TNBC both at transcriptional and translational levels compared to the other BC molecular subtypes. Gene expression profiling showed that high expression levels of BCL11A and its isoforms (BCL11A-XL, BCL11A-L and BCL11A-S) significantly correlated with TNBC pathology. Additionally, tumors positively immunostained showed high BCL11A mRNA levels compared with those with negative immunostaining. Our results confirmed recent data correlating BCL11A overexpression and TNBC subtype (Khaled et al. ). We found BCL11A protein expression in 26% of TNBCs in our cohort, likewise to the 29.6% reported by Chen et al. (Chen et al. ), in contrast with Khaled et al. (67% of BCL11A expression in TNBC with basal-like features) (Khaled et al. ) and Wang et al. (100% of BCL11A expression in TNBC using a different score to define BCL11A overexpression) (Wang et al. ). The lower percentage of BCL11A protein expression detected in our cohort could depend on several factors: the definition of BCL11A expression by several operators, the cut-off values used, or the analysis performed on all TNBCs despite classification into molecular sub-classes. Regarding the prognostic significance, we showed that BCL11A protein expression acts as an unfavorable prognostic factor in TNBC patients. Metaplastic and medullary histotypes, absence of LIV and AR downregulation can be considered prognostic factors in patients with BCL11A overexpressing TNBC. Moreover, BCL11A overexpressing TNBCs were associated with a higher proliferation index (> 35%). Among TNBC histotypes, the medullary type of pattern is often associated with variable immunohistochemical expression of basal markers (Rakha et al. ). Our previous findings confirmed that medullary and metaplastic carcinomas exhibit higher grades (G3) and higher proliferation index (Ki67 > 30%), while LVI was detected in only 7.4% of medullary carcinomas. Metaplastic carcinoma had poor 5 and 10 year survival in comparison with other histologic types (Sanges et al. ). We found a negative relationship between LVI and BCL11A expression, in contrast with previous results that gave no significant differences (Shen et al. ). However, Ugras et al. demonstrated that LVI and nodal metastases were less frequent in TNBC vs other BC subtypes (Ugras et al. ). Based on previous findings we could speculate that in BCL11A overexpressing TNBC the worse prognosis is not related to LVI rate. Our data showed an inverse association between BCL11A overexpression and AR expression levels in TNBCs. Considering that patients with LAR TNBC showed the best OS compared to the other TNBCs subtypes (Masuda et al. ), our results might suggest that BCL11A can be a biomarker for more aggressive non luminal TNBCs subgroups. Choi et al. findings could support previous hypothesis, showing that the inhibition of BCL11A and HDAC1/2 effectively reprogramming basal like cancer cells into luminal A cells, increasing ER expression and leading to tamoxifen sensitivity (Choi et al. ). In contrast with our results, Wang et al. identified a positive correlation between AR and BCL11A expression by analyzing all BC molecular subtypes (Wang et al. ). Our survival analysis did not show any relationship between BCL11A gene and/or protein expression and patient outcomes. Khaled et al. demonstrated that patients with copy number (CN) gains of BCL11A had a higher rate of relapse and metastasis and a lower rate of survival (Khaled et al. ). The differences could be related to the selection of TNBC with basal like phenotype included in the Khaled’s study, compared to our study in which all TNBC phenotypes, included LAR, were all considered. No nucleotide variants were found in BCL11A exon 4. The literature data demonstrates the presence of different genomic alterations for this gene in malignant diseases, as well as CV amplification, epigenetic deregulation, translocation or abnormal activation upon viral integration (Boelens et al. ; Jiang et al. ; Yin et al. ). We recognize that our study does have some limitations mainly related to its retrospective nature: key clinical follow-up data were unfortunately not found in medical records.
Our study highlights the role of BCL11A and its correlation with clinicopathological features of TNBC. BCL11A expression seems to be a poor prognostic factor in TNBC patients. BCL11A may become a prognostic factor for more aggressive non luminal TNBCs subgroups, with the worse prognosis of BCL11A overexpressing TNBC not related to LVI. Furthermore, BCL11A was overexpressed in more aggressive histologic types, such as metaplastic and medullary carcinomas. These results may provide a new paradigm for TNBC classification and a better treatment strategy.
Below is the link to the electronic supplementary material. Supplementary file1 (DOCX 18 KB)
|
Solomon Islands Oncology Unit: Sustainability in Terms of Outcomes | 841865c8-390d-48c0-971f-3bdafc4efd0d | 10830091 | Internal Medicine[mh] | A recent article reported on the establishment of the first oncology unit at the National Referral Hospital (NRH) in Solomon Islands. This project was accomplished with support from the Australian Government Department of Foreign Affairs and Trade. In their article, the authors write that Solomon Islands now has a sustainable oncology unit delivering chemotherapy treatments. Yip et al detail the efforts of Australian and Solomon Islands medical staff to improve oncology services. We applaud the establishment of the oncology unit at the NRH. However, we take this as a heuristic opportunity to raise and reiterate the ongoing critical challenge associated with the sustainability of donor-supported international health initiatives. In their article, the authors do not demonstrate how the new oncology unit has affected clinical outcomes beyond increasing case load. Moreover, it is unclear what the authors mean by sustainability. Recent reports from the head of the NRH oncology unit highlight continued shortages of medication, space, and manpower that negatively affect clinical care. Thus we ask, how is this an example of a sustainable project? Although debate continues about what sustainability means in global health, patient outcomes are an important measure of success and a definitive component of sustainability. Additionally, outcomes and reductions in health inequities—which require ongoing monitoring of clinical outcomes—must continue to improve after outside support decreases or ceases. A recent study by Cancedda et al argues that an outcome-focused, tangible, dynamic, and integrated approach is key to achieving sustainability in global health. Such an approach requires international funders and partners to provide longer-term and more integrated investments that allow for programmatic flexibility and adaptation to the local context. The WHO's recent Health Systems Resilience Toolkit defines resilient health systems in part by their ability to bring about positive health outcomes. Although the training of local staff and the creation of a dedicated oncology unit are important accomplishments, they do little to improve patient outcomes if chemotherapeutic agents are unavailable. Although the authors mention a 2018 visit to the Solomon Islands National Medical Store to discuss issues with medication supply chain, they do not explain what, if any, efforts were made to resolve or alleviate these deficits. Furthermore, while the authors point to an increased case load (39 new cases in 2019, 99 new cases in 2023) as evidence of the program's success, it is unknown whether these patients received a full, uninterrupted course of chemotherapy. Existing research suggests a strong link between uninterrupted anticancer therapy and response to therapy. - Reports from the oncology unit indicate that interruptions in chemotherapy are commonplace, with patients waiting up to 3 months to reinitiate therapy because of drug shortages. The establishment of an oncology unit is an important first step in improving cancer treatment in Solomon Islands. To make oncological services more sustainable, further efforts should be made to improve the provision of chemotherapeutic medications and monitoring of clinical outcomes. This requires continued international financial support and a more integrated, adaptive, longer-term approach. Only then will the goal of a resilient and sustainable health system in Solomon Islands manifest.
|
Blockade of insulin receptor signaling in the medullary cardiovascular centers impairs open‐loop arterial baroreflex function via attenuated neural arc in healthy male rats | 730f488f-d3c2-47e7-88da-abb207c85076 | 11867053 | Cardiovascular System[mh] | INTRODUCTION Insulin in the periphery is well known to play an important role in the regulation of glucose and lipid metabolism. Increasing evidence demonstrates that insulin also modulates neural activity in the central nervous system. For example, exogenous insulin delivery into the brain activates the sympathetic nervous system. , , , However, it remains to be fully elucidated how brain insulin modifies autonomic cardiovascular regulation. The arterial baroreflex is a negative feedback system that modulates autonomic nerve activity to maintain arterial pressure (AP) at rest and during physical activity. Arterial baroreflex dysfunction is known to induce AP lability, making high AP variability a risk factor for the development of cardiovascular disease and mortality. Evidence suggests that brain insulin availability may acutely modulate arterial baroreflex function. For example, Pricher et al. reported that lateral intracerebroventricular (ICV) injection of insulin potentiates arterial baroreflex function. Additionally, Daubert et al. found impaired arterial baroreflex function and decreased cerebrospinal fluid (CSF) insulin levels in pregnant animals. Furthermore, a follow‐up study demonstrated that ICV infusion of insulin improves the impaired arterial baroreflex function in pregnant rats. Moreover, it is well known that the action of insulin depends not only on the amount of insulin available, but also on the downstream signaling pathway at the insulin receptor (IR). , In support of this concept, impaired brain IR signaling, observed in diabetes mellitus (DM), , has been implicated in the generation of AP lability due to baroreflex impairment. However, to date, it remains to be elucidated whether changes in downstream IR signaling (achieved experimentally by blocking endogenous IR directly) alter baroreflex function. The arterial baroreflex is divided into the following two principal subsystems: the neural arc, which determines sympathetic nerve activity (SNA) in response to a baroreceptor pressure input, and the peripheral arc, which determines AP in response to SNA. Under normal physiological conditions, the arterial baroreflex operates as a closed‐loop negative feedback system. However, the neural and peripheral arcs usually cannot be evaluated under the baroreflex closed‐loop conditions since AP cannot be separated into the input and output pressures. Furthermore, for the same reason, the baroreflex closed‐loop conditions usually do not allow evaluation of baroreflex control of AP (i.e., the total reflex arc that determines the AP in response to the baroreceptor pressure input). With these experimental considerations in mind, to precisely assess arterial baroreflex function, a baroreflex open‐loop analysis must be performed. This can be established by surgically isolating the carotid sinus baroreceptor area from the systemic circulation. The purpose of this study was therefore to investigate the effects of an ICV injection of IR antagonist on arterial baroreflex function under open‐loop conditions in healthy male rats. We hypothesized that the blockade of IR signaling in the brain acutely impairs arterial baroreflex function via alterations in the neural arc, but not the peripheral arc, independent of circulating insulin and glucose levels. It is well known in the sympathetic baroreflex that carotid sinus baroreceptors project via sensory afferents to the nucleus tractus solitarius (NTS) followed by the caudal ventrolateral medulla (CVLM) and then the rostral ventrolateral medulla (RVLM) in sympathetic baroreflex function. , Moreover, insulin receptors are known to be expressed in numerous areas throughout the brainstem involved in regulating the cardiovascular system. Thus, we further investigated 1) whether IR‐positive neurons co‐localize with c‐Fos (a marker of neuronal activation)‐positive neurons activated by baroreflex stimulation in the NTS, the CVLM, and the RVLM by using immunofluorescence staining, and 2) whether blockade of IR signaling potentiates AP lability against hypertensive/hypotensive stress using baroreflex equilibrium diagram simulation.
METHODS 2.1 Ethical approval This study was performed in accordance with the U.S. Department of Health and Human Services NIH Guide for the Care and Use of Laboratory Animals . The Institutional Animal Care and Use Committee of the University of Texas Southwestern Medical Center approved this study (no. 2019–102 849), and all experimental procedures were conducted following approved institutional guidelines and regulations. 2.2 Animals Thirty‐five healthy male Sprague–Dawley rats (12–16 weeks, body weight: 384 ± 26 g) (Inotiv, West Lafayette, IN, USA) were used for the fluorescence immunohistochemistry and the baroreflex open‐loop in vivo experiment. The animals had free access to food and clean water and were kept one to four per cage under a 12 h light–dark cycle in an air‐conditioned room (22–24°C) until required for terminal experiments. 2.3 Fluorescence immunohistochemistry In six anesthetized rats, the bilateral carotid sinus baroreceptor regions were surgically isolated from the systemic circulation. The surgical and anesthesia procedures for the preparation of the baroreflex stimulation under the open‐loop conditions were described in the “ Open‐loop baroreflex in vivo experiment ” section below. We confirmed the adequacy of anesthesia by lack of a withdrawal response to tail pinch. In the baroreflex stimulation treated group ( n = 3), following the 60‐min post‐surgery recovery period, repetitive baroreflex stimulation was performed. By the procedure described below, carotid sinus pressure (CSP) was first decreased to 60 mm Hg for 4 min and then increased in a stepwise manner up to 180 mmHg in increments of 20 mmHg every minute. Peak c‐Fos expression is known to occur approximately 60–120 min following the neural activation. , Thus, the CSP input cycle was repeated for 120 min, followed by a 60‐min resting period under the closed‐loop conditions where CSP was matched to AP. In contrast, the control group ( n = 3) was under the closed‐loop conditions for the same period (i.e., 3 h) after the 60‐min post‐surgery recovery period. Immediately after the 60‐min resting period, transcardial perfusion with physiological saline was performed for 15 min followed by 15‐min 4% paraformaldehyde for tissue fixation. Thus, the tissue fixation was achieved within approximately 90 min after the baroreflex stimulation period. The brain tissue was harvested and post‐fixed overnight in 4% paraformaldehyde followed by dehydrated sequentially in 10% and 20% sucrose. The present immunofluorescence staining was adopted from our previous study. Briefly, after submerging in an optimal cutting temperature medium and freezing on dry ice, the brainstem tissue was sectioned at 35 μm using a cryostat (2800 Frigocut, Reichert Jung/Leica, Deer Park, IL, USA). The sections were washed in phosphate‐buffered saline (PBS). After blocking with an antibody buffer solution (PBS, 0.5% Triton‐X100, 5% normal goat serum [NGS], and 10 mg/mL bovine serum albumin [BSA]) for 60 min, co‐immunostaining for IR and c‐Fos was performed by incubation with the primary antibodies (mouse anti‐IRβ, 1:100, cat. no. sc‐57 342, RRID#AB_784102, Santa Cruz Biotechnology, Dallas, TX, USA; rabbit anti‐c‐Fos, 1:1000, cat. no. 2250, RRID#AB_2247211, Cell Signaling Technology, Danvers, MA, USA) overnight at room temperature. The sections were rinsed in PBS with 5% NGS and 10 mg/mL BSA and then incubated with fluorescence‐conjugated secondary antibodies (goat anti‐mouse Cy3, 1:500, cat. no. 115–165‐146, RRID#AB_2338690, Jackson ImmunoResearch Laboratories, West Grove, PA, USA; goat anti‐rabbit Alexa Fluor Plus 488, 1:500, cat. no. A32731, RRID#AB_2633280, Thermo Fisher Scientific, Waltham, MA, USA) for 60 min at room temperature. We previously used the same primary and secondary antibody combinations to perform the triple labeling of IR, c‐Fos, and DAPI in the NTS. Then, sections were rinsed with PBS with 5% NGS and 10 mg/mL BSA and pure water, and then fixed onto charged slides. The mounting medium with 4′,6‐diamidino‐2‐phenylindole (DAPI) (Mounting Medium With DAPI, cat. no. ab104139, Abcam, Cambridge, MA, USA) was applied to charged slides before coverslipping with micro cover glass. We performed negative control experiments to ensure the specificity of the antibodies. A fluorescence microscope system (Axio Imager A2, Zeiss, Oberkochen, Germany) was used for observing and obtaining the fluorescence images as described previously. , To evaluate whether c‐Fos‐positive neurons activated by baroreflex stimulation were specifically observed in medullary cardiovascular centers (the NTS, CVLM, and RVLM), we additionally investigated the degree of IR, c‐Fos, and DAPI overlap in the spinal trigeminal nucleus (Vsp) which is believed not to be activated by arterial baroreflex. The NTS, CVLM, RVLM, and Vsp were, respectively, determined as the regions from about −15.4 to −13.6 mm, −13.4 to −13.0 mm, −12.4 to −12.0 mm, and −15.4 to −12.0 mm to the bregma based on the rat brain atlas and previous studies. , , For quantitative analysis, the sections of the NTS (10 sections each rat), CVLM (3 sections each rat), and RVLM (3 sections each rat) were serially selected at approximately 2‐mm intervals from each rat. As for the Vsp, we collected three sections from −15.4 to −13.6 mm to the bregma region and one section each from −12.4 to −12.0 mm to the bregma and from −13.4 to −13.0 mm to the bregma regions (i.e., total 5 sections each rat). A series of grayscale photomicrographs at 10× magnification were acquired, and then we performed triple overlap analysis using image analysis software (FIJI; National Institute of Health, Maryland, USA) as we did previously. Briefly, the image background was subtracted within the region of interest. Subsequently, individualized thresholds (Triangle threshold, c‐Fos; Otsu threshold, IR and DAPI), and then particle size filters were applied for each binary image. Finally, IR + , c‐Fos + , and DAPI + areas were each calculated, as well as calculating the overlap between IR + /c‐Fos + /DAPI + areas. 2.4 Open‐loop baroreflex in vivo experiment 2.4.1 General surgical experimental procedures Rats were initially anesthetized with an intraperitoneal injection of a cocktail consisting of urethane (800 mg/kg) and α‐chloralose (65 mg/kg), and then mechanically ventilated with 100% oxygen gas after intubation. To maintain the depth of anesthesia, a 7‐fold dilution of the above anesthetic cocktail was continuously infused via the right femoral vein at a rate of 3–5 mL/h/kg. The depth of anesthesia was evaluated and monitored by tail pinch until all protocol was completed. Body temperature was maintained at approximately 36.5–37°C using a heating pad and a heat lamp. A catheter was inserted into the right femoral artery and connected to a pressure transducer (MLT0670, ADInstruments, Sydney, Australia) to measure AP. Electrocardiogram (ECG) recordings were obtained using needle electrodes. For recording renal SNA (RSNA), a branch of the isolated left renal nerve was attached to the electrodes (Platinum, AM Systems, Sequim, WA, USA) and covered with silicone glue (Kwik‐Sil, World Precision Instruments, Sarasota, FL, USA) for insulation and fixation. To validate that most of the RSNA recording was obtained from postganglionic renal sympathetic fibers, hexamethonium bromide (60 mg/kg) was administered intravenously after all experiments were completed. Additionally, RSNA background noise was measured over a 30‐min period after animals were euthanized by intravenous injection of saturated potassium chloride (4 M, 2 mL/kg) under anesthesia. For ICV injection, animals were placed on a stereotaxic head unit (David Kopf Instruments, Tujunga, CA, USA). A small hole was first drilled into the skull, and then a 35‐gauge needle (WPI, Sarasota, FL, USA) was implanted in the right lateral ventricle using coordinates of 0.9 mm caudal, 1.8 mm lateral, and 3.6 mm ventral to the bregma. To create an open‐loop baroreflex condition, the bilateral carotid sinus baroreceptor regions were isolated from the systemic circulation using previously described methods. , , In brief, the external carotid artery was ligated close to the carotid bifurcation. Subsequently, the internal carotid artery was embolized with 0.8‐mm diameter steel balls (SBM‐SUJ‐0.8, Tsubaki Nakashima, Nara, Japan), and the common carotid artery was catheterized. Physiological saline was then used to fill the isolated carotid sinuses and catheters. Using a servo‐pump system (ET‐126, Labworks, Costa Mesa, CA, USA), CSP was controlled via catheters inserted into the common carotid arteries. To minimize cardiovascular reflexes from the cardiopulmonary region and aortic arch, bilateral vagal and aortic depressor nerves were sectioned at the neck. Following the 60 min post‐surgery recovery period, CSP was first decreased to 60 mmHg for 4 min, and then increased stepwise from 60 to 180 mmHg in increments of 20 mmHg every minute according to previous studies. , , The stepwise CSP input cycle was repeated throughout the protocol. , Before evaluating AP, RSNA, and heart rate (HR) responses to the stepwise CSP input, blood was collected from the tail vein for assessing blood glucose and plasma insulin levels in some animals. Blood was obtained from the tail vein to reduce the potential negative impact that collecting blood from the jugular vein could have on arterial baroreflex function. Then, ICV injection of either 1 mM GSK1838705 (Sigma‐Aldrich, St. Louis, MO, USA) (1 μL/1 min), an IR antagonist, or artificial cerebrospinal fluid (aCSF, Harvard Apparatus, Holliston, MA, USA) (1 μL/1 min) as a control solution was commenced by using a microsyringe (Hamilton syringe; VWR, Missouri City, TX, USA) mounted on a micropump (UMP3, WPI). The final concentration of GSK1838705 was 1 mM, diluted in a solution of aCSF containing 50% dimethyl sulfoxide (DMSO, Sigma‐Aldrich) (1:1 aCSF and DMSO). Since it has been reported that RSNA increases approximately 120 min after the ICV injection of insulin, we anticipated that it would take several hours before significant effects of the ICV injection of the IR antagonist on arterial baroreflex function would manifest. Thus, we recorded AP, RSNA, and HR responses to the CSP input at 30, 60, 90, and 120 min after ICV injection. As a set of corollary experiments, we tested whether the 1:1 aCSF and DMSO solution changes the open‐loop baroreflex function. Again, venous blood was collected from the tail vein after recording the 120‐min data after ICV injection in some animals. The collected blood sample was assessed for blood glucose by using a handheld glucose meter (FreeStyle Precision Neo; Abbott, Chicago, IL, USA). Plasma insulin was assayed using an enzyme‐linked immunosorbent assay kit (Ultra‐Sensitive Rat Insulin ELISA Kit, catalog no. 90060, Crystal Chem, Elk Grove Village, IL, USA). 2.4.2 Data analysis RSNA, AP, CSP, and ECG signals were amplified, filtered, and continuously recorded on a computer at a 1 kHz sampling rate via the analog‐to‐digital converter (PowerLab 8/30, ADInstruments). Data analysis was performed by LabChart 8 application software (ADInstruments). HR was calculated from the ECG recording. For analyzing RSNA, the preamplified nerve signal was band‐pass filtered at 100–1000 Hz (Neuro Amp EX; ADInstruments) and then low‐pass filtered with a cutoff frequency of 30 Hz using the LabChart 8 application software. Full‐wave rectified signals of RSNA were subsequently used for quantification. AP, RSNA, and HR data were averaged for the last 10 s at each CSP level. To quantify the RSNA response to the stepwise CSP input, the RSNA value of the last 10 s at a CSP of 60 mmHg before ICV injection was designated as 100% baseline, and the stimulation‐induced changes in RSNA were expressed as a percentage of this baseline. The input–output relationships between CSP and RSNA (the neural arc), between CSP and AP (the total reflex arc), and between CSP and HR (HR control) were described by fitting the following four‐parameter logistic function to the data points as follows. , y = P 1 1 + exp P 2 CSP − P 3 + P 4 where y represents the output value (RSNA, AP, or HR), P 1 is the response range, P 2 is the slope coefficient, P 3 is the midpoint pressure on the CSP axis, and P 4 is the minimum value of output. The maximum gain of the logistic function was calculated from − P 1 P 2 /4. The input–output relationship between SNA and AP (the peripheral arc) was described by linear regression as follows : AP = b 0 + b 1 × RSNA where b 0 and b 1 represent the intercept and slope of the regression line, respectively. To estimate the operating point, using RSNA as the common abscissa and CSP or AP as the ordinate, the baroreflex equilibrium diagram was drawn by plotting data from the neural and peripheral arcs averaged for the last 10 s at each CSP level. , The operating‐point RSNA and AP were determined from the intersection of the fitted neural and peripheral arcs on the baroreflex equilibrium diagram. 2.5 Baroreflex equilibrium diagram simulation under hypertensive and hypotensive stresses Following previously described methods, the imposition of external disturbance (hypertensive stress [+5, +10, +15, and +20 mmHg] and hypotensive stress [−5, −10, −15, and −20 mmHg]) was simulated using the data from the open‐loop baroreflex in vivo experiment. The peripheral arc in the baroreflex equilibrium diagram was shifted upward (+5, +10, +15, and +20 mmHg) to mimic hypertensive stress and downward (−5, −10, −15, and −20 mmHg) to simulate hypotensive stress. Then, the operating‐point AP was estimated by the intersection of the neural arc and peripheral arc whose intercept was changed under hypertensive/hypotensive stress before and after ICV injection of control or IR antagonist solution. We determined the operating‐point AP rise and fall (Δoperating‐point AP) by calculating the difference in operating‐point AP from before to after hypertensive or hypotensive stress for evaluating AP lability against the imposition of external disturbance. 2.6 Statistical analysis The Shapiro–Wilk test was first performed to confirm data normality. In the histological data, we performed an unpaired t‐test or the Mann–Whitney U‐test, appropriately. Additionally, for evaluating the distribution of IR, c‐Fos, and DAPI expression, as well as the overlap area between IR, c‐Fos, and DAPI expression in the NTS, a two‐way repeated measures analysis of variance (ANOVA) was performed (group‐by‐coordinate). If an interaction and/or a main effect was significant, Bonferroni's multiple comparison test was conducted. In the physiological data, RSNA, AP, and HR at each CSP level before and 30, 60, 90, and 120 min after ICV injection of each solution were analyzed using a two‐way repeated measures ANOVA (time‐by‐CSP). If an interaction and/or a main effect was significant, Bonferroni's multiple comparison test was performed. For the analysis of parameters of static characteristics of the neural arc, peripheral arc, total reflex arc, heart HR control, operating‐point AP and RSNA, blood glucose, and plasma insulin before and 120 min after ICV injection control and GSK1838705 solutions, a two‐way repeated measures ANOVA was performed (solution‐by‐time). If an interaction and/or a main effect was significant, we performed Bonferroni's multiple comparison test. A one‐way repeated measures ANOVA or Friedman test was utilized for the parameters of static characteristics before and 30, 60, 90, and 120 min after ICV injection of 1:1 aCSF and DMSO solution. If a one‐way repeated measures ANOVA or Friedman test detected significance, Bonferroni's or Dunn's multiple comparisons test was performed. A paired t‐test or Wilcoxon signed‐rank test was used for analyzing blood glucose and plasma insulin before and 120 min after ICV injection of the 1:1 aCSF and DMSO solution as appropriate. In the simulation data, a two‐way repeated measures ANOVA was performed (time‐by‐pressure). If an interaction and/or a main effect was significant, Bonferroni's multiple comparison test was used. Moreover, we calculated the changes in ΔAP from before to 120 min after ICV injection of control or IR antagonist, and a two‐way repeated measures ANOVA was performed (solution‐by‐pressure). If an interaction and/or a main effect was significant, we performed Bonferroni's multiple comparison test. All statistical analyses were computed using statistical software (SPSS Statistics 28, IBM, Armonk, NY, USA). Statistical significance was defined as p < .05. Data are presented as the mean ± SD.
Ethical approval This study was performed in accordance with the U.S. Department of Health and Human Services NIH Guide for the Care and Use of Laboratory Animals . The Institutional Animal Care and Use Committee of the University of Texas Southwestern Medical Center approved this study (no. 2019–102 849), and all experimental procedures were conducted following approved institutional guidelines and regulations.
Animals Thirty‐five healthy male Sprague–Dawley rats (12–16 weeks, body weight: 384 ± 26 g) (Inotiv, West Lafayette, IN, USA) were used for the fluorescence immunohistochemistry and the baroreflex open‐loop in vivo experiment. The animals had free access to food and clean water and were kept one to four per cage under a 12 h light–dark cycle in an air‐conditioned room (22–24°C) until required for terminal experiments.
Fluorescence immunohistochemistry In six anesthetized rats, the bilateral carotid sinus baroreceptor regions were surgically isolated from the systemic circulation. The surgical and anesthesia procedures for the preparation of the baroreflex stimulation under the open‐loop conditions were described in the “ Open‐loop baroreflex in vivo experiment ” section below. We confirmed the adequacy of anesthesia by lack of a withdrawal response to tail pinch. In the baroreflex stimulation treated group ( n = 3), following the 60‐min post‐surgery recovery period, repetitive baroreflex stimulation was performed. By the procedure described below, carotid sinus pressure (CSP) was first decreased to 60 mm Hg for 4 min and then increased in a stepwise manner up to 180 mmHg in increments of 20 mmHg every minute. Peak c‐Fos expression is known to occur approximately 60–120 min following the neural activation. , Thus, the CSP input cycle was repeated for 120 min, followed by a 60‐min resting period under the closed‐loop conditions where CSP was matched to AP. In contrast, the control group ( n = 3) was under the closed‐loop conditions for the same period (i.e., 3 h) after the 60‐min post‐surgery recovery period. Immediately after the 60‐min resting period, transcardial perfusion with physiological saline was performed for 15 min followed by 15‐min 4% paraformaldehyde for tissue fixation. Thus, the tissue fixation was achieved within approximately 90 min after the baroreflex stimulation period. The brain tissue was harvested and post‐fixed overnight in 4% paraformaldehyde followed by dehydrated sequentially in 10% and 20% sucrose. The present immunofluorescence staining was adopted from our previous study. Briefly, after submerging in an optimal cutting temperature medium and freezing on dry ice, the brainstem tissue was sectioned at 35 μm using a cryostat (2800 Frigocut, Reichert Jung/Leica, Deer Park, IL, USA). The sections were washed in phosphate‐buffered saline (PBS). After blocking with an antibody buffer solution (PBS, 0.5% Triton‐X100, 5% normal goat serum [NGS], and 10 mg/mL bovine serum albumin [BSA]) for 60 min, co‐immunostaining for IR and c‐Fos was performed by incubation with the primary antibodies (mouse anti‐IRβ, 1:100, cat. no. sc‐57 342, RRID#AB_784102, Santa Cruz Biotechnology, Dallas, TX, USA; rabbit anti‐c‐Fos, 1:1000, cat. no. 2250, RRID#AB_2247211, Cell Signaling Technology, Danvers, MA, USA) overnight at room temperature. The sections were rinsed in PBS with 5% NGS and 10 mg/mL BSA and then incubated with fluorescence‐conjugated secondary antibodies (goat anti‐mouse Cy3, 1:500, cat. no. 115–165‐146, RRID#AB_2338690, Jackson ImmunoResearch Laboratories, West Grove, PA, USA; goat anti‐rabbit Alexa Fluor Plus 488, 1:500, cat. no. A32731, RRID#AB_2633280, Thermo Fisher Scientific, Waltham, MA, USA) for 60 min at room temperature. We previously used the same primary and secondary antibody combinations to perform the triple labeling of IR, c‐Fos, and DAPI in the NTS. Then, sections were rinsed with PBS with 5% NGS and 10 mg/mL BSA and pure water, and then fixed onto charged slides. The mounting medium with 4′,6‐diamidino‐2‐phenylindole (DAPI) (Mounting Medium With DAPI, cat. no. ab104139, Abcam, Cambridge, MA, USA) was applied to charged slides before coverslipping with micro cover glass. We performed negative control experiments to ensure the specificity of the antibodies. A fluorescence microscope system (Axio Imager A2, Zeiss, Oberkochen, Germany) was used for observing and obtaining the fluorescence images as described previously. , To evaluate whether c‐Fos‐positive neurons activated by baroreflex stimulation were specifically observed in medullary cardiovascular centers (the NTS, CVLM, and RVLM), we additionally investigated the degree of IR, c‐Fos, and DAPI overlap in the spinal trigeminal nucleus (Vsp) which is believed not to be activated by arterial baroreflex. The NTS, CVLM, RVLM, and Vsp were, respectively, determined as the regions from about −15.4 to −13.6 mm, −13.4 to −13.0 mm, −12.4 to −12.0 mm, and −15.4 to −12.0 mm to the bregma based on the rat brain atlas and previous studies. , , For quantitative analysis, the sections of the NTS (10 sections each rat), CVLM (3 sections each rat), and RVLM (3 sections each rat) were serially selected at approximately 2‐mm intervals from each rat. As for the Vsp, we collected three sections from −15.4 to −13.6 mm to the bregma region and one section each from −12.4 to −12.0 mm to the bregma and from −13.4 to −13.0 mm to the bregma regions (i.e., total 5 sections each rat). A series of grayscale photomicrographs at 10× magnification were acquired, and then we performed triple overlap analysis using image analysis software (FIJI; National Institute of Health, Maryland, USA) as we did previously. Briefly, the image background was subtracted within the region of interest. Subsequently, individualized thresholds (Triangle threshold, c‐Fos; Otsu threshold, IR and DAPI), and then particle size filters were applied for each binary image. Finally, IR + , c‐Fos + , and DAPI + areas were each calculated, as well as calculating the overlap between IR + /c‐Fos + /DAPI + areas.
Open‐loop baroreflex in vivo experiment 2.4.1 General surgical experimental procedures Rats were initially anesthetized with an intraperitoneal injection of a cocktail consisting of urethane (800 mg/kg) and α‐chloralose (65 mg/kg), and then mechanically ventilated with 100% oxygen gas after intubation. To maintain the depth of anesthesia, a 7‐fold dilution of the above anesthetic cocktail was continuously infused via the right femoral vein at a rate of 3–5 mL/h/kg. The depth of anesthesia was evaluated and monitored by tail pinch until all protocol was completed. Body temperature was maintained at approximately 36.5–37°C using a heating pad and a heat lamp. A catheter was inserted into the right femoral artery and connected to a pressure transducer (MLT0670, ADInstruments, Sydney, Australia) to measure AP. Electrocardiogram (ECG) recordings were obtained using needle electrodes. For recording renal SNA (RSNA), a branch of the isolated left renal nerve was attached to the electrodes (Platinum, AM Systems, Sequim, WA, USA) and covered with silicone glue (Kwik‐Sil, World Precision Instruments, Sarasota, FL, USA) for insulation and fixation. To validate that most of the RSNA recording was obtained from postganglionic renal sympathetic fibers, hexamethonium bromide (60 mg/kg) was administered intravenously after all experiments were completed. Additionally, RSNA background noise was measured over a 30‐min period after animals were euthanized by intravenous injection of saturated potassium chloride (4 M, 2 mL/kg) under anesthesia. For ICV injection, animals were placed on a stereotaxic head unit (David Kopf Instruments, Tujunga, CA, USA). A small hole was first drilled into the skull, and then a 35‐gauge needle (WPI, Sarasota, FL, USA) was implanted in the right lateral ventricle using coordinates of 0.9 mm caudal, 1.8 mm lateral, and 3.6 mm ventral to the bregma. To create an open‐loop baroreflex condition, the bilateral carotid sinus baroreceptor regions were isolated from the systemic circulation using previously described methods. , , In brief, the external carotid artery was ligated close to the carotid bifurcation. Subsequently, the internal carotid artery was embolized with 0.8‐mm diameter steel balls (SBM‐SUJ‐0.8, Tsubaki Nakashima, Nara, Japan), and the common carotid artery was catheterized. Physiological saline was then used to fill the isolated carotid sinuses and catheters. Using a servo‐pump system (ET‐126, Labworks, Costa Mesa, CA, USA), CSP was controlled via catheters inserted into the common carotid arteries. To minimize cardiovascular reflexes from the cardiopulmonary region and aortic arch, bilateral vagal and aortic depressor nerves were sectioned at the neck. Following the 60 min post‐surgery recovery period, CSP was first decreased to 60 mmHg for 4 min, and then increased stepwise from 60 to 180 mmHg in increments of 20 mmHg every minute according to previous studies. , , The stepwise CSP input cycle was repeated throughout the protocol. , Before evaluating AP, RSNA, and heart rate (HR) responses to the stepwise CSP input, blood was collected from the tail vein for assessing blood glucose and plasma insulin levels in some animals. Blood was obtained from the tail vein to reduce the potential negative impact that collecting blood from the jugular vein could have on arterial baroreflex function. Then, ICV injection of either 1 mM GSK1838705 (Sigma‐Aldrich, St. Louis, MO, USA) (1 μL/1 min), an IR antagonist, or artificial cerebrospinal fluid (aCSF, Harvard Apparatus, Holliston, MA, USA) (1 μL/1 min) as a control solution was commenced by using a microsyringe (Hamilton syringe; VWR, Missouri City, TX, USA) mounted on a micropump (UMP3, WPI). The final concentration of GSK1838705 was 1 mM, diluted in a solution of aCSF containing 50% dimethyl sulfoxide (DMSO, Sigma‐Aldrich) (1:1 aCSF and DMSO). Since it has been reported that RSNA increases approximately 120 min after the ICV injection of insulin, we anticipated that it would take several hours before significant effects of the ICV injection of the IR antagonist on arterial baroreflex function would manifest. Thus, we recorded AP, RSNA, and HR responses to the CSP input at 30, 60, 90, and 120 min after ICV injection. As a set of corollary experiments, we tested whether the 1:1 aCSF and DMSO solution changes the open‐loop baroreflex function. Again, venous blood was collected from the tail vein after recording the 120‐min data after ICV injection in some animals. The collected blood sample was assessed for blood glucose by using a handheld glucose meter (FreeStyle Precision Neo; Abbott, Chicago, IL, USA). Plasma insulin was assayed using an enzyme‐linked immunosorbent assay kit (Ultra‐Sensitive Rat Insulin ELISA Kit, catalog no. 90060, Crystal Chem, Elk Grove Village, IL, USA). 2.4.2 Data analysis RSNA, AP, CSP, and ECG signals were amplified, filtered, and continuously recorded on a computer at a 1 kHz sampling rate via the analog‐to‐digital converter (PowerLab 8/30, ADInstruments). Data analysis was performed by LabChart 8 application software (ADInstruments). HR was calculated from the ECG recording. For analyzing RSNA, the preamplified nerve signal was band‐pass filtered at 100–1000 Hz (Neuro Amp EX; ADInstruments) and then low‐pass filtered with a cutoff frequency of 30 Hz using the LabChart 8 application software. Full‐wave rectified signals of RSNA were subsequently used for quantification. AP, RSNA, and HR data were averaged for the last 10 s at each CSP level. To quantify the RSNA response to the stepwise CSP input, the RSNA value of the last 10 s at a CSP of 60 mmHg before ICV injection was designated as 100% baseline, and the stimulation‐induced changes in RSNA were expressed as a percentage of this baseline. The input–output relationships between CSP and RSNA (the neural arc), between CSP and AP (the total reflex arc), and between CSP and HR (HR control) were described by fitting the following four‐parameter logistic function to the data points as follows. , y = P 1 1 + exp P 2 CSP − P 3 + P 4 where y represents the output value (RSNA, AP, or HR), P 1 is the response range, P 2 is the slope coefficient, P 3 is the midpoint pressure on the CSP axis, and P 4 is the minimum value of output. The maximum gain of the logistic function was calculated from − P 1 P 2 /4. The input–output relationship between SNA and AP (the peripheral arc) was described by linear regression as follows : AP = b 0 + b 1 × RSNA where b 0 and b 1 represent the intercept and slope of the regression line, respectively. To estimate the operating point, using RSNA as the common abscissa and CSP or AP as the ordinate, the baroreflex equilibrium diagram was drawn by plotting data from the neural and peripheral arcs averaged for the last 10 s at each CSP level. , The operating‐point RSNA and AP were determined from the intersection of the fitted neural and peripheral arcs on the baroreflex equilibrium diagram.
General surgical experimental procedures Rats were initially anesthetized with an intraperitoneal injection of a cocktail consisting of urethane (800 mg/kg) and α‐chloralose (65 mg/kg), and then mechanically ventilated with 100% oxygen gas after intubation. To maintain the depth of anesthesia, a 7‐fold dilution of the above anesthetic cocktail was continuously infused via the right femoral vein at a rate of 3–5 mL/h/kg. The depth of anesthesia was evaluated and monitored by tail pinch until all protocol was completed. Body temperature was maintained at approximately 36.5–37°C using a heating pad and a heat lamp. A catheter was inserted into the right femoral artery and connected to a pressure transducer (MLT0670, ADInstruments, Sydney, Australia) to measure AP. Electrocardiogram (ECG) recordings were obtained using needle electrodes. For recording renal SNA (RSNA), a branch of the isolated left renal nerve was attached to the electrodes (Platinum, AM Systems, Sequim, WA, USA) and covered with silicone glue (Kwik‐Sil, World Precision Instruments, Sarasota, FL, USA) for insulation and fixation. To validate that most of the RSNA recording was obtained from postganglionic renal sympathetic fibers, hexamethonium bromide (60 mg/kg) was administered intravenously after all experiments were completed. Additionally, RSNA background noise was measured over a 30‐min period after animals were euthanized by intravenous injection of saturated potassium chloride (4 M, 2 mL/kg) under anesthesia. For ICV injection, animals were placed on a stereotaxic head unit (David Kopf Instruments, Tujunga, CA, USA). A small hole was first drilled into the skull, and then a 35‐gauge needle (WPI, Sarasota, FL, USA) was implanted in the right lateral ventricle using coordinates of 0.9 mm caudal, 1.8 mm lateral, and 3.6 mm ventral to the bregma. To create an open‐loop baroreflex condition, the bilateral carotid sinus baroreceptor regions were isolated from the systemic circulation using previously described methods. , , In brief, the external carotid artery was ligated close to the carotid bifurcation. Subsequently, the internal carotid artery was embolized with 0.8‐mm diameter steel balls (SBM‐SUJ‐0.8, Tsubaki Nakashima, Nara, Japan), and the common carotid artery was catheterized. Physiological saline was then used to fill the isolated carotid sinuses and catheters. Using a servo‐pump system (ET‐126, Labworks, Costa Mesa, CA, USA), CSP was controlled via catheters inserted into the common carotid arteries. To minimize cardiovascular reflexes from the cardiopulmonary region and aortic arch, bilateral vagal and aortic depressor nerves were sectioned at the neck. Following the 60 min post‐surgery recovery period, CSP was first decreased to 60 mmHg for 4 min, and then increased stepwise from 60 to 180 mmHg in increments of 20 mmHg every minute according to previous studies. , , The stepwise CSP input cycle was repeated throughout the protocol. , Before evaluating AP, RSNA, and heart rate (HR) responses to the stepwise CSP input, blood was collected from the tail vein for assessing blood glucose and plasma insulin levels in some animals. Blood was obtained from the tail vein to reduce the potential negative impact that collecting blood from the jugular vein could have on arterial baroreflex function. Then, ICV injection of either 1 mM GSK1838705 (Sigma‐Aldrich, St. Louis, MO, USA) (1 μL/1 min), an IR antagonist, or artificial cerebrospinal fluid (aCSF, Harvard Apparatus, Holliston, MA, USA) (1 μL/1 min) as a control solution was commenced by using a microsyringe (Hamilton syringe; VWR, Missouri City, TX, USA) mounted on a micropump (UMP3, WPI). The final concentration of GSK1838705 was 1 mM, diluted in a solution of aCSF containing 50% dimethyl sulfoxide (DMSO, Sigma‐Aldrich) (1:1 aCSF and DMSO). Since it has been reported that RSNA increases approximately 120 min after the ICV injection of insulin, we anticipated that it would take several hours before significant effects of the ICV injection of the IR antagonist on arterial baroreflex function would manifest. Thus, we recorded AP, RSNA, and HR responses to the CSP input at 30, 60, 90, and 120 min after ICV injection. As a set of corollary experiments, we tested whether the 1:1 aCSF and DMSO solution changes the open‐loop baroreflex function. Again, venous blood was collected from the tail vein after recording the 120‐min data after ICV injection in some animals. The collected blood sample was assessed for blood glucose by using a handheld glucose meter (FreeStyle Precision Neo; Abbott, Chicago, IL, USA). Plasma insulin was assayed using an enzyme‐linked immunosorbent assay kit (Ultra‐Sensitive Rat Insulin ELISA Kit, catalog no. 90060, Crystal Chem, Elk Grove Village, IL, USA).
Data analysis RSNA, AP, CSP, and ECG signals were amplified, filtered, and continuously recorded on a computer at a 1 kHz sampling rate via the analog‐to‐digital converter (PowerLab 8/30, ADInstruments). Data analysis was performed by LabChart 8 application software (ADInstruments). HR was calculated from the ECG recording. For analyzing RSNA, the preamplified nerve signal was band‐pass filtered at 100–1000 Hz (Neuro Amp EX; ADInstruments) and then low‐pass filtered with a cutoff frequency of 30 Hz using the LabChart 8 application software. Full‐wave rectified signals of RSNA were subsequently used for quantification. AP, RSNA, and HR data were averaged for the last 10 s at each CSP level. To quantify the RSNA response to the stepwise CSP input, the RSNA value of the last 10 s at a CSP of 60 mmHg before ICV injection was designated as 100% baseline, and the stimulation‐induced changes in RSNA were expressed as a percentage of this baseline. The input–output relationships between CSP and RSNA (the neural arc), between CSP and AP (the total reflex arc), and between CSP and HR (HR control) were described by fitting the following four‐parameter logistic function to the data points as follows. , y = P 1 1 + exp P 2 CSP − P 3 + P 4 where y represents the output value (RSNA, AP, or HR), P 1 is the response range, P 2 is the slope coefficient, P 3 is the midpoint pressure on the CSP axis, and P 4 is the minimum value of output. The maximum gain of the logistic function was calculated from − P 1 P 2 /4. The input–output relationship between SNA and AP (the peripheral arc) was described by linear regression as follows : AP = b 0 + b 1 × RSNA where b 0 and b 1 represent the intercept and slope of the regression line, respectively. To estimate the operating point, using RSNA as the common abscissa and CSP or AP as the ordinate, the baroreflex equilibrium diagram was drawn by plotting data from the neural and peripheral arcs averaged for the last 10 s at each CSP level. , The operating‐point RSNA and AP were determined from the intersection of the fitted neural and peripheral arcs on the baroreflex equilibrium diagram.
Baroreflex equilibrium diagram simulation under hypertensive and hypotensive stresses Following previously described methods, the imposition of external disturbance (hypertensive stress [+5, +10, +15, and +20 mmHg] and hypotensive stress [−5, −10, −15, and −20 mmHg]) was simulated using the data from the open‐loop baroreflex in vivo experiment. The peripheral arc in the baroreflex equilibrium diagram was shifted upward (+5, +10, +15, and +20 mmHg) to mimic hypertensive stress and downward (−5, −10, −15, and −20 mmHg) to simulate hypotensive stress. Then, the operating‐point AP was estimated by the intersection of the neural arc and peripheral arc whose intercept was changed under hypertensive/hypotensive stress before and after ICV injection of control or IR antagonist solution. We determined the operating‐point AP rise and fall (Δoperating‐point AP) by calculating the difference in operating‐point AP from before to after hypertensive or hypotensive stress for evaluating AP lability against the imposition of external disturbance.
Statistical analysis The Shapiro–Wilk test was first performed to confirm data normality. In the histological data, we performed an unpaired t‐test or the Mann–Whitney U‐test, appropriately. Additionally, for evaluating the distribution of IR, c‐Fos, and DAPI expression, as well as the overlap area between IR, c‐Fos, and DAPI expression in the NTS, a two‐way repeated measures analysis of variance (ANOVA) was performed (group‐by‐coordinate). If an interaction and/or a main effect was significant, Bonferroni's multiple comparison test was conducted. In the physiological data, RSNA, AP, and HR at each CSP level before and 30, 60, 90, and 120 min after ICV injection of each solution were analyzed using a two‐way repeated measures ANOVA (time‐by‐CSP). If an interaction and/or a main effect was significant, Bonferroni's multiple comparison test was performed. For the analysis of parameters of static characteristics of the neural arc, peripheral arc, total reflex arc, heart HR control, operating‐point AP and RSNA, blood glucose, and plasma insulin before and 120 min after ICV injection control and GSK1838705 solutions, a two‐way repeated measures ANOVA was performed (solution‐by‐time). If an interaction and/or a main effect was significant, we performed Bonferroni's multiple comparison test. A one‐way repeated measures ANOVA or Friedman test was utilized for the parameters of static characteristics before and 30, 60, 90, and 120 min after ICV injection of 1:1 aCSF and DMSO solution. If a one‐way repeated measures ANOVA or Friedman test detected significance, Bonferroni's or Dunn's multiple comparisons test was performed. A paired t‐test or Wilcoxon signed‐rank test was used for analyzing blood glucose and plasma insulin before and 120 min after ICV injection of the 1:1 aCSF and DMSO solution as appropriate. In the simulation data, a two‐way repeated measures ANOVA was performed (time‐by‐pressure). If an interaction and/or a main effect was significant, Bonferroni's multiple comparison test was used. Moreover, we calculated the changes in ΔAP from before to 120 min after ICV injection of control or IR antagonist, and a two‐way repeated measures ANOVA was performed (solution‐by‐pressure). If an interaction and/or a main effect was significant, we performed Bonferroni's multiple comparison test. All statistical analyses were computed using statistical software (SPSS Statistics 28, IBM, Armonk, NY, USA). Statistical significance was defined as p < .05. Data are presented as the mean ± SD.
RESULTS 3.1 Fluorescence immunohistochemistry Figure shows the individual representative images and co‐localization of IR, c‐Fos, and DAPI in neurons in the NTS, CVLM, RVLM, and Vsp. The representative images were pseudo‐colored from grayscale for illustration. The area of DAPI‐ and IR‐positive neurons in the NTS, CVLM, and RVLM did not differ significantly between baroreflex stimulation and control groups (Figure ). The 120‐min repetitive baroreflex stimulation significantly increased c‐Fos protein in the NTS, CVLM, and RVLM, but not in the Vsp (Figure ). Furthermore, the overlap area between IR + /c‐Fos + /DAPI + in the NTS, CVLM, and RVLM was significantly increased by the baroreflex stimulation (Figure ). The IR, c‐Fos, and DAPI overlap area in the Vsp, as the negative control area, was not significantly different among trials (Figure ). Figure shows the distribution of IR + , c‐Fos + , and DAPI + , as well as the overlap between IR + , c‐Fos + , and DAPI + in the NTS. In DAPI‐ and IR‐positive neurons, there were no significant differences between the groups at any coordinates in the NTS. Areas of c‐Fos‐positive neurons and the triple overlap in the baroreflex stimulation group were significantly greater than those in the control group at several coordinates in the NTS. 3.2 Open‐loop baroreflex in vivo experiment As compared to the baseline, blood glucose and plasma insulin were not significantly changed 120 min after ICV injection of control (aCSF) and GSK1838705 solutions (blood glucose; before: 132 ± 30 mg/dL, 120 min: 132 ± 21 mg/dL in control solution [ n = 5] vs. before: 122 ± 9 mg/dL, 120 min: 125 ± 14 mg/dL in GSK1838705 solution [ n = 5]; solution effect: p = .436; time effect: p = .775; interaction: p = .775; plasma insulin; before: 2.0 ± 0.8 ng/mL, 120 min: 1.2 ± 0.3 ng/mL in control solution [ n = 5] vs. before: 3.3 ± 1.6 ng/mL, 120 min: 2.6 ± 2.5 ng/mL in GSK1838705 solution [ n = 3]; solution effect: p = .168; time effect: p = .086; interaction: p = .903). ICV injection of aCSF containing 50% DMSO (GSK1838705 vehicle solution) also did not significantly change the blood glucose (before: 119 ± 22 mg/dL vs. after: 125 ± 30 mg/dL, p = .22 [ n = 5]) and plasma insulin (before: 1.8 ± 0.7 ng/mL vs. after: 1.1 ± 0.3 ng/mL, p = .067 [ n = 5]). Figure shows representative recordings of AP, RSNA, and HR responses to stepwise CSP input before and 30, 60, 90, and 120 min after ICV injection of control and GSK1838705 solutions. We successfully measured AP and HR in 24 rats and RSNA in 22 rats. A stepwise increase in CSP decreased RSNA, AP, and HR. Analysis of trial‐averaged static characteristics of the baroreflex demonstrated that the interactions (time‐by‐CSP) of the neural arc, total reflex arc, and HR control were significant in GSK1838705 but not in the control trial (Figure ). Moreover, ICV injection of GSK1838705 but not that of control significantly decreased AP, RSNA, and HR at each CSP level (Figure ). Since the effects of GSK1838705 were more pronounced at 120 min after the ICV injection, we used the values before and 120 min after ICV injection for comparisons of each parameter of the neural arc, peripheral arc, total reflex arc, and HR control. Table shows the parameters of static characteristics of the neural arc, peripheral arc, total reflex arc, and HR control before and 120 min after ICV injection of control and GSK1838705 solutions. In the neural arc, while the response range ( P 1 ), slope coefficient ( P 2 ), midpoint pressure ( P 3 ), and minimum value ( P 4 ) were not significantly changed by ICV injection of the GSK1838705 solution, the maximum gain ( G max ) was significantly decreased after ICV injection of GSK1838705. In the peripheral arc, intercept ( b 0 ) and slope ( b 1 ) were not significantly altered by ICV injection of either test solution. In the total reflex arc, although the response range ( P 1 ) and midpoint pressure ( P 3 ) were significantly decreased regardless of the test solutions, the significant interaction between solution (control vs. GSK1838705) and time (before vs. 120 min) was not observed in any parameters of the total reflex arc. In HR control, ICV injection of both solutions significantly decreased the response range ( P 1 ) without significantly changing the slope coefficient ( P 2 ) and midpoint pressure ( P 3 ). The minimum value ( P 4 ) after ICV injection of GSK1838705 was significantly lower than that after ICV injection of the control solution. GSK1838705, but not the control, significantly decreased the maximum gain ( G max ). Operating‐point RSNA and AP were determined from the baroreflex equilibrium diagram constructed from the fitted neural and peripheral arcs (Figure ). ICV injection of GSK1838705, but not the control solution, significantly decreased both operating‐point RSNA and AP (Figure ). Figure shows the trial‐averaged static characteristics of the baroreflex before and 30, 60, 90, and 120 min after ICV injection of a vehicle solution containing DMSO. We succeeded in measuring AP and HR in 5 rats and RSNA in 4 rats. The vehicle solution did not significantly change RSNA, AP, and HR responses to the stepwise CSP input. Moreover, none of the parameters of static characteristics of the neural arc, peripheral arc, total reflex arc, and HR arc, and operating‐point RSNA and AP were significantly altered after ICV injection of the vehicle solution (Table ). 3.3 Baroreflex equilibrium diagram simulation under hypertensive and hypotensive stresses Figure shows the simulation of hypertensive stress before and 120 min after ICV injection of control and IR antagonist solutions. Although Δoperating‐point AP was not significantly changed by ICV injection of control solution under hypertensive stress, there was a significant interaction in the GSK1838705 trial. Furthermore, the changes in Δoperating‐point AP from before to 120 min after ICV injection in the GSK1838705 trial were significantly higher than those in the control trial under all levels of hypertensive stress (+5, +10, +15, and +20 mmHg) (Figure ). Δoperating‐point AP was not significantly altered by ICV injection of control and GSK1838705 solutions under any levels of hypotensive stress (−5, −10, −15, and −20 mmHg) (Figure ).
Fluorescence immunohistochemistry Figure shows the individual representative images and co‐localization of IR, c‐Fos, and DAPI in neurons in the NTS, CVLM, RVLM, and Vsp. The representative images were pseudo‐colored from grayscale for illustration. The area of DAPI‐ and IR‐positive neurons in the NTS, CVLM, and RVLM did not differ significantly between baroreflex stimulation and control groups (Figure ). The 120‐min repetitive baroreflex stimulation significantly increased c‐Fos protein in the NTS, CVLM, and RVLM, but not in the Vsp (Figure ). Furthermore, the overlap area between IR + /c‐Fos + /DAPI + in the NTS, CVLM, and RVLM was significantly increased by the baroreflex stimulation (Figure ). The IR, c‐Fos, and DAPI overlap area in the Vsp, as the negative control area, was not significantly different among trials (Figure ). Figure shows the distribution of IR + , c‐Fos + , and DAPI + , as well as the overlap between IR + , c‐Fos + , and DAPI + in the NTS. In DAPI‐ and IR‐positive neurons, there were no significant differences between the groups at any coordinates in the NTS. Areas of c‐Fos‐positive neurons and the triple overlap in the baroreflex stimulation group were significantly greater than those in the control group at several coordinates in the NTS.
Open‐loop baroreflex in vivo experiment As compared to the baseline, blood glucose and plasma insulin were not significantly changed 120 min after ICV injection of control (aCSF) and GSK1838705 solutions (blood glucose; before: 132 ± 30 mg/dL, 120 min: 132 ± 21 mg/dL in control solution [ n = 5] vs. before: 122 ± 9 mg/dL, 120 min: 125 ± 14 mg/dL in GSK1838705 solution [ n = 5]; solution effect: p = .436; time effect: p = .775; interaction: p = .775; plasma insulin; before: 2.0 ± 0.8 ng/mL, 120 min: 1.2 ± 0.3 ng/mL in control solution [ n = 5] vs. before: 3.3 ± 1.6 ng/mL, 120 min: 2.6 ± 2.5 ng/mL in GSK1838705 solution [ n = 3]; solution effect: p = .168; time effect: p = .086; interaction: p = .903). ICV injection of aCSF containing 50% DMSO (GSK1838705 vehicle solution) also did not significantly change the blood glucose (before: 119 ± 22 mg/dL vs. after: 125 ± 30 mg/dL, p = .22 [ n = 5]) and plasma insulin (before: 1.8 ± 0.7 ng/mL vs. after: 1.1 ± 0.3 ng/mL, p = .067 [ n = 5]). Figure shows representative recordings of AP, RSNA, and HR responses to stepwise CSP input before and 30, 60, 90, and 120 min after ICV injection of control and GSK1838705 solutions. We successfully measured AP and HR in 24 rats and RSNA in 22 rats. A stepwise increase in CSP decreased RSNA, AP, and HR. Analysis of trial‐averaged static characteristics of the baroreflex demonstrated that the interactions (time‐by‐CSP) of the neural arc, total reflex arc, and HR control were significant in GSK1838705 but not in the control trial (Figure ). Moreover, ICV injection of GSK1838705 but not that of control significantly decreased AP, RSNA, and HR at each CSP level (Figure ). Since the effects of GSK1838705 were more pronounced at 120 min after the ICV injection, we used the values before and 120 min after ICV injection for comparisons of each parameter of the neural arc, peripheral arc, total reflex arc, and HR control. Table shows the parameters of static characteristics of the neural arc, peripheral arc, total reflex arc, and HR control before and 120 min after ICV injection of control and GSK1838705 solutions. In the neural arc, while the response range ( P 1 ), slope coefficient ( P 2 ), midpoint pressure ( P 3 ), and minimum value ( P 4 ) were not significantly changed by ICV injection of the GSK1838705 solution, the maximum gain ( G max ) was significantly decreased after ICV injection of GSK1838705. In the peripheral arc, intercept ( b 0 ) and slope ( b 1 ) were not significantly altered by ICV injection of either test solution. In the total reflex arc, although the response range ( P 1 ) and midpoint pressure ( P 3 ) were significantly decreased regardless of the test solutions, the significant interaction between solution (control vs. GSK1838705) and time (before vs. 120 min) was not observed in any parameters of the total reflex arc. In HR control, ICV injection of both solutions significantly decreased the response range ( P 1 ) without significantly changing the slope coefficient ( P 2 ) and midpoint pressure ( P 3 ). The minimum value ( P 4 ) after ICV injection of GSK1838705 was significantly lower than that after ICV injection of the control solution. GSK1838705, but not the control, significantly decreased the maximum gain ( G max ). Operating‐point RSNA and AP were determined from the baroreflex equilibrium diagram constructed from the fitted neural and peripheral arcs (Figure ). ICV injection of GSK1838705, but not the control solution, significantly decreased both operating‐point RSNA and AP (Figure ). Figure shows the trial‐averaged static characteristics of the baroreflex before and 30, 60, 90, and 120 min after ICV injection of a vehicle solution containing DMSO. We succeeded in measuring AP and HR in 5 rats and RSNA in 4 rats. The vehicle solution did not significantly change RSNA, AP, and HR responses to the stepwise CSP input. Moreover, none of the parameters of static characteristics of the neural arc, peripheral arc, total reflex arc, and HR arc, and operating‐point RSNA and AP were significantly altered after ICV injection of the vehicle solution (Table ).
Baroreflex equilibrium diagram simulation under hypertensive and hypotensive stresses Figure shows the simulation of hypertensive stress before and 120 min after ICV injection of control and IR antagonist solutions. Although Δoperating‐point AP was not significantly changed by ICV injection of control solution under hypertensive stress, there was a significant interaction in the GSK1838705 trial. Furthermore, the changes in Δoperating‐point AP from before to 120 min after ICV injection in the GSK1838705 trial were significantly higher than those in the control trial under all levels of hypertensive stress (+5, +10, +15, and +20 mmHg) (Figure ). Δoperating‐point AP was not significantly altered by ICV injection of control and GSK1838705 solutions under any levels of hypotensive stress (−5, −10, −15, and −20 mmHg) (Figure ).
DISCUSSION The major findings from this investigation are as follows: 1) IR and c‐Fos activated via baroreflex stimulation by stepwise CSP input were observed to co‐localize in the NTS, CVLM, and RVLM; 2) ICV injection of an IR antagonist acutely decreased RSNA, AP, and HR during stepwise CSP input; 3) blockade of IR in the brain acutely impaired neural arc function but not peripheral arc function; 4) the operating‐point RSNA and AP were acutely decreased by IR blockade in the brain under open‐loop conditions; and 5) ICV injection of an IR antagonist increased the operating‐point AP rise by hypertensive stress using numerical simulation. Importantly, the effects of the IR antagonist were independent of circulating insulin and glucose levels. To date, brain insulin levels have been suggested to modulate arterial baroreflex function. The present findings further suggest that not only the availability of insulin but also brain IR signaling is important for arterial baroreflex control of AP. It is well known that the signal from carotid sinus baroreceptors projects via sensory afferents to the NTS followed by CVLM and then RVLM in the sympathetic baroreflex function. , As such, we examined whether repetitive baroreflex stimulation activates c‐Fos in central neurons, and whether c‐Fos co‐localizes with IR. Activation of the arterial baroreflex has been reported to increase c‐Fos in the NTS, CVLM, and RVLM. , , , , Consistent with these findings, we observed that c‐Fos‐positive neurons in the NTS, CVLM, and RVLM were increased by 120‐min stepwise CSP input stimulation under open‐loop conditions. To the best of our knowledge, this is the first study showing that the NTS, CVLM, and RVLM are populated with IR‐positive neurons co‐expressing c‐Fos that are activated by baroreflex stimulation. 4.1 Possible mechanism of impairing arterial baroreflex function by brain IR blockade It has long been appreciated that acute ICV insulin infusion increases SNA. , , , Rahmouni et al. demonstrated that ICV administration of a high dose of insulin (500 mU) increases RSNA in rats. We found that ICV injection of an IR antagonist significantly reduced RSNA during stepwise CSP input. Furthermore, the operating‐point RSNA was decreased by the IR antagonist. It is well known that changes in the circulating levels of glucose and insulin alter SNA. , , We found that ICV injection of GSK1838705 did not significantly change blood glucose and plasma insulin levels in the present study. Thus, it is plausible that the decrease in RSNA is induced by impairment of IR signaling in the brain, independent of circulating glucose and insulin levels. The most important finding of this study is that the maximum gain of the neural arc was decreased by ICV injection of an IR antagonist, indicating that the neural arc function was impaired by IR blockade in the brain. In contrast, the slope and intercept of the regression line in the peripheral arc were not altered by the IR antagonist GSK1838705. The neural arc determines the SNA response to baroreceptor pressure input, while the peripheral arc determines AP as a result of cardiovascular responses to SNA input. Thus, these results suggest that IR signaling in the brain is involved in arterial baroreflex function via central nervous system mechanisms without affecting peripheral SNA‐induced vasoconstrictive function. It is noted that the operating‐point RSNA was significantly decreased by ICV injection of the IR antagonist. This suggests that the blockade of IR in the brain reduces baseline RSNA. Thus, it is challenging to exclude the possibility that the reduction of baseline RSNA per se would attenuate the neural arc gain. However, this is beyond the scope of the present investigation, and further research is warranted. We recently found that IRs are highly expressed in the NTS, and that microinjection of an IR antagonist (GSK1838705) into the NTS acutely increases the pressor response to electrically induced muscle contractions in baroreceptor‐intact rats but not in denervated rats. The results suggest that acute blockade of IR signaling in the NTS increases the exercise pressor reflex through interactions with baroreflex neurons in the NTS. Sensory information from the carotid sinus nerve projects to the NTS and further to the CVLM within the medulla oblongata. These neurons then project to the RVLM. , Critical neurotransmission from the CVLM to the RVLM is mediated through the inhibitory amino acid GABA, which inhibits sympathoexcitatory neurons in the RVLM, resulting in a decrease in SNA and AP. , The binding of insulin to IR activates phosphoinositide 3‐kinase, which converts phosphorylated phosphatidylinositol 4,5‐bisphosphate into (3,4,5)‐trisphosphate (PIP3) through phosphorylation. , PIP3 is known to activate ATP‐dependent potassium channels, resulting in hyperpolarization of neurons and decreased neuronal firing rate. Furthermore, it has been demonstrated that the microinjection of insulin into the NTS decreases the spontaneous discharge of baroreflex‐sensitive NTS neurons in anesthetized rats. Therefore, although speculative in nature, ICV injection of an IR antagonist may inhibit IR signaling‐induced decreases in NTS neuron activation by reducing hyperpolarization, which in turn reduces the activation of sympathoexcitatory neurons in the RVLM, resulting in the inhibition of the reflex response to changes in CSP. Furthermore, astrocytes in the NTS have been suggested to play a role in the arterial baroreflex. , NTS astrocytes might be involved in the brain IR blockade‐induced impairment of arterial baroreflex function. Alternatively, as Cassaglia et al. reported that the arcuate nucleus is the primary site of action where insulin enhances baroreflex function, it is therefore possible that blockade of IR signaling in the arcuate nucleus could also lead to impairment of arterial baroreflex function. In contrast, it has been reported that the nanoinjection of an IR antagonist into the arcuate nucleus does not significantly decrease lumbar SNA, AP, and HR in anesthetized healthy rats. Thus, it is unlikely that the IR antagonist‐induced arterial baroreflex dysfunction can be fully explained by its action in the arcuate nucleus. Of note, however, our study did not identify the site of action of the IR antagonist in the brain. Future studies should investigate this point. Blockade of IR in the brain decreased both AP during CSP input and operating‐point AP, which is consistent with results of the neural arc and operating‐point RSNA. Since previous studies examined the effects of brain insulin on the baroreflex under closed‐loop conditions, the arterial baroreflex control of AP has not been investigated. To the best of our knowledge, therefore, our study is the first to demonstrate that brain IR signaling has a crucial role in AP regulation via arterial baroreflex control of SNA. The operating point under open‐loop conditions is determined from the intersection between the neural and peripheral arcs on the baroreflex equilibrium diagram. , The present study showed that ICV injection of IR antagonist altered the neural arc by attenuating the maximum gain while the peripheral arc did not change. Therefore, the brain IR blockade‐induced reduction in AP may occur by impairing the neural arc function in arterial baroreflex control of SNA. IR antagonist injection in the brain impairs HR control by reducing the maximum gain. Since bilateral vagal and aortic depressor nerves were sectioned in this study, the changes in HR would mainly result from its response to CSP via SNA. Although we did not measure cardiac SNA, a previous study reported that the open‐loop static characteristics of baroreflex control of cardiac SNA parallel those of RSNA. Hence, IR blockade may impair arterial baroreflex control of HR via cardiac SNA. Furthermore, this is consistent with an earlier study, which found that baroreflex gain of HR was decreased in pregnant rabbits with a decrease in CSF insulin. Taken together, our findings suggest that arterial baroreflex control of HR is impaired by decreasing brain IR signaling. RSNA during stepwise CSP input was significantly decreased at 120 min after ICV injection of the IR antagonist, while AP and HR were significantly reduced from 30 to 60 min after the injection. It has been shown that the activation of RSNA is slowly increased by ICV injection of insulin compared to lumbar SNA. In addition, Pricher et al. demonstrated that the gain of baroreflex control of HR and lumbar SNA was improved within 60 min after brain insulin infusion. Therefore, it is plausible, albeit speculative, that decreased lumbar and cardiac SNA might induce attenuation of AP and HR before a remarkable decrease in RSNA. That being said, 120 min after IR antagonist injection, the changes in the total reflex arc were similar to those of the neural arc. Additionally, both operating‐point RSNA and AP were decreased. Therefore, there is no doubt that the decrease in AP is at least partially attributable to the attenuation of RSNA. 4.2 Clinical implications Our findings suggest that IR signaling in the brain plays a crucial role in AP regulation via modulation of arterial baroreflex function. Arterial baroreflex dysfunction has been well documented to induce AP lability. AP lability is often observed in patients with diabetes mellitus (DM), leading to orthostatic hypotension and exercise intolerance. Moreover, evidence suggests that high short‐term AP variability increases the risk of cardiovascular disease in DM patients. , Circulating insulin is transported into the brain across the blood–brain barrier. Hyperinsulinemia and peripheral insulin resistance, particularly in the early stage of type 2 DM, have been reported to induce a deficit in insulin transport across the blood–brain barrier. , , Increasing evidence suggests that reduced brain insulin levels impair arterial baroreflex function. , Notably, earlier studies have demonstrated that type 1 and type 2 DM animals have decreased brain IR signaling, suggesting that DM induces an impaired IR signaling pathway in the brain. , Taken together with the findings of the present study, the decrease not only in brain insulin availability but also in IR signaling in the medullary cardiovascular centers may, at least in part, induce AP lability in this disease. In this study, the numerical simulation using the baroreflex equilibrium diagram revealed that brain IR antagonist raised the operating‐point AP in response to hypertensive stress. This suggests that brain IR blockade may reduce the responsiveness of sympathetic baroreflex regulation against external perturbation, especially in the high AP range. The exaggerated AP response to exercise has often been observed in patients , , as well as animals , , , , with DM. Therefore, IR signaling dysfunction in the brain seen in DM might partially explain the AP lability in the high AP range. Since we studied only normal healthy rats in the present investigation, future investigations should track the causal relationship between brain IR signaling dysfunction and baroreflex failure using animals and patients with DM. 4.3 Limitations We also acknowledge several limitations in the present study. First, the acute effect of IR blockade in the brain on the arterial baroreflex was investigated in anesthetized rats. Thus, the present results may not be directly applicable to conscious individuals. Second, as discussed above, because we performed ICV injection for delivering IR antagonist into the brain, the site of action in the brain was not identified. Third, only male rats were used in this study. Because the menstrual/estrous cycle has been suggested to influence baroreflex function, , it is unknown whether the interpretation of our results applies to females. Further studies are required to elucidate these points. Fourth, the stepwise CSP input was repeated for 120 min in the present study. Prolonged baroreflex activation has been reported to acutely decrease AP levels. Although open‐loop baroreflex function and operating‐point AP and RSNA did not significantly change over 120 min in the control trial, the potential impact of repeated CSP input on arterial baroreflex function, as well as the operating‐point AP or RSNA, cannot be fully excluded. Lastly, GSK1838705 was diluted in aCSF containing 50% DMSO. To investigate the effect of DMSO, aCSF with a 50% DMSO solution was administered intraventricularly. As a result, the significant effects seen with ICV injection of GSK1838705 were not observed. Additionally, given the total CSF volume in adult rats, the concentration of DMSO after dilution in brain CSF would be expected to be less than about 0.5%. As such, it is unlikely that DMSO caused transient impairments in arterial baroreflex function.
Possible mechanism of impairing arterial baroreflex function by brain IR blockade It has long been appreciated that acute ICV insulin infusion increases SNA. , , , Rahmouni et al. demonstrated that ICV administration of a high dose of insulin (500 mU) increases RSNA in rats. We found that ICV injection of an IR antagonist significantly reduced RSNA during stepwise CSP input. Furthermore, the operating‐point RSNA was decreased by the IR antagonist. It is well known that changes in the circulating levels of glucose and insulin alter SNA. , , We found that ICV injection of GSK1838705 did not significantly change blood glucose and plasma insulin levels in the present study. Thus, it is plausible that the decrease in RSNA is induced by impairment of IR signaling in the brain, independent of circulating glucose and insulin levels. The most important finding of this study is that the maximum gain of the neural arc was decreased by ICV injection of an IR antagonist, indicating that the neural arc function was impaired by IR blockade in the brain. In contrast, the slope and intercept of the regression line in the peripheral arc were not altered by the IR antagonist GSK1838705. The neural arc determines the SNA response to baroreceptor pressure input, while the peripheral arc determines AP as a result of cardiovascular responses to SNA input. Thus, these results suggest that IR signaling in the brain is involved in arterial baroreflex function via central nervous system mechanisms without affecting peripheral SNA‐induced vasoconstrictive function. It is noted that the operating‐point RSNA was significantly decreased by ICV injection of the IR antagonist. This suggests that the blockade of IR in the brain reduces baseline RSNA. Thus, it is challenging to exclude the possibility that the reduction of baseline RSNA per se would attenuate the neural arc gain. However, this is beyond the scope of the present investigation, and further research is warranted. We recently found that IRs are highly expressed in the NTS, and that microinjection of an IR antagonist (GSK1838705) into the NTS acutely increases the pressor response to electrically induced muscle contractions in baroreceptor‐intact rats but not in denervated rats. The results suggest that acute blockade of IR signaling in the NTS increases the exercise pressor reflex through interactions with baroreflex neurons in the NTS. Sensory information from the carotid sinus nerve projects to the NTS and further to the CVLM within the medulla oblongata. These neurons then project to the RVLM. , Critical neurotransmission from the CVLM to the RVLM is mediated through the inhibitory amino acid GABA, which inhibits sympathoexcitatory neurons in the RVLM, resulting in a decrease in SNA and AP. , The binding of insulin to IR activates phosphoinositide 3‐kinase, which converts phosphorylated phosphatidylinositol 4,5‐bisphosphate into (3,4,5)‐trisphosphate (PIP3) through phosphorylation. , PIP3 is known to activate ATP‐dependent potassium channels, resulting in hyperpolarization of neurons and decreased neuronal firing rate. Furthermore, it has been demonstrated that the microinjection of insulin into the NTS decreases the spontaneous discharge of baroreflex‐sensitive NTS neurons in anesthetized rats. Therefore, although speculative in nature, ICV injection of an IR antagonist may inhibit IR signaling‐induced decreases in NTS neuron activation by reducing hyperpolarization, which in turn reduces the activation of sympathoexcitatory neurons in the RVLM, resulting in the inhibition of the reflex response to changes in CSP. Furthermore, astrocytes in the NTS have been suggested to play a role in the arterial baroreflex. , NTS astrocytes might be involved in the brain IR blockade‐induced impairment of arterial baroreflex function. Alternatively, as Cassaglia et al. reported that the arcuate nucleus is the primary site of action where insulin enhances baroreflex function, it is therefore possible that blockade of IR signaling in the arcuate nucleus could also lead to impairment of arterial baroreflex function. In contrast, it has been reported that the nanoinjection of an IR antagonist into the arcuate nucleus does not significantly decrease lumbar SNA, AP, and HR in anesthetized healthy rats. Thus, it is unlikely that the IR antagonist‐induced arterial baroreflex dysfunction can be fully explained by its action in the arcuate nucleus. Of note, however, our study did not identify the site of action of the IR antagonist in the brain. Future studies should investigate this point. Blockade of IR in the brain decreased both AP during CSP input and operating‐point AP, which is consistent with results of the neural arc and operating‐point RSNA. Since previous studies examined the effects of brain insulin on the baroreflex under closed‐loop conditions, the arterial baroreflex control of AP has not been investigated. To the best of our knowledge, therefore, our study is the first to demonstrate that brain IR signaling has a crucial role in AP regulation via arterial baroreflex control of SNA. The operating point under open‐loop conditions is determined from the intersection between the neural and peripheral arcs on the baroreflex equilibrium diagram. , The present study showed that ICV injection of IR antagonist altered the neural arc by attenuating the maximum gain while the peripheral arc did not change. Therefore, the brain IR blockade‐induced reduction in AP may occur by impairing the neural arc function in arterial baroreflex control of SNA. IR antagonist injection in the brain impairs HR control by reducing the maximum gain. Since bilateral vagal and aortic depressor nerves were sectioned in this study, the changes in HR would mainly result from its response to CSP via SNA. Although we did not measure cardiac SNA, a previous study reported that the open‐loop static characteristics of baroreflex control of cardiac SNA parallel those of RSNA. Hence, IR blockade may impair arterial baroreflex control of HR via cardiac SNA. Furthermore, this is consistent with an earlier study, which found that baroreflex gain of HR was decreased in pregnant rabbits with a decrease in CSF insulin. Taken together, our findings suggest that arterial baroreflex control of HR is impaired by decreasing brain IR signaling. RSNA during stepwise CSP input was significantly decreased at 120 min after ICV injection of the IR antagonist, while AP and HR were significantly reduced from 30 to 60 min after the injection. It has been shown that the activation of RSNA is slowly increased by ICV injection of insulin compared to lumbar SNA. In addition, Pricher et al. demonstrated that the gain of baroreflex control of HR and lumbar SNA was improved within 60 min after brain insulin infusion. Therefore, it is plausible, albeit speculative, that decreased lumbar and cardiac SNA might induce attenuation of AP and HR before a remarkable decrease in RSNA. That being said, 120 min after IR antagonist injection, the changes in the total reflex arc were similar to those of the neural arc. Additionally, both operating‐point RSNA and AP were decreased. Therefore, there is no doubt that the decrease in AP is at least partially attributable to the attenuation of RSNA.
Clinical implications Our findings suggest that IR signaling in the brain plays a crucial role in AP regulation via modulation of arterial baroreflex function. Arterial baroreflex dysfunction has been well documented to induce AP lability. AP lability is often observed in patients with diabetes mellitus (DM), leading to orthostatic hypotension and exercise intolerance. Moreover, evidence suggests that high short‐term AP variability increases the risk of cardiovascular disease in DM patients. , Circulating insulin is transported into the brain across the blood–brain barrier. Hyperinsulinemia and peripheral insulin resistance, particularly in the early stage of type 2 DM, have been reported to induce a deficit in insulin transport across the blood–brain barrier. , , Increasing evidence suggests that reduced brain insulin levels impair arterial baroreflex function. , Notably, earlier studies have demonstrated that type 1 and type 2 DM animals have decreased brain IR signaling, suggesting that DM induces an impaired IR signaling pathway in the brain. , Taken together with the findings of the present study, the decrease not only in brain insulin availability but also in IR signaling in the medullary cardiovascular centers may, at least in part, induce AP lability in this disease. In this study, the numerical simulation using the baroreflex equilibrium diagram revealed that brain IR antagonist raised the operating‐point AP in response to hypertensive stress. This suggests that brain IR blockade may reduce the responsiveness of sympathetic baroreflex regulation against external perturbation, especially in the high AP range. The exaggerated AP response to exercise has often been observed in patients , , as well as animals , , , , with DM. Therefore, IR signaling dysfunction in the brain seen in DM might partially explain the AP lability in the high AP range. Since we studied only normal healthy rats in the present investigation, future investigations should track the causal relationship between brain IR signaling dysfunction and baroreflex failure using animals and patients with DM.
Limitations We also acknowledge several limitations in the present study. First, the acute effect of IR blockade in the brain on the arterial baroreflex was investigated in anesthetized rats. Thus, the present results may not be directly applicable to conscious individuals. Second, as discussed above, because we performed ICV injection for delivering IR antagonist into the brain, the site of action in the brain was not identified. Third, only male rats were used in this study. Because the menstrual/estrous cycle has been suggested to influence baroreflex function, , it is unknown whether the interpretation of our results applies to females. Further studies are required to elucidate these points. Fourth, the stepwise CSP input was repeated for 120 min in the present study. Prolonged baroreflex activation has been reported to acutely decrease AP levels. Although open‐loop baroreflex function and operating‐point AP and RSNA did not significantly change over 120 min in the control trial, the potential impact of repeated CSP input on arterial baroreflex function, as well as the operating‐point AP or RSNA, cannot be fully excluded. Lastly, GSK1838705 was diluted in aCSF containing 50% DMSO. To investigate the effect of DMSO, aCSF with a 50% DMSO solution was administered intraventricularly. As a result, the significant effects seen with ICV injection of GSK1838705 were not observed. Additionally, given the total CSF volume in adult rats, the concentration of DMSO after dilution in brain CSF would be expected to be less than about 0.5%. As such, it is unlikely that DMSO caused transient impairments in arterial baroreflex function.
CONCLUSION The data demonstrate that the blockade of IR in the brain acutely alters arterial baroreflex function and decreases operating‐point RSNA and AP by impairing the neural arc function under open‐loop conditions. These results suggest that IR signaling in the brain plays an important role in AP regulation via the neural arc function of arterial baroreflex.
AH and MM conceived and designed experiments. AH performed experiments. AH and MM analyzed data. AH, TK, NH, AF, JAE, HKK, GAI, SAS, WV, and MM interpreted the results of experiments. AH and MM prepared figures. AH and MM drafted the manuscript or revised it critically for important intellectual content. AH, TK, NH, AF, JAE, HKK, GAI, SAS, WV, and MM have read and approved the final version of this manuscript.
TK received a consulting fee from NTT Research, Inc.
|
TUBERCULOUS SPONDYLODISCITIS IN A RENAL TRANSPLANT RECIPIENT – A CASE REPORT | 660b6884-717d-4b42-9168-e05f210a7fe6 | 11759119 | Surgical Procedures, Operative[mh] | Tuberculosis (TB) in transplant recipients is a rare but serious opportunistic infection with a mortality rate as high as 40% . About one-third to one-half of all cases of active TB after transplantation are disseminated or occur at extrapulmonary sites . Skeletal involvement occurs in 1%-3% of TB infections, half of these affecting the spine, usually the lower thoracic and upper lumbar region . To the best of our knowledge, this is the first reported case of spondylodiscitis in a renal transplant recipient.
A 56-year-old man diagnosed with chronic renal disease in 2009 reached end-stage renal disease in August 2012 and started hemodialysis. He had a history of diabetes mellitus, arterial hypertension and coronary artery disease. In November 2014, he underwent renal transplantation from a deceased donor. He was treated with cyclosporin, mycophenolate mofetil and corticosteroids. From April to July 2015, he was hospitalized three times due to fever of unknown origin and was treated with antibiotics (cefepime, meropenem). In September 2015, he was admitted to Department due to low-grade fever and suspicion of left side pneumonia. His plasma laboratory investigations showed C-reactive protein 37 mg/L and leukocytes 12x10 9 /L. On admission, empirical antibiotic therapy with ceftriaxone and azithromycin was started. On examination, the patient was afebrile without lymphadenopathy. Weakened noise was noted in the right basal lung area and cardiac examination was normal. Neurological examination revealed no focal motor weakness. During hospitalization, he started to feel radicular pain radiating to the upper abdomen from thoracolumbar spine, and he became paraplegic soon. Imaging evaluation with multi slice computed tomography (MSCT) and magnetic resonance imaging showed Th8-Th9 spondylodiscitis and osteolytic lesion from Th9-Th10 to L4-L5 level. Broad investigation was initiated. MSCT imaging of the chest showed a few nodal changes in lungs similar to metastasis. QuantiFERON test, as well as serology for cytomegalovirus and Epstein-Barr virus were all negative. Lesion biopsies of thoracic vertebra (Th8-Th9) were negative for TB by Gram stain and routine cultures were also negative. Endoscopic examination of gastrointestinal tract did not prove malignancy. Vertebral lesion biopsy showed chronic granulomatous inflammation. Anti-TB treatment was started with isoniazid 400 mg, pyrazinamide 2000 mg, ethambutol 1200 mg and Rimactan 600 mg. Because of continued neurological deterioration of the patient, repeated biopsy of the lesion between the Th8-L1 vertebral bodies was performed and spinal cord decompression was also made. Polymerase chain reaction (PCR) testing of the specimen for the Mycobacterium tuberculosis complex based on 16SrRNA was positive. After initiation of quadruple anti-TB therapy, the patient was pain free and showed significant improvement of his lower extremity neurological deficits. Renal graft function was stable, and creatinine values were 139-155 µmol/L. In October 2015, the patient was discharged and referred to a rehabilitation hospital, and later received isoniazid 400 mg and Rimactan 600 mg.
Pott disease after kidney transplantation is rarely reported in medical literature. We describe a kidney allograft recipient who presented with unexplained upper abdominal pain and fever one year after transplantation, which were proven to be due to tuberculous spondylodiscitis. Ozisik et al . report a case of spinal TB after heart transplantation, and the earliest symptom was back pain . Diagnosis of tuberculous spondylodiscitis was based on imaging methods and isolation of Mycobacterium tuberculosis from biopsy samples by PCR method although first biopsy showed granulomatous inflammation. A combination of surgery and anti-TB therapy is the most common form of treatment that produced good results in our patient. It is well known that rifampicin increases P-450 cytochrome activity and decreases cyclosporin level. Therefore, daily cyclosporin dose was increased in our patient and we measured cyclosporin level frequently. Some studies report that the duration of anti-TB therapy in renal recipients should be the same as in the general population .
Diagnosis of tuberculous spondylodiscitis is difficult because clinical findings usually are nonspecific and radiological features may mimic those of other diseases such as bacterial, fungal, inflammatory and neoplastic disease. Mycobacterial infections including extrapulmonary manifestations should be considered in all renal transplant recipients presenting with unexplained fever.
|
Impact of simulated animated video education on patients’ disease uncertainty, anxiety, and sleep quality in digestive endoscopy examination | 39ec4fc0-ffd4-4547-813d-2dbd246b2828 | 11871668 | Patient Education as Topic[mh] | Gastrointestinal endoscopy examinations were common procedures used for the diagnosis and management of various upper and lower gastrointestinal disorders . Patients undergoing gastrointestinal endoscopies commonly report feelings of uncertainty and anxiety related to their medical condition and the impending procedure . Disease uncertainty encompasses factors such as ambiguity, complexity, information deficit, and unpredictability, all of which can significantly influence patients’ emotional well-being and overall experience with medical care . Similarly, anxiety related to medical procedures was a well-documented phenomenon, characterized by symptoms of nervousness, apprehension, and distress, which can lead to suboptimal patient experiences and may even impact procedural outcomes. Furthermore, sleep disturbances were prevalent among individuals scheduled for medical procedures, potentially exacerbating pre-procedural stress and affecting overall well-being . It was within this context of patient psychosocial well-being that the impact of educational interventions, particularly simulated animated video education, was of great interest. The emergence of simulated animated video education as a patient education tool in the healthcare setting presents a novel approach to addressing patients’ informational and psychological needs before medical procedures. Fee-ley TH et al.’s research illustrates the advantages of analog animation education . Simulated animated videos offer a visually engaging and interactive platform through which complex medical information can be effectively conveyed to patients in a clear and understandable manner . Simulated animated video education has emerged as a potential intervention to address these psychosocial aspects in the healthcare setting. However, for patients with gastrointestinal endoscopy, the effect of simulated animation video education still needs to be studied. This manuscript discusses the results of a retrospective cohort study conducted to investigate the impact of simulated animated video education in comparison to conventional education on patients’ disease uncertainty, anxiety, and sleep quality in the context of digestive endoscopy examinations. Study design and population This study was a retrospective cohort study that selected 166 patients who underwent digestive endoscopy examinations at Zhongshan City People’s Hospital, Guangdong Province, China from June 2023 to December 2023 and met the inclusion and exclusion criteria as the study subjects. Inclusion criteria: Patients with upper gastrointestinal problems such as upper gastric discomfort, bloating, pain, heartburn, swallowing discomfort, belching, regurgitation, reduced appetite, weight loss, anemia, as well as lower gastrointestinal issues including chronic diarrhea, constipation, abdominal pain, and bloating were included. Furthermore, patients with abnormal findings in gastrointestinal barium meal examinations that could not exclude tumors at the end of the colon or ileum, those with unexplained lower intestinal obstruction requiring confirmation of the extent and severity of inflammatory bowel disease, and individuals over 45 years old undergoing routine health examinations were also included. Exclusion criteria: Patients with severe heart, brain, lung, or liver diseases; those with uncontrolled acute gastrointestinal bleeding; individuals in the acute phase of gastrointestinal perforation, aortic aneurysm, or vascular rupture leading to significant bleeding; those who had undergone multiple laparotomies or had intestinal adhesions; patients with severe stenosis obstructing endoscopy passage, inability to tolerate gastric tube insertion, or those with mental disorders were excluded (Fig. ). This study was approved by the Ethics Committee of Zhongshan People’s Hospital. The procedures were conducted in accordance with the ethical standards set forth by the Committee on Human Experimentation and the Helsinki Declaration of 1964, as revised in 2013. Informed consent was waived by the Ethics Committee of Zhongshan People’s Hospital for this retrospective study due to the exclusive use of de-identified patient data, which posed no potential harm or impact on patient care. This waiver was approved by the institutional review board and ethics committee of our institution in accordance with regulatory and ethical guidelines pertaining to retrospective studies. Intervention method Patients were divided into the conventional education group ( n = 77) and the simulated animated video education group ( n = 89) based on the educational mode. Conventional education involved communication between medical staff and patients scheduled for gastroscopy, informing the patients about the purpose of the examination, the procedure, and potential risks. Patients were advised about dietary restrictions before gastroscopy, discontinuation of anticoagulants and antiplatelet drugs, arranging for a companion, and attire considerations. Simulated animated video education involved organizing conventional educational content into an educational animation. Patients were instructed to watch this animation before the examination to fulfill the educational objectives: The video includes key topics such as preoperative diet and water ban requirements, surgical steps, expected feelings, risks and complications, postoperative diet and activity guidance, and how to communicate with doctors about doubts and concerns before and after surgery, which lasts for two minutes. The two-minute video was created using Adobe Animate by the Department of Gastroenterology, Endoscopy Centre of Zhongshan People’s Hospital (Fig. ). Data collection Patient demographic information, including age, Gender, BMI, Education Level, Hypertension, Diabetes, Hyperlipidemia was collected and recorded from the medical records system. Gastrointestinal endoscopy examination Prior to undergoing gastrointestinal endoscopy, patients were required to fast for 6–8 h and take oral anesthesia to anesthetize the mucous membrane of the throat to alleviate reflex nausea and vomiting during esophagogastroduodenoscopy. The physician inserts the endoscope through the oropharynx into the esophagus, sequentially enters the esophagus, stomach, and duodenum, re-examines these areas during the withdrawal of the endoscope, and takes biopsies at appropriate sites. The examination was performed using the GASTROINTESTINAL VIDEOSCOPE (Registration No. 3220720, OLYMPUS OPTICAL Co., Ltd., Japan). For colonoscopy, patients begin taking bowel-cleansing agents the day before the procedure, as the colonic mucosa cannot be adequately observed in the presence of feces. After insertion through the anus into the terminal ileum, the colon was examined internally for lesions, and the entire mucosa was observed during the withdrawal of the endoscope. The examination was conducted using the Evis Colono Videoscope (Registration No. 3220887, OLYMPUS OPTICAL Co., Ltd., Japan). Patient examination sites and disease conditions were documented. Disease uncertainty score Questionnaires were administered before and after the simulated video education, as well as one month after the examination. The Mishel’s Uncertainty in Illness Scale (MUIS) was used for assessment. This scale primarily measures patients’ levels of uncertainty regarding disease treatment, care, disease information, and prognosis during hospitalization. The scale has been translated into Chinese by domestic scholars, and its Cronbach’s α above 0.8. The scale consists of 4 dimensions: disease ambiguity, information deficit, complexity, and unpredictability, with a total of 33 items. The overall score ranges from 32.0 to 160.0, categorizable into low (32.0-74.7), moderate (74.8-117.4), and high (117.5–160.0) uncertainty levels. Higher scores indicate a higher level of disease uncertainty in patients . Hamilton anxiety rating A questionnaire survey was conducted to assess anxiety levels using the Hamilton Anxiety Scale (HAMA) before and after the simulated video education and one month after the examination. The HAMA, developed by Hamilton in 1959, was commonly utilized in clinical psychiatry and comprises 14 items. It was primarily employed to evaluate the severity of anxiety symptoms in patients with neurosis and other conditions. Each item in the HAMA was rated on a 5-point scale (0–4), representing: 0: none; 1: mild; 2: moderate; 3: severe; 4: very severe. The HAMA has demonstrated good internal consistency (Cronbach’s alpha = 0.893). The psychological scores (Hamilton Anxiety Scale) before and after treatment were compared between the two groups . PSQI sleep quality assessment A survey on sleep quality was conducted before and after the simulated video education, as well as one month after the examination. The Pittsburgh Sleep Quality Index (PSQI), developed by Buysse Dj and colleagues from the Sleep and Circadian Rhythms Center at the University of Pittsburgh Medical Center in 1993, was specifically designed to assess the subjective sleep quality of participants over the past month. The PSQI consists of 19 self-rated questions and 5 questions rated by sleep partners, with only the 19 self-rated questions being scored. These 19 self-rated questions form seven components, each scored from 0 to 3. The sum of the scores from each component yields the total score of the PSQI, ranging from 0 to 21, with higher scores indicating poorer sleep quality. The Chinese version of the PSQI demonstrated a Cronbach’s alpha of 0.71 . Statistical analysis SPSS 29.0 statistical software (SPSS Inc, Chicago, IL, USA) was used for data analysis. The classification data was expressed in the form of [n]. When the sample size was ≥ 40 and the theoretical frequency T ≥ 5, the basic formula was used to perform chi-square test. When the sample size is ≥ 40 but the theoretical frequency 1 ≤ T < 5, the Chi-square test of the correction formula is used. The Shapiro-Wilk method was used to test the normal distribution of continuous variables. For normally distributed continuous variables, they are presented as mean ± standard deviation (SD) and tested using the independent sample T-test. P < 0.05 for the two-sided test was considered statistically significant. This study was a retrospective cohort study that selected 166 patients who underwent digestive endoscopy examinations at Zhongshan City People’s Hospital, Guangdong Province, China from June 2023 to December 2023 and met the inclusion and exclusion criteria as the study subjects. Inclusion criteria: Patients with upper gastrointestinal problems such as upper gastric discomfort, bloating, pain, heartburn, swallowing discomfort, belching, regurgitation, reduced appetite, weight loss, anemia, as well as lower gastrointestinal issues including chronic diarrhea, constipation, abdominal pain, and bloating were included. Furthermore, patients with abnormal findings in gastrointestinal barium meal examinations that could not exclude tumors at the end of the colon or ileum, those with unexplained lower intestinal obstruction requiring confirmation of the extent and severity of inflammatory bowel disease, and individuals over 45 years old undergoing routine health examinations were also included. Exclusion criteria: Patients with severe heart, brain, lung, or liver diseases; those with uncontrolled acute gastrointestinal bleeding; individuals in the acute phase of gastrointestinal perforation, aortic aneurysm, or vascular rupture leading to significant bleeding; those who had undergone multiple laparotomies or had intestinal adhesions; patients with severe stenosis obstructing endoscopy passage, inability to tolerate gastric tube insertion, or those with mental disorders were excluded (Fig. ). This study was approved by the Ethics Committee of Zhongshan People’s Hospital. The procedures were conducted in accordance with the ethical standards set forth by the Committee on Human Experimentation and the Helsinki Declaration of 1964, as revised in 2013. Informed consent was waived by the Ethics Committee of Zhongshan People’s Hospital for this retrospective study due to the exclusive use of de-identified patient data, which posed no potential harm or impact on patient care. This waiver was approved by the institutional review board and ethics committee of our institution in accordance with regulatory and ethical guidelines pertaining to retrospective studies. Patients were divided into the conventional education group ( n = 77) and the simulated animated video education group ( n = 89) based on the educational mode. Conventional education involved communication between medical staff and patients scheduled for gastroscopy, informing the patients about the purpose of the examination, the procedure, and potential risks. Patients were advised about dietary restrictions before gastroscopy, discontinuation of anticoagulants and antiplatelet drugs, arranging for a companion, and attire considerations. Simulated animated video education involved organizing conventional educational content into an educational animation. Patients were instructed to watch this animation before the examination to fulfill the educational objectives: The video includes key topics such as preoperative diet and water ban requirements, surgical steps, expected feelings, risks and complications, postoperative diet and activity guidance, and how to communicate with doctors about doubts and concerns before and after surgery, which lasts for two minutes. The two-minute video was created using Adobe Animate by the Department of Gastroenterology, Endoscopy Centre of Zhongshan People’s Hospital (Fig. ). Patient demographic information, including age, Gender, BMI, Education Level, Hypertension, Diabetes, Hyperlipidemia was collected and recorded from the medical records system. Prior to undergoing gastrointestinal endoscopy, patients were required to fast for 6–8 h and take oral anesthesia to anesthetize the mucous membrane of the throat to alleviate reflex nausea and vomiting during esophagogastroduodenoscopy. The physician inserts the endoscope through the oropharynx into the esophagus, sequentially enters the esophagus, stomach, and duodenum, re-examines these areas during the withdrawal of the endoscope, and takes biopsies at appropriate sites. The examination was performed using the GASTROINTESTINAL VIDEOSCOPE (Registration No. 3220720, OLYMPUS OPTICAL Co., Ltd., Japan). For colonoscopy, patients begin taking bowel-cleansing agents the day before the procedure, as the colonic mucosa cannot be adequately observed in the presence of feces. After insertion through the anus into the terminal ileum, the colon was examined internally for lesions, and the entire mucosa was observed during the withdrawal of the endoscope. The examination was conducted using the Evis Colono Videoscope (Registration No. 3220887, OLYMPUS OPTICAL Co., Ltd., Japan). Patient examination sites and disease conditions were documented. Questionnaires were administered before and after the simulated video education, as well as one month after the examination. The Mishel’s Uncertainty in Illness Scale (MUIS) was used for assessment. This scale primarily measures patients’ levels of uncertainty regarding disease treatment, care, disease information, and prognosis during hospitalization. The scale has been translated into Chinese by domestic scholars, and its Cronbach’s α above 0.8. The scale consists of 4 dimensions: disease ambiguity, information deficit, complexity, and unpredictability, with a total of 33 items. The overall score ranges from 32.0 to 160.0, categorizable into low (32.0-74.7), moderate (74.8-117.4), and high (117.5–160.0) uncertainty levels. Higher scores indicate a higher level of disease uncertainty in patients . A questionnaire survey was conducted to assess anxiety levels using the Hamilton Anxiety Scale (HAMA) before and after the simulated video education and one month after the examination. The HAMA, developed by Hamilton in 1959, was commonly utilized in clinical psychiatry and comprises 14 items. It was primarily employed to evaluate the severity of anxiety symptoms in patients with neurosis and other conditions. Each item in the HAMA was rated on a 5-point scale (0–4), representing: 0: none; 1: mild; 2: moderate; 3: severe; 4: very severe. The HAMA has demonstrated good internal consistency (Cronbach’s alpha = 0.893). The psychological scores (Hamilton Anxiety Scale) before and after treatment were compared between the two groups . A survey on sleep quality was conducted before and after the simulated video education, as well as one month after the examination. The Pittsburgh Sleep Quality Index (PSQI), developed by Buysse Dj and colleagues from the Sleep and Circadian Rhythms Center at the University of Pittsburgh Medical Center in 1993, was specifically designed to assess the subjective sleep quality of participants over the past month. The PSQI consists of 19 self-rated questions and 5 questions rated by sleep partners, with only the 19 self-rated questions being scored. These 19 self-rated questions form seven components, each scored from 0 to 3. The sum of the scores from each component yields the total score of the PSQI, ranging from 0 to 21, with higher scores indicating poorer sleep quality. The Chinese version of the PSQI demonstrated a Cronbach’s alpha of 0.71 . SPSS 29.0 statistical software (SPSS Inc, Chicago, IL, USA) was used for data analysis. The classification data was expressed in the form of [n]. When the sample size was ≥ 40 and the theoretical frequency T ≥ 5, the basic formula was used to perform chi-square test. When the sample size is ≥ 40 but the theoretical frequency 1 ≤ T < 5, the Chi-square test of the correction formula is used. The Shapiro-Wilk method was used to test the normal distribution of continuous variables. For normally distributed continuous variables, they are presented as mean ± standard deviation (SD) and tested using the independent sample T-test. P < 0.05 for the two-sided test was considered statistically significant. Baseline characteristics Based on the table, the simulated animated video education group had 89 participants, while the regular education group had 77 participants. There were no statistically significant differences between the two groups in terms of age (52.14 ± 6.87 vs. 51.75 ± 7.21, t = 0.361, P = 0.718), gender distribution ( P = 0.982), BMI (23.57 ± 3.45 vs. 23.67 ± 3.31, t = 0.200, P = 0.842), hypertension prevalence ( P = 0.681), diabetes prevalence ( P = 0.681), and hyperlipidemia prevalence ( P = 0.878). The groups also did not differ significantly in education level ( P = 0.406). However, there was a statistically significant difference in the prevalence of hyperlipidemia (regular education: 8 (10.39%) vs. simulated animated video education: 11 (12.36%, P = 0.023) (Table ). Overall, the baseline characteristics were well balanced between the two education groups, ensuring comparability for the subsequent analysis. Examination areas and disease types The distribution of examination areas and disease types exhibited no statistically significant differences between patients who received regular education and those who received simulated animated video education (Table ). Specifically, examination areas including esophagus, stomach, duodenum, colon, and other areas showed no significant differences between the two educational intervention groups ( P = 0.616). Similarly, disease types such as superficial gastritis, atrophic gastritis, ulcer, and cancerous changes did not demonstrate significant differences between the two groups ( P = 0.525). These results indicate that the distribution of examination areas and disease types was similar in both groups, allowing for a balanced assessment of the impact of educational interventions on disease uncertainty, anxiety, and sleep quality. Disease uncertainty The Mean Uncertainty in Illness Scale (MUIS) scores demonstrated no statistically significant differences at the pre-education assessment between patients who received regular education and those who received simulated animated video education. However, following the educational interventions, the MUIS scores at post-education and one-month follow-up showed statistically significant decreases in both groups, indicating reduced disease uncertainty over time (Table ). These findings indicate that both interventions can reduce disease uncertainty after education and at one-month follow-up. However, the Simulated Animated Video Education has a greater impact on disease uncertainty in patients undergoing gastrointestinal endoscopy. Anxiety scores Comparison of HAMA scores revealed no statistically significant differences at the pre-education assessment between patients who received regular education and those who received simulated animated video education. However, following the educational interventions, the HAMA scores at post-education and the one-month follow-up demonstrated statistically significant decreases in both groups (Table ), indicating a reduction in anxiety levels after the educational interventions. Sleep quality The PQSI scores did not demonstrate statistically significant differences between patients who received regular education and those who received simulated animated video education at the pre-education assessment. However, following the educational interventions, the PQSI scores at the post-education and the one-month follow-up showed statistically significant decreases in both groups (Table ), indicating improved sleep quality after the educational interventions. Based on the table, the simulated animated video education group had 89 participants, while the regular education group had 77 participants. There were no statistically significant differences between the two groups in terms of age (52.14 ± 6.87 vs. 51.75 ± 7.21, t = 0.361, P = 0.718), gender distribution ( P = 0.982), BMI (23.57 ± 3.45 vs. 23.67 ± 3.31, t = 0.200, P = 0.842), hypertension prevalence ( P = 0.681), diabetes prevalence ( P = 0.681), and hyperlipidemia prevalence ( P = 0.878). The groups also did not differ significantly in education level ( P = 0.406). However, there was a statistically significant difference in the prevalence of hyperlipidemia (regular education: 8 (10.39%) vs. simulated animated video education: 11 (12.36%, P = 0.023) (Table ). Overall, the baseline characteristics were well balanced between the two education groups, ensuring comparability for the subsequent analysis. The distribution of examination areas and disease types exhibited no statistically significant differences between patients who received regular education and those who received simulated animated video education (Table ). Specifically, examination areas including esophagus, stomach, duodenum, colon, and other areas showed no significant differences between the two educational intervention groups ( P = 0.616). Similarly, disease types such as superficial gastritis, atrophic gastritis, ulcer, and cancerous changes did not demonstrate significant differences between the two groups ( P = 0.525). These results indicate that the distribution of examination areas and disease types was similar in both groups, allowing for a balanced assessment of the impact of educational interventions on disease uncertainty, anxiety, and sleep quality. The Mean Uncertainty in Illness Scale (MUIS) scores demonstrated no statistically significant differences at the pre-education assessment between patients who received regular education and those who received simulated animated video education. However, following the educational interventions, the MUIS scores at post-education and one-month follow-up showed statistically significant decreases in both groups, indicating reduced disease uncertainty over time (Table ). These findings indicate that both interventions can reduce disease uncertainty after education and at one-month follow-up. However, the Simulated Animated Video Education has a greater impact on disease uncertainty in patients undergoing gastrointestinal endoscopy. Comparison of HAMA scores revealed no statistically significant differences at the pre-education assessment between patients who received regular education and those who received simulated animated video education. However, following the educational interventions, the HAMA scores at post-education and the one-month follow-up demonstrated statistically significant decreases in both groups (Table ), indicating a reduction in anxiety levels after the educational interventions. The PQSI scores did not demonstrate statistically significant differences between patients who received regular education and those who received simulated animated video education at the pre-education assessment. However, following the educational interventions, the PQSI scores at the post-education and the one-month follow-up showed statistically significant decreases in both groups (Table ), indicating improved sleep quality after the educational interventions. The impact of educational interventions on patients undergoing medical procedures was an important aspect of healthcare delivery . This study aimed to investigate the influence of simulated animated video education compared to conventional education on patients’ disease uncertainty, anxiety, and sleep quality in the context of digestive endoscopy examinations. The findings of this study highlight the potential benefits of simulated animated video education in reducing disease uncertainty, alleviating anxiety, and improving sleep quality among patients undergoing gastrointestinal endoscopy. Disease uncertainty was a common concern among patients undergoing medical examinations and procedures . The results of this study demonstrated that both conventional and simulated animated video education interventions led to a reduction in disease uncertainty over time. However, the simulated animated video education had a greater impact on reducing disease uncertainty in patients undergoing gastrointestinal endoscopy. This finding suggests that providing visual and interactive educational content through animated videos may be more effective in addressing patients’ concerns and uncertainties regarding their medical condition and the impending procedure. The decrease in MUIS after educational intervention may be attributed to improved patient understanding and awareness, and the significant decrease in HAMA score could reflect patient improvement in mood, cognition, and somatic symptoms. Visual aids, such as animated videos, have the potential to effectively convey complex medical information in a clear and engaging manner, thereby empowering patients with knowledge and reducing ambiguity associated with their illness and the upcoming procedure . By addressing information deficit, complexity, unpredictability, and ambiguity related to the disease, the simulated animated video education may have positively influenced patients’ perceptions and expectations, ultimately contributing to reduced uncertainty levels. Moreover, the impact of educational interventions on anxiety levels among patients undergoing gastrointestinal endoscopy was another significant aspect evaluated in this study. Both conventional and simulated animated video education interventions were associated with a reduction in anxiety levels post-education and at the one-month follow-up. A study reported by Idrees demonstrated that playing mixed animated videos during self-examination among female patients could reduce the stigma associated with cancer and exhibited good efficiency and practicality in the early diagnosis of breast cancer. Animated simulation videos for educational purposes combine images, text, and audio-visual elements to vividly and innovatively present the entire process of gastrointestinal endoscopy. By enhancing patients’ understanding and reducing fear stemming from the unknown, these videos help lower anxiety levels. This finding underscores the potential of innovative educational methods, such as animated simulation videos, in alleviating procedural anxiety and improving patients’ overall mental health. The reduction in anxiety levels following the educational interventions may be linked to the ability of animated videos to enhance patient preparedness and mental readiness for the upcoming procedure . Visual depictions of the endoscopy process, along with information about the purpose of the examination and potential risks, may have helped alleviate fears and apprehensions related to the unknown aspects of the procedure . Preoperative anxiety is one of the primary causes of insomnia among patients. A notable decrease in PQSI scores was observed in both patient groups one month after surgery, with the group that watched the simulated animation video outperforming the conventional education group. The simulated animation video had a positive impact on the sleep quality of patients undergoing gastroscopy by providing intuitive understanding, enhancing psychological preparation, reducing anxiety-induced insomnia, and increasing a sense of security. As a result, patients experienced reduced anxiety, thereby lowering the risk of insomnia caused by preoperative anxiety. Additionally, the interactive nature of the animated videos may have engaged patients more effectively, providing them with a sense of involvement and control, ultimately contributing to reduced anxiety levels . The present study has several implications for clinical practice and patient care. The results underscore the value of incorporating innovative educational tools, such as simulated animated videos, into the pre-procedural education of patients undergoing gastrointestinal endoscopy. Healthcare providers can consider integrating visual and interactive educational content into their patient education strategies to address uncertainty, alleviate anxiety, and improve sleep quality among individuals scheduled for medical procedures. Moreover, the findings emphasize the importance of personalized and patient-centered approaches to healthcare delivery, recognizing the varying needs and preferences of patients when it comes to educational interventions. Our simulated animation videos adopt advanced animation techniques and design to show the whole process of gastrointestinal endoscopy in a vivid, vivid way. This expression not only enhances the visualization of information, but also makes the educational content easier to understand and remember. While the study provides valuable insights into the impact of simulated animated video education on patient outcomes, several limitations should be acknowledged. As a retrospective cohort study, bias may occur with the introduction of data collection and patient selection, and lack of long-term follow-up to assess sustained effects; to avoid selection bias, we ensured the reliability of data sources during data collection, adopted standardized data collection procedures, data verification and validation before data entry, and applied sound statistical analysis methods to avoid the impact of potential confounding factors on study findings. Moreover, the study focused on a specific patient population undergoing digestive endoscopy in one hospital, with small sample size and small differences in results that may not fully reflect the characteristics of all clinical patient groups, but its clinical significance and practical application value cannot be denied. However, the clinical significance and practical application value of the study findings cannot be denied. Based on the results of this study, future research should explore the long-term impact of animated educational videos on patient outcomes, consider different types of patient populations, and conduct multi-site, multi-center studies to collect broader and more representative sample data. And investigate the cost-effectiveness of implementing a simulated animation educational intervention in healthcare settings due to different cultural or regional differences. As an intuitive and easy to understand way of information transmission, simulated animation video education is also of great value in high-risk cardiac surgery and neurosurgery, requiring complex medical operations, or patients’ understanding. For long-term medical processes such as chronic disease management and rehabilitation patient education, simulated animation video education can also play an important role. In conclusion, the findings of this study highlight the potential benefits of simulated animated video education in reducing disease uncertainty, alleviating anxiety, and improving sleep quality among patients undergoing gastrointestinal endoscopy. The results support the integration of innovative educational approaches into pre-procedural patient education strategies as a means of enhancing patient experiences and well-being in the healthcare setting. By addressing patients’ psychological and informational needs using visually engaging and interactive content, healthcare providers can contribute to improved patient outcomes and satisfaction, ultimately fostering a more holistic and patient-centered approach to healthcare delivery. |
Two types of microorganisms isolated from petroleum hydrocarbon pollutants: Degradation characteristics and metabolic pathways analysis of petroleum hydrocarbons | f943dfbc-e303-4995-bb83-1e8a0141e01a | 11559972 | Biochemistry[mh] | With the rapid development of the economy and society, the one-time energy consumption of China is substantial, with oil accounting for the largest proportion. Simultaneously, oil production, storage, and transportation are becoming increasingly developed, leading to frequent offshore oil spills and serious pollution of the marine environment. According to the 2018 China Marine Ecological Environment Status Bulletin released by the Ministry of Ecology and Environmental Protection, the area of the four types of poor water quality sea areas in the summer of 2018 was approximately 33000 km 2 , with the main exceeding substances being active phosphates, inorganic nitrogen, and petroleum. Petroleum hydrocarbons are a mixture of various hydrocarbons, such as n-alkanes, branched alkanes, cycloalkanes, and aromatics, and a small amount of other organic substances, such as sulphides, nitrides, and cycloalkanes, 95% of which comprise carbon and hydrogen compounds . Numerous studies have shown that the chemical toxicity of oil to marine organisms varies depending on its type and composition . In general, the toxicity of refined oil is higher than that of crude oil, and the toxicity of low molecular weight hydrocarbons is higher than that of high molecular weight hydrocarbons . The toxicity of various hydrocarbons is generally in the order aromatic hydrocarbons > olefins > cyclohydrocarbons > chain hydrocarbons. Petroleum hydrocarbons are highly toxic to marine organisms, mainly because they disrupt the normal structure and permeability of cell membranes, interfere with the enzyme system of organisms, and thus affect their normal physiological and biochemical processes. Bioaccumulation in the food chain causes great harm to human health, including stomach diseases and even cancer. Petroleum hydrocarbons have been listed as key monitoring targets by the United Nations Environment Programme. The commonly used methods for the degradation of marine petroleum hydrocarbon pollutants can be divided into physical, chemical, and biological methods . Physical methods are direct methods for oil repair and recovery that do not change the physical and chemical properties of the oil . For example, adding 10 g/L commercial synthetic resin (XAD7) can remove 96.3% of emulsified oil in 25–50 mg/L simulated emulsified wastewater , and adding 20 g/L chitosan and acrylamide combined with a new hydrogel can remove 97% of 0.25 g/L crude oil wastewater . These methods are simple and safe but have limitations, such as incomplete treatment and recycling, and are often suitable for controlling and recycling sudden oil spills. Chemical methods mainly involve burning the spilled oil onsite and adding materials to oil-contaminated water . Chemical combustion can produce large amounts of particulate matter, CO 2 , and SO 2 that can cause serious air pollution. Altaş et al found that removal efficiency of chemical oxygen demand (COD) (50–80%) of petroleum refinery wastewater was obtained by using FeSO 4 ·7H 2 O together with Ca(OH) 2 as precipitant-aid. However, Fe 2+ may be easily oxidized by oxygen in air, while Fe 3+ easily forms hydroxide in water. In order to reduce the secondary pollution of the environment caused by the use of chemical agents during the oil removal process, environmentally friendly and renewable materials are being used. Peng et al. summarized the uses of cellulose-based materials for crude oil spill cleaning. These materials have shown promising results in oil removal. However, the main issues such as the material preparation, used oil sorbents disposal still need to be further addressed. Biological methods primarily involve bioremediation . Currently, anaerobic degradation of pollutants in common bioremediation can be achieved through two pathways. One way is to generate energy by coupling substrate oxidation to respiration via the reduction of a non-oxygen terminal electron acceptor, and the other way is to generate energy via a fermentation pathway. These organisms, some bacteria (e.g., Pseudomonas , Thiobacillus , Geobacter ) and some archaea (e.g., Desulfobacterales , Methanosaeteta sp ., Methanobacterium beijingense ), are called fermentative degraders . Qaderi F et al. built a moving-bed biofilm reactor (MBBR), in which a biofilm containing various microorganisms suitable for treating petroleum pollutants was prepared from the reflux sludge flow to the activated sludge unit at the Tehran refinery treatment plant. The results showed that retention time of 23 h, influent total petroleum hydrocarbon of 164.78 mg/L, and media filling ratio of 45% yielded, the highest efficiency of 97% in removing petroleum pollutants. A laboratory-scale study was conducted to evaluate the feasibility of a modified rotating biological contactor with polyurethane foam (RBC-PUF) attached to a disk as a porous support medium for the biodegradation of refinery wastewater. The RBC-PUF was inoculated with a seed culture obtained from an activated sludge unit treating the refinery wastewater, then a thin biofilm containing multiple microorganisms was formed for treating refinery wastewater. The results showed that the total COD and oil removal efficiency exceeded 87% and 80%, respectively . Compared with traditional physicochemical methods, biological methods have the advantages of low cost, small environmental impact, and no secondary pollution, and can be used for fixed-point remediation, making them promising methods for petroleum hydrocarbon treatment . Considering the complexity of petroleum hydrocarbon wastewater and the increasing concern about the toxicity of organic pollutants entering the environment, it is crucial to establish a predictive model that can quickly and accurately determine the biological toxicity of organic pollutants. Quantitative Structure Activity Relationship (QSAR) refers to the quantitative relationship between the molecular structure of organic pollutants and their biological toxicity . QSAR is such an effective prediction model, and the development of the Toxicity Estimation Software Tool (TEST) conforms to the trend of QSAR research, including the development of large QSAR packages, mutual penetration with large databases, and the application of toxicity models, making this software of great significance in the field of toxicity assessment and successfully opening up a new path for QSAR research . Therefore, this study aimed to investigate the effects of selected strains, strain dosage, petroleum hydrocarbon concentration, mixed strains, and mixing ratio on the degradation efficiency of petroleum hydrocarbons and explore the maximum degradation efficiency of petroleum hydrocarbons by strains. Exploring the biodegradation mechanisms provides a theoretical basis for improving the degradation efficiency of petroleum hydrocarbon wastewater. The results of this study will provide technical support and solutions for effective treatment of oil pollution. This technology can quickly and efficiently reduce the pollution caused by oil to the environment. Moreover, it can not only control point-source pollution and surface-source pollution, but also be extended to oil spill emergency response, thereby promoting the development of environmental protection.
Experimental materials The main reagents used in the experiment, LB medium, NaCl, and H 2 SO 4 , were purchased from Sinopharm Chemical Reagent Co., Ltd. (Beijing, China). Commercial engine oil, used as simulated petroleum hydrocarbon wastewater, was obtained from an automobile maintenance company in Changzhou, China. Two bacterial strains were isolated from the oil reservoir environment and named W01 and W02. Strains W01 and W02 were identified as Stenotrophomonas acidaminiphila (GenBank: EU352763.1) and Ochrobactrum (GenBank: KY471630.1), respectively. The methods for screening, isolation, and purification of strains can be found in . Growth and degradation of petroleum hydrocarbons by strains W01 and W02 Individual cultivation of strains: Strains W01 and W02 were inoculated into 100 mL of sterilised LB liquid medium and activated in a shaking incubator (ZHLY-300S, Shanghai Zhichu Instrument Co., Ltd., China) for 20 h at 30°C and 140 rpm. The activated two strains were then inoculated into two 100 mL LB liquid culture media at a volume ratio of 2%, and incubated for 48 h at 30°C, 140 rpm, and salinity of 1% until the logarithmic phase of strain growth. Engine oil at a volume ratio of 2% was added to a conical flask containing the liquid culture medium. The above solution was cultured under shaking conditions at 30°C and 140 rpm for 5 days to investigate the degradation efficiency of the strain on oil. Strain combination culture: Strains W01 and W02 were inoculated into 100 mL of sterilised LB liquid medium and activated for 20 h at 30°C and 140 rpm. The activated two strains were then inoculated into two 100 mL LB liquid culture media at a volume ratio of 2% and cultivated for 48 h under the conditions of 30°C, 140 rpm, and 1% salinity until the logarithmic phase of strain growth. W01 and W02 bacterial liquids were absorbed to form 100 mL of mixed bacterial liquid, with a combination ratio of W01: W02 = 1:1, W01: W02 = 2:1, and W01: W02 = 1:2, and 2% oil was added to the conical flask of liquid culture media. The above solution was cultured under shaking conditions at 30°C and 140 rpm for 5 days to investigate the degradation efficiency of the mixed strains on oil. Analytical methods The analysis methods of total petroleum hydrocarbon (TPH) and hydrocarbon composition were same as described by Chang et al. . Briefly, the TPH extractions were conducted in an automatic Soxhlet extractor (SOX 402 Micro, Gerhard Soxtherm, Germany) and analyzed using a gas chromatography-flame ionisation detector (GC-FID, 6890N, Agilent, USA). Then, the relative contents of the petroleum hydrocarbon components were analysed and determined using gas chromatography-mass spectrometry (GC-MS, 7890A-5975C, Agilent, USA). The testing method for GC-MS can be found in . The toxicity of the intermediate products is the basis for determining whether the treatment method is feasible. Based on the median lethal dose (LD 50 ) of oral rat as the toxicity parameter, the hierarchical clustering in TEST software, which is one of the QSAR model methods, was used to evaluate the oxicity of the products produced during the degradation of petroleum hydrocarbons . The toxicity analysis method for intermediate products can be found in .
The main reagents used in the experiment, LB medium, NaCl, and H 2 SO 4 , were purchased from Sinopharm Chemical Reagent Co., Ltd. (Beijing, China). Commercial engine oil, used as simulated petroleum hydrocarbon wastewater, was obtained from an automobile maintenance company in Changzhou, China. Two bacterial strains were isolated from the oil reservoir environment and named W01 and W02. Strains W01 and W02 were identified as Stenotrophomonas acidaminiphila (GenBank: EU352763.1) and Ochrobactrum (GenBank: KY471630.1), respectively. The methods for screening, isolation, and purification of strains can be found in .
Individual cultivation of strains: Strains W01 and W02 were inoculated into 100 mL of sterilised LB liquid medium and activated in a shaking incubator (ZHLY-300S, Shanghai Zhichu Instrument Co., Ltd., China) for 20 h at 30°C and 140 rpm. The activated two strains were then inoculated into two 100 mL LB liquid culture media at a volume ratio of 2%, and incubated for 48 h at 30°C, 140 rpm, and salinity of 1% until the logarithmic phase of strain growth. Engine oil at a volume ratio of 2% was added to a conical flask containing the liquid culture medium. The above solution was cultured under shaking conditions at 30°C and 140 rpm for 5 days to investigate the degradation efficiency of the strain on oil. Strain combination culture: Strains W01 and W02 were inoculated into 100 mL of sterilised LB liquid medium and activated for 20 h at 30°C and 140 rpm. The activated two strains were then inoculated into two 100 mL LB liquid culture media at a volume ratio of 2% and cultivated for 48 h under the conditions of 30°C, 140 rpm, and 1% salinity until the logarithmic phase of strain growth. W01 and W02 bacterial liquids were absorbed to form 100 mL of mixed bacterial liquid, with a combination ratio of W01: W02 = 1:1, W01: W02 = 2:1, and W01: W02 = 1:2, and 2% oil was added to the conical flask of liquid culture media. The above solution was cultured under shaking conditions at 30°C and 140 rpm for 5 days to investigate the degradation efficiency of the mixed strains on oil.
The analysis methods of total petroleum hydrocarbon (TPH) and hydrocarbon composition were same as described by Chang et al. . Briefly, the TPH extractions were conducted in an automatic Soxhlet extractor (SOX 402 Micro, Gerhard Soxtherm, Germany) and analyzed using a gas chromatography-flame ionisation detector (GC-FID, 6890N, Agilent, USA). Then, the relative contents of the petroleum hydrocarbon components were analysed and determined using gas chromatography-mass spectrometry (GC-MS, 7890A-5975C, Agilent, USA). The testing method for GC-MS can be found in . The toxicity of the intermediate products is the basis for determining whether the treatment method is feasible. Based on the median lethal dose (LD 50 ) of oral rat as the toxicity parameter, the hierarchical clustering in TEST software, which is one of the QSAR model methods, was used to evaluate the oxicity of the products produced during the degradation of petroleum hydrocarbons . The toxicity analysis method for intermediate products can be found in .
The effect of pure bacteria on the degradation of petroleum hydrocarbons The effect of strain species on the degradation efficiency of petroleum hydrocarbons components The effect of strain type on the degradation of petroleum hydrocarbons was studied. The degradation of petroleum hydrocarbons by strains W01 and W02 is shown in Figs and . As shown in , although the strain W01 can consume both short- and medium-long carbon chain C10-C30 components in petroleum hydrocarbons, the degradation efficiencies of different petroleum hydrocarbon components by the strain are also different. The highest degradation efficiency of C27 by the strain W01 was 79.95%, whereas the lowest degradation effect was observed for C24, at only 21.71%. The average degradation efficiency of the strain W01 of petroleum hydrocarbon components was 62.98%. In the report of Gregorio et al. , the Stenotrophomonas sp. M1 strain was used to degrade diesel oil. After 6 days, the degradation efficiencies of C27 and C24 were 39.45% and 23.65%, respectively. C28, C29, and C30 achieved high degradation efficiencies of 46.17%, 45.99%, and 43.06%, respectively, while in this study, they were 37.15%, 60.74% and 66.86%, respectively, during 5 days operation ( ). Overall, the average degradation efficiency in this study was higher than that of Gregorio et al. , indicating that the strain W01 had a good degradation effect on petroleum hydrocarbons. As shown in , the degradation effect of the strain W02 on different petroleum hydrocarbon components was also different. After 5 days, the strain W02 exhibited the best degradation efficiency of C17, reaching 81.93%, whereas the lowest degradation efficiency for C21 (29.97%). The average degradation efficiency of the strain W02 of petroleum hydrocarbon components was 59.14%. Kshirsagar et al. found that in 15 days, Ochrobactrum intermedium could degrade hydrocarbons in crude oil fractions 500 N, lube oil, and de-asphalted oil. The hydrocarbons removal varied from 20%-30%. Thus, strain W02 also had the ability to degrade petroleum hydrocarbons. The effect of strain dosage on the degradation efficiency of petroleum hydrocarbons To investigate the effect of different bacterial liquid dosages on the degradation of petroleum hydrocarbons under optimised conditions for the individual growth of the strains W01 and W02, 50, 100, 150, and 200 mL of the logarithmic growth phase bacterial liquid was added to 100 mL of wastewater with a petroleum hydrocarbon concentration of 1 mL/L while maintaining the same cultivation conditions. The above solution was cultured under shaking conditions at 30°C and 140 rpm for 5 days to investigate the degradation efficiency of TPH. The results are shown in . As shown in , under the conditions of different dosages of 50, 100, 150, and 200 mL of bacterial solution, the degradation efficiencies of both strains for 1 mL/L of petroleum hydrocarbon wastewater showed a rapid initial increase, followed by a slow and stable trend. The degradation efficiency of the strain W01 on TPH was 50.93% at 50 mL and 55.26% at 100 mL. The degradation efficiency increased by 28.07% when the dosage increased to 150 mL, reaching 70.77%. Subsequently, it stabilised, and the degradation efficiency reached 71.69% at 200 mL. The degradation efficiency of the strain W02 increased by 16.13% when the strain dosage was increased from 50 mL (50.77%) to 100 mL (58.96%). Then, the degradation efficiency of TPH increased by 29.38% when the dosage was increased from 100 mL to 150 mL, reaching 76.28%. Continue to increase the strain dosage to 200 mL, the degradation efficiency increased by 2.70%, reaching 78.34%. It can be seen that the degradation efficiency of TPH increases as the amount of strain dosage increases. However, it is evident that as the strain dosage increases from 150 mL to 200 mL, the increase in degradation efficiency is not significant. This may be because as the amount of strains added increases, the increase in biosurfactant glycolipopeptidal will inhibit biodegradation, as previously studied . Although the versatility of Stenotrophomonas and Ochrobactrum in degrading petroleum hydrocarbons is known [ – ], our findings further indicate that the isolate Stenotrophomonas and Ochrobactrum exhibited good engine oil degradation capacities. Notably, Stenotrophomonas and Ochrobactrum were responsible for degrading petroleum hydrocarbons, but it is difficult to have a comparative assessment since oil type and concentration varied across the studies ( ). The effect of petroleum hydrocarbon concentration on bacterial degradation The 200 mL of wastewater with initial petroleum hydrocarbon concentrations were set to 0.5, 1, 2, 3, and 4 mL/L, while maintaining the same conditions as other cultures, to investigate the degradation of different petroleum hydrocarbon concentrations by the growth of the strains W01 and W02 under optimised conditions. The samples were incubated under shaking conditions at 30°C and 140 rpm for 5 days. The results are shown in . As shown in , at initial petroleum hydrocarbon concentrations of 0.5, 1, 2, 3, and 4 mL/L, the degradation efficiency of TPH by both strains first increases and then decreases with an increase in the initial petroleum hydrocarbon concentration. Moreover, at a concentration of 2 mL/L, the degradation efficiency was optimal, reaching 81.22% and 85.88%. The reason for its decline may be due to the increase in petroleum hydrocarbon concentration inhibiting the destruction of the bacterial growth environment, thereby inhibiting the growth rate of bacteria and leading to a decrease in degradation efficiency, which has also been mentioned in a previous study . The effect of mixed bacteria on the degradation of petroleum hydrocarbons According to previous studies, there are differences in the consumption of petroleum hydrocarbons when W01 and W02 bacteria grow alone. Strains W01 and W02 were inoculated at a 1:1 ratio into 100 mL of wastewater with a petroleum hydrocarbon concentration of 1 mL/L and cultured for 5 days to investigate the effect of co-growth of the combined strains on the degradation of petroleum hydrocarbons. The results are shown in . shows that the combined strains can consume short- and medium-carbon chain C10-C30 components. The degradation efficiencies of C10, C12, C15, C17, C21, and C26 by mixed bacteria are all not below 80%. However, the degradation efficiencies of C10-C30 by strain W01 are all below 80%, while strain W02 only obtained degradation efficiencies above 80% for C10 and C17, respectively (as shown in Figs and ). The average degradation efficiency of the mixed strain of petroleum hydrocarbon components was 73.30%, which was 10.32% and 14.16% higher than that of strain W01 and W02 alone, respectively. This result was consistent with report of Kshirsagar et al. . Kshirsagar et al. discovered that mixing Ochrobactrum intermedium with other bacterial strains ( Pseudomonas putida , Pseudomonas mendocina , Bacillus cereus , Bacillus marisflavi , Lysinibacillus fusiformis ) could effectively degrade hydrocarbons in crude oil fractions 500 N, lube oil, and de-asphalted oil. The hydrocarbons removal varied from 40%-60%, which showed an increase of approximately 50% in biodegradation compared to Ochrobactrum intermedium . The levels of the C21-C30, C31-C35, and C36-C60 decreased by 13%–54%, 49%–70%, and 39%–60%, respectively. To investigate the consumption of petroleum hydrocarbons by the growth of the two bacterial strains in different proportions under optimised conditions, the two strains were mixed and cultured in different combinations according to W01: W02 = 1:1/1:2/2:1. The results are presented in . shows that, compared to the 1:1 ratio, when the mixing ratios are 1:2 and 2:1, the combined strain has a higher degradation efficiency of TPH during growth, which is approximately 12.0% higher. However, the degradation efficiency of different mixing ratios can be maintained above 60%, which also indicates that the consumption of petroleum hydrocarbons by the mixed strains under the same growth conditions is stronger than that of W01 and W02 bacteria growing alone. Similar situations have also occurred in previous study . Teng et al. constructed a mixed cold-adapted bacteria flora, which contained Stenotrophomonas sp., to treat contaminated water (1 g oil/L). The mixed flora demonstrated 53.68% of TPH removal after 30 days of incubation under 10°C. This may be because the characteristic of oil pollution is the presence of saturated components composed of alkanes and more stubborn unsaturated components. In this case, a combination of multiple bacteria has a wider and different enzymatic ability, thereby improving the degradation level of pollution . Moreover, different strains of bacteria grow and reproduce in the same environment, competing for nutrients and substrates. At the same time, their own metabolites and other factors can have a comprehensive impact on microbial growth . Analysis of the degradation mechanism of petroleum hydrocarbons GC-MS analysis of metabolic products in the degradation of petroleum hydrocarbons Four intermediates, namely n-eicosane, 1-neneneba hexadecane alcohol, 2-octyldodecanol, and 2,4-di-tert-butylphenol, were detected using GC-MS during the degradation of petroleum hydrocarbons by strain W01 ( and – Figs). During petroleum hydrocarbon degradation by the strain W02, four intermediates were detected using GC-MS, namely diphenylamine, lauryl methacrylate, 2-Ethylhexanol, and ethyl hexyl benzoate ( and – Figs). Metabolomics analysis of differentially expressed genes The correlation of differentially expressed genes between the two samples was tested using the original number of reads. According to the correlation heat map of the samples in , the correlation between the samples is strong. The Pearson product-moment correlation coefficient between the two biological replicates of the sample was > 0.9, and the differential gene expression levels between the sample and blank groups tended to be the same. According to the heat map analysis, the biological repeatability of sequencing was good, and the data were reliable. and Figs showed that there are 397 differentially expressed genes between the strain W01 sample and the blank sample, of which 44 are significantly upregulated, such as N-Butyryl-L-homoserine lactone, Glycyl-leucine, and 73 are significantly downregulated, such as eryyhro-3-Hydroxy-Ls-aspartate, 2-Deoxyguanosine, Propazine, Dimethylglycine. There were 397 differentially expressed genes between the W02 and blank samples, of which 32 are significantly upregulated, such as Procollagen 5-hydroxy-L-lysine, Propazine, N6-Acetyl-N6-hydroxy-L-lysine, Betaine, Pantothenic acid, and 27 are significantly downregulated, such as Ureidopropionic acid, 4-Hydroxycinnamic acid, Pantothenic acid, (R)-3-Hydroxybutyric acid, Deoxyinosine. shows the differential expression heatmaps of the strain W01 and the blank control group. The treated samples showed the following substances: L-Alanyl-gamma-D-glutamyl-L lysine (C14), 1- (3,4-Dihydroxyphenyl) -5-hydroxy-3-decanone (C16), Reversrol (C14), and other components. Simultaneously, the differential expression heatmaps of the strain W02 and the blank control group were analysed. As shown in , the result also indicated the presence of the following substances: L-Alanyl-gamma-D-glutamyl-L-lysine (C14), Lenacil (C13), 1- (3,4-Dihydroxyphenyl) -5-hydroxy-3-decane1 (C16), Labelalol (C19), and other components, which were mutually confirmed from the intermediate products detected using GC-MS. According to Sajna et al. and Zhang et al. , benzoate and phenylacetic acid are the upstream and downstream products of catechol, respectively, and heptylcyclohexane is broken into alkanes by intermediate enzymes or bases. Preliminary derivation of degradation pathways As strains that can grow and reproduce using petroleum hydrocarbons as the sole carbon source, the mechanisms by which strains W01 and W02 degrade petroleum hydrocarbons have become important research targets, and multiple scholars have explored their degradation mechanisms . According to the GC-MS detection and metabolomic determination, n-eicosane was present in the wastewater before strain W01 degradation, whereas n-eicosane was not detected in the wastewater after degradation, and the two intermediate products, 1-nenenebc hexadecane alcohol and 2-octyldodecanol, were obtained. Diphenylamine and lauryl methacrylate were present in the wastewater before the strain W02 degradation, whereas diphenylamine and lauryl methacrylate were not detected in the wastewater after degradation, and 2-Ethylhexanol and ethyl hexyl benzoate were obtained. In the degradation of petroleum hydrocarbons by the strain W01, n-eicosane was first oxidized to 2-octyldodecanol, the hydroxyl group of 2-octyldodecanol was further oxidized to the aldehyde group, the aldehyde group was further oxidized to the ester group, and finally the carbon oxygen bond broke to produce 1-nenenebb hexadecane alcohol, and the C-O bond in 7,9-di-tert-butyl-1-oxa [4.5] deca-6,9-dien-2,8-dione cracked to carboxylate, and finally degraded to 2,4-di-tert-butylphenol ( ) [ – ]. In the degradation of petroleum hydrocarbons by the strain W02, diphenylamine may react with hydroxyl radicals to generate benzene; the C-C bond of lauryl methacrylate broke, and benzene generated ethyl hexyl benzoate, followed by the C-O bond breaking to generate 2-Ethylhexanol ( ) . However, owing to the complexity and variability of microbial degradation processes, and the considerable impact of environmental factors, further refinement of degradation pathways requires more in-depth research. Toxicity assessment of intermediate products in petroleum hydrocarbon wastewater degradation by strains presents the predicted results for the intermediate product LD 50 detected in the W01 and W02 bacterial petroleum hydrocarbon-degrading wastewater systems. In the strain W01-degrading petroleum hydrocarbon wastewater system, the decrease in the content of n-eicosane indicates that it can be degraded by strain W01 into small molecule substances. Meanwhile, LD 50 of intermediate products such as 2-octyldodecanol and 2,4-di-tert-butylphenol, is significantly lower than that of n-eicosane. Besides, although the LD 50 value of 1-neneneba hexadecane alcohol is high, its content in the degraded wastewater is much lower than other substances. In the strain W02-degrading petroleum hydrocarbon wastewater system, the LD 50 values of the two newly generated substances, 2-Ethylhexanol and ethyl hexyl benzoate, inferred through degradation pathways significantly lower than that of lauryl methacrylate in the original petroleum hydrocarbon wastewater. This indicates a decreasing trend in the toxicity of the reaction system. Therefore, from the perspective of toxicity assessment, strains W01 and W02 have great potential for remediating petroleum hydrocarbon wastewater.
The effect of strain species on the degradation efficiency of petroleum hydrocarbons components The effect of strain type on the degradation of petroleum hydrocarbons was studied. The degradation of petroleum hydrocarbons by strains W01 and W02 is shown in Figs and . As shown in , although the strain W01 can consume both short- and medium-long carbon chain C10-C30 components in petroleum hydrocarbons, the degradation efficiencies of different petroleum hydrocarbon components by the strain are also different. The highest degradation efficiency of C27 by the strain W01 was 79.95%, whereas the lowest degradation effect was observed for C24, at only 21.71%. The average degradation efficiency of the strain W01 of petroleum hydrocarbon components was 62.98%. In the report of Gregorio et al. , the Stenotrophomonas sp. M1 strain was used to degrade diesel oil. After 6 days, the degradation efficiencies of C27 and C24 were 39.45% and 23.65%, respectively. C28, C29, and C30 achieved high degradation efficiencies of 46.17%, 45.99%, and 43.06%, respectively, while in this study, they were 37.15%, 60.74% and 66.86%, respectively, during 5 days operation ( ). Overall, the average degradation efficiency in this study was higher than that of Gregorio et al. , indicating that the strain W01 had a good degradation effect on petroleum hydrocarbons. As shown in , the degradation effect of the strain W02 on different petroleum hydrocarbon components was also different. After 5 days, the strain W02 exhibited the best degradation efficiency of C17, reaching 81.93%, whereas the lowest degradation efficiency for C21 (29.97%). The average degradation efficiency of the strain W02 of petroleum hydrocarbon components was 59.14%. Kshirsagar et al. found that in 15 days, Ochrobactrum intermedium could degrade hydrocarbons in crude oil fractions 500 N, lube oil, and de-asphalted oil. The hydrocarbons removal varied from 20%-30%. Thus, strain W02 also had the ability to degrade petroleum hydrocarbons.
The effect of strain type on the degradation of petroleum hydrocarbons was studied. The degradation of petroleum hydrocarbons by strains W01 and W02 is shown in Figs and . As shown in , although the strain W01 can consume both short- and medium-long carbon chain C10-C30 components in petroleum hydrocarbons, the degradation efficiencies of different petroleum hydrocarbon components by the strain are also different. The highest degradation efficiency of C27 by the strain W01 was 79.95%, whereas the lowest degradation effect was observed for C24, at only 21.71%. The average degradation efficiency of the strain W01 of petroleum hydrocarbon components was 62.98%. In the report of Gregorio et al. , the Stenotrophomonas sp. M1 strain was used to degrade diesel oil. After 6 days, the degradation efficiencies of C27 and C24 were 39.45% and 23.65%, respectively. C28, C29, and C30 achieved high degradation efficiencies of 46.17%, 45.99%, and 43.06%, respectively, while in this study, they were 37.15%, 60.74% and 66.86%, respectively, during 5 days operation ( ). Overall, the average degradation efficiency in this study was higher than that of Gregorio et al. , indicating that the strain W01 had a good degradation effect on petroleum hydrocarbons. As shown in , the degradation effect of the strain W02 on different petroleum hydrocarbon components was also different. After 5 days, the strain W02 exhibited the best degradation efficiency of C17, reaching 81.93%, whereas the lowest degradation efficiency for C21 (29.97%). The average degradation efficiency of the strain W02 of petroleum hydrocarbon components was 59.14%. Kshirsagar et al. found that in 15 days, Ochrobactrum intermedium could degrade hydrocarbons in crude oil fractions 500 N, lube oil, and de-asphalted oil. The hydrocarbons removal varied from 20%-30%. Thus, strain W02 also had the ability to degrade petroleum hydrocarbons.
To investigate the effect of different bacterial liquid dosages on the degradation of petroleum hydrocarbons under optimised conditions for the individual growth of the strains W01 and W02, 50, 100, 150, and 200 mL of the logarithmic growth phase bacterial liquid was added to 100 mL of wastewater with a petroleum hydrocarbon concentration of 1 mL/L while maintaining the same cultivation conditions. The above solution was cultured under shaking conditions at 30°C and 140 rpm for 5 days to investigate the degradation efficiency of TPH. The results are shown in . As shown in , under the conditions of different dosages of 50, 100, 150, and 200 mL of bacterial solution, the degradation efficiencies of both strains for 1 mL/L of petroleum hydrocarbon wastewater showed a rapid initial increase, followed by a slow and stable trend. The degradation efficiency of the strain W01 on TPH was 50.93% at 50 mL and 55.26% at 100 mL. The degradation efficiency increased by 28.07% when the dosage increased to 150 mL, reaching 70.77%. Subsequently, it stabilised, and the degradation efficiency reached 71.69% at 200 mL. The degradation efficiency of the strain W02 increased by 16.13% when the strain dosage was increased from 50 mL (50.77%) to 100 mL (58.96%). Then, the degradation efficiency of TPH increased by 29.38% when the dosage was increased from 100 mL to 150 mL, reaching 76.28%. Continue to increase the strain dosage to 200 mL, the degradation efficiency increased by 2.70%, reaching 78.34%. It can be seen that the degradation efficiency of TPH increases as the amount of strain dosage increases. However, it is evident that as the strain dosage increases from 150 mL to 200 mL, the increase in degradation efficiency is not significant. This may be because as the amount of strains added increases, the increase in biosurfactant glycolipopeptidal will inhibit biodegradation, as previously studied . Although the versatility of Stenotrophomonas and Ochrobactrum in degrading petroleum hydrocarbons is known [ – ], our findings further indicate that the isolate Stenotrophomonas and Ochrobactrum exhibited good engine oil degradation capacities. Notably, Stenotrophomonas and Ochrobactrum were responsible for degrading petroleum hydrocarbons, but it is difficult to have a comparative assessment since oil type and concentration varied across the studies ( ). The effect of petroleum hydrocarbon concentration on bacterial degradation The 200 mL of wastewater with initial petroleum hydrocarbon concentrations were set to 0.5, 1, 2, 3, and 4 mL/L, while maintaining the same conditions as other cultures, to investigate the degradation of different petroleum hydrocarbon concentrations by the growth of the strains W01 and W02 under optimised conditions. The samples were incubated under shaking conditions at 30°C and 140 rpm for 5 days. The results are shown in . As shown in , at initial petroleum hydrocarbon concentrations of 0.5, 1, 2, 3, and 4 mL/L, the degradation efficiency of TPH by both strains first increases and then decreases with an increase in the initial petroleum hydrocarbon concentration. Moreover, at a concentration of 2 mL/L, the degradation efficiency was optimal, reaching 81.22% and 85.88%. The reason for its decline may be due to the increase in petroleum hydrocarbon concentration inhibiting the destruction of the bacterial growth environment, thereby inhibiting the growth rate of bacteria and leading to a decrease in degradation efficiency, which has also been mentioned in a previous study .
The 200 mL of wastewater with initial petroleum hydrocarbon concentrations were set to 0.5, 1, 2, 3, and 4 mL/L, while maintaining the same conditions as other cultures, to investigate the degradation of different petroleum hydrocarbon concentrations by the growth of the strains W01 and W02 under optimised conditions. The samples were incubated under shaking conditions at 30°C and 140 rpm for 5 days. The results are shown in . As shown in , at initial petroleum hydrocarbon concentrations of 0.5, 1, 2, 3, and 4 mL/L, the degradation efficiency of TPH by both strains first increases and then decreases with an increase in the initial petroleum hydrocarbon concentration. Moreover, at a concentration of 2 mL/L, the degradation efficiency was optimal, reaching 81.22% and 85.88%. The reason for its decline may be due to the increase in petroleum hydrocarbon concentration inhibiting the destruction of the bacterial growth environment, thereby inhibiting the growth rate of bacteria and leading to a decrease in degradation efficiency, which has also been mentioned in a previous study .
According to previous studies, there are differences in the consumption of petroleum hydrocarbons when W01 and W02 bacteria grow alone. Strains W01 and W02 were inoculated at a 1:1 ratio into 100 mL of wastewater with a petroleum hydrocarbon concentration of 1 mL/L and cultured for 5 days to investigate the effect of co-growth of the combined strains on the degradation of petroleum hydrocarbons. The results are shown in . shows that the combined strains can consume short- and medium-carbon chain C10-C30 components. The degradation efficiencies of C10, C12, C15, C17, C21, and C26 by mixed bacteria are all not below 80%. However, the degradation efficiencies of C10-C30 by strain W01 are all below 80%, while strain W02 only obtained degradation efficiencies above 80% for C10 and C17, respectively (as shown in Figs and ). The average degradation efficiency of the mixed strain of petroleum hydrocarbon components was 73.30%, which was 10.32% and 14.16% higher than that of strain W01 and W02 alone, respectively. This result was consistent with report of Kshirsagar et al. . Kshirsagar et al. discovered that mixing Ochrobactrum intermedium with other bacterial strains ( Pseudomonas putida , Pseudomonas mendocina , Bacillus cereus , Bacillus marisflavi , Lysinibacillus fusiformis ) could effectively degrade hydrocarbons in crude oil fractions 500 N, lube oil, and de-asphalted oil. The hydrocarbons removal varied from 40%-60%, which showed an increase of approximately 50% in biodegradation compared to Ochrobactrum intermedium . The levels of the C21-C30, C31-C35, and C36-C60 decreased by 13%–54%, 49%–70%, and 39%–60%, respectively. To investigate the consumption of petroleum hydrocarbons by the growth of the two bacterial strains in different proportions under optimised conditions, the two strains were mixed and cultured in different combinations according to W01: W02 = 1:1/1:2/2:1. The results are presented in . shows that, compared to the 1:1 ratio, when the mixing ratios are 1:2 and 2:1, the combined strain has a higher degradation efficiency of TPH during growth, which is approximately 12.0% higher. However, the degradation efficiency of different mixing ratios can be maintained above 60%, which also indicates that the consumption of petroleum hydrocarbons by the mixed strains under the same growth conditions is stronger than that of W01 and W02 bacteria growing alone. Similar situations have also occurred in previous study . Teng et al. constructed a mixed cold-adapted bacteria flora, which contained Stenotrophomonas sp., to treat contaminated water (1 g oil/L). The mixed flora demonstrated 53.68% of TPH removal after 30 days of incubation under 10°C. This may be because the characteristic of oil pollution is the presence of saturated components composed of alkanes and more stubborn unsaturated components. In this case, a combination of multiple bacteria has a wider and different enzymatic ability, thereby improving the degradation level of pollution . Moreover, different strains of bacteria grow and reproduce in the same environment, competing for nutrients and substrates. At the same time, their own metabolites and other factors can have a comprehensive impact on microbial growth .
GC-MS analysis of metabolic products in the degradation of petroleum hydrocarbons Four intermediates, namely n-eicosane, 1-neneneba hexadecane alcohol, 2-octyldodecanol, and 2,4-di-tert-butylphenol, were detected using GC-MS during the degradation of petroleum hydrocarbons by strain W01 ( and – Figs). During petroleum hydrocarbon degradation by the strain W02, four intermediates were detected using GC-MS, namely diphenylamine, lauryl methacrylate, 2-Ethylhexanol, and ethyl hexyl benzoate ( and – Figs). Metabolomics analysis of differentially expressed genes The correlation of differentially expressed genes between the two samples was tested using the original number of reads. According to the correlation heat map of the samples in , the correlation between the samples is strong. The Pearson product-moment correlation coefficient between the two biological replicates of the sample was > 0.9, and the differential gene expression levels between the sample and blank groups tended to be the same. According to the heat map analysis, the biological repeatability of sequencing was good, and the data were reliable. and Figs showed that there are 397 differentially expressed genes between the strain W01 sample and the blank sample, of which 44 are significantly upregulated, such as N-Butyryl-L-homoserine lactone, Glycyl-leucine, and 73 are significantly downregulated, such as eryyhro-3-Hydroxy-Ls-aspartate, 2-Deoxyguanosine, Propazine, Dimethylglycine. There were 397 differentially expressed genes between the W02 and blank samples, of which 32 are significantly upregulated, such as Procollagen 5-hydroxy-L-lysine, Propazine, N6-Acetyl-N6-hydroxy-L-lysine, Betaine, Pantothenic acid, and 27 are significantly downregulated, such as Ureidopropionic acid, 4-Hydroxycinnamic acid, Pantothenic acid, (R)-3-Hydroxybutyric acid, Deoxyinosine. shows the differential expression heatmaps of the strain W01 and the blank control group. The treated samples showed the following substances: L-Alanyl-gamma-D-glutamyl-L lysine (C14), 1- (3,4-Dihydroxyphenyl) -5-hydroxy-3-decanone (C16), Reversrol (C14), and other components. Simultaneously, the differential expression heatmaps of the strain W02 and the blank control group were analysed. As shown in , the result also indicated the presence of the following substances: L-Alanyl-gamma-D-glutamyl-L-lysine (C14), Lenacil (C13), 1- (3,4-Dihydroxyphenyl) -5-hydroxy-3-decane1 (C16), Labelalol (C19), and other components, which were mutually confirmed from the intermediate products detected using GC-MS. According to Sajna et al. and Zhang et al. , benzoate and phenylacetic acid are the upstream and downstream products of catechol, respectively, and heptylcyclohexane is broken into alkanes by intermediate enzymes or bases. Preliminary derivation of degradation pathways As strains that can grow and reproduce using petroleum hydrocarbons as the sole carbon source, the mechanisms by which strains W01 and W02 degrade petroleum hydrocarbons have become important research targets, and multiple scholars have explored their degradation mechanisms . According to the GC-MS detection and metabolomic determination, n-eicosane was present in the wastewater before strain W01 degradation, whereas n-eicosane was not detected in the wastewater after degradation, and the two intermediate products, 1-nenenebc hexadecane alcohol and 2-octyldodecanol, were obtained. Diphenylamine and lauryl methacrylate were present in the wastewater before the strain W02 degradation, whereas diphenylamine and lauryl methacrylate were not detected in the wastewater after degradation, and 2-Ethylhexanol and ethyl hexyl benzoate were obtained. In the degradation of petroleum hydrocarbons by the strain W01, n-eicosane was first oxidized to 2-octyldodecanol, the hydroxyl group of 2-octyldodecanol was further oxidized to the aldehyde group, the aldehyde group was further oxidized to the ester group, and finally the carbon oxygen bond broke to produce 1-nenenebb hexadecane alcohol, and the C-O bond in 7,9-di-tert-butyl-1-oxa [4.5] deca-6,9-dien-2,8-dione cracked to carboxylate, and finally degraded to 2,4-di-tert-butylphenol ( ) [ – ]. In the degradation of petroleum hydrocarbons by the strain W02, diphenylamine may react with hydroxyl radicals to generate benzene; the C-C bond of lauryl methacrylate broke, and benzene generated ethyl hexyl benzoate, followed by the C-O bond breaking to generate 2-Ethylhexanol ( ) . However, owing to the complexity and variability of microbial degradation processes, and the considerable impact of environmental factors, further refinement of degradation pathways requires more in-depth research.
Four intermediates, namely n-eicosane, 1-neneneba hexadecane alcohol, 2-octyldodecanol, and 2,4-di-tert-butylphenol, were detected using GC-MS during the degradation of petroleum hydrocarbons by strain W01 ( and – Figs). During petroleum hydrocarbon degradation by the strain W02, four intermediates were detected using GC-MS, namely diphenylamine, lauryl methacrylate, 2-Ethylhexanol, and ethyl hexyl benzoate ( and – Figs).
The correlation of differentially expressed genes between the two samples was tested using the original number of reads. According to the correlation heat map of the samples in , the correlation between the samples is strong. The Pearson product-moment correlation coefficient between the two biological replicates of the sample was > 0.9, and the differential gene expression levels between the sample and blank groups tended to be the same. According to the heat map analysis, the biological repeatability of sequencing was good, and the data were reliable. and Figs showed that there are 397 differentially expressed genes between the strain W01 sample and the blank sample, of which 44 are significantly upregulated, such as N-Butyryl-L-homoserine lactone, Glycyl-leucine, and 73 are significantly downregulated, such as eryyhro-3-Hydroxy-Ls-aspartate, 2-Deoxyguanosine, Propazine, Dimethylglycine. There were 397 differentially expressed genes between the W02 and blank samples, of which 32 are significantly upregulated, such as Procollagen 5-hydroxy-L-lysine, Propazine, N6-Acetyl-N6-hydroxy-L-lysine, Betaine, Pantothenic acid, and 27 are significantly downregulated, such as Ureidopropionic acid, 4-Hydroxycinnamic acid, Pantothenic acid, (R)-3-Hydroxybutyric acid, Deoxyinosine. shows the differential expression heatmaps of the strain W01 and the blank control group. The treated samples showed the following substances: L-Alanyl-gamma-D-glutamyl-L lysine (C14), 1- (3,4-Dihydroxyphenyl) -5-hydroxy-3-decanone (C16), Reversrol (C14), and other components. Simultaneously, the differential expression heatmaps of the strain W02 and the blank control group were analysed. As shown in , the result also indicated the presence of the following substances: L-Alanyl-gamma-D-glutamyl-L-lysine (C14), Lenacil (C13), 1- (3,4-Dihydroxyphenyl) -5-hydroxy-3-decane1 (C16), Labelalol (C19), and other components, which were mutually confirmed from the intermediate products detected using GC-MS. According to Sajna et al. and Zhang et al. , benzoate and phenylacetic acid are the upstream and downstream products of catechol, respectively, and heptylcyclohexane is broken into alkanes by intermediate enzymes or bases.
As strains that can grow and reproduce using petroleum hydrocarbons as the sole carbon source, the mechanisms by which strains W01 and W02 degrade petroleum hydrocarbons have become important research targets, and multiple scholars have explored their degradation mechanisms . According to the GC-MS detection and metabolomic determination, n-eicosane was present in the wastewater before strain W01 degradation, whereas n-eicosane was not detected in the wastewater after degradation, and the two intermediate products, 1-nenenebc hexadecane alcohol and 2-octyldodecanol, were obtained. Diphenylamine and lauryl methacrylate were present in the wastewater before the strain W02 degradation, whereas diphenylamine and lauryl methacrylate were not detected in the wastewater after degradation, and 2-Ethylhexanol and ethyl hexyl benzoate were obtained. In the degradation of petroleum hydrocarbons by the strain W01, n-eicosane was first oxidized to 2-octyldodecanol, the hydroxyl group of 2-octyldodecanol was further oxidized to the aldehyde group, the aldehyde group was further oxidized to the ester group, and finally the carbon oxygen bond broke to produce 1-nenenebb hexadecane alcohol, and the C-O bond in 7,9-di-tert-butyl-1-oxa [4.5] deca-6,9-dien-2,8-dione cracked to carboxylate, and finally degraded to 2,4-di-tert-butylphenol ( ) [ – ]. In the degradation of petroleum hydrocarbons by the strain W02, diphenylamine may react with hydroxyl radicals to generate benzene; the C-C bond of lauryl methacrylate broke, and benzene generated ethyl hexyl benzoate, followed by the C-O bond breaking to generate 2-Ethylhexanol ( ) . However, owing to the complexity and variability of microbial degradation processes, and the considerable impact of environmental factors, further refinement of degradation pathways requires more in-depth research.
presents the predicted results for the intermediate product LD 50 detected in the W01 and W02 bacterial petroleum hydrocarbon-degrading wastewater systems. In the strain W01-degrading petroleum hydrocarbon wastewater system, the decrease in the content of n-eicosane indicates that it can be degraded by strain W01 into small molecule substances. Meanwhile, LD 50 of intermediate products such as 2-octyldodecanol and 2,4-di-tert-butylphenol, is significantly lower than that of n-eicosane. Besides, although the LD 50 value of 1-neneneba hexadecane alcohol is high, its content in the degraded wastewater is much lower than other substances. In the strain W02-degrading petroleum hydrocarbon wastewater system, the LD 50 values of the two newly generated substances, 2-Ethylhexanol and ethyl hexyl benzoate, inferred through degradation pathways significantly lower than that of lauryl methacrylate in the original petroleum hydrocarbon wastewater. This indicates a decreasing trend in the toxicity of the reaction system. Therefore, from the perspective of toxicity assessment, strains W01 and W02 have great potential for remediating petroleum hydrocarbon wastewater.
Two strains, W01 and W02, were cultured and identified as Stenotrophomonas acidaminiphila and Ochrobactrum , respectively. The degradation ability of the two strains to petroleum hydrocarbons was studied by changing the strain type, strain dosage, petroleum hydrocarbon concentration, mixed strains, and mixed ratio on the degradation effect of petroleum hydrocarbons. The results showed that the degradation efficiency of petroleum hydrocarbons by the mixed strains was higher than that when the strains grew alone. When the ratio was 1:2 or 2:1, the degradation efficiency of petroleum hydrocarbons by the mixed strains during growth was relatively high, reaching 80%. Intermediate products were identified through GC-MS and differential expression gene metabolomics analyses, and potential degradation pathways for petroleum hydrocarbon wastewater were proposed. In the degradation of petroleum hydrocarbons by the strain W01, n-eicosane was first oxidized to 2-octyldodecanol, the hydroxyl group of 2-octyldodecanol was further oxidized to the aldehyde group, the aldehyde group was further oxidized to the ester group, and finally the carbon oxygen bond broke to produce 1-nenenebb hexadecane alcohol, and the C-O bond in 7,9-di-tert-butyl-1-oxa [4.5] deca-6,9-dien-2,8-dione was cracked and then carboxylated, and finally degraded to 2,4-di-tert-butylphenol. In the degradation of petroleum hydrocarbons by the strain W02, diphenylamine may react with hydroxyl radicals to generate benzene; the C-C bond of lauryl methacrylate broke with benzene to generate ethyl hexyl benzoate, and the C-O bond broke to generate 2-Ethylhexanol. The toxicity assessment of the intermediate products showed a substantial decrease in the LD 50 of the final intermediate products, demonstrating the potential of W01 and W02 for degrading petroleum hydrocarbons.
S1 Fig Mass spectrum of n-eicosane. (DOCX) S2 Fig Mass spectrum of 1-neneneba hexadecane alcohol. (DOCX) S3 Fig Mass spectrum of 2-octyldodecanol. (DOCX) S4 Fig Mass spectrum of 2,4-ditert-butylphenol. (DOCX) S5 Fig Mass spectrum of diphenylamine. (DOCX) S6 Fig Mass spectrum of lauryl methacrylate. (DOCX) S7 Fig Mass spectrum of 2-Ethylhexanol. (DOCX) S8 Fig Mass spectrum of ethyl hexyl benzoate. (DOCX) S9 Fig Heatmap of the correlation of mRNA expression between the samples (a: strain W01, b: strain W02). (DOCX) S10 Fig Volcano plot of gene expression (a: strain W01, b: strain W02). (DOCX) S11 Fig Statistic of differently expressed metabolite. (DOCX) S1 Section Screening, isolation, and purification of bacterial strains. (DOCX) S2 Section The testing method of GC-MS. (DOCX) S3 Section The toxicity analysis method for intermediate products. (DOCX) S1 Data Datasets used in the research—Compressed files archive. (ZIP)
|
A qualitative study exploring strategies to improve the inter-professional management of diabetes and periodontitis | 5c7d14df-699f-4d54-9639-3a542ae6d8ba | 7059110 | Health Communication[mh] | Introduction Diabetes has been recognised as a risk factor for periodontitis (advanced gum disease) since the early 1990s, with the risk of periodontitis being increased 2–3 times in individuals with poorly controlled diabetes compared to individuals without . Periodontitis is a distressing chronic inflammatory disease of the gums and other supporting tissues of the teeth (including the alveolar jaw bone) which results in progressive tissue damage and ultimately tooth loss if untreated . Notwithstanding the effects of periodontitis on quality of life , it also impacts on a number of systemic conditions, including diabetes and cardiovascular disease . Severe periodontitis has been reported to be the sixth most prevalent disease globally and UK prevalence data have shown that 8% of the adult population have advanced periodontitis . The pathogenic mechanisms linking periodontitis and diabetes are incompletely understood but the level of glycaemic control is key in determining risk . Similar to the other complications of diabetes, the risk for periodontitis increases with poorer glycaemic control . Evidence has emerged to support a bidirectional relationship between diabetes and periodontitis; that is, diabetes increases risk for periodontitis, and periodontitis increases risk of diabetes complications and renders glycaemic control more difficult. Furthermore, there is compelling evidence that there is potential to improve glycaemic control through the treatment of periodontitis . Meta-analyses and Cochrane reviews have confirmed reductions in HbA1c of 3−4 mmol/mol (0.3–0.4%) following effective periodontal therapy up to 3–4 months after treatment . Over the last decade, guidance documents have been published by various professional and scientific organisations to improve inter-professional working in the context of diabetes and periodontitis, examples of which are summarised in . More recently (2018), the European Federation of Periodontology (EFP) and the International Diabetes Federation (IDF) held a joint workshop on diabetes and periodontitis. They published identical papers in both a dental journal ( Journal of Clinical Periodontology ) and a medical journal ( Diabetes Research & Clinical Practice ) aiming to improve inter-professional awareness of the links between the diseases . The publications included a suite of guidelines for dental and medical professionals, patients (whether being seen in the context of the medical practice or the dental practice), pharmacists, policymakers, and universities and research centres. The guidelines are all freely available to download from the EFP website . Key recommendations in these publications and others include that: - medical and dental healthcare professionals should inform patients of the bidirectional relationship between periodontitis and diabetes; - medical professionals should recommend that patients with diabetes visit a dental professional for assessment, and consider collaborating with the dental team; - dental professionals should consider liaising with the patient’s physician regarding their patient’s diabetes control (in the case of patients with known diabetes) or suspected diabetes (in the case of patients who do not currently have a diagnosis of diabetes). Mixed methods research exploring current practice and views of dental clinicians relating to guidance in the context of diabetes and periodontitis has shown that there is good uptake of informing patients about the bidirectional relationship, but contacting the patient’s doctor is not reported as happening to any great extent . A similar study exploring medical clinicians’ current practice and motivation relating to the guidance has shown that the evidence base and published guidance is not widely known and best practice recommendations are not being followed . Notwithstanding, this study found that the evidence for the bidirectional relationship between diabetes and periodontitis was valued by medical clinicians, and informing patients was considered legitimate by the medical team, particularly to the role of nurses . As difficulties with collaborative working between dental and medical clinicians have been reported previously in the literature , and dissemination of guidelines alone is insufficient in promoting a change in clinical practice , this study aimed to explore potential ways to enable improved inter-professional working as outlined in extant diabetes and periodontitis guidance documents.
Methods 2.1 Setting and study sample Qualitative research design with six iterative workshops each lasting between 30−60 min conducted with staff in two medical and two dental primary care practices in the North of England, and two workshops were held with people with diabetes at Newcastle University . Conducting the separate workshops for people with diabetes, the medical staff and dental practice staff was intended to create a comfortable (familiar) and uninhibited space for discussion, as suggested in multidisciplinary healthcare and social research . Recruitment of the medical and dental practice staff was facilitated by the North East and North Cumbria (NENC) Clinical Research Network (CRN), who distributed a study summary (inviting potential participants to email an expression of interest to the researcher) to research-active dental and medical practices in their region. Workshops took place at lunchtime in the practices, and participants were remunerated in accordance with the Department of Health and Social Care AcoRD guidance . The people with diabetes were recruited from a patient and public involvement (PPI) group at Newcastle University’s School of Dental Sciences and the workshops were held in a seminar room at the university. Travel expenses were refunded and the participants were given a gift card for their participation. All participants were provided with written and verbal information about the study prior to signing consent forms. The recruitment period ran from September 2017 until January 2018. A favourable ethical opinion was obtained from North West-Greater Manchester West National Health Service (NHS) Research Ethics Committee (16/NW/0030). 2.2 Workshop delivery At the beginning of the workshops, a summary of key components of the association between diabetes and periodontitis was given, together with results of the previous workshop to provide context for discussion. The workshops followed a topic guide, however the participants were encouraged to talk freely and the discussion was participant-led. The discussion concluded with a recap of the main discussion findings (delivered by the workshop facilitator, SMB), ensuring an accurate account of the discussion, which also allowed the participants an opportunity to reflect on their contributions and refine their comments should they wish to do so. 2.3 Analysis Consent was obtained from the participants to audio-record the workshops and reflective notes were made by the researcher (SMB). The audio recordings were transcribed, anonymised and subsequently checked for accuracy against the recording. Thematic analysis was used to identify common attributes within the data . Notable discussion points and specific comments of interest were identified from the transcripts and supporting reflective notes, and codes or key words were applied by the researcher (SMB), and subsequently discussed with the research team. Following completion of the sixth workshop, the transcripts were revisited and a process of re-reading (whilst listening to the audio recordings) enabled application of the constant comparison method to revise the codes . Emergent patterns and resultant themes were formulated via an inductive approach to the data analysis . Quotes which illustrated concepts relating to a particular theme were considered in detail and unpacked to explore meaning and develop better understanding. Analytical discussion during meetings of the research team provided the opportunity to further explore and clarify the emergent themes.
Setting and study sample Qualitative research design with six iterative workshops each lasting between 30−60 min conducted with staff in two medical and two dental primary care practices in the North of England, and two workshops were held with people with diabetes at Newcastle University . Conducting the separate workshops for people with diabetes, the medical staff and dental practice staff was intended to create a comfortable (familiar) and uninhibited space for discussion, as suggested in multidisciplinary healthcare and social research . Recruitment of the medical and dental practice staff was facilitated by the North East and North Cumbria (NENC) Clinical Research Network (CRN), who distributed a study summary (inviting potential participants to email an expression of interest to the researcher) to research-active dental and medical practices in their region. Workshops took place at lunchtime in the practices, and participants were remunerated in accordance with the Department of Health and Social Care AcoRD guidance . The people with diabetes were recruited from a patient and public involvement (PPI) group at Newcastle University’s School of Dental Sciences and the workshops were held in a seminar room at the university. Travel expenses were refunded and the participants were given a gift card for their participation. All participants were provided with written and verbal information about the study prior to signing consent forms. The recruitment period ran from September 2017 until January 2018. A favourable ethical opinion was obtained from North West-Greater Manchester West National Health Service (NHS) Research Ethics Committee (16/NW/0030).
Workshop delivery At the beginning of the workshops, a summary of key components of the association between diabetes and periodontitis was given, together with results of the previous workshop to provide context for discussion. The workshops followed a topic guide, however the participants were encouraged to talk freely and the discussion was participant-led. The discussion concluded with a recap of the main discussion findings (delivered by the workshop facilitator, SMB), ensuring an accurate account of the discussion, which also allowed the participants an opportunity to reflect on their contributions and refine their comments should they wish to do so.
Analysis Consent was obtained from the participants to audio-record the workshops and reflective notes were made by the researcher (SMB). The audio recordings were transcribed, anonymised and subsequently checked for accuracy against the recording. Thematic analysis was used to identify common attributes within the data . Notable discussion points and specific comments of interest were identified from the transcripts and supporting reflective notes, and codes or key words were applied by the researcher (SMB), and subsequently discussed with the research team. Following completion of the sixth workshop, the transcripts were revisited and a process of re-reading (whilst listening to the audio recordings) enabled application of the constant comparison method to revise the codes . Emergent patterns and resultant themes were formulated via an inductive approach to the data analysis . Quotes which illustrated concepts relating to a particular theme were considered in detail and unpacked to explore meaning and develop better understanding. Analytical discussion during meetings of the research team provided the opportunity to further explore and clarify the emergent themes.
Results 3.1 Participant characteristics Participant characteristics (n = 43 in total) are shown in . Two medical practices were recruited, one with a below-average percentage of patients with diabetes (4%) and one with an above-average percentage (16%). Participants were medical practice staff members with a range of job roles including general practitioner (GP), nurse, practice manager and administrator. Two dental practices were recruited, one located in an urban area whilst the other was in a semi-rural area. Participants were dental practice staff members from a range of job roles including dentist (GDP), dental hygienist/therapist (DHT) and dental nurse. Two-thirds of the participants in the workshops that recruited people with diabetes were female, two-thirds were retired and participant ages ranged from 22 to 75 years. 3.2 Themes Major themes and illustrative quotes are cross-referenced in the text to . The discussions were prompted by the topic guide and focused on two broad areas: accommodation of evidence and guidelines; and interaction, both experienced and planned. 3.3 Accommodation of evidence and guidelines During the workshops, participants engaged with the evidence and recommendations, and discussion focused on accommodating new knowledge into the context of their existing views and experience. Medical and dental professionals reported a lack of awareness of various aspects of the published guidance for management of patients with periodontitis and diabetes (Quote 1 (Q1), ). The medical teams had no knowledge of the guidelines or published evidence in the scientific literature but they were surprised to learn, and enthused by, the evidence regarding the effect of periodontitis treatment in improving glycaemic control. Whilst dental professionals were aware of the evidence, some of them were unfamiliar regarding the units used to measure glycaemic control (e.g. mmol/mol or % HbA1c values) and others were unsure of the reliability of patient self-report regarding diabetes control, both of which could reduce effective communication (Q2 and Q3, ). Both the medical and dental staff members expressed doubt in relation to each other’s knowledge on the topic, stating that they felt the lack of knowledge was mutual across both professions (medical and dental) (Q4 and Q5, ). Patient participants (diagnosed with diabetes) reported that they had never been informed about the links between diabetes and periodontitis by either their GP or GDP; and they felt the probability of receiving that information in the future was low due to the rushed nature of appointments. Furthermore, there was a suggestion that an oral health educational intervention delivered around the time of diabetes diagnosis, alongside information regarding the complications of diabetes, would ensure that all newly diagnosed patients were informed (Q6, ). 3.4 Experienced and planned interaction A key factor that led the medical practice participants to doubt the knowledge of dental professionals on the subject area was the absence of referrals from dental professionals in this context, as reported by the medical practice staff (Q7, ). Whilst it was clear that inter-professional communication did exist in other contexts between medicine and dentistry, albeit rarely, there were numerous accounts of negative experiences from both the medical and dental professionals. For example, some medical professionals reported that they were only contacted by dental professionals in relation to queries regarding a dose adjustment of anti-coagulant medication prior to certain dental procedures (such as tooth extractions); which they (the medical professionals) considered was inappropriate. They felt that dental professionals should seek relevant advice from a dental regulatory or advisory source (Q8, ). In addition, another concern among medical professionals arose when patients with toothache attended the medical practice on the advice of the dental practice (who may not have been able to schedule a timely appointment for the patient), suggesting that the GP would be able to prescribe antibiotics (Q9, ). Issuing antibiotics to patients with toothache was looked upon poorly by these medical practice staff as it was effectively asking the GP to work outside of their scope of practice. Dental professionals also noted poor inter-professional communication and stated that whilst they considered their enquiries to be legitimate, they felt they were often ignored or dismissed by their medical practice colleagues (Q10 and Q11, ). Patients with diabetes wanted their healthcare professionals to take time and explain the relationship between their diabetes and periodontitis, however they reported being rushed ‘in and out’ of consultations, particularly dental appointments. One patient had encountered an aggressive response from their GP in relation to a dental matter and therefore expressed doubt that communication could ever be improved (Q12, ). Medical professionals outlined that although they would not recommend direct communication via letters or phone calls due to operational time constraints, they would welcome an indirect referral via the patient (Q13, ). Furthermore, they reported that signposting a patient to the GP was common practice and used by a whole range of people, including hairdressers (Q14, ). Whilst it was noted that not all patients would act on signposting, dental professionals concurred that in the context of diabetes and periodontitis, signposting an individual with suspected diabetes to their GP for investigation was perceived and experienced to be acceptable (Q15 and Q16, ).
Participant characteristics Participant characteristics (n = 43 in total) are shown in . Two medical practices were recruited, one with a below-average percentage of patients with diabetes (4%) and one with an above-average percentage (16%). Participants were medical practice staff members with a range of job roles including general practitioner (GP), nurse, practice manager and administrator. Two dental practices were recruited, one located in an urban area whilst the other was in a semi-rural area. Participants were dental practice staff members from a range of job roles including dentist (GDP), dental hygienist/therapist (DHT) and dental nurse. Two-thirds of the participants in the workshops that recruited people with diabetes were female, two-thirds were retired and participant ages ranged from 22 to 75 years.
Themes Major themes and illustrative quotes are cross-referenced in the text to . The discussions were prompted by the topic guide and focused on two broad areas: accommodation of evidence and guidelines; and interaction, both experienced and planned.
Accommodation of evidence and guidelines During the workshops, participants engaged with the evidence and recommendations, and discussion focused on accommodating new knowledge into the context of their existing views and experience. Medical and dental professionals reported a lack of awareness of various aspects of the published guidance for management of patients with periodontitis and diabetes (Quote 1 (Q1), ). The medical teams had no knowledge of the guidelines or published evidence in the scientific literature but they were surprised to learn, and enthused by, the evidence regarding the effect of periodontitis treatment in improving glycaemic control. Whilst dental professionals were aware of the evidence, some of them were unfamiliar regarding the units used to measure glycaemic control (e.g. mmol/mol or % HbA1c values) and others were unsure of the reliability of patient self-report regarding diabetes control, both of which could reduce effective communication (Q2 and Q3, ). Both the medical and dental staff members expressed doubt in relation to each other’s knowledge on the topic, stating that they felt the lack of knowledge was mutual across both professions (medical and dental) (Q4 and Q5, ). Patient participants (diagnosed with diabetes) reported that they had never been informed about the links between diabetes and periodontitis by either their GP or GDP; and they felt the probability of receiving that information in the future was low due to the rushed nature of appointments. Furthermore, there was a suggestion that an oral health educational intervention delivered around the time of diabetes diagnosis, alongside information regarding the complications of diabetes, would ensure that all newly diagnosed patients were informed (Q6, ).
Experienced and planned interaction A key factor that led the medical practice participants to doubt the knowledge of dental professionals on the subject area was the absence of referrals from dental professionals in this context, as reported by the medical practice staff (Q7, ). Whilst it was clear that inter-professional communication did exist in other contexts between medicine and dentistry, albeit rarely, there were numerous accounts of negative experiences from both the medical and dental professionals. For example, some medical professionals reported that they were only contacted by dental professionals in relation to queries regarding a dose adjustment of anti-coagulant medication prior to certain dental procedures (such as tooth extractions); which they (the medical professionals) considered was inappropriate. They felt that dental professionals should seek relevant advice from a dental regulatory or advisory source (Q8, ). In addition, another concern among medical professionals arose when patients with toothache attended the medical practice on the advice of the dental practice (who may not have been able to schedule a timely appointment for the patient), suggesting that the GP would be able to prescribe antibiotics (Q9, ). Issuing antibiotics to patients with toothache was looked upon poorly by these medical practice staff as it was effectively asking the GP to work outside of their scope of practice. Dental professionals also noted poor inter-professional communication and stated that whilst they considered their enquiries to be legitimate, they felt they were often ignored or dismissed by their medical practice colleagues (Q10 and Q11, ). Patients with diabetes wanted their healthcare professionals to take time and explain the relationship between their diabetes and periodontitis, however they reported being rushed ‘in and out’ of consultations, particularly dental appointments. One patient had encountered an aggressive response from their GP in relation to a dental matter and therefore expressed doubt that communication could ever be improved (Q12, ). Medical professionals outlined that although they would not recommend direct communication via letters or phone calls due to operational time constraints, they would welcome an indirect referral via the patient (Q13, ). Furthermore, they reported that signposting a patient to the GP was common practice and used by a whole range of people, including hairdressers (Q14, ). Whilst it was noted that not all patients would act on signposting, dental professionals concurred that in the context of diabetes and periodontitis, signposting an individual with suspected diabetes to their GP for investigation was perceived and experienced to be acceptable (Q15 and Q16, ).
Discussion Inter-professional communication and collaborative working in the context of diabetes and periodontitis have been recommended in various best practice guidance publications over the last decade to improve patient care and diabetes outcomes. This study suggests that there may be currently little-to-no interaction between dental and medical clinicians in the context of diabetes and periodontitis, and there appears to be little appetite for improved (direct) communication by clinicians. Successful introduction and implementation of clinical guidelines have been shown to vary and whilst knowledge is important, previous studies looking at the uptake of diabetes recommendations have identified that contextual and motivational barriers can affect implementation . Behavioural change is complex and multifactorial. Whilst the workshops revealed a lack of knowledge regarding various elements of the guidelines, previous negative interactions such as inappropriate enquiries and stymied inter-professional communication were the focus of much discussion. Furthermore, a lack of referrals from dental professionals in the context of diabetes and periodontitis, caused medical teams to challenge dental professional’s ownership of the guidance. For their part, dental participants knew of the evidence, but described a history of non-replies or dismissive responses from GPs. Miscommunication between dental and medical clinicians has previously been reported elsewhere in the literature. A study of German GPs and GDPs showed problematic collaborative working with allegations of poor knowledge, uncertainty of role and previous difficult interactions . Holzinger et al. also reported GPs’ frustration relating to requests from GDPs for advice on anticoagulant therapy and dose adjustment for dental procedures, suggesting that this practice is widespread . Cope et al. found UK GPs to be equally frustrated when faced with prescribing antibiotics to treat dental pain (particularly given concerns regarding antimicrobial resistance) . Miscommunication between physicians and other allied health professionals in the context of diabetes has also been reported, with barriers relating to uncertainty of role and distrust of inter-professional working . Schweizer studied inter-professional collaboration and diabetes care in Switzerland and suggested that perception about collaboration is important and the negative experiences of communication are likely to influence team-working . The findings of the current study are consistent with the literature and suggest that despite the continued publication of international guidance documents advocating the benefits of inter-professional communication, the implementation of these recommendations offer a significant challenge for dental and medical clinicians and additional strategies are needed to change clinical practice. With time constraints being common in healthcare, minimal disruption has been reported to be important in the implementation of clinical behaviours . Active signposting has been shown to be effective in the context of reducing the amount of inappropriate GP consultations by triaging patients to a more appropriate healthcare professional ; and it is recommended in the UK NHS Year of Care initiative for managing long term conditions . During the workshops, previous occurrences of indirect referral were described as acceptable by dental professionals in the case of suspected diabetes; and medical professionals suggested that indirect referral from a variety of sources was not only commonplace, but actually preferred over a letter or a telephone call. Furthermore, NHS England have (August 2019) published a commissioning standard for the dental care of people with diabetes that recommends signposting. The document states that considerable NHS savings can be made by informing patients with diabetes (in the medical practice) about the links between periodontitis and diabetes and signposting them to a dentist for periodontal screening . In the event of patients with diabetes and periodontitis, particularly those with poorly controlled diabetes who do not have a dentist, engaging with and acting on information and signposting by their family physician (or nurse), they stand to benefit from periodontal treatment that has the potential to improve their glycaemic control. HbA1c reductions of up to 3−4 mmol/mol seen up to 3–4 months after periodontitis treatment could be a significant incentive for improved inter-professional communication. Whilst it is uncertain whether the implementation of the NHS commissioning standard will be more successful than that of previous diabetes and periodontitis guidance, further research is needed to explore whether signposting in the context of diabetes and periodontitis offers an effective substitute to direct referral and an important solution to the problems associated with inter-professional communication. 4.1 Strengths and weaknesses Six workshops enabled exploration of the perspectives of patients, medical and dental professionals to interprofessional communication in the context of diabetes and periodontitis. Workshops were considered appropriate methodology as they would enable learning, broad discussion and problem solving. However, the availability of interested healthcare professionals and patients made recruitment problematic. This lead to the decision to recruit at a practice level and through a PPI group. To facilitate recruitment further, the workshops were held either at lunchtime or as part of an established meeting, and refreshments and remuneration provided. Future research should aim to conduct an integrated workshop with people with diabetes and healthcare professionals together as it may stimulate alternative discussion on this topic.
Strengths and weaknesses Six workshops enabled exploration of the perspectives of patients, medical and dental professionals to interprofessional communication in the context of diabetes and periodontitis. Workshops were considered appropriate methodology as they would enable learning, broad discussion and problem solving. However, the availability of interested healthcare professionals and patients made recruitment problematic. This lead to the decision to recruit at a practice level and through a PPI group. To facilitate recruitment further, the workshops were held either at lunchtime or as part of an established meeting, and refreshments and remuneration provided. Future research should aim to conduct an integrated workshop with people with diabetes and healthcare professionals together as it may stimulate alternative discussion on this topic.
Conclusion Whilst inter-professional collaboration in the context of diabetes and periodontitis is a key recommendation that features in numerous published best practice guidance documents, it is clearly complex and challenging to implement. We consider that it is important for academics and specialists involved in guideline publication to consider the implementation of their recommendations as part of the process of developing guidance. Indirect referral, whereby the patient is signposted to a healthcare professional, was suggested by medical and dental professionals as a useful alternative to the traditional (and time consuming) letter or telephone call and has recently been recommended in an NHS commissioning standard outlining dental care for people with diabetes. Further research is necessary to evaluate signposting in this context to establish whether it is an effective, albeit indirect, communication tool.
No potential conflict of interest was reported by the authors.
This research was funded by a UK National Institute for Health Research Doctoral Research Fellowship (DRF-2014-07-023).
|
Borderline Personality Symptoms, Body Modification, and Emotional Regulation | 83e34f2d-c86a-4fc9-82a3-613a1078bc7d | 11764733 | Surgical Procedures, Operative[mh] | About 1.6% of the global population is estimated to meet the diagnostic criteria for borderline personality disorder (BPD) . The Diagnostic and Statistical Manual of Mental Disorders-5-TR (DSM-5-TR) defines BPD as “a pervasive pattern of instability of interpersonal relationships, self-image, and affects, and marked impulsivity, beginning by early adulthood and present in a variety of contexts” . Diagnostic criteria for BPD include desperate efforts to avoid abandonment, volatile interpersonal relationships, identity disturbance, impulsivity, suicidal ideation and behavior, mood instability, feelings of chronic emptiness, intense anger, paranoid ideation, and non-suicidal self-injury (NSSI) . In particular, BPD is a disorder characterized by emotional dysregulation, which may lead individuals to seek any release from overwhelming feelings, even if that release is maladaptive in the long term. Clinicians are advised to look for cutting scars and other evidence of NSSI when evaluating a patient with BPD . Treatments for BPD typically focus on distress tolerance, mindfulness, emotional regulation, and interpersonal skills . The phenomenon of NSSI involves intentional harm to oneself without suicidal motivations. While NSSI occurs in the general population and with other psychiatric disorders (estimated lifetime prevalence = 22.0%) , its prevalence is particularly elevated in BPD , with prevalence estimates of 65–80% in BPD samples . Furthermore, many patients report “cravings” to engage in NSSI, similar to cravings seen in addictions . The notable difference between cravings for substances and NSSI is that NSSI is craved primarily as a response to negative emotions rather than also as a means of eliciting positive emotions . Nonetheless, other research has supported addictive-like features of NSSI . In addition to emotional regulation, motives to engage in NSSI include self-derogation and interpersonal functions . To assess whether NSSI provided temporary relief from negative emotions for those with personality disorders such as BPD, Snir et al. prompted participants with BPD and avoidant personality disorder (APD) at five random times throughout the day to complete an experience-sampling digital diary, in which they described their affective, interpersonal, and behavioral (including NSSI) experiences at the current moment. With both BPD and APD patients, feelings of perceived rejection/isolation increased prior to engaging in NSSI and decreased afterward. Although participants with BPD and APD endorsed using NSSI for relief from unwanted emotions, participants with BPD and APD reported no significant change in general negative affect after engaging in NSSI . This may speak to the function of NSSI across various personality disorders. Furthermore, in individuals who had previously engaged in NSSI and people with BPD, physiological arousal decreased during and after listening to an audio clip on the topic of NSSI . Thus, people may engage in NSSI to cope with emotional dysregulation, perceiving that it aids in emotion regulation, whether it succeeds in doing so. It may be possible that some with BPD might elect to get body modifications (i.e., tattoos or piercings) for a similar emotional regulation purpose. Studies have investigated the link between symptoms of BPD and body modifications in clinical and non-clinical populations. When surveying adolescents in a socio-rehab center, individuals with a greater number of body modifications were more likely to engage in alcohol/drug use, aggressive behavior, and self-destructive behavior , similar to BPD symptoms of impulsivity and difficulty controlling anger . Wohlrab et al. listed broad categories that may explain motivations for obtaining body modifications in a non-clinical sample. These categories include beauty/art/fashion, individuality, personal narrative, physical endurance, group affiliations/commitment, resistance, spirituality/cultural tradition, addiction, sexual motivation, and no specific reason. Although not clearly stated, some of these categories may be broadly extrapolated to underlying motives for NSSI and may relate to BPD symptoms. For example, individuality may relate to the concept of self-expression. Since many people with BPD experience an unstable and unclear self-concept, permanently altering the body in a way that reflects one’s uniqueness may combat discomfort from an unclear self-concept. The personal narrative motive may also be considered a form of emotional regulation; ‘reclaiming’ a part of one’s body that was once violated may aid in sustaining a more positive emotional outlook. Those with BPD, however, may experience different motives for body modification than in non-clinical samples, such as emotional regulation. Emotional regulation and body modification were investigated in victims of childhood abuse, many of whom have BPD . Participants who had reported childhood abuse/neglect on a survey were more likely to report body modification, and participants who had endorsed more severe childhood abuse/neglect reported more piercings and tattoos than those with less severe abuse/neglect . Although Ernst et al. did not assess individual motivations for body modification, researchers suggested that coping with prior adversity and expressing autonomy may be a motivation for body modification. Similarly, physical endurance motivation refers to being able to withstand the pain of the body modification process , which may be used in a manner similar to NSSI. Lastly, addiction refers to the potential addictive component of getting new body modifications , which may relate to proneness to addictive behaviors seen among those with BPD or cravings for engaging in NSSI . Body modifications, NSSI, and emotional regulation may be uniquely interconnected in people with BPD. Blay et al. evaluated BPD patients’ symptoms, symptom severity, NSSI, and body modifications. Approximately 69.8% of BPD patients had at least one tattoo , compared to a general population rate of around 27.3% . Furthermore, the number of piercings an individual had was significantly positively correlated with their NSSI scores . Similarly, among participants recruited through Craigslist advertisements, there was a positive correlation between the total number of body modifications a participant had (piercing, tattooing, scarification, etc.) and their degree of BPD symptoms . Finally, some who engage in NSSI obtain tattoos to cover up scars or to prevent further NSSI , as they may not want to scar their tattoos. Body modifications may even be an explicit alternative to NSSI or a protective factor against NSSI. In a study by Claes et al. , patients with eating disorders were asked to complete a self-report questionnaire, including assessments of self-injury habits and tattoos and body piercings. A negative correlation was found between body modifications and self-injury, and patients with piercings exhibited less severe eating disorder symptoms . Ernst et al. and Claes et al. suggest that body modifications may serve as a socially acceptable form of altering the body for the sake of regulating aversive emotional states. Among participants subscribing to a tattooing and body modification magazine, those who reported cutting themselves had a greater number of piercings than those who did not cut themselves, were more likely to report being addicted to body modifications, and more frequently reported the expected physical pain as motivation for getting body modifications . However, this study did not evaluate whether there was a linear relationship between the number and surface area of body modifications and the amount or frequency of NSSI. Consistent with the proposal that body modifications may serve as a protective factor against NSSI , participants recruited from internet piercing communities, in which 50% of participants reported a history of NSSI, stated that NSSI decreased after obtaining piercings, and 25% reported that their self-harm ceased altogether . Thus, body modifications may replace or correlate with decreased self-injury over time. The concept of body modifications as a method for emotional regulation and an alternative to NSSI is exemplified in a case study by Anderson and Sansone . The case details a 19-year-old man with depression who used tattooing to regulate undesirable emotional states and to ward off thoughts of self-harm. He reported that the physical pain of tattooing provided a temporary distraction from emotional pain that would ‘take [his] mind off it.’ The young man asserted that scars from cutting would have been considered embarrassing, whereas his tattoos were considered “cool”. He also explained that more intense emotional pain would lead him to choose a more sensitive area for tattoo placement . Thus, it is clear that emotional regulation, NSSI, and body modifications relate; however, it remains unclear whether body modifications are being used as a more adaptive or socially acceptable emotional regulation strategy for those who would otherwise engage in NSSI. The current study aimed to extend the work of Blay et al. . Rather than recruiting individuals diagnosed with BPD, the current study investigated BPD symptom severity in the general population. Specifically, we hypothesized that (H1) NSSI craving would be significantly associated with body modifications craving scores among participants who had body modifications; (H2) body modifications (e.g., number of tattoos/piercings, percentage of tattoo body coverage, locations of tattoos/piercings) would account for significant variance in BPD symptom scores; and (H3) self-expression and emotional regulation motivations for body modifications would account for significant variance in BPD symptom scores.
2.1. Participants Participants ( N = 199) were recruited through the researchers’ departmental subject pool and flyers distributed across college campuses, posted on social media, and posted in local tattoo parlors and hair salons in a mid-sized metropolitan area in the Northeast U.S. to survey a wide variety of individuals. This recruitment strategy was selected to maximize variability within the sample in terms of age and presence/absence of body modifications, which should improve the generalizability of the results. We were specifically targeting individuals who would have higher numbers of body modifications by recruiting from tattoo and piercing parlors in person and via their social media platforms and recruiting in hair salons to balance the sample with individuals who may be less likely to have body modifications. Individuals 18 and older were eligible to participate in the study and were able to enter a raffle for one of three cash prizes (USD 200, USD 100, and USD 50) following participation. The sample size was determined by the time frame available for completion of this study by the internal funding mechanism, attempting to recruit as many participants as possible within that time frame for adequate statistical power. Responses were collected from October 2023 to January 2024. Participants’ M ( SD ) AGE was 26.87 (10.74) years old, ranging from 18–66. Participants’ demographic information is presented in (due to researcher error, demographics were not collected for the first 95 participants. Demographic information is presented for the remaining 104 participants). 2.2. Measures 2.2.1. Borderline Symptom List The Borderline Symptom List (BSL) is a 23-item, Likert-type scale with response options ranging from 0 (not at all) to 4 (very strong). Responses are averaged together, with higher scores reflecting greater BPD symptomatology. Examples of scale items include “My mood rapidly cycled in terms of anxiety, anger, and depression”, “Criticism had a devastating effect on me”, and “I felt disgusted by myself”. The authors reported a Cronbach’s α of 0.97 and strong evidence of convergent and discriminant validity. The Cronbach’s α of the BSL for the current study was 0.95. 2.2.2. Difficulties in Emotion Regulation Scale The Difficulties in Emotion Regulation Scale (DERS) is a 36-item, Likert-type scale ranging from 1 (almost never) to 5 (almost always). Responses are averaged after reverse scoring select items, and higher scores indicate increased difficulty with emotional regulation. The scale is comprised of 6 factors, labeled Nonacceptance of Emotional Responses, Difficulties Engaging in Goal-Directed Behavior, Impulse Control Difficulties, Lack of Emotional Awareness, Limited Access to Emotion Regulation Strategies, and Lack of Emotional Clarity. Questions include: “I experience my emotions as overwhelming and out of control”, “When I’m upset, I become angry with myself for feeling that way”, and “When I’m upset, I have difficulty controlling my behaviors”. The authors reported a Cronbach’s α of 0.93, and construct validity was supported in the validation sample. For the current study, the DERS Cronbach’s α was 0.95 for the total scale. Subscale Cronbach’s α’s were as follows: Nonacceptance of Emotional Responses α = 94, Difficulties Engaging in Goal-Directed Behavior α = 0.89, Impulse Control Difficulties α = 0.90, Lack of Emotional Awareness α = 0.63, Limited Access to Emotion Regulation Strategies α = 0.90, and Lack of Emotional Clarity α = 0.84. 2.2.3. Suicidal Behavior and Body Modification The Suicidal Behavior and Body Damage and Modifications Scale (SBBDM-S) evaluates the lifetime presence of suicidal behavior, NSSI, and body modifications. The scale consists of 10 items, with 1–4 investigating suicidal behavior, 5–8 investigating body modifications, and 9–10 investigating NSSI. Within the body modifications portion, three sub-scores indicated the total number of piercings, the percentage of tattoo body coverage, and the total body modifications score (sum of the previous two subscales). The SBBDM-S was adapted from its original form for the purpose of the current study. This adapted version consists only of the body modifications portion. Some additional questions were added to investigate the number of body modifications an individual has, their locations, and how long it has been since obtaining their most recent body modifications. 2.2.4. Craving Experience Questionnaire The Craving Experience Questionnaire (CEQ) is a 10-item scale originally designed to assess the urgency of cravings in smokers and nicotine users, with response options on a spectrum from 0 (not at all) to 10 (extremely). Example items include “At that time, how strong was the urge to have …?” and “At that time, how vividly did you imagine how your body would feel?” and “At that time, how intrusive were the thoughts?” Crobach’s α was 0.91 for the strength form of the scale and 0.94 for the frequency form of the scale in the validation sample. The CEQ’s validity was supported among smokers and in a modified version assessing alcohol cravings. The current study adapted this scale to investigate cravings for NSSI and body modifications, removing two items asking about sensory stimuli associated with nicotine use. Cronbach’s α for the CEQ in the current study was 0.95 for the NSSI adaptation and 0.90 for the body modifications adaptation. 2.2.5. Motivations for Body Modification Scale (MBMS) Currently, no measures exist that assess a wide range of motivations for obtaining body modifications, especially pertaining to emotional regulation and self-expression. In a literature review, Wohlrab et al. reported important motivations for body modifications such as beauty/art/fashion, individuality, personal narrative, physical endurance against pain, group affiliations/commitment, social resistance/protest, spirituality/cultural tradition, addiction to body modifications, sexual motivation, and no specific reason. Thus, we developed the MBMS to include items assessing these categories, along with emotional regulation and self-expression motivations for obtaining body modifications. The original MBMS consisted of 17 questions on a Likert-type scale ranging from 1 (strongly disagree) to 5 (strongly agree). Information regarding the development and initial validity of the scale and its factor structure are provided in . The final MBMS included 15 total items, comprised of three factors assessing Emotion Regulation (α = 0.89), Expression/Autonomy (α = 0.81), and Social Identity (α = 0.63). 2.3. Procedure This study was preregistered at https://osf.io/x5wkb/?view_only=f1dae1f78c3a4bf4b2d158fb60d12d1a (accessed on 9 December 2024), and responses were collected via Qualtrics. After consenting to the survey, participants completed questionnaires consisting of the BSL and the modified CEQ-NSSI. Participants were then asked if they had any body modifications (tattoos and piercings excluding first earlobe piercings). Participants who endorsed body modifications were asked to complete the body modification portion of the adapted SBBDM-S, the body modifications adaptation of the CEQ, and the MBMS (designed for this study). After completing the study, all participants were provided with a debriefing statement that further detailed the purpose of the study and contained a link to enter the raffle. 2.4. Statistical Analyses We planned two linear regression analyses to evaluate (1) if the number of tattoos and piercings participants reported and tattoo coverage (predictor variables, entered simultaneously) accounted for variance in BSL scores (outcome variable); (2) if emotional regulation and self-expression (MBMS Emotion Regulation and Expression/Autonomy subscale scores, predictor variables entered simultaneously) accounted for variance in BSL scores (outcome variable); and (3) if NSSI craving scores significantly positively correlated with body modification craving scores. Correlations and descriptive statistics are also reported. Data and analyses were conducted using SPSS software version 28. Using G*Power version 3.1.9.5, sensitivity analysis showed that for a sample of 199, with a power of 0.90, with three tested predictors of seven total possible predictors, the analysis would have the power to detect an effect size of f 2 = 0.073, which represents a small-to-medium effect size, suggesting sufficient sample size for the planned regression analyses in the current study.
Participants ( N = 199) were recruited through the researchers’ departmental subject pool and flyers distributed across college campuses, posted on social media, and posted in local tattoo parlors and hair salons in a mid-sized metropolitan area in the Northeast U.S. to survey a wide variety of individuals. This recruitment strategy was selected to maximize variability within the sample in terms of age and presence/absence of body modifications, which should improve the generalizability of the results. We were specifically targeting individuals who would have higher numbers of body modifications by recruiting from tattoo and piercing parlors in person and via their social media platforms and recruiting in hair salons to balance the sample with individuals who may be less likely to have body modifications. Individuals 18 and older were eligible to participate in the study and were able to enter a raffle for one of three cash prizes (USD 200, USD 100, and USD 50) following participation. The sample size was determined by the time frame available for completion of this study by the internal funding mechanism, attempting to recruit as many participants as possible within that time frame for adequate statistical power. Responses were collected from October 2023 to January 2024. Participants’ M ( SD ) AGE was 26.87 (10.74) years old, ranging from 18–66. Participants’ demographic information is presented in (due to researcher error, demographics were not collected for the first 95 participants. Demographic information is presented for the remaining 104 participants).
2.2.1. Borderline Symptom List The Borderline Symptom List (BSL) is a 23-item, Likert-type scale with response options ranging from 0 (not at all) to 4 (very strong). Responses are averaged together, with higher scores reflecting greater BPD symptomatology. Examples of scale items include “My mood rapidly cycled in terms of anxiety, anger, and depression”, “Criticism had a devastating effect on me”, and “I felt disgusted by myself”. The authors reported a Cronbach’s α of 0.97 and strong evidence of convergent and discriminant validity. The Cronbach’s α of the BSL for the current study was 0.95. 2.2.2. Difficulties in Emotion Regulation Scale The Difficulties in Emotion Regulation Scale (DERS) is a 36-item, Likert-type scale ranging from 1 (almost never) to 5 (almost always). Responses are averaged after reverse scoring select items, and higher scores indicate increased difficulty with emotional regulation. The scale is comprised of 6 factors, labeled Nonacceptance of Emotional Responses, Difficulties Engaging in Goal-Directed Behavior, Impulse Control Difficulties, Lack of Emotional Awareness, Limited Access to Emotion Regulation Strategies, and Lack of Emotional Clarity. Questions include: “I experience my emotions as overwhelming and out of control”, “When I’m upset, I become angry with myself for feeling that way”, and “When I’m upset, I have difficulty controlling my behaviors”. The authors reported a Cronbach’s α of 0.93, and construct validity was supported in the validation sample. For the current study, the DERS Cronbach’s α was 0.95 for the total scale. Subscale Cronbach’s α’s were as follows: Nonacceptance of Emotional Responses α = 94, Difficulties Engaging in Goal-Directed Behavior α = 0.89, Impulse Control Difficulties α = 0.90, Lack of Emotional Awareness α = 0.63, Limited Access to Emotion Regulation Strategies α = 0.90, and Lack of Emotional Clarity α = 0.84. 2.2.3. Suicidal Behavior and Body Modification The Suicidal Behavior and Body Damage and Modifications Scale (SBBDM-S) evaluates the lifetime presence of suicidal behavior, NSSI, and body modifications. The scale consists of 10 items, with 1–4 investigating suicidal behavior, 5–8 investigating body modifications, and 9–10 investigating NSSI. Within the body modifications portion, three sub-scores indicated the total number of piercings, the percentage of tattoo body coverage, and the total body modifications score (sum of the previous two subscales). The SBBDM-S was adapted from its original form for the purpose of the current study. This adapted version consists only of the body modifications portion. Some additional questions were added to investigate the number of body modifications an individual has, their locations, and how long it has been since obtaining their most recent body modifications. 2.2.4. Craving Experience Questionnaire The Craving Experience Questionnaire (CEQ) is a 10-item scale originally designed to assess the urgency of cravings in smokers and nicotine users, with response options on a spectrum from 0 (not at all) to 10 (extremely). Example items include “At that time, how strong was the urge to have …?” and “At that time, how vividly did you imagine how your body would feel?” and “At that time, how intrusive were the thoughts?” Crobach’s α was 0.91 for the strength form of the scale and 0.94 for the frequency form of the scale in the validation sample. The CEQ’s validity was supported among smokers and in a modified version assessing alcohol cravings. The current study adapted this scale to investigate cravings for NSSI and body modifications, removing two items asking about sensory stimuli associated with nicotine use. Cronbach’s α for the CEQ in the current study was 0.95 for the NSSI adaptation and 0.90 for the body modifications adaptation. 2.2.5. Motivations for Body Modification Scale (MBMS) Currently, no measures exist that assess a wide range of motivations for obtaining body modifications, especially pertaining to emotional regulation and self-expression. In a literature review, Wohlrab et al. reported important motivations for body modifications such as beauty/art/fashion, individuality, personal narrative, physical endurance against pain, group affiliations/commitment, social resistance/protest, spirituality/cultural tradition, addiction to body modifications, sexual motivation, and no specific reason. Thus, we developed the MBMS to include items assessing these categories, along with emotional regulation and self-expression motivations for obtaining body modifications. The original MBMS consisted of 17 questions on a Likert-type scale ranging from 1 (strongly disagree) to 5 (strongly agree). Information regarding the development and initial validity of the scale and its factor structure are provided in . The final MBMS included 15 total items, comprised of three factors assessing Emotion Regulation (α = 0.89), Expression/Autonomy (α = 0.81), and Social Identity (α = 0.63).
The Borderline Symptom List (BSL) is a 23-item, Likert-type scale with response options ranging from 0 (not at all) to 4 (very strong). Responses are averaged together, with higher scores reflecting greater BPD symptomatology. Examples of scale items include “My mood rapidly cycled in terms of anxiety, anger, and depression”, “Criticism had a devastating effect on me”, and “I felt disgusted by myself”. The authors reported a Cronbach’s α of 0.97 and strong evidence of convergent and discriminant validity. The Cronbach’s α of the BSL for the current study was 0.95.
The Difficulties in Emotion Regulation Scale (DERS) is a 36-item, Likert-type scale ranging from 1 (almost never) to 5 (almost always). Responses are averaged after reverse scoring select items, and higher scores indicate increased difficulty with emotional regulation. The scale is comprised of 6 factors, labeled Nonacceptance of Emotional Responses, Difficulties Engaging in Goal-Directed Behavior, Impulse Control Difficulties, Lack of Emotional Awareness, Limited Access to Emotion Regulation Strategies, and Lack of Emotional Clarity. Questions include: “I experience my emotions as overwhelming and out of control”, “When I’m upset, I become angry with myself for feeling that way”, and “When I’m upset, I have difficulty controlling my behaviors”. The authors reported a Cronbach’s α of 0.93, and construct validity was supported in the validation sample. For the current study, the DERS Cronbach’s α was 0.95 for the total scale. Subscale Cronbach’s α’s were as follows: Nonacceptance of Emotional Responses α = 94, Difficulties Engaging in Goal-Directed Behavior α = 0.89, Impulse Control Difficulties α = 0.90, Lack of Emotional Awareness α = 0.63, Limited Access to Emotion Regulation Strategies α = 0.90, and Lack of Emotional Clarity α = 0.84.
The Suicidal Behavior and Body Damage and Modifications Scale (SBBDM-S) evaluates the lifetime presence of suicidal behavior, NSSI, and body modifications. The scale consists of 10 items, with 1–4 investigating suicidal behavior, 5–8 investigating body modifications, and 9–10 investigating NSSI. Within the body modifications portion, three sub-scores indicated the total number of piercings, the percentage of tattoo body coverage, and the total body modifications score (sum of the previous two subscales). The SBBDM-S was adapted from its original form for the purpose of the current study. This adapted version consists only of the body modifications portion. Some additional questions were added to investigate the number of body modifications an individual has, their locations, and how long it has been since obtaining their most recent body modifications.
The Craving Experience Questionnaire (CEQ) is a 10-item scale originally designed to assess the urgency of cravings in smokers and nicotine users, with response options on a spectrum from 0 (not at all) to 10 (extremely). Example items include “At that time, how strong was the urge to have …?” and “At that time, how vividly did you imagine how your body would feel?” and “At that time, how intrusive were the thoughts?” Crobach’s α was 0.91 for the strength form of the scale and 0.94 for the frequency form of the scale in the validation sample. The CEQ’s validity was supported among smokers and in a modified version assessing alcohol cravings. The current study adapted this scale to investigate cravings for NSSI and body modifications, removing two items asking about sensory stimuli associated with nicotine use. Cronbach’s α for the CEQ in the current study was 0.95 for the NSSI adaptation and 0.90 for the body modifications adaptation.
Currently, no measures exist that assess a wide range of motivations for obtaining body modifications, especially pertaining to emotional regulation and self-expression. In a literature review, Wohlrab et al. reported important motivations for body modifications such as beauty/art/fashion, individuality, personal narrative, physical endurance against pain, group affiliations/commitment, social resistance/protest, spirituality/cultural tradition, addiction to body modifications, sexual motivation, and no specific reason. Thus, we developed the MBMS to include items assessing these categories, along with emotional regulation and self-expression motivations for obtaining body modifications. The original MBMS consisted of 17 questions on a Likert-type scale ranging from 1 (strongly disagree) to 5 (strongly agree). Information regarding the development and initial validity of the scale and its factor structure are provided in . The final MBMS included 15 total items, comprised of three factors assessing Emotion Regulation (α = 0.89), Expression/Autonomy (α = 0.81), and Social Identity (α = 0.63).
This study was preregistered at https://osf.io/x5wkb/?view_only=f1dae1f78c3a4bf4b2d158fb60d12d1a (accessed on 9 December 2024), and responses were collected via Qualtrics. After consenting to the survey, participants completed questionnaires consisting of the BSL and the modified CEQ-NSSI. Participants were then asked if they had any body modifications (tattoos and piercings excluding first earlobe piercings). Participants who endorsed body modifications were asked to complete the body modification portion of the adapted SBBDM-S, the body modifications adaptation of the CEQ, and the MBMS (designed for this study). After completing the study, all participants were provided with a debriefing statement that further detailed the purpose of the study and contained a link to enter the raffle.
We planned two linear regression analyses to evaluate (1) if the number of tattoos and piercings participants reported and tattoo coverage (predictor variables, entered simultaneously) accounted for variance in BSL scores (outcome variable); (2) if emotional regulation and self-expression (MBMS Emotion Regulation and Expression/Autonomy subscale scores, predictor variables entered simultaneously) accounted for variance in BSL scores (outcome variable); and (3) if NSSI craving scores significantly positively correlated with body modification craving scores. Correlations and descriptive statistics are also reported. Data and analyses were conducted using SPSS software version 28. Using G*Power version 3.1.9.5, sensitivity analysis showed that for a sample of 199, with a power of 0.90, with three tested predictors of seven total possible predictors, the analysis would have the power to detect an effect size of f 2 = 0.073, which represents a small-to-medium effect size, suggesting sufficient sample size for the planned regression analyses in the current study.
Under half of our sample (40.5%) reported having tattoos. Of those who reported tattoos, a majority (64.2%) reported four or fewer tattoos. Over half of our sample (55%) did not have piercings. Of those who have piercings, half (50%) of the sample reported having less than seven piercings. Thus, our sample had more tattoos than a large Pew research sample of American participants (15–24%) and more piercings than a national probability sample of 500 adults aged 18–50 (14%) . Descriptive information and inter-correlations are provided in . 3.1. Body Modification and NSSI Craving: H1 Supporting H1, NSSI craving was strongly positively correlated with body modification craving among participants who had body modifications ( r = 0.56, p < 0.001). See . 3.2. Extent of Body Modifications: H2 The model was statistically significant: the number and extent of tattoos and piercings explained a significant proportion of variance in BSL scores, F (3, 59) = 4.04, p = 0.011, R 2 = 0.17. In contrast to hypotheses, the number of tattoos an individual had, along with the approximate percentage of tattoo body coverage, did not significantly account for the variance in BSL scores ( p = 0.32 and p = 0.80, respectively). Supporting hypotheses, BSL score variance was significantly accounted for by the number of piercings an individual had ( p = 0.001), such that each additional piercing was associated with a 0.10 increase in BSL score. See for results. 3.3. Motivations for Body Modifications: H3 The model was not statistically significant, F (3, 129) = 1.36, p = 0.26, R 2 = 0.03. Emotion Regulation, Expression/Autonomy, and Social Identity scores were not significantly associated with BSL scores ( p = 0.16–0.72), contrary to hypotheses. There was some evidence of heteroscedasticity; however, transformations of the variables yielded similar non-significant results. Thus, untransformed data are reported for ease of interpretation. See for results.
Supporting H1, NSSI craving was strongly positively correlated with body modification craving among participants who had body modifications ( r = 0.56, p < 0.001). See .
The model was statistically significant: the number and extent of tattoos and piercings explained a significant proportion of variance in BSL scores, F (3, 59) = 4.04, p = 0.011, R 2 = 0.17. In contrast to hypotheses, the number of tattoos an individual had, along with the approximate percentage of tattoo body coverage, did not significantly account for the variance in BSL scores ( p = 0.32 and p = 0.80, respectively). Supporting hypotheses, BSL score variance was significantly accounted for by the number of piercings an individual had ( p = 0.001), such that each additional piercing was associated with a 0.10 increase in BSL score. See for results.
The model was not statistically significant, F (3, 129) = 1.36, p = 0.26, R 2 = 0.03. Emotion Regulation, Expression/Autonomy, and Social Identity scores were not significantly associated with BSL scores ( p = 0.16–0.72), contrary to hypotheses. There was some evidence of heteroscedasticity; however, transformations of the variables yielded similar non-significant results. Thus, untransformed data are reported for ease of interpretation. See for results.
The purpose of this study was to examine the relationship between BPD symptomatology, body modification, and motivations to acquire additional body modifications. This included assessing whether BPD symptoms were associated with body modifications, whether body modifications are motivated by emotional regulation and self-expression, and whether cravings to engage in NSSI were correlated with cravings to obtain additional body modifications. Blay et al. investigated emotional regulation as a motivator for body modifications in a clinical population with BPD, finding that NSSI correlated with the total number of body modifications obtained by this population. The current study extends these results by looking at a non-clinical population, investigating self-expression motivations for body modifications, and examining the relationship between cravings for NSSI and body modification. We hypothesized that the extent of body modifications an individual had would be associated with their level of BPD symptomatology, such that an increased amount of body modification would indicate greater BPD psychopathology. This had been found previously . Contrary to our initial hypotheses, neither the number of tattoos nor the percentage of body coverage of tattoos was associated with BPD symptoms. However, the number of piercings an individual had did significantly explain variance in BPD symptomatology, as hypothesized. These results are consistent with findings from previous studies, in which individuals who reported engaging in NSSI, often related to BPD, had a greater number of piercings . However, the lack of correlation between BPD features and tattoos is partly inconsistent with Blay et al. . This may be due to our non-clinical sample, as Blay et al. collected data directly from an outpatient sample with BPD . Blay and colleagues found significant associations between body modifications and total BPD scores, as measured by the Structured Clinical Interview for the DSM-5, but a non-significant association was found between body modification and BSL scores , which was the measure used in the current study. Thus, it is possible that the BSL is less sensitive to BPD symptoms that are associated with body modification than the Structured Clinical Interview for the DSM-5. Importantly, Blay et al. found that the number of piercings was significantly associated with emotional dysregulation, whereas the percentage of the body covered by tattoos was significantly associated with sensation-seeking . Similarly, recent research among 762 non-clinical adults found a stronger relationship between piercings and symptoms of attention-deficit hyperactivity disorder than between tattoos and attention-deficit hyperactivity disorder symptoms, though both were statistically significant . It is possible that having tattoos may be regarded as more socially acceptable than having piercings, especially when concerning more ‘extreme’ piercings . This is supported by the rise in tattoo popularity, with an online Ipsos poll reporting a rise in tattoos from 21% of the population in 2012 to 30% in 2019. Minimal research has been conducted on societal acceptance of individuals with multiple tattoos versus multiple piercings. However, based on the rise in tattoo acceptance broadly, individuals with and without BPD symptoms may be equally likely to get tattoos due to increased social acceptance, such that they may be less predictive of BPD psychopathology. We hypothesized that participants with body modifications would be more likely to report emotional regulation and self-expression as motivations for obtaining their body modifications if they were higher in BPD pathology. In contrast to the hypotheses, neither emotional regulation nor self-expression motivations were significantly associated with BPD pathology. Emotional dysregulation is heightened in individuals who engage in NSSI, and NSSI appears to serve as an emotional regulator . If NSSI and body modifications are interconnected among individuals high in BPD symptomatology, then emotional regulation may serve as a motivational factor for obtaining body modifications, similar to its motivating NSSI. Although emotional regulation motivations for body modifications were not significantly associated with BPD psychopathology, as hypothesized, they were significantly positively correlated with NSSI cravings ( r = 0.24, p < 0.001), body modification craving ( r = 0.49, p < 0.001), and the percent of body coverage by tattoos ( r = 0.29, p < 0.001), the number of tattoos ( r = 0.30, p < 0.001), and the number of piercings ( r = 0.24, p < 0.001), strongly suggesting the interrelationships of these motivations for body modification and both craving for NSSI and further body modifications (see ). Additionally, although self-expression/autonomy motivations for body modifications were not significantly associated with BPD symptoms, as hypothesized, they were significantly positively correlated with body modification cravings ( r = 0.5, p < 0.001), and the number of tattoos ( r = 0.23, p < 0.05) and piercings ( r = 0.21, p < 0.05). As hypothesized, NSSI craving scores were significantly associated with body modification craving scores. Importantly, both NSSI and body modification cravings were significantly correlated with almost all DERS subscales as well, even though the DERS was not associated with the number and percentage coverage of body modifications. Given that emotional regulation motivations for body modifications were significantly associated with NSSI craving scores, people may use body modifications as an alternative form of emotional regulation that is more socially acceptable than engaging in self-harming behaviors. This is supported by case studies, including the young man who reported using the pain of tattooing as a distraction from NSSI cravings and emotional fluctuations . Similarly, a veteran who met the criteria for multiple psychiatric disorders, including BPD, disclosed that he had engaged in typically observed NSSI behaviors but preferred the pain of tattooing due to its social acceptance. He reported that the tattooing process combatted negative emotional states, tension, and feelings of numbness . The connection between NSSI and body modification may, therefore, be driven by emotional regulation for individuals with BPD symptoms, with body modifications being endorsed by some individuals due to their greater social acceptance over other NSSI methods. However, some people with BPD may not view body modification as a means of regulating their emotions, but emotional regulation may be a side effect of body modification. This is likely due to the endorphin release from self-injury: Directly after engaging in NSSI, individuals demonstrated a significant increase in salivary β-endorphin levels . These endorphin releases have not been explicitly studied with regard to body modifications, although Wohlrab et al. and Weiler et al. interpolate that this may be a motivation for body modifications among some individuals. It is likely that many individuals without significant psychopathology are motivated to obtain additional body modifications as a means of self-expression and autonomy and for social identity reasons, as our scale suggested. Additional research is needed to better understand under what circumstances and for whom differing motivations may guide decisions to obtain body modifications. 4.1. Limitations Despite the important findings in the current study, there are some notable limitations. First, the study was cross-sectional in nature; our data is therefore only correlational and cannot be used to infer causality. Second, there is also a potential for self-report and selection bias despite efforts to recruit a diverse sample. Third, demographic information was missing for half of the participants due to researcher error. This diminishes the ability to draw conclusions regarding differences between demographic groups, rendering the generalizability of the current findings unknown. Fourth, although the study was advertised in multiple locations to attract a range of respondents, many participants were younger adults, based on the available demographic information, who may have only recently become old enough to legally obtain body modifications without parental consent. Additionally, we did not question participants regarding cultural–historical motivations for body modifications. Two recent reviews of motivations for body modifications do speak in greater detail about cultural, historical, and social motivations for body modifications . Furthermore, we did not account for recruitment sources in our survey and are therefore unable to provide a breakdown of how many participants found the study from various recruitment sources and are therefore unable to with confidence confirm heterogeneity in our sample beyond the existing demographic information. Additionally, the emotional regulation and self-expression motivations for body modification scales were developed solely for this study and were not previously validated measures. The Social Identity subscale was brief, consisting of only three items, and internal consistency was inadequate. We only measured NSSI craving using a scale initially designed to measure substance use craving, though some evidence suggests these cravings differ from one another . While the BSL scale had many questions on thoughts of NSSI, there were no subscales examining the incidence of NSSI itself. These omissions, while limitations, were in pursuit of survey brevity, as shorter surveys have been shown to increase survey completion . Finally, we did not collect longitudinal data on the progression of body modifications along with BPD symptoms and NSSI and body modification craving over time. 4.2. Future Directions Future researchers should survey a larger group of participants with more body modifications so that participants have a wide range of body modification coverage for analyses. Additionally, researchers can determine if age is a factor in the relationship between BPD symptoms and body modification. Body modifications may be a more modern form of coping or self-expression endorsed by younger individuals, whereas older adults may have developed more generationally sanctioned coping skills or other adaptive coping skills over time. This is supported by the finding that the severity of BPD symptoms wanes with age . Future research should, as previously mentioned, examine endorphin release and body modifications. Additionally, future research should survey people with BPD and investigate if their level of concern with social acceptability moderates their preference for either NSSI or body modification. Individuals with BPD who are seeking a more socially acceptable method of inflicting pain on themselves to regulate their emotions may elect to obtain body modifications, which are more likely to be perceived as “cool” rather than living with self-harm scars that may result in stigma . Lastly, longitudinal and experience-sampling methods may better identify the temporal relationships between NSSI craving, body modification craving, and engaging in NSSI and body modification. 4.3. Conclusions and Implications The implications of this study are important for furthering understanding of BPD and related psychopathology. Clinicians may benefit from assessing whether a patient is using body modification as an NSSI alternative and if it is adaptive or maladaptive. Overall, there are many research gaps involving body modifications and their possible emotion regulation function, considering their recent rise in popularity.
Despite the important findings in the current study, there are some notable limitations. First, the study was cross-sectional in nature; our data is therefore only correlational and cannot be used to infer causality. Second, there is also a potential for self-report and selection bias despite efforts to recruit a diverse sample. Third, demographic information was missing for half of the participants due to researcher error. This diminishes the ability to draw conclusions regarding differences between demographic groups, rendering the generalizability of the current findings unknown. Fourth, although the study was advertised in multiple locations to attract a range of respondents, many participants were younger adults, based on the available demographic information, who may have only recently become old enough to legally obtain body modifications without parental consent. Additionally, we did not question participants regarding cultural–historical motivations for body modifications. Two recent reviews of motivations for body modifications do speak in greater detail about cultural, historical, and social motivations for body modifications . Furthermore, we did not account for recruitment sources in our survey and are therefore unable to provide a breakdown of how many participants found the study from various recruitment sources and are therefore unable to with confidence confirm heterogeneity in our sample beyond the existing demographic information. Additionally, the emotional regulation and self-expression motivations for body modification scales were developed solely for this study and were not previously validated measures. The Social Identity subscale was brief, consisting of only three items, and internal consistency was inadequate. We only measured NSSI craving using a scale initially designed to measure substance use craving, though some evidence suggests these cravings differ from one another . While the BSL scale had many questions on thoughts of NSSI, there were no subscales examining the incidence of NSSI itself. These omissions, while limitations, were in pursuit of survey brevity, as shorter surveys have been shown to increase survey completion . Finally, we did not collect longitudinal data on the progression of body modifications along with BPD symptoms and NSSI and body modification craving over time.
Future researchers should survey a larger group of participants with more body modifications so that participants have a wide range of body modification coverage for analyses. Additionally, researchers can determine if age is a factor in the relationship between BPD symptoms and body modification. Body modifications may be a more modern form of coping or self-expression endorsed by younger individuals, whereas older adults may have developed more generationally sanctioned coping skills or other adaptive coping skills over time. This is supported by the finding that the severity of BPD symptoms wanes with age . Future research should, as previously mentioned, examine endorphin release and body modifications. Additionally, future research should survey people with BPD and investigate if their level of concern with social acceptability moderates their preference for either NSSI or body modification. Individuals with BPD who are seeking a more socially acceptable method of inflicting pain on themselves to regulate their emotions may elect to obtain body modifications, which are more likely to be perceived as “cool” rather than living with self-harm scars that may result in stigma . Lastly, longitudinal and experience-sampling methods may better identify the temporal relationships between NSSI craving, body modification craving, and engaging in NSSI and body modification.
The implications of this study are important for furthering understanding of BPD and related psychopathology. Clinicians may benefit from assessing whether a patient is using body modification as an NSSI alternative and if it is adaptive or maladaptive. Overall, there are many research gaps involving body modifications and their possible emotion regulation function, considering their recent rise in popularity.
|
Post-translational modification-centric base editor screens to assess phosphorylation site functionality in high-throughput | 1b6588a5-ad07-40a1-b1d1-ab0a15bc9430 | 11804830 | Biochemistry[mh] | Nearly every eukaryotic cellular process is controlled by post-translational modifications (PTMs), which can modulate protein subcellular localization, protein-biomolecular interactions, enzymatic activity, stability, etc. Protein phosphorylation is arguably the best characterized PTM . The human genome encodes for roughly 500 protein kinases and 200 phosphatases that control the coupling and hydrolysis of phosphates, respectively, on substrates in a rapid and dynamic fashion , . These signaling cascades organize into elaborate biochemical networks that allow the cell to process information about its intra- and extracellular environmental changes. Mass spectrometry-based proteomics has revolutionized our ability to map global signaling pathways at the phosphorylation site (phosphosite)-specific level, across time and cellular space. Current phosphoproteomics experiments can quantify tens of thousands of phosphosites, tracking their dynamics upon cell stimulation, drug treatment, or mutational status – . Unfortunately, fewer than 3% of the nearly quarter million phosphorylation sites identified have an ascribed function , . Since functionally characterizing novel modification sites is laborious and resource intensive, we often resort to describing complex biological systems by the limited number of well-characterized phosphosites for which we have good reagents. Functional genomics has greatly increased our throughput for associating genes with specific cellular phenotypes. Genome-wide CRISPR/Cas9 technology coupled with phenotypic screens allow researchers to identify which genes or non-coding regions are important for a specific function such as gene expression , cytokine secretion , cell proliferation , or cell survival , . More recently, CRISPR/Cas9-mediated base editors, which introduce specific nucleotide substitutions in genomic DNA rather than double stranded DNA breaks , have been used for mutational scanning across protein coding genes and regulatory elements – . Base editor technology holds immense promise to study PTM site function in high-throughput by mutating specific amino acids, bypassing the need to create site-specific homology-directed repair templates. Here, we describe an experimental workflow to study phosphorylation site functionality in high-throughput. By coupling quantitative phosphoproteomics with “proteome-wide” base editing of individual phosphosites and phenotypic screens, we are able to functionally evaluate a large number of previously unstudied phosphosites that are involved in cell proliferation or the transcriptional responses following T cell activation. T cell activation via stimulation of the T cell receptor (TCR) and co-stimulation of CD28 activates multiple kinases that form both negative and positive feedback loops, as well as several transcription factors including NFAT, NFκB, and AP1. Applying PTM-centric base editor screens to T cell activation, we show that we can recapitulate many known aspects of the pathway, while discovering novel kinase activities and specific phosphorylation events that control different aspects of the transcriptional response. This allowed us to identify a specific phosphosite on the phosphatase PHLPP1 as a novel regulator of T cell activation-induced NFAT and NFκB activities. Transcriptional profiling of PHLPP1 phosphosite mutant T cells shows that individual phosphorylation events differentially impact downstream gene expression in subtle yet distinct patterns, creating the potential to map causal links between signaling and gene expression. PTM-centric base editor screens provide an experimental framework to functionally interrogate and systematically decode the vast network of biochemical signaling events to their downstream phenotypes. Optimization of base editing for experimentally-derived phosphorylation sites To develop a system by which we could profile and then functionally assess signaling pathways and their effects on gene expression, we focused on a classic model of T cell activation in the human T cell leukemia line Jurkat E6–1, during which multiple kinases and downstream transcription factors are activated . We performed a temporally-resolved quantitative phosphoproteomic experiment, assaying global phosphorylation patterns for 0, 3, 9 and 27 minutes of T cell activation using α-CD3 and α-CD28 agonist antibodies , . Of the 26,037 quantified phosphopeptides, 899 were significantly differentially regulated (moderated F test, FDR < 0.05) during this time series ( , , and ). Replicates showed strong correlation and that the three-to-nine-minute transition showed the largest changes according to principle component analysis . PTM-SEA , a PTM site-centric analog to GSEA (gene set enrichment analysis), showed that various kinase activities were temporally regulated during the first 30 minutes of T cell activation . This analysis culminated with the perturbation signatures for “anti-CD3” and “phorbol esters”, indicating the temporally-regulated phosphopeptides reflected the appropriate T cell activation pathways. Using a custom bioinformatics pipeline, we queried which of the ~19,000 confidently localized phosphosites we could target using SpCas9-mediated C-to-T or A-to-G editors. We included all detected phosphosites, rather than only temporally regulated ones since there are likely to be phosphosites that were not statistically significant but still contribute to T cell activation. Considering editing windows and targetable locations, we found 7,618 unique phosphosites were targetable with 11,392 distinct sgRNAs using the SpCas9-A-to-G editor ABE8e, while 7,063 unique phosphosites could be targeted by the SpCas9-C-to-T editor BE4 ( and ) , . Roughly half of the editable phosphosites overlapped between the two base editors. The amino acid side chain representation of targetable phosphosites reflected those detected and statistically regulated . ABE8e appears to make more structurally conservative missense mutations, and unlike BE4, can target tyrosine-encoding codons . To develop a flexible genomic engineering approach, we based our base editing strategy on previous genome-wide CRISPR/Cas9 screens in primary T cells where sgRNAs are delivered via lentivirus, followed by electroporation of Cas9 protein . We nucleofected several different types of biomolecules to determine the most efficient base editors . Using Jurkat cells stably expressing a sgRNA targeting the model HEK3 site in humans, we nucleofected either plasmid DNA, chemically synthesized and capped mRNA, or recombinant protein of different base editor versions. We found that purified recombinant ABE8e protein (NGG PAM,) properly edited over 95% of adenosines in the base editing window . To test the reproducibility of the ABE8e protein, we re-expressed and purified the protein . Again, we found that 92% of the adenosines in the base editing window were mutated to guanosine via Sanger sequencing . These results demonstrate that we can reproducibly achieve sufficiently high base editing efficiency with ABE8e protein for high-throughput screens. Base editing phosphosites that promote or inhibit markers of T cell activation We tested whether mutating phosphosites in proteins known to be involved in the TCR signaling pathway with ABE8e protein would affect markers of T cell activation. We base edited the activating tyrosine of MAPK1 (also known as ERK2) Y187, and two targets in the TCR associated kinase ZAP70 Y292 and Y315 in Jurkat E6.1 cells. MAPK1 Y187 acts as a positive control as it is important for proper T cell activation signaling and transcriptional responses . Mutant or control cells were activated with α-CD3/CD28 agonist antibodies for 12 hours, stained for the early activation marker surface CD69 levels and analyzed by flow cytometry. Mutation of an inhibitory phosphotyrosine on ZAP70 Y292H increased CD69 surface expression whereas MAPK1 Y187C showed diminished surface CD69 ( and ). ZAP70 Y315H showed no effect, consistent with previous reports . Together these data establish that phosphosite mutations can have positive or negative effects on T cell activation levels. Phosphosite-centric functional phenotypic screens using pooled base editing for cell proliferation To assess base editing efficiency in pooled format, we created a lentiviral library consisting of roughly 11,000 phosphosite-targeting sgRNAs for missense mutations, 250 non-targeting controls, and 250 intergenic controls as negative controls. We also included 250 guides that introduce terminating edits in essential genes via mRNA splice site disruption, effectively knocking out the gene. TPR (triple parameter reporter) Jurkat cells, which have individual fluorescent reporters driven by separate NFAT, NFκB, and AP1 transcriptional response elements , were transduced at a multiplicity of infection of 0.3. After puromycin selection, a 500x library coverage aliquot of cells was collected and the rest were electroporated with ABE8e protein. To confirm our base editing was efficient we analyzed the representation of sgRNAs, comparing pre- and six days post-ABE8e protein electroporation . The representation of sgRNAs disrupting splice junctions in 250 essential genes was significantly lower six days post-base editing compared to all sgRNAs in the library indicating our base editing approach was working efficiently. The relative representation of intergenic and non-targeting controls was not affected by introduction of ABE8e protein. We next examined whether mutation of phosphosites important for cell division would affect cell viability or proliferation in our pooled format without specific stimuli or selective pressure . Examination of phosphorylation sites on CDK1, a kinase whose activity is necessary for proper cell division, showed that Y160C;T161A and Y15C reduced cell viability post-ABE8e electroporation similar in magnitude and direction to levels previously seen using homology-directed recombination . The silent mutation of Y19Y showed no effect. Analysis of the whole phosphosite-mutant dataset using the Model-based Analysis of Genome-wide CRISPR/Cas9 Knockout (MAGeCK) resulted in 89 sgRNAs that were significantly enriched pre-ABE compared to post-editing ( and ). sgRNAs targeting phosphosites that were depleted after ABE8e protein introduction were enriched for genes involved in the term “Cell Cycle” amongst the top three Reactome pathways . 58 sgRNAs introducing phosphosite mutations were enriched post-base editing, suggesting a proliferative advantage . These genes belonged to the Reactome signatures for “membrane trafficking”, “metabolism of RNA” and “transcriptional regulation by RUNX1”, pathways involved in the proliferation of glioblastoma and acute myeloid leukemia cells – . GO analysis of the genes knocked out through splice site disruption showed an enrichment for mRNA processing and cell cycle . PTM site-centric pathway analyses can identify groups of phosphosites (in this case, mutation of phosphosites) enriched in the pre- or post-edited pools. Kinase Library , which uses primary sequence motifs derived from biochemical kinase reactions to predict kinase activity through motif enrichment, identified CDK1/4/5/6/13/18 motifs from aggregated phosphosite mutants depleted relative to pre-edited cells . MELK1, NIM1, DYRK2/4, and YANK2/3 motifs were enriched in the pre-edited cells. Together, these data show that our approach of base editing phosphorylation sites in a pooled format can identify phosphorylated residues and putative kinases important for proper cell cycle proliferation or survival. Coupling functional phosphosite screens with transcriptional reporters identifies novel regulators of NFAT transcriptional activity To screen for phosphosites that are functionally linked to transcriptional outputs, we performed a “proteome-wide” base editor screen in activated TPR Jurkat cells, utilizing a GFP transcriptional reporter driven by the 4x ARRE2 sequence derived from the murine IL2 promoter . TPR Jurkat cells stably integrated with the sgRNA library described above were electroporated with ABE8e protein, stimulated with α-CD3/CD28 antibodies for 12 hours, and sorted for high and low GFP (NFAT activity) levels . Genomic DNA was collected from the cells sorted into high and low bins, sgRNA identifiers were amplified by PCR, and were sequenced by next-generation sequencing (NGS). Total read-normalized, log transformed sgRNA counts were moderately to strongly correlated indicating sufficient data quality between quadruplicates, and the mean correlation between sgRNA counts in GFP high and low bins was strong ( - and ). We used MAGeCK to identify and rank which phosphosite mutations, the targets of the sgRNAs, regulate the NFAT transcriptional reporter. Phosphosites with multiple sgRNAs were combined in MAGeCK and the high and low GFP bins were compared. We identified 411 sgRNAs enriched in the GFP high bin, and 293 in the GFP low bin ( and ). Rolling up our phosphosite level perturbations to the gene level (gene-centric), we performed pathway analyses, using multiple tools, to assess the fidelity of our approach. Enrichment analysis using MAGeCKFlute identified TCR pathways as enriched in the GFP low . g:Profiler , a Gene Ontology-based analysis, also identified TCR pathway in the GFP low bin . GSEA identified the “TCR Calcium Pathway” signature in the GFP high bin (Supp Figure 4E), likely due, in part, to dephosphorylation of NFAT as a regulatory mechanism . Edits containing bystander mutations next to the target phosphosite were not enriched in the GFP high and low bins compared to the library as a whole, suggesting they do not confer extra effects . Utilizing PTM site-centric analyses, Kinase Library identified several kinases implicated in NFAT activity regulation, such as JNK , NLK , and CAMK2G , enriched in the GFP low bin . The GFP high bin was enriched for the CDK9 motif, a kinase involved in general transcriptional regulation , as well as motifs of kinases known to be involved in T cell activation such as PAK2 and CDK5 . CLK3, PAK4, SRPK1 and MYLK4 were also enriched in the GFP high bin but have poorly characterized roles in controlling NFAT transcriptional activity . PTM-SEA agreed with the Kinase Library results for PAK2 and CDK9 . However, PTM-SEA also identified mutations of MAPK1, MTOR, and DYRK1A/2/3 substrates to be enriched in the GFP low bins, corroborating these kinases’ involvement in T cell activation . As expected, these results demonstrate that phosphosite mutations can directly implicate their involvement in regulating NFAT transcriptional activity, recapitulating known signaling pathways that were constructed from various studies of general T cell activation. These results also strongly suggest that our approach of base editing phosphosites is not only capable of rediscovering crucial signaling molecules in T cell activation, but provides new insights by identifying novel, regulatory kinases and phosphorylation events. We next asked how phosphopeptide abundance dynamics related to functional readouts in our NFAT-GFP screen. As phosphopeptide abundances could have increased and/or decreased in our time course data, we plotted the F statistic from the mass spectrometric analyses against the log2 fold change in the GFP high and low bins calculated by MAGeCK. We found that there was an overlap between the two datasets, where MAPK1 Y817C was amongst the largest changes in phosphopeptide levels and GFP expression . We also tested whether phosphosites predicted to be functional by Ochoa et al. 2020 correlated with our empirical data for NFAT-GFP activity. Although there was a positive trend between our phenotypic screen data and predicted functional scores, the correlation was not statistically significant . These results underscore the need for PTM-centric phenotypic screens. PTM-centric base editor screening identifies the uncharacterized phosphorylation-mediated nuclear localization of PHLPP1 PHLPP1 is a protein phosphatase most-widely implicated in AKT signaling in cancer , . In macrophages, PHLPP1 attenuates the JAK/STAT axis by dephosphorylating STAT1 . The post-translational mechanisms controlling PHLPP1 function, and its involvement in T cell biology , are poorly understood. We identified the mutation PHLPP1 S118P in our base editing screen to have a strong, negative impact on NFAT transcriptional activity, on par with MAPK1 Y187C . We nucleofected ribonucleoproteins comprised of in vitro transcribed sgRNA coupled to ABE8e protein to validate our screen results. The intergenic mutation at the HEK3 site was used as an editing control. For positive effect controls, we introduced a terminating edit in the scaffolding protein LCP2 (SLP76), a critical molecule for proper T cell activation or the MAPK1 Y187C mutation which prevented or altered T cell activation, respectively. After bulk editing was assessed by amplicon sequencing (x̅ = 70%, ), four to eight validated single cell clones were combined to avoid clone-specific effects. Editing at the top two predicted off-target sites for all clones was within the noise of the assay, about <3% editing. MAPK1 Y187C and PHLPP1 S118P both showed diminished NFAT transcriptional activity . However, unlike MAPK1 Y187C, PHLPP1 S118P showed a small but statistically significant increase in CFP levels (NFκB activity reporter). To extend this analysis, we performed transcriptional profiling of clonal Jurkat T cells harboring the homozygous HEK 3 intergenic edit, LCP2 terminating edit, the MAPK1 Y187C, or the PHLPP1 S118P mutations for zero and six hours post CD3/CD28 stimulation. LCP2 -terminated cells showed no effect of transcriptional activation six hours after activation, whereas MAPK Y187C showed an intermediate pattern compared to the HEK 3 controls ( , , and ). The PHLPP1 mutant cells were similar in their transcription patterns compared to the HEK 3 editing controls, though differences were apparent . We highlighted the genes differentially expressed between HEK 3, MAPK1 Y187C, and PHLPP1 S118P, and plotted them alongside the LCP2 terminating edit cells . After k-means clustering and GO analysis of differentially expressed genes, clusters one, five, and six, which were higher in the PHLPP1 mutant compared to the HEK3 control identified multiple terms associated with NFκB signaling (NF-kappaB complex, TNFR signaling), corroborating our transcriptional reporter results . The MAPK1 Y187C mutant cells were enriched for genes in sterol and isoprenoid biosynthesis (cluster 4 of ). These results suggest different phosphorylation sites in the T cell activation pathway can regulate downstream gene expression in disparate ways. S118 lies within the bipartite nuclear localization sequences (NLSs) at the N-terminus of PHLPP1 . To test the hypothesis that phosphorylation controls PHLPP1 subcellular localization, we ectopically expressed N-terminal extension (NTE) constructs with the wild type sequence, both halves of the NLS mutated, or S118P, the result of A-to-G editing. We also included the more archetypal amino acid substitution, S118A, which removes the phosphorylatable residue without potential structural changes introduced by the rotationally constrained amino acid proline, and the S118E mutation as a potential phosphomimetic. We found that both S118P and S118A reduced PHLPP1 NTE nuclear localization to comparable levels, but less severe than the full NLS mutant . S118E showed a similar loss of nuclear localization compared to S118A or S118P, indicating in this case the glutamate may not accurately reflect a bona fide phosphorylation . These results suggest that phosphorylation of the NTE of PHLPP1 regulates nuclear localization. These data also suggest that serine to proline substitutions are reasonable proxies for loss of a phosphorylatable residue for screening purposes. The T cell activation transcriptional program can be dissected through precise, base editor-mediated signaling modulation Transcriptional profiling HEK3 (intergenic editing control) , MAPK1 and PHLPP1 mutant Jurkat T cells showed more differentially expressed genes six hours post stimulus than in resting conditions (BH corrected p value <0.05) , suggesting the mutated phosphosites indeed affect T cell activation-induced transcriptional responses. Inspection of the differentially expressed genes at six hours post T cell activation showed specific T cell-related genes are expressed at subtle but different levels between phosphosite-mutant genotypes ( and ). For example, BCL11A and THEMIS were activated to a higher extent in MAPK1 Y187C mutants compared to PHLPP1 S118P or control. In contrast, NR4A3, ZFP36L1, and IL21R were highest in the PHLPP1 S118P cells, whereas GZMA, TNFSF14, NFKBIA, and JUND were highest in the editing control cells (intergenic HEK3 ). Intracellular GZMB staining 24 hours after T cell activation corroborated the gene expression results, showing a loss of GZMB protein in MAPK1 Y187C but not PHLPP1 S118P cells ( and ). These results suggest that mutating different phosphosites in ostensibly the same signaling pathway can alter transcriptional responses, and may provide a means for fine-tuning gene expression. To develop a system by which we could profile and then functionally assess signaling pathways and their effects on gene expression, we focused on a classic model of T cell activation in the human T cell leukemia line Jurkat E6–1, during which multiple kinases and downstream transcription factors are activated . We performed a temporally-resolved quantitative phosphoproteomic experiment, assaying global phosphorylation patterns for 0, 3, 9 and 27 minutes of T cell activation using α-CD3 and α-CD28 agonist antibodies , . Of the 26,037 quantified phosphopeptides, 899 were significantly differentially regulated (moderated F test, FDR < 0.05) during this time series ( , , and ). Replicates showed strong correlation and that the three-to-nine-minute transition showed the largest changes according to principle component analysis . PTM-SEA , a PTM site-centric analog to GSEA (gene set enrichment analysis), showed that various kinase activities were temporally regulated during the first 30 minutes of T cell activation . This analysis culminated with the perturbation signatures for “anti-CD3” and “phorbol esters”, indicating the temporally-regulated phosphopeptides reflected the appropriate T cell activation pathways. Using a custom bioinformatics pipeline, we queried which of the ~19,000 confidently localized phosphosites we could target using SpCas9-mediated C-to-T or A-to-G editors. We included all detected phosphosites, rather than only temporally regulated ones since there are likely to be phosphosites that were not statistically significant but still contribute to T cell activation. Considering editing windows and targetable locations, we found 7,618 unique phosphosites were targetable with 11,392 distinct sgRNAs using the SpCas9-A-to-G editor ABE8e, while 7,063 unique phosphosites could be targeted by the SpCas9-C-to-T editor BE4 ( and ) , . Roughly half of the editable phosphosites overlapped between the two base editors. The amino acid side chain representation of targetable phosphosites reflected those detected and statistically regulated . ABE8e appears to make more structurally conservative missense mutations, and unlike BE4, can target tyrosine-encoding codons . To develop a flexible genomic engineering approach, we based our base editing strategy on previous genome-wide CRISPR/Cas9 screens in primary T cells where sgRNAs are delivered via lentivirus, followed by electroporation of Cas9 protein . We nucleofected several different types of biomolecules to determine the most efficient base editors . Using Jurkat cells stably expressing a sgRNA targeting the model HEK3 site in humans, we nucleofected either plasmid DNA, chemically synthesized and capped mRNA, or recombinant protein of different base editor versions. We found that purified recombinant ABE8e protein (NGG PAM,) properly edited over 95% of adenosines in the base editing window . To test the reproducibility of the ABE8e protein, we re-expressed and purified the protein . Again, we found that 92% of the adenosines in the base editing window were mutated to guanosine via Sanger sequencing . These results demonstrate that we can reproducibly achieve sufficiently high base editing efficiency with ABE8e protein for high-throughput screens. We tested whether mutating phosphosites in proteins known to be involved in the TCR signaling pathway with ABE8e protein would affect markers of T cell activation. We base edited the activating tyrosine of MAPK1 (also known as ERK2) Y187, and two targets in the TCR associated kinase ZAP70 Y292 and Y315 in Jurkat E6.1 cells. MAPK1 Y187 acts as a positive control as it is important for proper T cell activation signaling and transcriptional responses . Mutant or control cells were activated with α-CD3/CD28 agonist antibodies for 12 hours, stained for the early activation marker surface CD69 levels and analyzed by flow cytometry. Mutation of an inhibitory phosphotyrosine on ZAP70 Y292H increased CD69 surface expression whereas MAPK1 Y187C showed diminished surface CD69 ( and ). ZAP70 Y315H showed no effect, consistent with previous reports . Together these data establish that phosphosite mutations can have positive or negative effects on T cell activation levels. To assess base editing efficiency in pooled format, we created a lentiviral library consisting of roughly 11,000 phosphosite-targeting sgRNAs for missense mutations, 250 non-targeting controls, and 250 intergenic controls as negative controls. We also included 250 guides that introduce terminating edits in essential genes via mRNA splice site disruption, effectively knocking out the gene. TPR (triple parameter reporter) Jurkat cells, which have individual fluorescent reporters driven by separate NFAT, NFκB, and AP1 transcriptional response elements , were transduced at a multiplicity of infection of 0.3. After puromycin selection, a 500x library coverage aliquot of cells was collected and the rest were electroporated with ABE8e protein. To confirm our base editing was efficient we analyzed the representation of sgRNAs, comparing pre- and six days post-ABE8e protein electroporation . The representation of sgRNAs disrupting splice junctions in 250 essential genes was significantly lower six days post-base editing compared to all sgRNAs in the library indicating our base editing approach was working efficiently. The relative representation of intergenic and non-targeting controls was not affected by introduction of ABE8e protein. We next examined whether mutation of phosphosites important for cell division would affect cell viability or proliferation in our pooled format without specific stimuli or selective pressure . Examination of phosphorylation sites on CDK1, a kinase whose activity is necessary for proper cell division, showed that Y160C;T161A and Y15C reduced cell viability post-ABE8e electroporation similar in magnitude and direction to levels previously seen using homology-directed recombination . The silent mutation of Y19Y showed no effect. Analysis of the whole phosphosite-mutant dataset using the Model-based Analysis of Genome-wide CRISPR/Cas9 Knockout (MAGeCK) resulted in 89 sgRNAs that were significantly enriched pre-ABE compared to post-editing ( and ). sgRNAs targeting phosphosites that were depleted after ABE8e protein introduction were enriched for genes involved in the term “Cell Cycle” amongst the top three Reactome pathways . 58 sgRNAs introducing phosphosite mutations were enriched post-base editing, suggesting a proliferative advantage . These genes belonged to the Reactome signatures for “membrane trafficking”, “metabolism of RNA” and “transcriptional regulation by RUNX1”, pathways involved in the proliferation of glioblastoma and acute myeloid leukemia cells – . GO analysis of the genes knocked out through splice site disruption showed an enrichment for mRNA processing and cell cycle . PTM site-centric pathway analyses can identify groups of phosphosites (in this case, mutation of phosphosites) enriched in the pre- or post-edited pools. Kinase Library , which uses primary sequence motifs derived from biochemical kinase reactions to predict kinase activity through motif enrichment, identified CDK1/4/5/6/13/18 motifs from aggregated phosphosite mutants depleted relative to pre-edited cells . MELK1, NIM1, DYRK2/4, and YANK2/3 motifs were enriched in the pre-edited cells. Together, these data show that our approach of base editing phosphorylation sites in a pooled format can identify phosphorylated residues and putative kinases important for proper cell cycle proliferation or survival. To screen for phosphosites that are functionally linked to transcriptional outputs, we performed a “proteome-wide” base editor screen in activated TPR Jurkat cells, utilizing a GFP transcriptional reporter driven by the 4x ARRE2 sequence derived from the murine IL2 promoter . TPR Jurkat cells stably integrated with the sgRNA library described above were electroporated with ABE8e protein, stimulated with α-CD3/CD28 antibodies for 12 hours, and sorted for high and low GFP (NFAT activity) levels . Genomic DNA was collected from the cells sorted into high and low bins, sgRNA identifiers were amplified by PCR, and were sequenced by next-generation sequencing (NGS). Total read-normalized, log transformed sgRNA counts were moderately to strongly correlated indicating sufficient data quality between quadruplicates, and the mean correlation between sgRNA counts in GFP high and low bins was strong ( - and ). We used MAGeCK to identify and rank which phosphosite mutations, the targets of the sgRNAs, regulate the NFAT transcriptional reporter. Phosphosites with multiple sgRNAs were combined in MAGeCK and the high and low GFP bins were compared. We identified 411 sgRNAs enriched in the GFP high bin, and 293 in the GFP low bin ( and ). Rolling up our phosphosite level perturbations to the gene level (gene-centric), we performed pathway analyses, using multiple tools, to assess the fidelity of our approach. Enrichment analysis using MAGeCKFlute identified TCR pathways as enriched in the GFP low . g:Profiler , a Gene Ontology-based analysis, also identified TCR pathway in the GFP low bin . GSEA identified the “TCR Calcium Pathway” signature in the GFP high bin (Supp Figure 4E), likely due, in part, to dephosphorylation of NFAT as a regulatory mechanism . Edits containing bystander mutations next to the target phosphosite were not enriched in the GFP high and low bins compared to the library as a whole, suggesting they do not confer extra effects . Utilizing PTM site-centric analyses, Kinase Library identified several kinases implicated in NFAT activity regulation, such as JNK , NLK , and CAMK2G , enriched in the GFP low bin . The GFP high bin was enriched for the CDK9 motif, a kinase involved in general transcriptional regulation , as well as motifs of kinases known to be involved in T cell activation such as PAK2 and CDK5 . CLK3, PAK4, SRPK1 and MYLK4 were also enriched in the GFP high bin but have poorly characterized roles in controlling NFAT transcriptional activity . PTM-SEA agreed with the Kinase Library results for PAK2 and CDK9 . However, PTM-SEA also identified mutations of MAPK1, MTOR, and DYRK1A/2/3 substrates to be enriched in the GFP low bins, corroborating these kinases’ involvement in T cell activation . As expected, these results demonstrate that phosphosite mutations can directly implicate their involvement in regulating NFAT transcriptional activity, recapitulating known signaling pathways that were constructed from various studies of general T cell activation. These results also strongly suggest that our approach of base editing phosphosites is not only capable of rediscovering crucial signaling molecules in T cell activation, but provides new insights by identifying novel, regulatory kinases and phosphorylation events. We next asked how phosphopeptide abundance dynamics related to functional readouts in our NFAT-GFP screen. As phosphopeptide abundances could have increased and/or decreased in our time course data, we plotted the F statistic from the mass spectrometric analyses against the log2 fold change in the GFP high and low bins calculated by MAGeCK. We found that there was an overlap between the two datasets, where MAPK1 Y817C was amongst the largest changes in phosphopeptide levels and GFP expression . We also tested whether phosphosites predicted to be functional by Ochoa et al. 2020 correlated with our empirical data for NFAT-GFP activity. Although there was a positive trend between our phenotypic screen data and predicted functional scores, the correlation was not statistically significant . These results underscore the need for PTM-centric phenotypic screens. PHLPP1 is a protein phosphatase most-widely implicated in AKT signaling in cancer , . In macrophages, PHLPP1 attenuates the JAK/STAT axis by dephosphorylating STAT1 . The post-translational mechanisms controlling PHLPP1 function, and its involvement in T cell biology , are poorly understood. We identified the mutation PHLPP1 S118P in our base editing screen to have a strong, negative impact on NFAT transcriptional activity, on par with MAPK1 Y187C . We nucleofected ribonucleoproteins comprised of in vitro transcribed sgRNA coupled to ABE8e protein to validate our screen results. The intergenic mutation at the HEK3 site was used as an editing control. For positive effect controls, we introduced a terminating edit in the scaffolding protein LCP2 (SLP76), a critical molecule for proper T cell activation or the MAPK1 Y187C mutation which prevented or altered T cell activation, respectively. After bulk editing was assessed by amplicon sequencing (x̅ = 70%, ), four to eight validated single cell clones were combined to avoid clone-specific effects. Editing at the top two predicted off-target sites for all clones was within the noise of the assay, about <3% editing. MAPK1 Y187C and PHLPP1 S118P both showed diminished NFAT transcriptional activity . However, unlike MAPK1 Y187C, PHLPP1 S118P showed a small but statistically significant increase in CFP levels (NFκB activity reporter). To extend this analysis, we performed transcriptional profiling of clonal Jurkat T cells harboring the homozygous HEK 3 intergenic edit, LCP2 terminating edit, the MAPK1 Y187C, or the PHLPP1 S118P mutations for zero and six hours post CD3/CD28 stimulation. LCP2 -terminated cells showed no effect of transcriptional activation six hours after activation, whereas MAPK Y187C showed an intermediate pattern compared to the HEK 3 controls ( , , and ). The PHLPP1 mutant cells were similar in their transcription patterns compared to the HEK 3 editing controls, though differences were apparent . We highlighted the genes differentially expressed between HEK 3, MAPK1 Y187C, and PHLPP1 S118P, and plotted them alongside the LCP2 terminating edit cells . After k-means clustering and GO analysis of differentially expressed genes, clusters one, five, and six, which were higher in the PHLPP1 mutant compared to the HEK3 control identified multiple terms associated with NFκB signaling (NF-kappaB complex, TNFR signaling), corroborating our transcriptional reporter results . The MAPK1 Y187C mutant cells were enriched for genes in sterol and isoprenoid biosynthesis (cluster 4 of ). These results suggest different phosphorylation sites in the T cell activation pathway can regulate downstream gene expression in disparate ways. S118 lies within the bipartite nuclear localization sequences (NLSs) at the N-terminus of PHLPP1 . To test the hypothesis that phosphorylation controls PHLPP1 subcellular localization, we ectopically expressed N-terminal extension (NTE) constructs with the wild type sequence, both halves of the NLS mutated, or S118P, the result of A-to-G editing. We also included the more archetypal amino acid substitution, S118A, which removes the phosphorylatable residue without potential structural changes introduced by the rotationally constrained amino acid proline, and the S118E mutation as a potential phosphomimetic. We found that both S118P and S118A reduced PHLPP1 NTE nuclear localization to comparable levels, but less severe than the full NLS mutant . S118E showed a similar loss of nuclear localization compared to S118A or S118P, indicating in this case the glutamate may not accurately reflect a bona fide phosphorylation . These results suggest that phosphorylation of the NTE of PHLPP1 regulates nuclear localization. These data also suggest that serine to proline substitutions are reasonable proxies for loss of a phosphorylatable residue for screening purposes. Transcriptional profiling HEK3 (intergenic editing control) , MAPK1 and PHLPP1 mutant Jurkat T cells showed more differentially expressed genes six hours post stimulus than in resting conditions (BH corrected p value <0.05) , suggesting the mutated phosphosites indeed affect T cell activation-induced transcriptional responses. Inspection of the differentially expressed genes at six hours post T cell activation showed specific T cell-related genes are expressed at subtle but different levels between phosphosite-mutant genotypes ( and ). For example, BCL11A and THEMIS were activated to a higher extent in MAPK1 Y187C mutants compared to PHLPP1 S118P or control. In contrast, NR4A3, ZFP36L1, and IL21R were highest in the PHLPP1 S118P cells, whereas GZMA, TNFSF14, NFKBIA, and JUND were highest in the editing control cells (intergenic HEK3 ). Intracellular GZMB staining 24 hours after T cell activation corroborated the gene expression results, showing a loss of GZMB protein in MAPK1 Y187C but not PHLPP1 S118P cells ( and ). These results suggest that mutating different phosphosites in ostensibly the same signaling pathway can alter transcriptional responses, and may provide a means for fine-tuning gene expression. Linking specific signaling events to their downstream functions is a fundamental goal of biology. Choosing novel phosphosites for mechanistic follow-up studies is often resource intensive and laborious. We aimed to create a screening platform to functionally assess and prioritize individual phosphosites and how they contribute to specific phenotypes in high-throughput by integrating mass spectrometry-based phosphoproteomics with CRISPR-mediated base editor screens. We found that mutation of phosphosites via base editing in a pooled format can assess multiple functional readouts, in positive and negative directions, for proliferative or transcriptional phenotypes. This novel experimental and computational framework will greatly enable future studies, allowing the community to address the complexity of signaling pathways. Our study focused on phosphosites empirically identified in a parallel experiment rather than from PTM repositories or databases, mitigating the introduction of indiscriminate coding mutations that can have PTM-independent effects , , , . Moreover, as phosphoproteomic analyses of novel, unstudied systems continue to grow (i.e. patient-specific cancers, primary mouse or human immune cells, etc.) the bioinformatic tools needed to identify base editor-targetable phosphosites from empirical mass spectrometry data will become increasingly important . All gene-centric pathway analyses from genes with mutated phosphosites that altered NFAT activity revealed that the TCR pathway was enriched, indicating the results from our phosphosite base editor screen can recapitulate known aspects of well-studied pathways. Interestingly, gene-level GSEA analysis found “TCR Calcium Pathway” enriched in the GFP high bin. The phosphosite mutations driving this result were all in NFAT isoforms, or molecules known to regulate NFAT dephosphorylation: NFATC1/2/3, CABIN1, and RCAN1. This pathway regulates gene expression, cytokine production, and T cell activation through the NFAT signaling pathway by dephosphorylation of NFAT molecules, replicating the effect of Calcineurin ( PPP3CC ) in the translocation of NFAT and T cell activation , . Site-centric pathway analyses of enriched phosphosite mutations revealed two major trends in the signatures identified in our Jurkat T cell activation model: TCR signaling and cell cycle. As Jurkat cells are rapidly dividing transformed cells it is not surprising that many of the phosphosites identified by mass spectrometry, then subsequently in the base editor screens, identified cell cycle genes and phosphorylation events. The DYRK family of kinases is a good example ( and ). DYRKs are well known to control NFAT transcriptional activity , , but also have roles in T cell proliferation . Our results suggest DYRKs have a stronger role in T cell activation compared to proliferation. We envision that applying base editor screens of PTM sites in primary immune cells will provide clearer connections within signaling-to-transcription networks. Our PTM-centric base editor phenotypic screen identified a novel mechanism for PHLPP1 as a regulator of T cell activation-mediated gene expression. PHLPP1’s function has been primarily studied in cancer contexts , , though its role in regulatory T cell development was identified through a genetic study . Our analyses originally identified it as a positive regulator of NFAT activity, and through transcriptional profiling we found that PHLPP1 can negatively regulate genes downstream of NFκB. This regulation is likely due to the phosphorylation-induced nuclear translocation of PHLPP1, though the precise substrates of dephosphorylation via PHLPP1 remain to be determined. It is worth noting that the phosphopeptide identifying PHLPP1 pS118 was induced at nine minutes 1.6-fold on average but was not statistically regulated in our phosphoproteomics data . Conversely, MAPK1 Y817C had both one of the strongest fold changes at the phosphopeptide abundance level and in the NFAT-GFP screen. The complimentary but orthogonal information between the phosphoproteomics and the base editor screen results is likely due to the specific nature of our chosen screening platform, an NFAT activity reporter. T cell activation is highly complex and not limited to transcriptional outputs, and thus, a substantial proportion of phosphorylation events may be completely unrelated to induction of NFAT activity. This underscores the need for functional screens of PTM sites as small changes to catalytic enzymes may have prominent downstream effects. Gene expression profiling of various phosphosite-mutant Jurkats after T cell activation revealed subtle but different transcriptional responses. For example, the MAPK1 phosphomutant showed decreased expression of GZMB compared to the PHLPP1 mutant, at the protein and mRNA level , despite having similar effects on NFAT reporter activity . MAPK1 phosphomutant cells expressed less NR4A1 and NR4A3 than the PHLPP1 phosphomutant cells. GZMB is the major effector molecule of the cytotoxic program and NR4As have been shown to be drivers of T cell exhaustion . These results together lead to an intriguing possibility of fine-tuning expression levels of specific genes, through genetic or small molecule inhibitor manipulation of signaling pathways. Proteome-wide, PTM-centric base editing coupled to phenotypic screens provides a powerful experimental framework to untangle the vast network of biochemical signaling reactions and how they lead to the control of specific cellular functions. This phosphosite-targeting base editing approach directly assesses the impact of a phosphosite mutation on a given phenotype rather than relying purely on evolutionary or structural conservation, and provides interpretable results with high confidence for further mechanistic studies , , , . With currently available base editors it should be possible to functionally screen other PTMs including acetylation/ubiquitination (lysine), O-GlcNAcylation (Ser/Thr), cysteine , or even specific proteolysis events (caspases), utilizing established infrastructure common to many research institutions. We envision this approach to be widely enabling to the cell biology community. Limitations and future steps for PTM-centric base editor screens We focused on ABE8e for our technology development because it can be readily expressed and purified from Escherichia coli , and its activity after electroporation into cells harboring sgRNA-expressing plasmids is highly efficacious , . Also, as an A-to-G editor it can mutate tyrosine residues, a small but important fraction of total phosphosites. This property is missing from C-to-T editors. The amino acid substitutions made by current base editors to study phosphorylation are not archetypal (S/T to A; Y to F), threonines targeted by ABE8e being the exception (Thr to Ala). Over half of the serines mutated in our study were to proline, a limitation of current base editors. However, since several kinase families are proline directed and the majority of phosphosites are in flexible loops, this is likely to be problematic in only a subset of cases – . Empirically, the vast majority of Pro substitutions in our data had no effect, arguing it is not inherently disruptive. In fact, our mutational analysis of the PHLPP1 phosphosite showed similar effects between the S118P and the archetypal S118A mutation used for proper validation, suggesting proline mutations are not invariably detrimental to screen for phosphosite function. Of course, it is prudent to validate the findings of a screen with orthogonal approaches. Prime editors, which can install specific, desired sequences, will likely become increasingly useful as they become adapted for genome-wide screens . They may also allow for installation of phosphomimetic substitutions, S/T/Y to D or E. However, carboxylic side chains can often not properly mimic a bona fide phosphorylated residue . Moreover, employing Cas molecules with less restrictive PAM sequences (NG rather than NGG), should double to triple the percentage of targetable phosphosites . Our nucleofection of purified ABE protein approach borrows from previous literature performing CRISPR-mediated gene knockouts in primary human myeloid and T cells, the latter of which can be performed at a genome-wide scale , . As our phosphosite-targeting sgRNA library is much smaller than standard genome-wide libraries, our PTM-centric workflow should be readily adapted to other systems. Moreover, as mass spectrometry-based phosphoproteomics becomes more sensitive and robust, and unstudied systems such as patient-specific cancers or a variety of human or mouse immune cells are starting to be analyzed, the tools developed here should enable the community to study new signaling pathways in a more comprehensive and non-biased manner. We focused on ABE8e for our technology development because it can be readily expressed and purified from Escherichia coli , and its activity after electroporation into cells harboring sgRNA-expressing plasmids is highly efficacious , . Also, as an A-to-G editor it can mutate tyrosine residues, a small but important fraction of total phosphosites. This property is missing from C-to-T editors. The amino acid substitutions made by current base editors to study phosphorylation are not archetypal (S/T to A; Y to F), threonines targeted by ABE8e being the exception (Thr to Ala). Over half of the serines mutated in our study were to proline, a limitation of current base editors. However, since several kinase families are proline directed and the majority of phosphosites are in flexible loops, this is likely to be problematic in only a subset of cases – . Empirically, the vast majority of Pro substitutions in our data had no effect, arguing it is not inherently disruptive. In fact, our mutational analysis of the PHLPP1 phosphosite showed similar effects between the S118P and the archetypal S118A mutation used for proper validation, suggesting proline mutations are not invariably detrimental to screen for phosphosite function. Of course, it is prudent to validate the findings of a screen with orthogonal approaches. Prime editors, which can install specific, desired sequences, will likely become increasingly useful as they become adapted for genome-wide screens . They may also allow for installation of phosphomimetic substitutions, S/T/Y to D or E. However, carboxylic side chains can often not properly mimic a bona fide phosphorylated residue . Moreover, employing Cas molecules with less restrictive PAM sequences (NG rather than NGG), should double to triple the percentage of targetable phosphosites . Our nucleofection of purified ABE protein approach borrows from previous literature performing CRISPR-mediated gene knockouts in primary human myeloid and T cells, the latter of which can be performed at a genome-wide scale , . As our phosphosite-targeting sgRNA library is much smaller than standard genome-wide libraries, our PTM-centric workflow should be readily adapted to other systems. Moreover, as mass spectrometry-based phosphoproteomics becomes more sensitive and robust, and unstudied systems such as patient-specific cancers or a variety of human or mouse immune cells are starting to be analyzed, the tools developed here should enable the community to study new signaling pathways in a more comprehensive and non-biased manner. Cell Culture Jurkat E6.1 and HEK293T cells were purchased through ATCC. Triple parameter reporter (TPR) Jurkat cells were purchased from Professor Peter Steinberger at the Medical University of Vienna. TPR Jurkat cells and E6.1 Jurkat cells were cultured and passaged in Roswell Park Memorial Institute medium 1640 (RPMI-1640) plus GlutaMax (Thermo Fisher Scientific) supplemented with 10% heat inactivated fetal bovine serum. Cells were passaged and maintained at cell densities between 1–5e5 cells per mL. HEK293T cells were cultured and passaged in Dulbecco’s Modified Eagle Medium (DMEM) supplemented with 10% heat inactivated fetal bovine serum. All cells were incubated and maintained at 37°C with 5% CO2. Phosphoproteomic analysis Jurkat E6.1 cells (ATCC) were activated in 96 well tissue culture plates for the stated times at 2e5 cells/mL. Plates were coated with 3.33 μg/mL α-CD3 (HITa3) and α-CD28 antibodies (Biolegend), in 100 μL PBS at 4°C overnight, followed by one cold PBS wash. To stop the reaction, the cells were transferred to ice cold PBS, washed twice, flash frozen, and stored at −80°C until processing. Phosphoproteomic analysis was performed as previously described . Data was analyzed using Spectrum Mill (Agilent and Broad Institute) for phosphopeptide identification and quantification. The TMT denominator for each sample was the mean of all TMT channels. After global median normalization and median absolute deviation scaling, a moderated F test (limma, R) was performed to identify “regulated” phosphopeptide levels across the time series. Statistically significant phosphopeptides (Benjamini-Hochberg adj. p value <0.05) were visualized using Morpheus ( https://software.broadinstitute.org/morpheus/ ). The .json file associated with this manuscript, , can be used to explore these data in Morpheus. ABE8e protein expression and purification Recombinant ABE8e was expressed and purified as previously described . Briefly, ABE8e was expressed as an 8xHis tagged protein from a rhamnose-inducible promoter in BL21-Star DE3 cells with low RNase activity (Thermo Fisher Scientific). At an OD600 of ~0.8, the 2X terrific broth Escherichia coli cultures were cold shocked on ice for one hour, and induced with 0.8% final concentration of rhamnose. Roughly 24 hours later, cells were lysed via lysozyme and sonication, and ABE8e protein was purified on Ni-NTA resin. After imidazole elution, ABE8e protein was further purified using cation exchange. Fractions were monitored by UV and SDS-PAGE. ABE8e protein containing fractions were pooled, concentrated to ~90 pmol/μL using 100 MW cutoff filters, aliquoted, flash frozen, and stored at −80°C. Base editing with arrayed lentivirus Oligonucleotides containing the protospacer were chosen by hand or from our original bioinformatic analysis, and ordered from IDT DNA technologies. CACCG was added to the forward oligo, where AAAC was added to the reverse complement of the protospacer sequence. A C on the 3’ end of the reverse complement oligo was also appended. pRDA118 (Addgene Plasmid # 133459) was digested with BsbmI_V2 (NEB) and FastAP (Thermo) for the last five minutes, followed by gel purification. Oligos were mixed 1:1 at 100 μM, phosphorylated via PNK (NEB) and annealed after a 5 minutes 95°C step at 5°C per five minutes until 25°C. Annealed oligos were diluted 1:200 and 1 μL was mixed with 25–50 ng of digested backbone. T4 ligation (NEB) was performed for 20 minutes at 37°C and the 5–10 μL ligation reaction was transformed into Stbl3 E. coli , made in house. Colonies were verified via Sanger sequencing. One μg of sgRNA plasmids were transfected with Lipofectamine 3000 (Invitrogen) 1 μg of pPAX2 and 0.1 μg VSVG into HEK293Ts seeded the night before at 2e5 cells per 6 well. After three hours the 2.5 mL media was replaced with 5 mLs media supplemented with 1% BSA. Supernatants were collected after 3 days, filtered to remove HEK293T cells, concentrated 10-fold with Lenti-X precipitation solution (Alstem), and aliquots were flash frozen and stored at −80°C. Wild-type or TPR Jurkat cells were spinfected for two hours at 666 x g. Two days later, 2 μg/mL puromycin was added until the non-transduced cells were all dead (2–4 days). To test base editor delivery molecules, only the HEK3 site-targeting sgRNA was used, and electroporation was performed using Lonza Nucleofection with the SE or P3 nucleofection solution. pCMV-BE4max was used for plasmid DNA. Base editor mRNAs were designed generated by in vitro transcription using the HiScribe T7 High-Yield RNA synthesis kit (NEB Cat No. E2040S) . NEBnext polymerase was used to PCR-amplify template plasmids and install a functional T7 promoter and a 120 nucleotide polyadenine tail. Transcription reactions were set up with complete substitution of uracil by N1-methylpseudouridine (Trilink BioTechnologies Cat No. N-1080) and co-transcriptional 5’ capping with the CleanCap AG analog (Trilink BioTechnologies Cat No. N-7113) to generate a 5’ Cap1 structure. mRNAs were purified using ethanol precipitation according to kit instructions, dissolved in nuclease-free water, and normalized to a concentration of 2 micrograms per microliter using Nanodrop RNA quantification of diluted test samples. Base editing with in vitro transcribed guides and ABE8e protein sgRNAs were transcribed in vitro using the EnGen sgRNA Synthesis kit (NEB) with oligonucleotides containing the protospacers . The sgRNAs were then purified (NEB Monarch RNA Cleanup kit) and quantified by nanodrop. 3 μg of purified sgRNA were then mixed with 1 μL of 90 μM ABE8e protein in Lonza P3 electroporation buffer in a total reaction volume of 10 μL and incubated at room temperature to allow ribonucleoprotein complex to form. TPR Jurkat cells were collected, spun, washed twice with 37°C PBS, and 2e5 cells were resuspended per cuvette well in Lonza electroporation buffer P3 with the ABE8e and sgRNA ribonucleoprotein complex to a total reaction volume of 20 μL. Electroporation and cell recovery was performed according to the manufacturer’s instructions. Two days post nucleofection, gDNA of the cells was extracted, and base editing efficiency of each population was determined through PCR amplification of the edited genomic region of interest and quantified through Sanger sequencing and EditR software analysis . To produce purely edited cell populations, single cell clones were isolated from the corresponding bulk edited cell populations by diluting 0.8 cells per well in 96-well plates. Once isolated, these cells were grown until confluence, then individually genotyped through extraction of its gDNA, PCR amplification of the edited genomic region of interest, and Sanger sequencing via EditR . Once single cell clones were validated by Sanger sequencing, 4–8 single cell clones were mixed together to avoid clone-specific effects. CD69 staining Cells activated as described above were washed once with cold Cell wash buffer (Biolegend) and incubated with 0.5 uL α-CD69-APC antibodies (Biolegend) at 4°C, wrapped in foil, for 20 minutes. Cells were washed twice with cold PBS and stored on ice prior to flow cytometry. Phosphosite library design and construction All phosphorylation sites determined through the accompanying phosphoproteomic analysis were filtered for phosphosites that were localized in the peptide with a confidence of 90% or greater. For all genes with phosphosites, we used an in-house base editor design tool to first design all possible guides targeting these genes. All adenines in the window of 4–8 of the sgRNA (where 1 is the most PAM-distal position, and positions 21–23 are the PAM) were considered to be edited. sgRNAs with cloning sites, poly Ts and greater than five perfect matches in the genome were excluded. sgRNAs targeting the phosphosite of interest were then picked. sgRNAs predicted to make silent edits at phosphosites but also predicted to have bystander edits in the window of 4–8 were excluded. We then included 250 non-targeting controls, 250 intergenic controls, 250 controls targeting splice sites of essential genes and 250 controls targeting splice sites of TCR genes. The code is available at https://github.com/mhegde/base-editor-design-tool . Proteome-wide base editor screen TPR Jurkat cells were spinfected in 4 µg/mL polybrene at 30°C for 45 minutes at 666 x g at a multiplicity of infection of 0.3, assuming 30% transduction efficiency, and maintained a 500x library coverage. Transduction quadruplicates were used for downstream replicates. Puromycin was added to 2 μg/mL and cells were selected for stable integrants for seven days. Three days after removal of puromycin, an aliquot of cells corresponding to 500x library coverage was saved as the pre-ABE8e inputs. Cells were washed with 37°C PBS twice, and resuspended in SE nucleofection reagent (Lonza) with 94 pmoles of ABE8e protein per well. Each well was 2e5 library-containing TPR cells in 19 µL of SE + 1 µL ABE8e protein, and all 16 wells were used simultaneously for a library coverage of 250x. Electroporation and cell recovery was performed according to the manufacturer’s instructions. Six days after ABE8e protein introduction, another 500x aliquot of cells was frozen for the post-ABE8e sample. 14 days after ABE8e protein introduction, the mutant library TPR Jurkat pool was washed in room temp PBS, and activated as described above. 2000x library coverage of TPR cells was prepared for FACS. FACS was performed on a Bigfoot Spectral Cell Sorter (ThermoFisher) and roughly 1e6 cells were collected per bin. The top one, and bottom two 12.5% bins were sorted. The two bottom bins (“bottom” and “low”) were sorted to avoid unactivated cells, which always was about 15% of all cells. Only the second lowest bin (“low”) was used for downstream analyses due to the superior performance of controls, and to ensure cells were activated. Collected cells were pelleted and stored at −80°C until further processing. Genomic DNA (gDNA) was isolated and PCR-amplified for barcode abundance determination as previously described . Standard Illumina adapters were added and stagger regions were introduced for base diversity, according to the protocol “sgRNA/shRNA/ORF PCR for Illumina Sequencing” (The Broad Institute GPP). Libraries were sequenced at LJI a depth of 10e6 reads or more. Differential sgRNA abundance analyses via MAGeCK and MAGeCKFlute Programs To identify the functional phosphorylation sites through mutational analyses, we analyzed raw reads of GFPhigh and GFPlow samples using the MAGeCK program (v0.5.9.5), used for analyzing CRISPR screens . To summarize the results of MAGeCK’s phosphosite and sgRNA data, we used MAGeCKFlute program (version 2.4.0) . Intergenic and non-targeting controls were used for normalization and size factor estimation. Before conducting the pathway enrichment and gene-centric GSEA analysis with MAGeCKFlute, we converted the phosphorylated and mutated gene products to their corresponding gene symbols. For site-centric analyses (PTM-SEA, Kinase Library), we used the MAGeCK output for ssGSEA2.0 R package with default settings . PTM-SEA included iKiP data but excluded the LINCS P100 terms for clarity. To investigate the putative kinases responsible for phosphorylating phosphosites abundant in post-base editing and NFAT signaling, we employed The Kinase Library Enrichment analysis, which offers an atlas of primary sequence substrate preferences for the human serine/threonine kinome, was used on the PhosphositePlus.org website . Data visualization is done using R 4.3.0, Graphpad Prism 10, Protigy (v1.1.7), iDEP 0.96. Dataset comparisons To compare the mass spectrometry-based phosphoproteomic data to the phenotypic screen results, we used the F statistic calculated during the moderated F-test as our variable. We reasoned that since we had multiple time points, where multiple pairwise comparisons could be made, the F statistic would capture the magnitude and reproducibility of a phosphopeptide’s abundance over time. Directionality is lost, though directionality may differ between any two different time point comparisons. To compare predicted functional scores, we downloaded the from Ochoa et al. 2020 and compared the column “functional_score” to the log2 fold change of the base editor screens calculated by MAGeCK. PHLPP1 localization Plasmids encoding the N-terminal extension (NTE) of PHLPP1 were HA tagged , and the appropriate codons were mutated using site-directed mutagenesis (Agilent). HeLa cells were transiently transfected with WT, NLS mutant, S118P, S118A, or S118E PHLPP1 NTEs, and after two days were fixed, and stained with anti-HA antibody(1:500 dilution) (Cell Signaling, 3742), Alexa Fluor-647-phalloidin (Invitrogen, A22285), and DAPI stains. Images were acquired using spinning disk confocal microscopy. Quantification was performed using individual cells and were statistically tested using a one-way ANOVA with Tukey’s multiple test corrections. Transcriptional profiling of phosphosite-mutant T cell lines The various phosphosite mutants were introduced into TPR Jurkat cells via in vitro transcribed sgRNAs described above. After single cell cloning 4–8 single cell clones were mixed together to avoid clone-specific effects. Cells were incubated and activated with α-CD3/CD28 agonist antibodies as described above, except that the antibody concentrations were 1 μg/mL. Cells were activated for 0 and six hours. Cells were stained with BioLegend TotalSeq-C Human Universal Cocktail and anti-human hashtag antibodies according to NYCG CITEseq protocols [ http://cite-seq.com/ ] and processed using 10X Genomics 5’ HT with Feature Barcode assay according to protocol, with the addition of selective transcript removal using Jumpcode Genomics CRISPRclean Single Cell Boost kit. Briefly, 100K cells from each phosphosite-mutant line per time point were blocked with BioLegend Human TruStain FcX Fc receptor blocking solution. Cells were then stained with BioLegend TotalSeq-C Human Universal Cocktail resuspended in BioLegend Cell Staining Buffer (CSB) at a concentration of 1 vial per 500K cells and anti-human hashtag antibodies at a concentration of 0.75 μg/1e6 cells for 30 minutes at 4°C. Cells were then washed 3x in CSB. Cell viability and concentration was assessed using the Moxi Go II and Moxi Cyte Viability Reagent containing propidium iodide (PI). Cells were pelleted and resuspended in 0.04% BSA in PBS for a final concentration of 1300–1600 cells/μl and processed using 10X Genomics Next GEM Single Cell 5’ HT v2 assay. Gene Expression (GEX) and Cell Surface Protein (CSP) libraries were constructed according to protocol (CG000424 Rev D) with the following deviation: Post-ligation product of the GEX library was subjected to Jumpcode Genomics CRISPRclean Single Cell Boost kit following protocol. Libraries were sequenced at a targeted depth of 50K reads/cell for GEX and 5k reads/cell for CSP on an Illumina NovaSeq 6000 and an Element Aviti. Cellranger count v7.1.0 was used to generate cloupe files for analysis. Data was analyzed using the Loupe Browser (10X Genomics). For , HEK3 , MAPK1 , LCP2 , and PHLPP1 phosphosite-mutant cells were selected for zero and six hours post-activation, and differential gene expression (log2 fold change) was determined using local (sample specific) expression. Genes with a p value less or equal to 0.1 were determined to be regulated, as recommended by the software. For , differentially expressed genes were determined using only HEK3 , MAPK1 , and PHLPP1 cells, though LCP2 terminating edit cells’ log2 fold change values were included in the plot for context. For , genes were selected from the plot in 5D through their known involvement in T cell signaling. The .json file associated with this analysis, , can be used to explore these data in Morpheus. Granzyme B staining 3.5e5 cells were incubated with fixable viability dye prior to fixation and permeabilization using Ghost Dye UV 450 (Cytek) at a 1:100 dilution, in a total volume of 50μL, for 15 minutes in PBS containing 2% FBS (PBS 2%) at 4°C and then washed once in the same medium at 500g for 4 min. Cells were fixed in 50 μL of BD Cytofix during 20 min at 4°C and then washed in PBS 2% at 500g for 4 min. After, cells were fixed and permeabilized using the eBioscience Foxp3 / Transcription Factor Staining Buffer Set (Thermo Fisher) for 40 minutes at 4°C and then washed in perm/wash buffer at 500g for 4 min. Expression of granzyme b was evaluated using the Granzyme B- A647 monoclonal antibody (clone GB11, BD biosciences) at a 1:100 dilution, in a total volume of 50μL. Cells were incubated 1 hour at RT followed by a 4°C incubation overnight. Cells were washed in perm/wash at 500g for 4 min and then resuspended for analysis. Analyses were performed on LSRII cytometer (BD Biosciences). 10,000 events were recorded and data analyses were performed in FlowJo software (Tree Star, Ashland, OR). Jurkat E6.1 and HEK293T cells were purchased through ATCC. Triple parameter reporter (TPR) Jurkat cells were purchased from Professor Peter Steinberger at the Medical University of Vienna. TPR Jurkat cells and E6.1 Jurkat cells were cultured and passaged in Roswell Park Memorial Institute medium 1640 (RPMI-1640) plus GlutaMax (Thermo Fisher Scientific) supplemented with 10% heat inactivated fetal bovine serum. Cells were passaged and maintained at cell densities between 1–5e5 cells per mL. HEK293T cells were cultured and passaged in Dulbecco’s Modified Eagle Medium (DMEM) supplemented with 10% heat inactivated fetal bovine serum. All cells were incubated and maintained at 37°C with 5% CO2. Jurkat E6.1 cells (ATCC) were activated in 96 well tissue culture plates for the stated times at 2e5 cells/mL. Plates were coated with 3.33 μg/mL α-CD3 (HITa3) and α-CD28 antibodies (Biolegend), in 100 μL PBS at 4°C overnight, followed by one cold PBS wash. To stop the reaction, the cells were transferred to ice cold PBS, washed twice, flash frozen, and stored at −80°C until processing. Phosphoproteomic analysis was performed as previously described . Data was analyzed using Spectrum Mill (Agilent and Broad Institute) for phosphopeptide identification and quantification. The TMT denominator for each sample was the mean of all TMT channels. After global median normalization and median absolute deviation scaling, a moderated F test (limma, R) was performed to identify “regulated” phosphopeptide levels across the time series. Statistically significant phosphopeptides (Benjamini-Hochberg adj. p value <0.05) were visualized using Morpheus ( https://software.broadinstitute.org/morpheus/ ). The .json file associated with this manuscript, , can be used to explore these data in Morpheus. Recombinant ABE8e was expressed and purified as previously described . Briefly, ABE8e was expressed as an 8xHis tagged protein from a rhamnose-inducible promoter in BL21-Star DE3 cells with low RNase activity (Thermo Fisher Scientific). At an OD600 of ~0.8, the 2X terrific broth Escherichia coli cultures were cold shocked on ice for one hour, and induced with 0.8% final concentration of rhamnose. Roughly 24 hours later, cells were lysed via lysozyme and sonication, and ABE8e protein was purified on Ni-NTA resin. After imidazole elution, ABE8e protein was further purified using cation exchange. Fractions were monitored by UV and SDS-PAGE. ABE8e protein containing fractions were pooled, concentrated to ~90 pmol/μL using 100 MW cutoff filters, aliquoted, flash frozen, and stored at −80°C. Oligonucleotides containing the protospacer were chosen by hand or from our original bioinformatic analysis, and ordered from IDT DNA technologies. CACCG was added to the forward oligo, where AAAC was added to the reverse complement of the protospacer sequence. A C on the 3’ end of the reverse complement oligo was also appended. pRDA118 (Addgene Plasmid # 133459) was digested with BsbmI_V2 (NEB) and FastAP (Thermo) for the last five minutes, followed by gel purification. Oligos were mixed 1:1 at 100 μM, phosphorylated via PNK (NEB) and annealed after a 5 minutes 95°C step at 5°C per five minutes until 25°C. Annealed oligos were diluted 1:200 and 1 μL was mixed with 25–50 ng of digested backbone. T4 ligation (NEB) was performed for 20 minutes at 37°C and the 5–10 μL ligation reaction was transformed into Stbl3 E. coli , made in house. Colonies were verified via Sanger sequencing. One μg of sgRNA plasmids were transfected with Lipofectamine 3000 (Invitrogen) 1 μg of pPAX2 and 0.1 μg VSVG into HEK293Ts seeded the night before at 2e5 cells per 6 well. After three hours the 2.5 mL media was replaced with 5 mLs media supplemented with 1% BSA. Supernatants were collected after 3 days, filtered to remove HEK293T cells, concentrated 10-fold with Lenti-X precipitation solution (Alstem), and aliquots were flash frozen and stored at −80°C. Wild-type or TPR Jurkat cells were spinfected for two hours at 666 x g. Two days later, 2 μg/mL puromycin was added until the non-transduced cells were all dead (2–4 days). To test base editor delivery molecules, only the HEK3 site-targeting sgRNA was used, and electroporation was performed using Lonza Nucleofection with the SE or P3 nucleofection solution. pCMV-BE4max was used for plasmid DNA. Base editor mRNAs were designed generated by in vitro transcription using the HiScribe T7 High-Yield RNA synthesis kit (NEB Cat No. E2040S) . NEBnext polymerase was used to PCR-amplify template plasmids and install a functional T7 promoter and a 120 nucleotide polyadenine tail. Transcription reactions were set up with complete substitution of uracil by N1-methylpseudouridine (Trilink BioTechnologies Cat No. N-1080) and co-transcriptional 5’ capping with the CleanCap AG analog (Trilink BioTechnologies Cat No. N-7113) to generate a 5’ Cap1 structure. mRNAs were purified using ethanol precipitation according to kit instructions, dissolved in nuclease-free water, and normalized to a concentration of 2 micrograms per microliter using Nanodrop RNA quantification of diluted test samples. in vitro transcribed guides and ABE8e protein sgRNAs were transcribed in vitro using the EnGen sgRNA Synthesis kit (NEB) with oligonucleotides containing the protospacers . The sgRNAs were then purified (NEB Monarch RNA Cleanup kit) and quantified by nanodrop. 3 μg of purified sgRNA were then mixed with 1 μL of 90 μM ABE8e protein in Lonza P3 electroporation buffer in a total reaction volume of 10 μL and incubated at room temperature to allow ribonucleoprotein complex to form. TPR Jurkat cells were collected, spun, washed twice with 37°C PBS, and 2e5 cells were resuspended per cuvette well in Lonza electroporation buffer P3 with the ABE8e and sgRNA ribonucleoprotein complex to a total reaction volume of 20 μL. Electroporation and cell recovery was performed according to the manufacturer’s instructions. Two days post nucleofection, gDNA of the cells was extracted, and base editing efficiency of each population was determined through PCR amplification of the edited genomic region of interest and quantified through Sanger sequencing and EditR software analysis . To produce purely edited cell populations, single cell clones were isolated from the corresponding bulk edited cell populations by diluting 0.8 cells per well in 96-well plates. Once isolated, these cells were grown until confluence, then individually genotyped through extraction of its gDNA, PCR amplification of the edited genomic region of interest, and Sanger sequencing via EditR . Once single cell clones were validated by Sanger sequencing, 4–8 single cell clones were mixed together to avoid clone-specific effects. Cells activated as described above were washed once with cold Cell wash buffer (Biolegend) and incubated with 0.5 uL α-CD69-APC antibodies (Biolegend) at 4°C, wrapped in foil, for 20 minutes. Cells were washed twice with cold PBS and stored on ice prior to flow cytometry. All phosphorylation sites determined through the accompanying phosphoproteomic analysis were filtered for phosphosites that were localized in the peptide with a confidence of 90% or greater. For all genes with phosphosites, we used an in-house base editor design tool to first design all possible guides targeting these genes. All adenines in the window of 4–8 of the sgRNA (where 1 is the most PAM-distal position, and positions 21–23 are the PAM) were considered to be edited. sgRNAs with cloning sites, poly Ts and greater than five perfect matches in the genome were excluded. sgRNAs targeting the phosphosite of interest were then picked. sgRNAs predicted to make silent edits at phosphosites but also predicted to have bystander edits in the window of 4–8 were excluded. We then included 250 non-targeting controls, 250 intergenic controls, 250 controls targeting splice sites of essential genes and 250 controls targeting splice sites of TCR genes. The code is available at https://github.com/mhegde/base-editor-design-tool . TPR Jurkat cells were spinfected in 4 µg/mL polybrene at 30°C for 45 minutes at 666 x g at a multiplicity of infection of 0.3, assuming 30% transduction efficiency, and maintained a 500x library coverage. Transduction quadruplicates were used for downstream replicates. Puromycin was added to 2 μg/mL and cells were selected for stable integrants for seven days. Three days after removal of puromycin, an aliquot of cells corresponding to 500x library coverage was saved as the pre-ABE8e inputs. Cells were washed with 37°C PBS twice, and resuspended in SE nucleofection reagent (Lonza) with 94 pmoles of ABE8e protein per well. Each well was 2e5 library-containing TPR cells in 19 µL of SE + 1 µL ABE8e protein, and all 16 wells were used simultaneously for a library coverage of 250x. Electroporation and cell recovery was performed according to the manufacturer’s instructions. Six days after ABE8e protein introduction, another 500x aliquot of cells was frozen for the post-ABE8e sample. 14 days after ABE8e protein introduction, the mutant library TPR Jurkat pool was washed in room temp PBS, and activated as described above. 2000x library coverage of TPR cells was prepared for FACS. FACS was performed on a Bigfoot Spectral Cell Sorter (ThermoFisher) and roughly 1e6 cells were collected per bin. The top one, and bottom two 12.5% bins were sorted. The two bottom bins (“bottom” and “low”) were sorted to avoid unactivated cells, which always was about 15% of all cells. Only the second lowest bin (“low”) was used for downstream analyses due to the superior performance of controls, and to ensure cells were activated. Collected cells were pelleted and stored at −80°C until further processing. Genomic DNA (gDNA) was isolated and PCR-amplified for barcode abundance determination as previously described . Standard Illumina adapters were added and stagger regions were introduced for base diversity, according to the protocol “sgRNA/shRNA/ORF PCR for Illumina Sequencing” (The Broad Institute GPP). Libraries were sequenced at LJI a depth of 10e6 reads or more. To identify the functional phosphorylation sites through mutational analyses, we analyzed raw reads of GFPhigh and GFPlow samples using the MAGeCK program (v0.5.9.5), used for analyzing CRISPR screens . To summarize the results of MAGeCK’s phosphosite and sgRNA data, we used MAGeCKFlute program (version 2.4.0) . Intergenic and non-targeting controls were used for normalization and size factor estimation. Before conducting the pathway enrichment and gene-centric GSEA analysis with MAGeCKFlute, we converted the phosphorylated and mutated gene products to their corresponding gene symbols. For site-centric analyses (PTM-SEA, Kinase Library), we used the MAGeCK output for ssGSEA2.0 R package with default settings . PTM-SEA included iKiP data but excluded the LINCS P100 terms for clarity. To investigate the putative kinases responsible for phosphorylating phosphosites abundant in post-base editing and NFAT signaling, we employed The Kinase Library Enrichment analysis, which offers an atlas of primary sequence substrate preferences for the human serine/threonine kinome, was used on the PhosphositePlus.org website . Data visualization is done using R 4.3.0, Graphpad Prism 10, Protigy (v1.1.7), iDEP 0.96. To compare the mass spectrometry-based phosphoproteomic data to the phenotypic screen results, we used the F statistic calculated during the moderated F-test as our variable. We reasoned that since we had multiple time points, where multiple pairwise comparisons could be made, the F statistic would capture the magnitude and reproducibility of a phosphopeptide’s abundance over time. Directionality is lost, though directionality may differ between any two different time point comparisons. To compare predicted functional scores, we downloaded the from Ochoa et al. 2020 and compared the column “functional_score” to the log2 fold change of the base editor screens calculated by MAGeCK. Plasmids encoding the N-terminal extension (NTE) of PHLPP1 were HA tagged , and the appropriate codons were mutated using site-directed mutagenesis (Agilent). HeLa cells were transiently transfected with WT, NLS mutant, S118P, S118A, or S118E PHLPP1 NTEs, and after two days were fixed, and stained with anti-HA antibody(1:500 dilution) (Cell Signaling, 3742), Alexa Fluor-647-phalloidin (Invitrogen, A22285), and DAPI stains. Images were acquired using spinning disk confocal microscopy. Quantification was performed using individual cells and were statistically tested using a one-way ANOVA with Tukey’s multiple test corrections. The various phosphosite mutants were introduced into TPR Jurkat cells via in vitro transcribed sgRNAs described above. After single cell cloning 4–8 single cell clones were mixed together to avoid clone-specific effects. Cells were incubated and activated with α-CD3/CD28 agonist antibodies as described above, except that the antibody concentrations were 1 μg/mL. Cells were activated for 0 and six hours. Cells were stained with BioLegend TotalSeq-C Human Universal Cocktail and anti-human hashtag antibodies according to NYCG CITEseq protocols [ http://cite-seq.com/ ] and processed using 10X Genomics 5’ HT with Feature Barcode assay according to protocol, with the addition of selective transcript removal using Jumpcode Genomics CRISPRclean Single Cell Boost kit. Briefly, 100K cells from each phosphosite-mutant line per time point were blocked with BioLegend Human TruStain FcX Fc receptor blocking solution. Cells were then stained with BioLegend TotalSeq-C Human Universal Cocktail resuspended in BioLegend Cell Staining Buffer (CSB) at a concentration of 1 vial per 500K cells and anti-human hashtag antibodies at a concentration of 0.75 μg/1e6 cells for 30 minutes at 4°C. Cells were then washed 3x in CSB. Cell viability and concentration was assessed using the Moxi Go II and Moxi Cyte Viability Reagent containing propidium iodide (PI). Cells were pelleted and resuspended in 0.04% BSA in PBS for a final concentration of 1300–1600 cells/μl and processed using 10X Genomics Next GEM Single Cell 5’ HT v2 assay. Gene Expression (GEX) and Cell Surface Protein (CSP) libraries were constructed according to protocol (CG000424 Rev D) with the following deviation: Post-ligation product of the GEX library was subjected to Jumpcode Genomics CRISPRclean Single Cell Boost kit following protocol. Libraries were sequenced at a targeted depth of 50K reads/cell for GEX and 5k reads/cell for CSP on an Illumina NovaSeq 6000 and an Element Aviti. Cellranger count v7.1.0 was used to generate cloupe files for analysis. Data was analyzed using the Loupe Browser (10X Genomics). For , HEK3 , MAPK1 , LCP2 , and PHLPP1 phosphosite-mutant cells were selected for zero and six hours post-activation, and differential gene expression (log2 fold change) was determined using local (sample specific) expression. Genes with a p value less or equal to 0.1 were determined to be regulated, as recommended by the software. For , differentially expressed genes were determined using only HEK3 , MAPK1 , and PHLPP1 cells, though LCP2 terminating edit cells’ log2 fold change values were included in the plot for context. For , genes were selected from the plot in 5D through their known involvement in T cell signaling. The .json file associated with this analysis, , can be used to explore these data in Morpheus. 3.5e5 cells were incubated with fixable viability dye prior to fixation and permeabilization using Ghost Dye UV 450 (Cytek) at a 1:100 dilution, in a total volume of 50μL, for 15 minutes in PBS containing 2% FBS (PBS 2%) at 4°C and then washed once in the same medium at 500g for 4 min. Cells were fixed in 50 μL of BD Cytofix during 20 min at 4°C and then washed in PBS 2% at 500g for 4 min. After, cells were fixed and permeabilized using the eBioscience Foxp3 / Transcription Factor Staining Buffer Set (Thermo Fisher) for 40 minutes at 4°C and then washed in perm/wash buffer at 500g for 4 min. Expression of granzyme b was evaluated using the Granzyme B- A647 monoclonal antibody (clone GB11, BD biosciences) at a 1:100 dilution, in a total volume of 50μL. Cells were incubated 1 hour at RT followed by a 4°C incubation overnight. Cells were washed in perm/wash at 500g for 4 min and then resuspended for analysis. Analyses were performed on LSRII cytometer (BD Biosciences). 10,000 events were recorded and data analyses were performed in FlowJo software (Tree Star, Ashland, OR). Supplementary Data 1 Supplementary Data 2 Supplementary Tables |
Dr. Google to Dr. ChatGPT: assessing the content and quality of artificial intelligence-generated medical information on appendicitis | d9d5e5d0-d31b-4d20-b989-2e93be331cc3 | 11078845 | Patient Education as Topic[mh] | Four widely used AI chatbots, ChatGPT-3.5 and ChatGPT-4 (OpenAI, San Francisco, California), Bard (Google, Mountain View, California), and Claude-2 (Anthropic, San Francisco, California), were queried to generate content about appendicitis. ChatGPT-3.5, Bard, and Claude-2 are all freely available, while ChatGPT-4 requires a subscription to use. Each chatbot was prompted using four standardized, sequenced questions on September 9th, 2023: “Please tell me about appendicitis” “Please tell me more” “What are the side effects and complications of the treatments?” “Please provide a list of sources” To assess quality and accuracy of the AI-generated text, we utilized a modified DISCERN instrument, which is an existing validated tool used to judge the quality of written consumer health information . Like the original DISCERN instrument, our modified tool includes 16 criteria, each scored on a 5-point Likert scale with a maximum score of 80. Criteria assessed include accuracy, listing disclosures, listing sources, verifying sources, balance and bias, areas of uncertainty, describing the condition, its diagnosis and treatment, benefits and risks of each treatment, risk of nontreatment, recommendation of expert medical opinion, presence of major errors that could cause harm, and overall professionalism/tone (Fig. ). Using the modified DISCERN tool, three investigators independently scored the generated texts blinded to the identity of the AI platforms. Prior to scoring, all three investigators comprehensively reviewed UpToDate and the American College of Surgeons information sheet on appendicitis to establish a standard of high-quality medical information for scoring AI-generated content . Inter-rater reliability for quality scores between investigators was calculated using Cohen’s kappa. These AI-generated texts were then assessed for readability via an online application ( https://readable.com , Added Bytes Ltd., Brighton, England) using two validated readability measures: Flesch Reading Ease (FRE) and Flesch Kincaid Grade Level (FKGL) . These readability measures are utilized commonly in health literacy literature . Their formulas are as follows : [12pt]{minimal}
$${}\, = \,206.835 - 1.015\,( {}\,{}\,{}\,{}}}{{{}\,{}\,{}\,{}}}} ) - \,84.6( {}\,{}\,{}\,{}}}{{{}\,{}\,{}\,{}}}} )$$ FRE score = 206.835 - 1.015 total number of words total number of sentences - 84.6 total number of syllables total number of words [12pt]{minimal}
$${}\,\, = \,\,0.39\,( {}\,{}\,{}\,{}}}{{{}\,{}\,{}\,{}}}} )\, + \,11.8( {}\,{}\,{}\,{}}}{{{}\,{}\,{}\,{}}}} )\, - \,15.59$$ FKGL = 0.39 total number of words total number of sentences + 11.8 total number of syllables total number of words - 15.59 The FRE score ranges from 0 to 100 and corresponds to an American educational level, with lower scores indicating more difficult reading material (0–10, extremely difficult/professional level; 10–30, very difficult/college graduate level; 30–50, difficult/college level; 50–60, fairly difficult/10th- to 12th-grade level; 60–70, plain English/8th- to 9th-grade level; 70–80, fairly easy to read/7th-grade level; 80–90, easy to read or conversational English/6th-grade level; 90–100, very easy to read/5th-grade level) . FKGL ranges from 0 to 18, and the score indicates the number of years of education required to comprehend the text . Thus, healthcare education material with FRE scores greater than 80, and FKGL scores of 7 or lower, would correspond with the recommended AMA and NIH readability levels . The primary endpoints of this study were the absolute difference in the quality (modified DISCERN) scores, and the difference in readability scores (FKGL and FRE) between the AI platforms tested. Mean quality scores were recorded and used for analysis. The present study was exempt from institutional review as it did not involve human subjects. Continuous variables were displayed as means ± standard deviation (SD) or medians with interquartile range if non-parametric. One-way ANOVA with Bonferroni correction was performed to evaluate differences in quality scores between the AI chatbots. A two-sided p-value ≤ 0.05 was considered statistically significant. Statistical analyses were performed using Stata version 18.0 (StataCorp, College Station, TX, USA).
ChatGPT-3.5, ChatGPT-4, Bard, and Claude-2 achieved overall mean (SD) quality scores of 60.7 (1.2), 62.0 (1.0), 62.3 (1.2), and 51.3 (2.3), respectively (Fig. ). Inter-rater reliability was 0.81, 0.75, 0.81, and 0.72, respectively, indicating substantial to near perfect agreement between investigators. Claude-2 demonstrated a significantly lower mean quality score compared to ChatGPT-4 ( p < 0.001), ChatGPT-3.5 ( p < 0.001), and Bard ( p < 0.001). There was no significant difference in mean quality scores between ChatGPT-3.5, ChatGPT-4, and Bard (Fig. ). When comparing specific criteria, Bard was the only AI platform that listed verifiable sources, while Claude-2 provided fabricated sources. ChatGPT-3.5 and GPT-4 did not provide any sources. ChatGPT-4 demonstrated higher quality scores on the bias criterion compared to Bard ( p < 0.001) and ChatGPT-3.5 ( p = 0.006), while Claude-2 scored higher than Bard ( p = 0.006) (Table ). ChatGPT-4 is the only chatbot that referred to the long-term failure rate of non-surgical treatment of appendicitis. None of the chatbots made any disclosures. However, all four chatbots discussed the risks of not treating appendicitis and provided accurate medical content regarding etiology, symptoms, and treatment with no major factual errors. All chatbots except for Claude-2 advised readers to consult a physician if experiencing symptoms. Regarding readability, FKGL and FRE scores of ChatGPT-3.5, ChatGPT-4, Bard, and Claude-2 were 14.6 and 23.8, 11.9 and 33.9, 8.6 and 52.8, 11.0 and 36.6, respectively (Fig. ). All scores indicated difficulty readability, ranging from a high school student to college graduate reading skill level. Text generated by Bard had the lowest reading difficulty, ranging between 8th and 10th grade reading level.
AI-powered LLMs have been the subject of medical research since their release to the public in November 2022. While healthcare workers have assessed AI for many uses, AI-powered tools, including chatbots, have become widely available to patients. For example, Woebot is a mental health AI-powered chatbot that uses cognitive behavioral therapy techniques to help patients manage anxiety, depression, and other mental health conditions. No such conversational tools exist for surgery yet, but existing heavily developed AI-chatbots such as ChatGPT, Claude-2, and Bard, offer appealing and adaptable choices for patients who use the internet. Recent literature suggests 74% of laypersons use the internet to search about health-related issues . Thus far, multiple studies have assessed the accuracy and quality of medical information generated by AI about bariatric surgery, hepatology, and ophthalmology [ – ]. Other studies assessed the ability of LLMs in clinical settings, for example in diagnosing clinical vignettes, and in medical education . To date, no study has compared the capability of multiple widely available AI platforms to generate medical information about appendicitis, suggesting the present study to be relevant to healthcare systems weighing the value of utilizing these tools for patient education on this common medical condition. In this study, we assessed the quality and readability of medical content about appendicitis produced by four AI platforms: ChatGPT-3.5, ChatGPT-4, Bard, and Claude-2. Claude-2 created content of lower quality than the rest, with significant concerns including fabricated sources and failing to advise readers to seek professional medical care if experiencing symptoms of acute appendicitis. Additionally, all four AI chatbots produced text with difficult readability at the college student reading skill level, which is well beyond the recommended 6th-grade level by the AMA and NIH. Our findings corroborate prior findings by Samaan et al., who demonstrated good overall quality when assessing ChatGPT-3.5’s information about bariatric surgery . Samaan et al. gathered questions from patient support groups such as Facebook, modified them into suitable prompts, then asked ChatGPT-3.5 a total of 151 questions. The generated text was assessed for accuracy and reproducibility by two bariatric surgeons, with only moderate agreement. In their analysis, ChatGPT-3.5 provided comprehensive responses to over 85% of questions asked about bariatric surgery eligibility, efficacy, procedure options, preoperative preparation, recovery, risks, and lifestyle changes. This indicates generally adequate detail about those procedures for patients seeking additional information before deciding to pursue bariatric surgery as a weight-loss option. We found similar results when prompting the tested LLMs about appendicitis and its etiology, symptoms, and treatment. However, our study interrogated four different AI-platforms, using general questions limited to appendicitis. Unlike bariatric surgery, most appendicitis cases do not require life-long follow-up and support, therefore patients would not usually have additional questions after the acute and recovery phases. Samaan et al. did not assess the readability of generated answers and did not ask the AI-platform to provide a list of sources. Similarly, Yeo et al. assessed information generated by ChatGPT-3.5 about hepatology topics such as cirrhosis and hepatocellular carcinoma, but from a physician’s point of view . They found that ChatGPT-3.5 offers extensive knowledge, but unlike bariatric surgery, was comprehensive in less than half of the questions asked. Additionally, it lacked the ability to specify decision-making cut-offs (such as MELD-Na scores) and treatment durations and was deficient in its knowledge about regional guideline variations, such as hepatocellular carcinoma screening criteria. They did note that it provided good advice to patients and caregivers. This suggests that AI-platforms in their current form could be more valuable to patients and their caregivers than physicians. Yeo et al.’s grading only assessed accuracy and reproducibility and did not measure readability or ask for sources. Momenaei et al. assessed ChatGPT-4 generated information about surgical treatment of retinal diseases . Of note, ChatGPT-4 use requires a subscription, unlike GPT-3.5. A list of common questions about definition, prevalence, visual impact, diagnostic methods, surgical and nonsurgical treatment options, postoperative information, surgery-related complications, and visual prognosis were curated. ChatGPT-4 was graded for appropriateness and readability was evaluated using FKGL and FRE scores. Nearly 90% of questions asked had appropriate responses overall. The average FKGL and FRE scores were between 14–15 and 28–35, respectively, indicating difficult or very difficult readability, at the level of a college graduate. Our investigation yielded similar results for ChatGPT-4 when asked about appendicitis, with FKGL and FRE scores of 11.9 and 33.9, respectively. Momenaei et al. did not ask ChatGPT-4 to generate sources, but their results mirror our study’s results, indicating good overall quality in answering questions about surgical specialties with difficult readability, but Yeo et al.’s results show further improvement is needed regarding specific details and cut-offs, especially in non-surgical topics. Our study highlights the potential these AI-LLMs have in providing patients with accurate information about a common health condition. Three out of the four tested AI-platforms scored favorably, suggesting they are not far away from being useful and a valid option for educating patients. However, progress must be made to make it clear to the reader that those AI-platforms cannot replace trained professionals, with disclosures reminding the reader that they are trained artificial intelligence large language models, with no innate comprehension of the world. This is an overlooked area that has been highlighted recently, as these AI platforms are impressionable by the prompt or information fed through them and can be influenced to produce biased but believable information . Thus, caution must be exercised when using those platforms. None of the aforementioned studies looked into this area when assessing these AI-platforms. Readability is another area that requires additional progress. The AMA and NIH recommend patient educational material to be at or below a sixth grade reading level to be accessible to the public . As highlighted by our study and Momenaei et al.’s study, these AI platforms produce content that is difficult to read in a wide variety of topics, not just gastrointestinal surgery . This in not unexpected, as these AI-platforms are designed to produce content at the level of an average high school student. Improving this area will facilitate health literacy significantly. Readability could potentially be improved if the original prompt specified that generated text must be at the 5–6th grade reading level. This has been previously studied. For example, Moons et al. prompted ChatGPT-3.5 and Bard to reduce the reading levels of existing patient educational materials . However, it is unlikely that this is how a lay person may use the AI platforms when searching for information, therefore it is less relevant in our study. Some platforms such as ChatGPT-4 can be programed by the user to adhere to a certain set of preferences. As such, they can be set by the user to only produce text with easy readability. Again, this may not be obvious or known to the average user of the AI platform. In fact, this skill is an emerging area of study coined “Prompt Engineering,” and requires some introduction at a minimum, for the end user to fully utilize the versatile nature of the AI platforms available . Regarding the scoring system used to assess the quality of generated texts, the DISCERN instrument is well validated in the literature, having been used to assess patient-directed healthcare information in hundreds of previously published studies . To adapt DISCERN to evaluate AI-generated text, we modified its criteria to include verifiability of sources and the tone of the produced material. Recent literature has revealed some AI platforms to fabricate otherwise highly believable sources, with convincing titles, authors lists, journals, and even Digital Object Identifier numbers that cannot be verified anywhere in the literature . Moreover, one study has shown the capabilities of AI chatbots in generating authentic-appearing abstracts of fictitious conditions using scientific jargon, and scientists were only able to identify 68% of those abstracts. Thus, it was crucial to include this area in our critical analysis of generated medical content to shed light on this phenomenon . This study was not without limitations. Recommendations and guidelines vary based on region and patient variables such as comorbidities. The AI-platforms tested did not elicit those differences, thus the information generated may not be relevant to all patients. Additionally, these AI-platforms are updated regularly. Minor updates occur without notice. This creates an element of inter-study and temporal variation that cannot be controlled for. One update could generate a technical glitch that would then be remedied in an update. Similarly, one version of the platform could outperform an updated version if the platform becomes adaptable, such that a healthcare worker’s adapted version of the platform becomes more detailed and healthcare oriented, while a layperson’s generated content remains less detailed. Furthermore, researchers may also test those AI-platforms at peak hours and receive different responses compared to off-peak hours, so repeatability may be impacted. Regarding the scoring system employed, DISCERN is well-known and recognized, but there exists an urgent need for a validated AI-oriented scoring system to be developed to ensure comparability between studies. Finally, even after vetting these AI-platforms, it would be imperative that regulators and healthcare administration to closely monitor their output once they are used in real patient interactions.
The present study compares the medical information generated by four popular AI platforms about acute appendicitis. Our results reveal that ChatGPT-3.5, ChatGPT-4, and Bard produce content of higher quality than Claude-2, although Bard is the only potentially credible chatbot tested, as it was the only platform to list verifiable sources. However, all four platforms produced content too difficult for the public to comprehend, thereby limiting their use in their current form but highlighting gaps for future versions of AI platforms to fill.
|
Delayed diagnosis in inflammatory bowel disease: Time to consider solutions | f728d3ec-a538-44f4-a791-40578323c967 | 11438659 | Patient Education as Topic[mh] | Inflammatory bowel disease (IBD) is a chronic, relapsing gastrointestinal disorder that encompasses two distinct forms: Crohn’s disease (CD) and ulcerative colitis (UC). Its prevalence peaks in early adulthood and then stabilizes, but the age-standardized prevalence continues to increase over time. Compared with men, women have a higher prevalence of intestinal diseases, more deaths, and more disability-adjusted life years. Over the past half-century, the incidence of IBD has increased in industrialized countries. While it has now stabilized or declined in developed Western countries, the incidence is rising dramatically in newly industrialized developing countries, such as China. Physicians play a crucial role in promoting the early diagnosis of CD to prevent the progression of severe disease and complications. The gap between the onset of IBD and its formal diagnosis, often referred to as delayed diagnosis, can have significant consequences. In CD, delayed diagnosis can result in cumulative intestinal damage, fibrosis, and disability. In UC, early diagnosis and effective treatment can lower the long-term risk of colorectal cancer and reduce the need for surgery. The initial diagnosis of IBD is based on a combination of history, physical examination, laboratory tests, and endoscopic, histologic, and radiologic findings. There is no single definitive test to establish a diagnosis, and current diagnostic methods do not always allow for early detection and treatment. Diagnosing IBD is challenging, and patients often experience significant delays in diagnosis owing to subclinical inflammation, which can lead to poor clinical outcomes. The National Institute for Health and Care Excellence (NICE) guidelines emphasize the importance of early detection and diagnosis of IBD, stating that “delays in assessment and diagnosis can lead to adverse outcomes such as clinical complications and impaired quality of life”. Delays in the diagnosis of IBD are common. A recent systematic review found a median diagnostic delay of 2 to 26 months for adults with CD and 2 to 12 months for those with UC. In a manuscript published in the World Journal of Gastroenterology , Blüthner et al reported that among the first German cohort of adult IBD patients, those with CD exhibited a higher risk of diagnostic delay than those with UC. The study noted that the time to diagnosis largely depended on the physician, with disease-specific symptoms and available diagnostic methods helping to shorten this time. The reasons for the delay in diagnosing IBD are not fully understood, but one possible explanation is the nonspecific nature of its symptoms, which can easily be mistaken for more common conditions such as irritable bowel syndrome (IBS) or hemorrhoids. The longer delay in diagnosing CD compared with UC may be because UC lesions are confined to the colon and often trigger rectal bleeding, which prompts patients to seek medical attention sooner. In contrast, patients with CD may present with a broader range of nonspecific symptoms, such as abdominal pain or constipation, which are more commonly associated with other conditions that have overlapping symptoms. Early detection of IBD is crucial in primary care. To better manage the impact of delayed diagnosis, it is essential to understand the barriers to diagnosis and develop strategies to overcome them. Early and rapid diagnosis can improve the initial treatment, reduce the duration of illness, and enhance patient satisfaction.
When chronic inflammation begins in the affected bowel, it may take time before noticeable clinical signs of IBD appear. Consequently, there can be a significant interval between the onset of inflammation and the emergence of symptoms that enable a correct and timely diagnosis. This subclinical inflammation can be difficult to recognize and treat, sometimes leading to intestinal damage even before clinically active inflammation develops. If clinically active inflammation can be diagnosed before complications arise, it may be possible to control and treat the disease within the “window of opportunity,” when interventions are most likely to be successful. While the most common symptoms of IBD are chronic diarrhea, abdominal pain, weight loss, and rectal bleeding, up to 30% of patients with IBD may exhibit no symptoms. This makes diagnosing IBD challenging. Additionally, limited knowledge of IBD among the general public, non-specialized gastroenterologists, or general practitioners (GPs) can lead to misinterpretation of potential IBD signs and symptoms, resulting in significant delays in diagnosis and referral. Medical diagnostic errors or delays stem from a complex interplay of systemic, organizational, clinical, and patient interaction factors. Gastrointestinal symptoms account for approximately 10% of all primary care visits. However, in high-income countries, the prevalence of IBD is only 0.3%. Blüthner et al demonstrated that the delay in diagnosing CD is significantly longer and largely dependent on the physician compared with UC in the German population. Physician awareness plays a crucial role, and since more than one-third of primary care physicians struggle to confidently recognize the major symptoms of IBD, we recommend developing a strategy to train primary care physicians and raise awareness of IBD within medical school curricula. One medical school in New York has integrated IBD patient panels into its curriculum as a supplement to traditional lectures-an initiative that could enhance theoretical knowledge and understanding of IBD in later medical careers. At the same time, data on the prevalence of IBD are limited, which may be attributed to differences in healthcare systems, data collection methods, and definitions of time to diagnosis (including symptom onset and the initial visit). Collecting more detailed health-related data and categorizing meaningful outcomes, such as health-related complications, could provide a more consistent framework for addressing these systemic challenges. To reduce the delay in diagnosing IBD, it is crucial not only to enhance the diagnostic skills of primary care physicians and other clinicians but also to develop more innovative and accurate diagnostic methods. Clinically available diagnostic approaches may include fecal examination, colonoscopy, radiology, and capsule endoscopy. These tests can be uncomfortable, painful, and sometimes even dangerous for patients. Additionally, genetic and serologic tests are not sufficiently specific or sensitive to be useful for diagnosing IBD. Non-invasive alternatives can assist in diagnosing IBD, with fecal markers showing promise. Fecal lactoferrin and calprotectin can help differentiate between IBD and non-inflammatory diseases. Additionally, fecal calprotectin has proven to be a cost-effective screening tool for identifying patients with subclinical symptoms of IBD. These screening modalities are now commonly used and are supported by NICE. Gastrointestinal ultrasound also demonstrates high sensitivity and specificity for diagnosing CD and for the initial evaluation of patients with suspected UC. Additionally, it is inexpensive, safe, and does not require bowel preparation. Recent studies have explored advances in diagnostic modalities for IBD by focusing on volatile organic compounds released in body fluids, including feces, urine, and breath. Apical space gases in feces or urine can be analyzed via gas chromatography/mass spectrometry. Modeling based on these compounds could help distinguish CD and UC from IBS and healthy populations. Similar studies have found that headspace gases in urine can be used to diagnose IBS. While fecal samples are relatively easy to obtain, patients often dislike or resist collecting them, making urine testing an attractive alternative. Early data from breath samples also show promise. Non-invasive diagnosis of IBD is becoming a reality, helping to avoid patient discomfort and embarrassment while reducing associated risks. This approach can also lead to significant savings in both money and resources for healthcare providers. Additionally, advancements in artificial intelligence (AI) are making substantial progress in improving IBD diagnosis. AI is increasingly used to enhance the interpretation of endoscopic images and assess disease severity. For example, a convolutional neural network model can automatically detect the degree of erosions and ulcers in small bowel mucosa from capsule endoscopy images with an accuracy of 95.6%, sensitivity of 90.8%, and specificity of 97.1%. This capability can help determine the risk of bleeding in IBD mucosal lesions. AI technology is expected to significantly impact clinical practice and trials, as it will aid clinicians and help patients manage their diseases more effectively.
IBD is a common chronic intestinal disease that includes CD and UC, both of which can present with a wide range of clinical manifestations. Early diagnosis of IBD is crucial for effective patient care and prognosis, yet delays in diagnosis remain common. Such delays can result in intestinal damage, fibrosis, and dysfunction, and may increase the likelihood of requiring surgical intervention. The reasons for delayed diagnosis of IBD are not fully understood, but they may be related to the nonspecific nature of clinical symptoms, the limited number of diagnostic modalities available, and the need to enhance the diagnostic skills of GPs and some clinicians. Improving the personal competence of GPs and clinicians could also be a significant factor in addressing these delays. Existing tests, such as endoscopy, may cause embarrassment and pose risks for patients, whereas fecal calprotectin has shown promise as an effective diagnostic method. Recent studies have explored the use of volatile organic compounds released in body fluids, including feces, urine, and breath. Given the current limitations in improving the diagnosis of IBD, it is important to consider additional solutions to address and reduce delays in diagnosis.
|
Dental care during COVID‐19 pandemic: Follow‐up survey of experts' opinion | 23c8030a-8774-4474-92ac-fbcc5d466ba4 | 8444799 | Dental[mh] | INTRODUCTION Severe acute respiratory syndrome coronavirus 2 (SARS‐CoV‐2) can be transmitted through saliva and respiratory droplets, as well as by contact with contaminated surfaces (Chan et al., ; Li et al., ; Xu et al., ; Yu et al., ). Its airborne transmission route among the high number of pre‐ or asymptomatic patients poses a significant challenge in dental settings, where face masks cannot be worn by patients. Furthermore, the risk of coronavirus disease 2019 (COVID‐19) transmission in dental settings may be further increased by the production and spread of contaminated aerosol and splatter (Epstein et al., ; Izzetti et al., ). During the first wave of pandemic, the risk of nosocomial transmission as well as of initial shortage of appropriate personal protective equipment (PPE) led several European countries to temporarily suspend elective dental treatments (Coulthard, ; Gurzawska‐Comis et al., ; Luo et al., ). Infection control and prevention (ICP) protocols and recommendations were then developed to support dental health professionals returning to work after practice closures and restrictions, advising mitigation factors in particular for aerosol‐generating procedures (AGP; Clarkson et al., ; Kumbargere Nagraj et al., ). Irrespective of COVID‐19 individual risk assessment, emergency dental treatments were always recommended to be carried out. During the first wave of pandemic, a survey of experts was performed asking for their opinion on appropriate PPE as well as assessment of risk associated with dental care. Overall, transmission risk of COVID‐19 was assumed to be high in dental settings, especially for AGP. Thus, maximum protection (i.e., FFP2/FFP3 masks, caps, gowns, and face protection) was recommended by the wide majority of European experts in oral and maxillofacial surgery and oral surgery participating in our survey on dental care provision during the first wave, especially for AGP (Becker et al., ). However, with the progression of the pandemic, it became clear that dental care could not be postponed any longer. The availability of PPE and understanding of disease transmission improved over time. Therefore, continuation of elective treatment has been recommended by international organizations, except for suspected or confirmed COVID‐19 cases or in areas with very high prevalence of COVID‐19 (ECDC, ; World Health Organization, ). Several patient‐related measures have been proposed so far to limit COVID‐19 cross‐infection in dental settings (e.g., limitation of the number of accompanying people, social distancing in waiting areas, hand disinfection, and mouth rinses; Gurzawska‐Comis et al., ). Additionally, national guidelines have been published by most of the countries including recommendations on ICP during pandemic. Frequently, they followed international guidelines (e.g., WHO, ECDC) which proposed use of appropriate PPE for dental staff based on patients' risk assessment for COVID‐19 and the type of dental treatment (ECDC, ; World Health Organization, ). Furthermore, most national guidelines recommended the use of preprocedural mouth rinse with antiseptic agents, even though this measure is still lacking scientific evidence regarding its efficacy and benefits (Clarkson et al., ). As an adjunct, preprocedural real‐time polymerase chain reaction (PCR)‐based test for low‐risk asymptomatic patients attending AGP elective dental treatments might also be considered (Umer & Arif, ), while the utilization of preprocedural rapid serological tests should be discouraged due to the high frequency of false‐negative results (Tysiąc‐Miśta & Bulanda, ). In addition, it is still controversial whether real‐time PCR testing could be beneficial during a pandemic in dental settings. The use of fixed or mobile filter systems and ventilation protocols is still debatable. As most guidelines were produced during the first wave of pandemic and as there is still not enough scientific evidence, solid and updated recommendations on ICP are required. As long as there is still a lack of solid evidence, the experts' opinions should be accounted. Therefore, the primary aim of the present study was to collect updated opinions of European experts, who participated in the survey proposed by our group in April–May 2020 (Becker et al., ) on ICP in dental settings, PPE and additional measures to minimize COVID‐19 transmission risk. Secondary aim was to analyze how experts' opinion changed overtime, considering the new scientific information available and the experience gathered at the frontline since the outbreak of COVID‐19 pandemic.
MATERIAL AND METHODS The study protocol was submitted to and approved by the Ethics Committee of the University of Dusseldorf (protocol no. 2020‐926). The study was conducted according to the Declaration of Helsinki and the European Medicines Agency Guidelines for Good Clinical Practice. This study was also conducted and reported following the “Good practice in the conduct and reporting of survey research” criteria (Kelly et al., ). 2.1 Study population The present survey‐based study was conducted via a questionnaire distributed among 27 academic experts in Oral and Maxillofacial Surgery or Oral Surgery from different European countries, who had responded to our previous survey (Becker et al., ). An invitation was sent by email explaining the purpose of the present follow‐up study. A link to the consent form and online survey was reported as well. All experts were either based in one of the 27 European Union (EU) countries or within the following states with strong connection to EU: Iceland, Norway, Moldovia, Switzerland, and United Kingdom (UK; Becker et al., ). As no response was received from France, Slovakia, Hungary, Bulgaria, and Lithuania during the first study (Becker et al., ), these respective countries were excluded from the follow‐up survey. The participation in the survey was voluntary and without any incentive. All responders signed an informed consent form before accessing the questionnaire which was provided through an online survey platform (SurveyMonkey ® ). Data collection of the follow‐up survey took place from 23 November to 4 February 2021, whereas the initial survey had been performed from 12 April to 22 May 2020. Data were stored anonymously. 2.2 Questionnaire A 14‐item structured questionnaire was developed (Attachment 1). It was created based on a published questionnaire, previously completed by all the invited experts (Becker et al., ). In order to allow the comparison between data recorded in the two waves of pandemic, the first 10 items included minor modifications and covered the same areas as the first survey. In order to explore emerging topics of interest, 4 new items were added. Furthermore, the option to leave a comment on a question was added. The questionnaire was organized as follows: participant working environment (two items, single choice); infection risk of dental health professionals (two items, single choice); facilities for dental treatment of COVID‐19 positive patients (one item single choice); measures to prevent infection of health professionals and nosocomial transmission in dental clinics (two items, matrix/single choice per row); information about where dental care was provided for COVID‐19 positive patients or at high risk (one item, multiple select); recommended PPE for dental health professionals (two items, matrix/single choice per row); antiseptic mouth rinse (one item, single choice); COVID‐19 testing of dental staff (two items, single choice); recommended type of treatment provision during the second wave (one item, single choice). As in the previous survey (Becker et al., ), they were asked if they treated their patients in the private and/or university dental clinic settings (question 1) and information on the number of staff members working in their departments was also explored (question 2). In question 3 and 4 (scoring of the risk from AGP, and non‐AGP), participants had an option to leave a comment to explain their response. Question 5 was added, asking where patients infected with COVID‐19 (COVID+) should be treated. Question 8 on room‐related preventive measures was moved to question 6, and two options for response were added (high‐efficiency particulate air [HEPA] filters, high vacuum evacuators). Question 9 was moved to 7 (patient‐related measures), and antigen or PCR test prior to treatment was added as a new response. Question 10 was moved to 8 (facilities performing treatment of COVID‐19 patients in the respective countries), and now it was possible to leave a comment and state, whether one was actually involved in treatment of COVID‐19 patients. Questions 5 and 6 were shifted to questions 9 and 10 (protective measures for non‐AGP/AGP), and now also antigen test before treatment could be selected as a preventive measure. Additionally, instead using the classifying answers among recommended for patients of unknown risk/high risk/very high risk/not recommended , we now specified recommended for every patient/patient with COVID‐19 symptoms/not recommended . Questions 11–14 were added to the survey and included aspects that were in general not relevant during first wave of pandemic (when elective treatments were postponed in most countries), that is, virucidal mouth rinse, regular testing of staff, and continuation of elective treatments. 2.3 Statistical analysis The data analysis was performed using Microsoft Excel ® for Mac version 16.37 (Microsoft ® ) and an online survey tool (Survey Monkey ® ). For each question, the absolute number of votes and the relative agreement (%) were calculated. As only one group was surveyed (experts in oral surgery), no comparative analyses were performed per time point. To compare expert votes between first and second survey, chi‐square test was used. Results were found significant if p < .05.
Study population The present survey‐based study was conducted via a questionnaire distributed among 27 academic experts in Oral and Maxillofacial Surgery or Oral Surgery from different European countries, who had responded to our previous survey (Becker et al., ). An invitation was sent by email explaining the purpose of the present follow‐up study. A link to the consent form and online survey was reported as well. All experts were either based in one of the 27 European Union (EU) countries or within the following states with strong connection to EU: Iceland, Norway, Moldovia, Switzerland, and United Kingdom (UK; Becker et al., ). As no response was received from France, Slovakia, Hungary, Bulgaria, and Lithuania during the first study (Becker et al., ), these respective countries were excluded from the follow‐up survey. The participation in the survey was voluntary and without any incentive. All responders signed an informed consent form before accessing the questionnaire which was provided through an online survey platform (SurveyMonkey ® ). Data collection of the follow‐up survey took place from 23 November to 4 February 2021, whereas the initial survey had been performed from 12 April to 22 May 2020. Data were stored anonymously.
Questionnaire A 14‐item structured questionnaire was developed (Attachment 1). It was created based on a published questionnaire, previously completed by all the invited experts (Becker et al., ). In order to allow the comparison between data recorded in the two waves of pandemic, the first 10 items included minor modifications and covered the same areas as the first survey. In order to explore emerging topics of interest, 4 new items were added. Furthermore, the option to leave a comment on a question was added. The questionnaire was organized as follows: participant working environment (two items, single choice); infection risk of dental health professionals (two items, single choice); facilities for dental treatment of COVID‐19 positive patients (one item single choice); measures to prevent infection of health professionals and nosocomial transmission in dental clinics (two items, matrix/single choice per row); information about where dental care was provided for COVID‐19 positive patients or at high risk (one item, multiple select); recommended PPE for dental health professionals (two items, matrix/single choice per row); antiseptic mouth rinse (one item, single choice); COVID‐19 testing of dental staff (two items, single choice); recommended type of treatment provision during the second wave (one item, single choice). As in the previous survey (Becker et al., ), they were asked if they treated their patients in the private and/or university dental clinic settings (question 1) and information on the number of staff members working in their departments was also explored (question 2). In question 3 and 4 (scoring of the risk from AGP, and non‐AGP), participants had an option to leave a comment to explain their response. Question 5 was added, asking where patients infected with COVID‐19 (COVID+) should be treated. Question 8 on room‐related preventive measures was moved to question 6, and two options for response were added (high‐efficiency particulate air [HEPA] filters, high vacuum evacuators). Question 9 was moved to 7 (patient‐related measures), and antigen or PCR test prior to treatment was added as a new response. Question 10 was moved to 8 (facilities performing treatment of COVID‐19 patients in the respective countries), and now it was possible to leave a comment and state, whether one was actually involved in treatment of COVID‐19 patients. Questions 5 and 6 were shifted to questions 9 and 10 (protective measures for non‐AGP/AGP), and now also antigen test before treatment could be selected as a preventive measure. Additionally, instead using the classifying answers among recommended for patients of unknown risk/high risk/very high risk/not recommended , we now specified recommended for every patient/patient with COVID‐19 symptoms/not recommended . Questions 11–14 were added to the survey and included aspects that were in general not relevant during first wave of pandemic (when elective treatments were postponed in most countries), that is, virucidal mouth rinse, regular testing of staff, and continuation of elective treatments.
Statistical analysis The data analysis was performed using Microsoft Excel ® for Mac version 16.37 (Microsoft ® ) and an online survey tool (Survey Monkey ® ). For each question, the absolute number of votes and the relative agreement (%) were calculated. As only one group was surveyed (experts in oral surgery), no comparative analyses were performed per time point. To compare expert votes between first and second survey, chi‐square test was used. Results were found significant if p < .05.
RESULTS A total of 26 out of 27 experts from different European countries responded to the survey. No response was received from Estonia. The mean duration of answering the questionnaire was 8 min and 45 s. Details on the adherence to the “Good practice in the conduct and reporting of survey research” criteria for questionnaire studies are reported in Attachment 2 from submission (Kelly et al., ). 3.1 Participant working environment Ten participants (38.5%) responded to be treating patients at the Dental University Hospital, and 14 (53.9%) indicated treatment in both private practice and Dental University Hospital. Beside their academic responsibility, two participants (7.7%) reported performing their clinical activity in private practice. Despite the minor changes in the responses, no significant differences were observed ( X 2 = 1.67, df = 2, p = .43). The number of staff members working in their departments was heterogenous: 0–10 (one expert, 3.9%), 11–20 (nine experts, 34.6%), 21–30 (six experts, 23.1%), 31–50 (two experts, 7.7%), 51–100 (six experts, 23.1%), >100 (two experts, 7.7%). Whereas number of staff members increased slightly compared to survey one, no significant differences were observed ( X 2 = 4.34, df = 5, p = .50). 3.2 Infection risk of dental health professionals For aerosol‐free treatments (non‐AGP), 12 experts (46.2%) found the risk to be low, whereas six rated the risk to be neutral (23.1%) and eight rated the transmission risk to be high (30.7%; Figure ). When comparing the findings with the first survey, risk associated with non‐AGP retrieved significantly lower scores ( X 2 = 6.25, df = 2, p = .04). However, it was commented that ICP needs to be followed, and proper protection has to be used, otherwise risk may be higher. For AGP, three experts rated the risk to be low (11.5%), eight experts scored the risk to be neutral (30.8%) and 15 (57.7%) to be high (Figure ). Thus, risk associated with AGP was scored significantly lower compared to the initial survey ( X 2 = 9.08, df = 2, p = .01). It was commented that the overall risk would be high, but risk may be neutralized by precautions and PPE, and following of ICP guidelines. 3.3 Facilities for dental treatment of COVID‐19 positive patients A total of nine experts (34.6%) responded that dental treatments of COVID‐19+ patients should be performed at Dental University Hospitals. The remaining 17 participants (65.4%) suggested private practice and Dental University Hospitals as eligible. None of the experts identified the private practice as the only recommended setting for urgent treatment of COVID‐19 positive patients. The responses were in line with the initial survey ( X 2 = 0.00, df = −1, p = 1.00). 3.4 Measures to prevent infection of health professionals and nosocomial transmission in dental clinics Findings regarding measures to prevent infection of health professionals and nosocomial transmission in dental clinics are presented in Figure . Comparable to the initial survey, most of the experts (23 experts, 88.5%) found treating COVID‐19+ patients in separate isolation rooms relevant ( X 2 = 2.07. df = 2, p = .36). In contrast, fewer experts recommended the reduction of AGP (19 experts, 73.1%), and this was significantly different from the initial survey ( X = 28.51, df = 2, p < .01). More than half of the participants also found the use of rubber dam relevant (16 experts, 61.5%), which was comparable to the initial survey ( X 2 = 2.12, df = 2, p = .35). Heterogenous opinions were found for extraoral radiographs (relevant: 11 experts, 42.3% neutral: eight experts, 30.8%, not relevant: seven experts, 26.9%), and these opinions were significantly different to the initial survey ( X 2 = 7.32, df = 2, p = .02). Air disinfection also received heterogenous scores (relevant: 13 experts, 50.0%; neutral: 12 experts, 46.2%; not relevant: one expert, 3.9%), which was comparable to survey one ( X 2 = 0.94, df = 2, p = .62). Additionally, heterogeneous recommendations were found for high vacuum evacuators (relevant: 10 experts, 38.5%, neutral: 15 experts, 57.7%, not relevant: one expert, 3.85%; new item). In contrast, limiting the contact among staff members was found relevant by the wide majority of experts in both surveys (22 experts, 84.6% in second survey, 23 experts 85.0% in first survey; X 2 = 0.01, df = 2, p = 1.00). Additionally, natural air ventilation (21 experts, 80.8%) and usage of fixed filter systems or mobile filtration units with HEPA filters (17 experts, 65.4%) were found to be relevant by most of the experts (new items). Several patient‐related measures were found important to limit the risk of COVID‐19 transmission (Figure ). In both surveys, the experts agreed that the number of patients in waiting area and the time they spend there should be minimized. Most experts also agreed to limit the number of accompanying people (23 experts, 88.5%), which was in line with previous survey. Phone interviews to assess the health status (COVID‐19 risk assessment) were scored to be highly relevant (19 experts, 73.1%) during both surveys ( X 2 = 2.33, df = 2, p = .31). In contrast, assessing patient treatment needs via phone was approved by 14 (53.9%) of the experts only, which was comparable to the previous survey ( X 2 = 1.76, df = 2, p = .41). The hand hygiene (25 experts, 96.2%) and surgical mask wear inside the clinic were considered to be crucial (24 experts, 92.3%; X 2 = 0.00, df = 2, p =.99 and X 2 = 4.23, df = 2, p = .12). Slightly fewer experts recommended temperature taking (15 experts, 57.7%; X 2 = 2.95, df = 2, p = .23) and mouth rinse (19 experts, 73.1%; X 2 = 0.65, df = 2, p = .72). In contrast, significantly fewer experts recommended to postpone elective treatments (relevant: 10 experts, 38.5%; X 2 = 12.45, df = 2, p < .01). Heterogenous recommendations were found regarding antigen or real‐time PCR testing prior to treatment (not relevant: six experts, 23.1%, neutral: nine experts, 34.6%; relevant: 11 experts, 42.3%; new item). 3.5 Information about where dental care was provided for patients with a high risk of COVID‐19 Results on where dental care was provided for patients with high risk of COVID‐19 are presented in Figure . Dental treatments of COVID‐19+ patients were performed at Department of Oral Surgery (18 experts, 69.2%) and/or at the Department of Oral and Maxillofacial Surgery (15 experts, 57.7%). In 13 countries, both departments were responsible in parallel. Private practice was mentioned by 11 experts (42.3%), whereas other facilities (e.g., emergency units) were highlighted by 10 experts (38.5%). A total of seven experts (26.9%) responded to be involved in the treatments of infected patients. 3.6 Recommended PPE for dental health professionals For non‐AGP procedures, the recommendations varied for the different PPE measures (Figure ). The majority of experts recommended FFP2/FFP3 masks for every patient (16 experts, 61.5%), whereas eight experts (30.8%) recommended these masks for patients with COVID symptoms only and two experts (7.7%) did not recommend them at all, so they were slightly less frequently recommended compared to the initial survey ( X 2 = 1.52, df = 2, p = .47). Face shields or goggles were recommended for every patient by the majority of experts (19 experts, 73.1%) likewise to the initial survey ( X 2 = 3.20, df = 2, p = .20). Heterogenous recommendations were found for overshoes (not recommended: nine experts, 34.6%, every patient: seven experts, 26.9%; symptomatic patients: 10 experts, 38.5%) likewise to the first survey ( X 2 = 1.51, df = 2, p = .56). Gowns were recommended for every patient by 12 experts (46.2%), while eight experts recommended them only for symptomatic patients (30.8%), by trend gowns were slightly less recommended as compared to the first survey ( X 2 = 2.27, df = 2, p = .32). Significantly less experts recommended caps (recommended for every patient by 12 experts, 46.2%; X 2 = 11.26, df = 2, p < .01) and slightly fewer experts recommended double gloves (recommended for every patient by four experts, 15.4%) ( X 2 = 4.34, df = 2, p = .11). Antigen test prior to treatment was recommended for symptomatic patients by 15 experts (57.7%), whereas eight experts (30.8%) did not recommend them at all (new item). For AGP, the recommended type of PPE is presented in Figure . The majority of experts recommended using FFP2/FFP3 masks (21 experts, 80.8%), similar to survey one. Face shields were recommended by 23 experts (88.5%), again almost identical to survey one. Overshoes were recommended by 11 experts (42.3%), comparable to survey one ( X 2 = 1.00, df = 2, p = .62). Gowns were recommended by most of the experts (16 experts, 61.5%), which was also in line with the initial survey ( X 2 = 1.46, df = 2, p = .48). Caps were recommended by 15 experts (57.7%), and this recommendation changed significantly compared to the initial survey where they were recommended more frequently ( X 2 = 6.83, df = 2, p = .03). Double gloves were recommended by only seven experts (26.9%) whereas they were by trend more often recommended during survey one ( X 2 = 3.49, df = 2, p = .18). Antigen test prior to treatment was recommended for every patient by seven experts (26.9%), whereas 12 experts considered this measure useful for symptomatic patients (46.2%). Seven experts did not find this measure useful at all (26.9%; new item). 3.7 Antiseptic mouth rinse Most of the experts (20 experts, 76.9%) advised antiseptic mouth rinse prior to treatment. Most frequently mentioned were hydrogen peroxide (H 2 O 2 ) (recommended by seven experts, varying concentration and rinsing time), chlorhexidine (CHX; two experts), combination of CHX and H 2 O 2 (one expert), chlorhexidine digluconate plus chlorobutanol (Eludryl ® ; one expert, 2–3 min), essential oils plus ethanol (Listerine ® ; one expert). 3.8 COVID‐19 testing of dental staff The majority of experts (14 experts, 53.9%) recommended testing of staff on a regular basis. Real‐time PCR test (five times, from weekly to monthly) and antigen test (four times, mostly on weekly basis) were proposed. One expert (3.9%) proposed testing only after contact with COVID+ people. A minority (eight experts, 30.8%) responded that regular tests should be performed, including real‐time PCR test (five times, i.e., every 3 months/every 20 days/weekly/time interval not specified), weekly antigen test (twice), or only when staff is symptomatic. 3.9 Recommended type of treatment provision during the second wave Most experts recommended that all treatments should be provided during the second wave of pandemic (20 experts, 76.9%). However, two experts commented that elective treatments in oral surgery should not be performed as long as the incidence of COVID‐19 is high, and one expert commented that COVID‐19 symptoms should be assessed prior to elective treatment. Only one expert suggested that elective treatments could be postponed.
Participant working environment Ten participants (38.5%) responded to be treating patients at the Dental University Hospital, and 14 (53.9%) indicated treatment in both private practice and Dental University Hospital. Beside their academic responsibility, two participants (7.7%) reported performing their clinical activity in private practice. Despite the minor changes in the responses, no significant differences were observed ( X 2 = 1.67, df = 2, p = .43). The number of staff members working in their departments was heterogenous: 0–10 (one expert, 3.9%), 11–20 (nine experts, 34.6%), 21–30 (six experts, 23.1%), 31–50 (two experts, 7.7%), 51–100 (six experts, 23.1%), >100 (two experts, 7.7%). Whereas number of staff members increased slightly compared to survey one, no significant differences were observed ( X 2 = 4.34, df = 5, p = .50).
Infection risk of dental health professionals For aerosol‐free treatments (non‐AGP), 12 experts (46.2%) found the risk to be low, whereas six rated the risk to be neutral (23.1%) and eight rated the transmission risk to be high (30.7%; Figure ). When comparing the findings with the first survey, risk associated with non‐AGP retrieved significantly lower scores ( X 2 = 6.25, df = 2, p = .04). However, it was commented that ICP needs to be followed, and proper protection has to be used, otherwise risk may be higher. For AGP, three experts rated the risk to be low (11.5%), eight experts scored the risk to be neutral (30.8%) and 15 (57.7%) to be high (Figure ). Thus, risk associated with AGP was scored significantly lower compared to the initial survey ( X 2 = 9.08, df = 2, p = .01). It was commented that the overall risk would be high, but risk may be neutralized by precautions and PPE, and following of ICP guidelines.
Facilities for dental treatment of COVID‐19 positive patients A total of nine experts (34.6%) responded that dental treatments of COVID‐19+ patients should be performed at Dental University Hospitals. The remaining 17 participants (65.4%) suggested private practice and Dental University Hospitals as eligible. None of the experts identified the private practice as the only recommended setting for urgent treatment of COVID‐19 positive patients. The responses were in line with the initial survey ( X 2 = 0.00, df = −1, p = 1.00).
Measures to prevent infection of health professionals and nosocomial transmission in dental clinics Findings regarding measures to prevent infection of health professionals and nosocomial transmission in dental clinics are presented in Figure . Comparable to the initial survey, most of the experts (23 experts, 88.5%) found treating COVID‐19+ patients in separate isolation rooms relevant ( X 2 = 2.07. df = 2, p = .36). In contrast, fewer experts recommended the reduction of AGP (19 experts, 73.1%), and this was significantly different from the initial survey ( X = 28.51, df = 2, p < .01). More than half of the participants also found the use of rubber dam relevant (16 experts, 61.5%), which was comparable to the initial survey ( X 2 = 2.12, df = 2, p = .35). Heterogenous opinions were found for extraoral radiographs (relevant: 11 experts, 42.3% neutral: eight experts, 30.8%, not relevant: seven experts, 26.9%), and these opinions were significantly different to the initial survey ( X 2 = 7.32, df = 2, p = .02). Air disinfection also received heterogenous scores (relevant: 13 experts, 50.0%; neutral: 12 experts, 46.2%; not relevant: one expert, 3.9%), which was comparable to survey one ( X 2 = 0.94, df = 2, p = .62). Additionally, heterogeneous recommendations were found for high vacuum evacuators (relevant: 10 experts, 38.5%, neutral: 15 experts, 57.7%, not relevant: one expert, 3.85%; new item). In contrast, limiting the contact among staff members was found relevant by the wide majority of experts in both surveys (22 experts, 84.6% in second survey, 23 experts 85.0% in first survey; X 2 = 0.01, df = 2, p = 1.00). Additionally, natural air ventilation (21 experts, 80.8%) and usage of fixed filter systems or mobile filtration units with HEPA filters (17 experts, 65.4%) were found to be relevant by most of the experts (new items). Several patient‐related measures were found important to limit the risk of COVID‐19 transmission (Figure ). In both surveys, the experts agreed that the number of patients in waiting area and the time they spend there should be minimized. Most experts also agreed to limit the number of accompanying people (23 experts, 88.5%), which was in line with previous survey. Phone interviews to assess the health status (COVID‐19 risk assessment) were scored to be highly relevant (19 experts, 73.1%) during both surveys ( X 2 = 2.33, df = 2, p = .31). In contrast, assessing patient treatment needs via phone was approved by 14 (53.9%) of the experts only, which was comparable to the previous survey ( X 2 = 1.76, df = 2, p = .41). The hand hygiene (25 experts, 96.2%) and surgical mask wear inside the clinic were considered to be crucial (24 experts, 92.3%; X 2 = 0.00, df = 2, p =.99 and X 2 = 4.23, df = 2, p = .12). Slightly fewer experts recommended temperature taking (15 experts, 57.7%; X 2 = 2.95, df = 2, p = .23) and mouth rinse (19 experts, 73.1%; X 2 = 0.65, df = 2, p = .72). In contrast, significantly fewer experts recommended to postpone elective treatments (relevant: 10 experts, 38.5%; X 2 = 12.45, df = 2, p < .01). Heterogenous recommendations were found regarding antigen or real‐time PCR testing prior to treatment (not relevant: six experts, 23.1%, neutral: nine experts, 34.6%; relevant: 11 experts, 42.3%; new item).
Information about where dental care was provided for patients with a high risk of COVID‐19 Results on where dental care was provided for patients with high risk of COVID‐19 are presented in Figure . Dental treatments of COVID‐19+ patients were performed at Department of Oral Surgery (18 experts, 69.2%) and/or at the Department of Oral and Maxillofacial Surgery (15 experts, 57.7%). In 13 countries, both departments were responsible in parallel. Private practice was mentioned by 11 experts (42.3%), whereas other facilities (e.g., emergency units) were highlighted by 10 experts (38.5%). A total of seven experts (26.9%) responded to be involved in the treatments of infected patients.
Recommended PPE for dental health professionals For non‐AGP procedures, the recommendations varied for the different PPE measures (Figure ). The majority of experts recommended FFP2/FFP3 masks for every patient (16 experts, 61.5%), whereas eight experts (30.8%) recommended these masks for patients with COVID symptoms only and two experts (7.7%) did not recommend them at all, so they were slightly less frequently recommended compared to the initial survey ( X 2 = 1.52, df = 2, p = .47). Face shields or goggles were recommended for every patient by the majority of experts (19 experts, 73.1%) likewise to the initial survey ( X 2 = 3.20, df = 2, p = .20). Heterogenous recommendations were found for overshoes (not recommended: nine experts, 34.6%, every patient: seven experts, 26.9%; symptomatic patients: 10 experts, 38.5%) likewise to the first survey ( X 2 = 1.51, df = 2, p = .56). Gowns were recommended for every patient by 12 experts (46.2%), while eight experts recommended them only for symptomatic patients (30.8%), by trend gowns were slightly less recommended as compared to the first survey ( X 2 = 2.27, df = 2, p = .32). Significantly less experts recommended caps (recommended for every patient by 12 experts, 46.2%; X 2 = 11.26, df = 2, p < .01) and slightly fewer experts recommended double gloves (recommended for every patient by four experts, 15.4%) ( X 2 = 4.34, df = 2, p = .11). Antigen test prior to treatment was recommended for symptomatic patients by 15 experts (57.7%), whereas eight experts (30.8%) did not recommend them at all (new item). For AGP, the recommended type of PPE is presented in Figure . The majority of experts recommended using FFP2/FFP3 masks (21 experts, 80.8%), similar to survey one. Face shields were recommended by 23 experts (88.5%), again almost identical to survey one. Overshoes were recommended by 11 experts (42.3%), comparable to survey one ( X 2 = 1.00, df = 2, p = .62). Gowns were recommended by most of the experts (16 experts, 61.5%), which was also in line with the initial survey ( X 2 = 1.46, df = 2, p = .48). Caps were recommended by 15 experts (57.7%), and this recommendation changed significantly compared to the initial survey where they were recommended more frequently ( X 2 = 6.83, df = 2, p = .03). Double gloves were recommended by only seven experts (26.9%) whereas they were by trend more often recommended during survey one ( X 2 = 3.49, df = 2, p = .18). Antigen test prior to treatment was recommended for every patient by seven experts (26.9%), whereas 12 experts considered this measure useful for symptomatic patients (46.2%). Seven experts did not find this measure useful at all (26.9%; new item).
Antiseptic mouth rinse Most of the experts (20 experts, 76.9%) advised antiseptic mouth rinse prior to treatment. Most frequently mentioned were hydrogen peroxide (H 2 O 2 ) (recommended by seven experts, varying concentration and rinsing time), chlorhexidine (CHX; two experts), combination of CHX and H 2 O 2 (one expert), chlorhexidine digluconate plus chlorobutanol (Eludryl ® ; one expert, 2–3 min), essential oils plus ethanol (Listerine ® ; one expert).
COVID‐19 testing of dental staff The majority of experts (14 experts, 53.9%) recommended testing of staff on a regular basis. Real‐time PCR test (five times, from weekly to monthly) and antigen test (four times, mostly on weekly basis) were proposed. One expert (3.9%) proposed testing only after contact with COVID+ people. A minority (eight experts, 30.8%) responded that regular tests should be performed, including real‐time PCR test (five times, i.e., every 3 months/every 20 days/weekly/time interval not specified), weekly antigen test (twice), or only when staff is symptomatic.
Recommended type of treatment provision during the second wave Most experts recommended that all treatments should be provided during the second wave of pandemic (20 experts, 76.9%). However, two experts commented that elective treatments in oral surgery should not be performed as long as the incidence of COVID‐19 is high, and one expert commented that COVID‐19 symptoms should be assessed prior to elective treatment. Only one expert suggested that elective treatments could be postponed.
DISCUSSION Transmission risk of COVID‐19 in dental settings has been reported to be high (Banakar et al., ) for the reason being that dentists work in proximity to the oral cavity, which is the natural reservoir of SARS‐CoV‐2. AGP are likely to further increase infection risk in dental environment (Nulty et al., ; United States Department of Labor, ). Due to the absence of evidence‐based recommendations on ICP during the first wave of COVID‐19 pandemic, the authors performed a survey gathering the opinion of European experts. This study revealed that adequate personal protective equipment (PPE) and reduction of AGP were considered to be crucial (Becker et al., ). However, as new scientific evidence on SARS‐CoV‐2 transmission became available, this led to the development of national and international COVID‐19 ICP guidelines (Becker et al., ). Despite this, these guidelines were found not to be uniform across Europe. Therefore, the present study aimed at providing follow‐up of experts' opinion‐based recommendations on crucial aspects of ICP for the continuation of dental treatment during COVID‐19 pandemic, and to compare them with the results from the first survey. In summary, the present survey revealed that experts rated the transmission risks in the dental settings to be significantly lower compared to the initial survey. Almost half of the experts now rated the risk of non‐AGP to be low, whereas the AGP were still considered to involve high infection risk for dental health professionals by most of the experts. Regarding the PPE, significantly fewer experts recommended FFP2/FFP3 masks for non‐AGP compared to the first survey, whereas opinion on the PPE for AGP did not change except for headwear. During the second survey, experts no longer suggested postponing elective treatments. No uniform responses were retrieved regarding testing of staff and patients. The majority of experts found antiseptic mouth rinse to be relevant and confirmed its use it in daily practice. A slightly higher number of dental staff working during the second wave was also reported, which might account for the resume of elective dental treatments. Reasons why the overall risk of SARS‐CoV‐2 transmission was rated lower in the second survey probably because of new scientific evidence and availability of adequate PPE. Additionally, epidemiological data revealed low infection rates among dental health professionals, thus showing effectiveness of the recommended measures (Estrich et al., ). The perception of risk might have changed also due to practical experience gained during pandemic or as a result of simple relaxation, after months of alertness. Allison et al. ( ) reported surface contamination was higher in the proximity to patients and operators, remaining high within a radius of 1–1.5 m. Surface contamination remained detectable at a maximum distance of 4 m. Sergis et al. ( ) showed that avoidance of premisting (mixing of coolant water and air prior to burr contact) might reduce the spread of small droplets from high‐speed hand pieces. The aerosol particles generated during AGP are associated with higher risk of transmission and nosocomial infection. However, there is still a lack of evidence to what extent aerosol‐generated particles are infectious once diluted in large amounts of water. This might reflect experts' opinion in the second survey. Risk from non‐AGP and AGP was rated significantly lower compared to first survey, however, the majority of experts rated the risk associated with AGP still to be high. Number of patient‐related measures, COVID‐19 ICP, and adequate PPE were introduced after the outbreak of pandemic and were frequently recommended by most of the experts. The patient journey was suggested to start with COVID‐19 risk assessment by telephone, even though remote identification of dental treatment need was no longer suggested. In this context, it has to be noted that only less than one third of experts recommended antigen testing prior to treatment for every patient. Thus, a high number of infected asymptomatic patients is expected to be missed by the proposed strategy. The reasons for the limited advice for rapid testing may be related to lack of scientific evidence, low sensitivity, and cost (Hirotsu et al., ; Mak et al., ; Scohy et al., ). However, the real‐time PCR testing was slightly more appreciated by the experts. To ensure patients’ safety, the experts in both surveys agreed that the time in the waiting area should be reduced and patients should possibly attend their appointments alone. Patients’ hand hygiene and wearing of masks were highlighted by all experts to be relevant, and the latter was by trend even more appreciated in the second survey. Body temperature check before the appointment is still debatable. The temperature screening has been described by the Clinical Evidence Assessment published in May 2020 (ECRI CEA, ) to be ineffective and will potentially miss more than half of infected individuals. The reason given was related to the low number of infected individuals who have fever at the time of screening and also inconsistent technique by operators. In addition, dental infection often presents with fever and if the patient would not be allowed to access the dental care due to the temperature screening, the consequence might become life‐threatening (i.e., sepsis, compromised airway in case of Ludwig angina). Additionally, virus load was reported to be high prior to the onset of symptoms including fever (Walsh, Jordan, et al., ). Thus, temperature check may not be the most effective measure, even though it was employed by more than half of the experts. A number of measures to prevent infection of health professionals and nosocomial transmission in dental clinics have been implemented since the start of pandemic. The experts in both surveys recognized the use of rubber dam and limiting contact between staff members to be relevant. Furthermore, the experts' opinion regarding the use of an isolation room for symptomatic patients did not change significantly, and isolation rooms were found useful by almost all experts also in the second survey. The opinion regarding air disinfection was not uniform, and it might be related to lack of evidence and also the spare information provided by national guidelines (Becker et al., ). Natural air ventilation, by contrast, was recommended by the vast majority. Proper ventilation depends on various factors as room volume, size of windows, air flow vectors, temperature, humidity, and characteristics of aerosol particles that have an impact on duration of fallow time (Sergis et al., ). In addition, shorter fallow times may be required when high‐volume suction and rubber dam are used (Scotttish Dental Clinical Effectiveness Programme, ). The use of suction was suggested for the reduction of contamination from AGP by 67%–75% at 0.5–1.5 m (Allison et al., ). Other methods for decreasing contamination from AGP were proposed, such as mechanical or hybrid filtration systems, which can be fixed or mobile. They can include HEPA filtration and may be used in conjunction with air disinfection (Kumbargere Nagraj et al., ). Approximately two thirds of the experts in the second survey considered filtration systems including HEPA filters to be relevant. A patient‐related measure that has been originally suggested by experts to reduce the risk of SARS‐CoV‐2 transmission was extraoral radiography (Meng et al., ). However, the results from second survey suggested that their indication is not as relevant as originally suggested. Another patient‐related measure is preprocedural an antiseptic mouth rinse, which has been also proposed to reduce viral load. Approximately two thirds of experts still found antiseptic mouth rinse to be relevant and reported to also utilize it in daily practice. An in vitro study suggested significant reduction of SARS‐CoV‐2 infectivity with dequalinium chloride, benzalkonium chloride, polyvidone‐iodine, ethanol, and essential oils (Meister et al., ). However, there is still a lack of clinical studies providing evidence regarding virucidal efficiency (Kumbargere Nagraj et al., ). There was a shortage of PPE during the first wave of the pandemic, and it was not clear which type of PPE was appropriate for non‐AGP/AGP. In contrast with our previous study (Becker et al., ), although the majority of experts still recommended that dentists should use FFP2/FFP3 masks for every patient undergoing non‐AGP, it was less frequently recommended compared to the first survey. This change cannot be explained by the improved availability of PPE and is therefore most likely related to the reduced risk assumed for non‐AGP. By contrast, for AGP, the majority still recommended usage of FFP2/FFP3 masks. Interestingly, headwear such as caps was recommended by less than half of experts for non‐AGP, and by a slightly more than half for AGP. This was a significant reduction compared to the initial survey, maybe because contact transmission is no longer considered to be the main transmission route of SARS‐CoV‐2. Indeed, the use of different type of PPE for each procedure for dental staff might be impractical as PPE has to be changed between the patients. Therefore, some dental hospitals/clinics introduced (Grossman et al., ). Slightly more than half of the experts found testing of staff members relevant, whereas only approximately one third also reported that they were performing it in clinical practice. This might be related to the start of vaccination program in European countries, which is expected to provide immunity among dental health professionals (Walsh, Jordan, et al., ). As emerged from the follow‐up survey, high‐risk patients/COVID+ patients still pose a significant challenge to the dental health professional whenever treatment needs are urgent and cannot be postponed. One third of the experts suggested that COVID+ patients should be treated in hospital settings only, whereas approximately two thirds recommended that treatments can be performed at both, clinics and private practice. A limitation of the survey relates to the small number of involved experts, all of whom have the same specialized profession. Our procedure was chosen during the initial survey to enable equal representation of every country involved. Due to the anonymous approach, pooling per country was not possible. Future survey studies, however, might include a larger number of participants from different specializations and different levels of experience. Another limitation of the present survey is that most experts responded before the widespread dissemination of mutants of SARS‐CoV‐2, which were reported to have a much higher infectivity (Galloway et al., ; Leung et al., ), and also the start of mass vaccination programs. In conclusion, this follow‐up study revealed that the present pandemic still poses significant challenges on dental health professionals. In particular, early identification of potentially infectious patients and proper protection during AGPs appear to be highly relevant. However, additional challenges may arise due to new variants of the virus which may be more infectious, thus possibly making dental practices a hotspot for virus transmission.
The authors declare that they have no conflict of interest related to this study.
Giulia Brunello: Conceptualization (equal); Formal analysis (equal); Methodology (equal); Resources (equal); Writing‐original draft (equal); Writing‐review & editing (equal). Katarzyna Gurzawska‐Comis: Conceptualization (equal); Investigation (equal); Methodology (equal); Writing‐original draft (equal); Writing‐review & editing (equal). Kathrin Becker: Conceptualization (equal); Data curation (equal); Formal analysis (equal); Methodology (equal); Project administration (equal); Resources (equal); Visualization (equal); Writing‐original draft (equal); Writing‐review & editing (equal). Stefano Sivolella: Conceptualization (equal); Methodology (equal); Supervision (equal); Writing‐review & editing (equal). Frank Schwarz: Conceptualization (equal); Methodology (equal); Supervision (equal); Writing‐review & editing (equal). Bjorn Klinge: Conceptualization (equal); Investigation (equal); Methodology (equal); Supervision (equal); Writing‐review & editing (equal).
Supplementary Material Click here for additional data file.
|
Arteriolosclerosis differs from venular collagenosis in relation to cerebrovascular parenchymal damages: an autopsy-based study | 3050b4b3-14c5-4110-b083-a73a20077823 | 10512076 | Forensic Medicine[mh] | Arteriolar sclerosis is closely correlated with brain tissue damage, while cerebral venules have been consistently ignored.
We found an important role of venular lesions in microvessel-related parenchymal damage which is different from that of arteriolosclerosis.
This finding may give a novel perspective in understanding of mechanism underlying microvessel-related parenchymal damage.
Cerebrovascular parenchymal damage, including lacunes, microinfarcts, myelin loss and haemosiderin deposits, is common during the autopsy of aged individuals. Previous neuropathological studies have found that arteriolar sclerosis is closely correlated with brain tissue damage, indicating a potential aetiology of cerebrovascular parenchymal lesions. Apart from the arteriolar injuries, histopathological studies also found that venular collagenosis, involving small veins and venules in periventricular regions, was associated with more severe myelin loss. In addition, studies on brain venules have reported that it is sometimes possible to confuse abnormal venules with hyalinised arterioles without special stains. Putting these findings together, questions on the roles and heterogeneities of arteriolar and venular abnormalities in the etiopathogenesis of parenchymal damage may be raised. However, no pathological study has compared the roles of arteriolosclerosis and venular collagenosis in parenchymal damage. This may be largely due to the lack of specific staining methods to distinguish venules from arterioles, especially those with diameters of <1 mm. In view of the accumulating evidence on the role of the cerebral venous system relative to the glymphatic pathway, arterioles and venules should be comprehensively investigated to further understand the pathophysiological mechanisms of tissue damage caused by or associated with vessel disease. To address these issues, we examined the topographical associations between cerebral microvascular injuries, including arteriolosclerosis and venular collagenosis, and different types of parenchymal damage associated with small vessel disease, including lacunes, microinfarcts, myelin loss, parenchymal haemosiderin deposits,and perivascular haemosiderin leakage in seven different cerebral regions from 27 postmortem cases, to explore the underlying pathophysiological mechanisms of different microvascular injuries.
Specimen selection Our study comprised 27 human postmortem brains (mean age:77.33±14.07 years; 13 males), and the clinicopathological profiles are shown in (details in ). Cases with stroke or brain tumour history according to their clinical records (obtained at donation from the donors’ family members, reviewed by W-YQ) were excluded from this study. Due to lack of detailed clinical records, vascular risk factors could not be accurately assessed for these cases. A more detailed description of the standardised protocol of the Human Brain Bank in China has been published elsewhere. Written informed consent was obtained from both the brain donors and their next of kin. 10.1136/svn-2022-001924.supp1 Supplementary data The number of lacunes was recorded during the gross examination of the right hemisphere. After the autopsy, the right hemisphere was immersion-fixed in 10% buffered aqueous formaldehyde solution for 2–6 weeks, following which it was dissected in coronal planes at approximately 1.0 cm intervals and was paraffin-embedded. During brain autopsy, tissue blocks were systematically taken in all cases from the frontal periventricular white matter (the distance to the ependyma <0.5 cm), occipital periventricular white matter (the distance to ependyma <0.5 cm), deep white matter in the parietal lobe (the distance to ependyma >1.5 cm), superior frontal cortex (Brodmann area 9), occipital cortex (Brodmann areas 17 and 18), hippocampus and putamen.
Our study comprised 27 human postmortem brains (mean age:77.33±14.07 years; 13 males), and the clinicopathological profiles are shown in (details in ). Cases with stroke or brain tumour history according to their clinical records (obtained at donation from the donors’ family members, reviewed by W-YQ) were excluded from this study. Due to lack of detailed clinical records, vascular risk factors could not be accurately assessed for these cases. A more detailed description of the standardised protocol of the Human Brain Bank in China has been published elsewhere. Written informed consent was obtained from both the brain donors and their next of kin. 10.1136/svn-2022-001924.supp1 Supplementary data The number of lacunes was recorded during the gross examination of the right hemisphere. After the autopsy, the right hemisphere was immersion-fixed in 10% buffered aqueous formaldehyde solution for 2–6 weeks, following which it was dissected in coronal planes at approximately 1.0 cm intervals and was paraffin-embedded. During brain autopsy, tissue blocks were systematically taken in all cases from the frontal periventricular white matter (the distance to the ependyma <0.5 cm), occipital periventricular white matter (the distance to ependyma <0.5 cm), deep white matter in the parietal lobe (the distance to ependyma >1.5 cm), superior frontal cortex (Brodmann area 9), occipital cortex (Brodmann areas 17 and 18), hippocampus and putamen.
Paraffin-embedded serial sections from seven tissue blocks were cut at a 5 µm thickness. Sections underwent histological staining with Luxol fast blue (LFB) to assess white matter pallor (as a measure of myelin loss), Prussian blue to demonstrate haemosiderin, Masson’s trichrome to evaluate the degree of collagen deposition in the microvascular walls and H&E for visualising arteriole/venule walls, as well as microinfarcts in the brain parenchyma. Immunohistochemistry was performed using antibodies against alpha-smooth muscular actin (αSMA, Cat# 19245, RRID: AB_2734735, rabbit monoclonal antibody, diluted 1:400; Cell Signaling Technology) and monocarboxylate transporter 1 (MCT1, Cat# 20 139-1-AP, RRID: AB_2878645, rabbit polyclonal antibody, diluted 1:800; Proteintech)to distinguish between venules and arterioles(details of primary antibodies and antigen retrieval protocols for immunohistochemistry are presented in ). Immunopositivity was detected using the VECTASTAIN ABC-HRP Kit (rabbit IgG, PK-4001; Vector Laboratories, Burlingame, California, USA) with 3,3 diaminobenzidine as a chromogen and hematoxylin as the counterstain. All histologically and immunohistochemically stained sections were subsequently dehydrated using a series of alcohols, cleared, and mounted in neutral balsam (ZLI-9555, ZSGB-BIO, Beijing, China). Image analysis All image analyses were performed by two trained neurologists (YC and M-YH), who were blinded to the neuropathological diagnosis, and analysis results were re-examined and confirmed by one pathologist (W-YQ). Whole stained sections were scanned using an Aperio digital pathology slide scanner (Leica Biosystems), and monochrome images were uploaded into the Image J software programme (Media Cybernetics, USA; V.6.3). Assessing microvascular wall pathology To evaluate the severity of arteriolosclerosis and venular collagenosis, 2400 μm×3800 µm single images were captured randomly in anti-αSMA and anti-MCT1 immunostained sections to differentiate arterioles and venules, and the corresponding visual fields on the delineated images were subsequently captured in the adjacent Masson’s trichrome sections to assess the microvascular walls. Based on our previous report, arterioles were identified as negative immunolabeled for MCT1 and generally positive immunolabeled for αSMA. According to the rating scheme for cerebrovascular lesions, arterioles with a score of 2 or 3 were assessed as arteriosclerotic arterioles in our study, following which the percentage of arteriolosclerosis was calculated. Similarly, for venular collagenosis evaluation, vessels ranging from 10 to 300 µm in diameter, which showed intense labelling for MCT1 in the endothelium with a thin anti-αSMA-stained pattern in the media tunica, were considered as venules, and those with severe stenosis (the lumen occupying more than 50% of the vessel diameter) or occlusion by thick collagenous walls were assessed as venular collagenosis ( ) according to previous scores by Moody et al . The ratio of the number of collagenised venules to the total number of venules in the visual field was calculated and expressed as a percentage of venular collagenosis. Assessing and quantifying vascular parenchymal damages Microinfarcts were identified as sharply delimited microscopic regions of cellular death or tissue necrosis, sometimes with cavitation (ie, a central fluid-filled cavity). For the quantification of myelin loss, the 8-bit grayscale threshold was manually adjusted to select only white matter regions. At each LFB-stained section, five images (1 mm × 1 mm) were captured randomly, and the inverted mean grey values of the LFB stain were measured using the Image J software. If necessary, the selected images were subjected to the manual setting of the regions of interest to exclude grey matter and meningeal structures. A pixel value of 0 represented white, and 255 represented black; therefore, a lower value signified a more severe white matter pallor, indicating myelin loss. The mean grey value was recorded for each of the five visual fields, and the mean value was calculated as the severity of myelin loss in the respective white matter region. To support the data accuracy by LFB stain, we also conducted anti-myelin basic protein (Cat# ab40390, RRID: AB_1141521, rabbit polyclonal antibody, diluted 1:500; Abcam) immunohistochemistry stain in the adjacent sections, and assessed the images in a similar way ( ). For haemosiderin evaluations, perivascular and brain parenchymal haemosiderin were assessed separately based on H&E-stained and Prussian blue-stained sections. For each selected region, the number of vessels with perivascular haemosiderin leakage on the entire slide was recorded, and the average number of haemosiderin deposits in the brain parenchyma was calculated from five 2 mm × 2 mm images. Statistics Variables were tested for normality using the Shapiro-Wilk test and visual inspection of variable histograms. Descriptive analyses were conducted using the median and 25%–75% percentiles for continuous variables and frequency and percentage for categorical variables. Comparisons of cerebral small vessel pathology in different brain regions were conducted using the Wilcoxon matched-pairs signed-rank test. Spearman’s correlation coefficients were used in the respective regions to investigate the association between the percentage of venular collagenosis or arteriolosclerosis and age. The Mann-Whitney U test for non-parametric analysis of the age was performed between patients with or without lacunes, and there was no significant difference in the age at death between groups (76.56±14.61 years for those with lacunes vs 76.44±14.66 years for those without lacunes, p=1.000; 72.67±13.09 years for those with microinfarcts vs 79.67±14.31 years for those without microinfarcts, p=0.2088). The Mann-Whitney U test was used for non-parametric analysis between patients with and without lacunes to compare the percentage of arteriolosclerosis or venular collagenosis in each region. Spearman’s partial correlation coefficients (adjusted for age) were used to assess associations between these microvascular lesions and the number of microinfarcts, the mean grey value of LFB, and the number of haemosiderin deposits in each region. The significance level was set at 0.05, and all tests of significance were two tailed. All statistical analyses were conducted using SAS V.9.4 (SAS Institute).
All image analyses were performed by two trained neurologists (YC and M-YH), who were blinded to the neuropathological diagnosis, and analysis results were re-examined and confirmed by one pathologist (W-YQ). Whole stained sections were scanned using an Aperio digital pathology slide scanner (Leica Biosystems), and monochrome images were uploaded into the Image J software programme (Media Cybernetics, USA; V.6.3). Assessing microvascular wall pathology To evaluate the severity of arteriolosclerosis and venular collagenosis, 2400 μm×3800 µm single images were captured randomly in anti-αSMA and anti-MCT1 immunostained sections to differentiate arterioles and venules, and the corresponding visual fields on the delineated images were subsequently captured in the adjacent Masson’s trichrome sections to assess the microvascular walls. Based on our previous report, arterioles were identified as negative immunolabeled for MCT1 and generally positive immunolabeled for αSMA. According to the rating scheme for cerebrovascular lesions, arterioles with a score of 2 or 3 were assessed as arteriosclerotic arterioles in our study, following which the percentage of arteriolosclerosis was calculated. Similarly, for venular collagenosis evaluation, vessels ranging from 10 to 300 µm in diameter, which showed intense labelling for MCT1 in the endothelium with a thin anti-αSMA-stained pattern in the media tunica, were considered as venules, and those with severe stenosis (the lumen occupying more than 50% of the vessel diameter) or occlusion by thick collagenous walls were assessed as venular collagenosis ( ) according to previous scores by Moody et al . The ratio of the number of collagenised venules to the total number of venules in the visual field was calculated and expressed as a percentage of venular collagenosis. Assessing and quantifying vascular parenchymal damages Microinfarcts were identified as sharply delimited microscopic regions of cellular death or tissue necrosis, sometimes with cavitation (ie, a central fluid-filled cavity). For the quantification of myelin loss, the 8-bit grayscale threshold was manually adjusted to select only white matter regions. At each LFB-stained section, five images (1 mm × 1 mm) were captured randomly, and the inverted mean grey values of the LFB stain were measured using the Image J software. If necessary, the selected images were subjected to the manual setting of the regions of interest to exclude grey matter and meningeal structures. A pixel value of 0 represented white, and 255 represented black; therefore, a lower value signified a more severe white matter pallor, indicating myelin loss. The mean grey value was recorded for each of the five visual fields, and the mean value was calculated as the severity of myelin loss in the respective white matter region. To support the data accuracy by LFB stain, we also conducted anti-myelin basic protein (Cat# ab40390, RRID: AB_1141521, rabbit polyclonal antibody, diluted 1:500; Abcam) immunohistochemistry stain in the adjacent sections, and assessed the images in a similar way ( ). For haemosiderin evaluations, perivascular and brain parenchymal haemosiderin were assessed separately based on H&E-stained and Prussian blue-stained sections. For each selected region, the number of vessels with perivascular haemosiderin leakage on the entire slide was recorded, and the average number of haemosiderin deposits in the brain parenchyma was calculated from five 2 mm × 2 mm images.
To evaluate the severity of arteriolosclerosis and venular collagenosis, 2400 μm×3800 µm single images were captured randomly in anti-αSMA and anti-MCT1 immunostained sections to differentiate arterioles and venules, and the corresponding visual fields on the delineated images were subsequently captured in the adjacent Masson’s trichrome sections to assess the microvascular walls. Based on our previous report, arterioles were identified as negative immunolabeled for MCT1 and generally positive immunolabeled for αSMA. According to the rating scheme for cerebrovascular lesions, arterioles with a score of 2 or 3 were assessed as arteriosclerotic arterioles in our study, following which the percentage of arteriolosclerosis was calculated. Similarly, for venular collagenosis evaluation, vessels ranging from 10 to 300 µm in diameter, which showed intense labelling for MCT1 in the endothelium with a thin anti-αSMA-stained pattern in the media tunica, were considered as venules, and those with severe stenosis (the lumen occupying more than 50% of the vessel diameter) or occlusion by thick collagenous walls were assessed as venular collagenosis ( ) according to previous scores by Moody et al . The ratio of the number of collagenised venules to the total number of venules in the visual field was calculated and expressed as a percentage of venular collagenosis.
Microinfarcts were identified as sharply delimited microscopic regions of cellular death or tissue necrosis, sometimes with cavitation (ie, a central fluid-filled cavity). For the quantification of myelin loss, the 8-bit grayscale threshold was manually adjusted to select only white matter regions. At each LFB-stained section, five images (1 mm × 1 mm) were captured randomly, and the inverted mean grey values of the LFB stain were measured using the Image J software. If necessary, the selected images were subjected to the manual setting of the regions of interest to exclude grey matter and meningeal structures. A pixel value of 0 represented white, and 255 represented black; therefore, a lower value signified a more severe white matter pallor, indicating myelin loss. The mean grey value was recorded for each of the five visual fields, and the mean value was calculated as the severity of myelin loss in the respective white matter region. To support the data accuracy by LFB stain, we also conducted anti-myelin basic protein (Cat# ab40390, RRID: AB_1141521, rabbit polyclonal antibody, diluted 1:500; Abcam) immunohistochemistry stain in the adjacent sections, and assessed the images in a similar way ( ). For haemosiderin evaluations, perivascular and brain parenchymal haemosiderin were assessed separately based on H&E-stained and Prussian blue-stained sections. For each selected region, the number of vessels with perivascular haemosiderin leakage on the entire slide was recorded, and the average number of haemosiderin deposits in the brain parenchyma was calculated from five 2 mm × 2 mm images.
Variables were tested for normality using the Shapiro-Wilk test and visual inspection of variable histograms. Descriptive analyses were conducted using the median and 25%–75% percentiles for continuous variables and frequency and percentage for categorical variables. Comparisons of cerebral small vessel pathology in different brain regions were conducted using the Wilcoxon matched-pairs signed-rank test. Spearman’s correlation coefficients were used in the respective regions to investigate the association between the percentage of venular collagenosis or arteriolosclerosis and age. The Mann-Whitney U test for non-parametric analysis of the age was performed between patients with or without lacunes, and there was no significant difference in the age at death between groups (76.56±14.61 years for those with lacunes vs 76.44±14.66 years for those without lacunes, p=1.000; 72.67±13.09 years for those with microinfarcts vs 79.67±14.31 years for those without microinfarcts, p=0.2088). The Mann-Whitney U test was used for non-parametric analysis between patients with and without lacunes to compare the percentage of arteriolosclerosis or venular collagenosis in each region. Spearman’s partial correlation coefficients (adjusted for age) were used to assess associations between these microvascular lesions and the number of microinfarcts, the mean grey value of LFB, and the number of haemosiderin deposits in each region. The significance level was set at 0.05, and all tests of significance were two tailed. All statistical analyses were conducted using SAS V.9.4 (SAS Institute).
The characteristics of the 27 participants are presented in . The average age at death was 77.3 years (SD: 14.1 years), and 48.7% (13/27) of the donors were males. The median postmortem delay was 8.4 hours (IQR: 4.5–17 hours). Parenchymal damages and microvascular wall pathology in different brain regions Cerebrovascular parenchymal damage was evaluated across all selected white matter and grey matter regions, except for myelin loss, which was assessed only in the white matter ( ). In gross specimens, lacunes were observed in 59.26% (16/27) of the brains, and microinfarcts were observed under a microscope in 66.67% (18/27) of the brains. The values of white matter pallor were lower in the periventricular white matter than that in the deep white matter (117.0±7.2 vs 119.1±7.5, p=0.05), indicating a higher degree of myelin loss. Both parenchymal hsemosiderin deposits and vessels with perivascular hsemosiderin leakage were commonly observed in the putamen (median: 27.4 (parenchymal) and 15 (perivascular)) and were evenly distributed in all other regions (median range: 4.2–5.8 (parenchymal) and 0–1 (perivascular)). Cerebral microvascular wall pathology was also observed in all brain regions ( ). In most brain areas, arteriolosclerosis and venular collagenosis were found in a similar distribution pattern, both of which were most frequently observed in the periventricular white matter, less so in the deep white matter and cortical grey matter, and least frequently observed in the hippocampus. However, in the putamen, arteriolosclerosis was as common as in the deep white matter, while venular lesions were relatively rare in some cases. When patients were divided into mild and severe groups according to the median of total arteriolosclerosis or venular collagenosis burden, the severity of arteriolosclerosis significantly increased with venular collagenosis in every brain region, although 20%–30% of the cases still had severe arteriolosclerosis and mild venular collagenosis simultaneously or vice versa ( ). We also found that the severity of venular collagenosis was significantly associated across the brain regions, whereas the severity of arteriolosclerosis was not balanced between the white matter and cortex. Arteriolosclerosis increased synchronously in the white matter and putamen ( ). Associations of arteriolosclerosis and venular collagenosis with age The severity of arteriolosclerosis and venular collagenosis significantly increased with age in all brain regions except the putamen, where venular collagenosis was not significantly associated with age ( ). Associations of arteriolosclerosis and venular collagenosis with parenchymal damage Associations between arteriolosclerosis and venular collagenosis with each type of parenchymal lesion were assessed in every brain region ( ). Compared with those without lacunes, those with lacunes had an increased proportion of arteriolosclerosis only in the putamen (p=0.004) but not in other regions, whereas no difference in venular collagenosis was found in any region (p>0.1 for all comparisons). In the hippocampus, venular collagenosis was significantly more common in the microinfarcts group than in the non-microinfarcts group in the hippocampus (p=0.036); however, in other regions, both arteriolosclerosis and venular collagenosis were not statistically different between two groups. No association between the degree of myelin loss and arteriolosclerosis was observed in both the periventricular and deep white matter; however, lower white matter pallor values (indicating myelin loss) were significantly associated with an increased proportion of venular collagenosis in these regions. After adjusting for age, venular collagenosis was still associated with white matter pallor in the occipital periventricular white matter (r=−0.469, p=0.016) and deep white matter (r=−0.437, p=0.025) but not in the frontal periventricular white matter (r=−0.344, p=0.086). Neither arteriolosclerosis nor venular collagenosis was associated with perivascular or brain parenchymal haemosiderin deposits in any region (p>0.1 for all comparisons, ), except in the superior frontal cortex, where arteriolosclerosis correlated with decreased brain parenchymal haemosiderin deposits (p=0.024).
Cerebrovascular parenchymal damage was evaluated across all selected white matter and grey matter regions, except for myelin loss, which was assessed only in the white matter ( ). In gross specimens, lacunes were observed in 59.26% (16/27) of the brains, and microinfarcts were observed under a microscope in 66.67% (18/27) of the brains. The values of white matter pallor were lower in the periventricular white matter than that in the deep white matter (117.0±7.2 vs 119.1±7.5, p=0.05), indicating a higher degree of myelin loss. Both parenchymal hsemosiderin deposits and vessels with perivascular hsemosiderin leakage were commonly observed in the putamen (median: 27.4 (parenchymal) and 15 (perivascular)) and were evenly distributed in all other regions (median range: 4.2–5.8 (parenchymal) and 0–1 (perivascular)). Cerebral microvascular wall pathology was also observed in all brain regions ( ). In most brain areas, arteriolosclerosis and venular collagenosis were found in a similar distribution pattern, both of which were most frequently observed in the periventricular white matter, less so in the deep white matter and cortical grey matter, and least frequently observed in the hippocampus. However, in the putamen, arteriolosclerosis was as common as in the deep white matter, while venular lesions were relatively rare in some cases. When patients were divided into mild and severe groups according to the median of total arteriolosclerosis or venular collagenosis burden, the severity of arteriolosclerosis significantly increased with venular collagenosis in every brain region, although 20%–30% of the cases still had severe arteriolosclerosis and mild venular collagenosis simultaneously or vice versa ( ). We also found that the severity of venular collagenosis was significantly associated across the brain regions, whereas the severity of arteriolosclerosis was not balanced between the white matter and cortex. Arteriolosclerosis increased synchronously in the white matter and putamen ( ).
The severity of arteriolosclerosis and venular collagenosis significantly increased with age in all brain regions except the putamen, where venular collagenosis was not significantly associated with age ( ).
Associations between arteriolosclerosis and venular collagenosis with each type of parenchymal lesion were assessed in every brain region ( ). Compared with those without lacunes, those with lacunes had an increased proportion of arteriolosclerosis only in the putamen (p=0.004) but not in other regions, whereas no difference in venular collagenosis was found in any region (p>0.1 for all comparisons). In the hippocampus, venular collagenosis was significantly more common in the microinfarcts group than in the non-microinfarcts group in the hippocampus (p=0.036); however, in other regions, both arteriolosclerosis and venular collagenosis were not statistically different between two groups. No association between the degree of myelin loss and arteriolosclerosis was observed in both the periventricular and deep white matter; however, lower white matter pallor values (indicating myelin loss) were significantly associated with an increased proportion of venular collagenosis in these regions. After adjusting for age, venular collagenosis was still associated with white matter pallor in the occipital periventricular white matter (r=−0.469, p=0.016) and deep white matter (r=−0.437, p=0.025) but not in the frontal periventricular white matter (r=−0.344, p=0.086). Neither arteriolosclerosis nor venular collagenosis was associated with perivascular or brain parenchymal haemosiderin deposits in any region (p>0.1 for all comparisons, ), except in the superior frontal cortex, where arteriolosclerosis correlated with decreased brain parenchymal haemosiderin deposits (p=0.024).
In this study, we demonstrated that arteriolosclerosis and venular collagenosis were significantly correlated with age and were commonly observed in the white matter regions. More severe myelin loss was significantly associated with increased collagenised venules but not sclerotic arterioles, whereas lacunes were only associated with more severe arteriolosclerosis in the putamen but not with venular collagenosis in any region. Neither of these microvascular pathologies was associated with perivascular or parenchymal haemosiderin. Our results indicate that arteriolosclerosis and venular collagenosis may be involved in the different subtypes of parenchymal damage. Unlike arterioles, cerebral venules have been consistently ignored in neuropathological studies. This is owing to the limited number of histological methods for identifying cerebral venules. The method of anti-αSMA immunohistochemical staining, which relies on the continuous wreath of concentric smooth muscle in the arteries, has been commonly used in previous pathological studies. However, arterioles may be misinterpreted as venules when arteriolosclerosis occurs because of smooth muscle cell loss and fibrohyalinised material deposits. Anti-MCT1 immunohistochemical staining, a sensitive and reliable method for distinguishing cerebral venules from arterioles, combined with anti-αSMA, was used in our study to overcome these problems. We found that although the severity of arteriolosclerosis significantly increased with venular collagenosis across the brain regions, an inverse severity distribution was found in 20%–30% of cases. These data suggest that, despite some shared risk factors, the mechanisms underlying the development of arteriolosclerosis and venous collagenosis might differ. Additionally, more severe sclerotic arterioles were distributed in the white matter and putamen than in the cortex, which is in line with previous knowledge. However, the burden of collagenised venules in all the brain regions was balanced. In recent brain venule studies using MRI, most researchers visually counted only deep medullary veins of regular arrangement perpendicular to the ventricles. Our findings suggest that venular lesions in one brain region can represent those of the whole brain. Arteriolosclerosis has been commonly recognised as a critical cause of leukoaraiosis, while recent pathological studies on collagenised veins/venules have also suggested a close relationship with white matter pallor. In this study, our findings challenge the widely accepted knowledge that arteriolosclerosis is the main vascular cause of leukoaraiosis. A potential explanation might be that collagenised venules could be misinterpreted as sclerotic arterioles by the common staining methods of anti-αSMA immunohistochemical staining and Masson’s trichrome staining ( ), which has also attracted the attention of researchers in previous studies. Our findings further support the important role of venules in aggravating white matter lesions. Narrowed venules accompanied by increased pressure and vascular resistance can result in blood–brain barrier leakage and exudation of fluid into interstitial spaces, leading to vasogenic oedema. Interstitial oedema impairs the neuropil nearby, aggravating demyelination and gliosis. The impaired white matter could also stimulate venular collagen deposition in response to weak mechanical support of the vessels. It is not surprising that lacunes are more likely to be associated with arteriolar diseases rather than venular diseases because arteriolosclerosis might result in occlusion of the supplying artery and complete ischaaemia as a consequence. In general, the lack of association between microvascular wall pathologies and brain haemosiderin deposits found in our study is inconsistent with previous reports. This might be related to the differences in the study population. Previous studies mainly investigated patients with cerebral amyloid angiopathy and found that small vessels could be involved in vascular amyloid-beta accumulation to aggravate the progression of haemosiderin deposits in vascular walls. In contrast, for the aged donors in our study, haemosiderin was much less common, which might limit the statistical power. Our findings imply different roles for arteriolosclerosis and venular collagenosis in relation to different subtypes of microvessel-related parenchymal damage, which may have been previously overlooked. The anatomical branching patterns, tortuosity and age-related pathologies differ significantly depending on the vessel type (arteriolar or venular). In particular, arterioles are the primary site of vascular resistance, where the greatest changes in blood pressure and flow velocity occur. The physiological function of the arterial system mainly includes blood and nutrient supply, and that of venules includes material exchange, inflammatory cell infiltration and fluid drainage. Hence, it is not challenging to understand the pathological changes in arterioles or venules that might result in different brain parenchymal lesions. The main limitation of our study is its small sample size. As in common pathological studies, the selected samples in our observational study could not represent the whole-brain status, although multiple regions were assessed. Second, although cases with stroke or brain tumour history were ruled out, the heterogeneous disease background of the samples can not be ignored. In addition, the information on risk factors and neuroimaging is crucial; however, the small sample size, incomplete premortem data and lack of postmortem imaging data hindered the further analysis of potential factors of arteriolosclerosis and venular collagenosis. In summary, our study highlights the importance of cerebral small venous disease in vascular parenchymal lesions. Further studies on brain arterioles and venules are warranted to improve our understanding of the heterogeneity of cerebral small vessel diseases.
|
Development of lifestyle assessment: A Delphi survey of multi-faceted health experts | 74a81878-441a-4b3f-8e09-7eb6992b6a6d | 11838917 | Health Promotion[mh] | Lifestyle is becoming a major subject in the development of academic meaning and socioeconomic aspects in fields such as health promotion, marketing, and sociology, focusing on human behavior [ – ]. Lifestyle is defined as a lifestyle formed according to external environmental factors that reflect an individual’s life consciousness, values, and personality while consuming time and money . Lifestyles that reflect such multi-faceted aspects are being studied to identify the characteristics and trends of society, groups, and individuals or small groups . Recently in the field of health promotion, interest in lifestyle has increased as it increases life expectancy while reducing the prevalence of chronic diseases, premature mortality, and non-communicable diseases . Such a lifestyle is reported to be an important determinant of health and well-being in human life, and the need for a healthy lifestyle approach has been emphasized (WHO, 2021). Recently, an academic approach to lifestyle, which determines health and well-being, has been used in the analysis of causal relationships, risk factors, interventions, and the development of evaluation tools [ , , ]. In the field of health promotion, the importance of an academic approach that can measure and evaluate a multi-dimensional and dynamic lifestyle has been emphasized to form and improve a healthy lifestyle . Previous studies have quantified the frequency and exposure of behaviors such as physical activity, smoking, and drinking to analyze lifestyle characteristics; however, tools that can measure habits, behaviors, management, and diversity in daily life are limited . Such a limitation makes it impossible to grasp the practice trend of a lifestyle that reflects the intensity of health-related routines and habits repeated in daily life and the diversity of activities . Previous studies have focused on behaviors such as obesity, physical activity, diet, drinking, and smoking as lifestyle factors for human health and well-being . These factors have previously been emphasized in the healthcare field . On the other hand, the importance of relationships and participation in social factors has been emphasized before , but the need for a comprehensive approach that includes them as key relatively unnoticed factors is increasing . Considering this research situation, it is essential to systematically measure lifestyles to achieve health and physical, mental, and social well-being . However, no evaluation tool can measure the diversity and degree of a comprehensive healthy lifestyle in daily life, and the need for development has been confirmed. This study aims to develop an assessment tool to measure the diversity and degree of lifestyle behaviors related to human health. Therefore, we attempted to develop quantitative measurement items that reflect the comprehensive concepts of health and lifestyle and to verify their validity.
The Delphi technique, developed by Dalkey and Helmer (1963) at the Rand Corporation in the 1950s , is designed to gather purposeful opinions from experts identified in each field . We used the modified Delphi method, which obtains consensus on the opinions of a group of experts through a series of structured open and closed questions . The study was approved by the Yonsei University Mirae Institutional Review Board, considering the ethical aspects of the participants (Approval number:1041849-202211-SB-207-02). The data collection for this study was conducted from March 1, 2023, to April 30, 2023. Also, this study was conducted through an online survey, and given the impracticability of obtaining written consent, it proceeded after obtaining approval from the research ethics committee for an exemption from written consent. The expert panel We aimed to recruit healthcare professionals who are considered experts in lifestyle and health. To this end, the researchers invited health professionals who had completed graduate master’s courses or had more than five years of experience in related fields. To increase the diversity of the expert panel, we targeted Korean professionals with recognized expertise in lifestyle and/or healthcare management within the scope of occupational therapy, physical therapy, nursing, and nutrition. The eligibility criteria were as follows: (1) ability to complete approximately three rounds of surveys within two months, (2) no restrictions on computer skills, and (3) no restrictions on access to the Internet and e-mail. After obtaining consent for participation, a Delphi survey was conducted. For those selected as potential experts, an invitation email was sent that included a letter introducing the Delphi survey, a consent form related to the study, and an IRB guide. The experts were informed that they had to respond to all rounds until a consensus was reached (approximately three rounds), each round was given an appropriate amount of time to respond, participation was voluntary, and they could withdraw at any time. They were also informed that their individual responses would not be shared with other expert panels and their contributions would be kept confidential. Delphi surveys were conducted only with experts who expressed consent, and as a reward for their time, participants were given a $50 gift card after completing all the Delphi surveys. The Delphi method The study was conducted with participants who provided consent to potential expert panels. Demographic information was collected from the participants. Participants were instructed to read the guidelines introducing the concept of multi-faceted lifestyle components developed to create lifestyle assessment tools in the introduction of the survey attached to the email (Appendix 1). The first section explains the lifestyle factors of physical activity, eating habits, activity participation, and activity diversity, based on a literature review and previous studies. Participants were then shown the following sections consisting of 72 questions:18 areas of physical activity, 18 areas of nutrition, 18 areas of social relationships, and 18 areas of social participation. These items were developed based on literature review [ , – ]. The Delphi survey consisted of three steps ( ). In the first stage, a literature review of the development of the questions was derived from previous studies. In the second stage, the Delphi was conducted with experts, and the Delphi had a total of two rounds. The research team provided information about the study and obtained consent from all participants from a panel of experts before the study began. The participants responded to the first- and second-round questionnaires delivered through e-mail. Each round of the survey was conducted for one week, and to maximize the response rate, participants who did not respond were given additional time to send a reminder and answer a day after the deadline. In the final stage, active, balanced, connected, diversity (ABCD) was developed that reflected the opinions of experts. Round 1 The survey was open and closed to accommodate the opinions of the expert panel. The first survey consisted of 72 closed-ended and four open-ended questions. Each participant was encouraged to submit their revised recommendations and additional comments regarding the questionnaire. Round 1 of the survey took approximately 30 minutes to complete. Round 2 The Round 2 survey was revised to reflect the responses of the expert panels in Round 1. It consisted of 72 closed-ended questions in four domains (physical activity, nutrition, social relationships, and social participation). Each participant was informed via e-mail and asked to answer each question using a 4-point Likert scale (strongly irrelevant, not relevant, relevant, or strongly relevant). Neutral (ordinary) scores were not included ( ) to obtain clear opinions from the expert panel. In the second round, four items from the first round were modified. Data analysis To derive opinions, the collected response data were analyzed for means, standard deviations, medians, content validity, convergence, and stability. For general information from the expert panel, the mean, standard deviation, and median were obtained using descriptive statistics. The convergence, stability, and consensus for the collected values were calculated to confirm the validity of the item content. The degree of convergence has a value of zero when all opinions converge at a point and increases when opinion deviations are large. This means that closer to zero, the more valid the item [23–8]. The calculation formula for the CVR value is as follows ( ): During repeated survey, the smaller the response difference of the panel, the higher the stability if the response is constant, the higher the stability if the response is 0.5 ~ 0.8, and the additional investigation is required if it is 0.8 or more , The degree of agreement has a value of 1 when the opinion is congruent, and the closer to 1, the more valid the item . For the fitness value of each item, the content validity ratio (CVR) value was analyzed based on the criteria of Lawshe (1975) , and the data were analyzed based on 0.51, the minimum value according to the 14 expert panels participating in this study.
We aimed to recruit healthcare professionals who are considered experts in lifestyle and health. To this end, the researchers invited health professionals who had completed graduate master’s courses or had more than five years of experience in related fields. To increase the diversity of the expert panel, we targeted Korean professionals with recognized expertise in lifestyle and/or healthcare management within the scope of occupational therapy, physical therapy, nursing, and nutrition. The eligibility criteria were as follows: (1) ability to complete approximately three rounds of surveys within two months, (2) no restrictions on computer skills, and (3) no restrictions on access to the Internet and e-mail. After obtaining consent for participation, a Delphi survey was conducted. For those selected as potential experts, an invitation email was sent that included a letter introducing the Delphi survey, a consent form related to the study, and an IRB guide. The experts were informed that they had to respond to all rounds until a consensus was reached (approximately three rounds), each round was given an appropriate amount of time to respond, participation was voluntary, and they could withdraw at any time. They were also informed that their individual responses would not be shared with other expert panels and their contributions would be kept confidential. Delphi surveys were conducted only with experts who expressed consent, and as a reward for their time, participants were given a $50 gift card after completing all the Delphi surveys.
The study was conducted with participants who provided consent to potential expert panels. Demographic information was collected from the participants. Participants were instructed to read the guidelines introducing the concept of multi-faceted lifestyle components developed to create lifestyle assessment tools in the introduction of the survey attached to the email (Appendix 1). The first section explains the lifestyle factors of physical activity, eating habits, activity participation, and activity diversity, based on a literature review and previous studies. Participants were then shown the following sections consisting of 72 questions:18 areas of physical activity, 18 areas of nutrition, 18 areas of social relationships, and 18 areas of social participation. These items were developed based on literature review [ , – ]. The Delphi survey consisted of three steps ( ). In the first stage, a literature review of the development of the questions was derived from previous studies. In the second stage, the Delphi was conducted with experts, and the Delphi had a total of two rounds. The research team provided information about the study and obtained consent from all participants from a panel of experts before the study began. The participants responded to the first- and second-round questionnaires delivered through e-mail. Each round of the survey was conducted for one week, and to maximize the response rate, participants who did not respond were given additional time to send a reminder and answer a day after the deadline. In the final stage, active, balanced, connected, diversity (ABCD) was developed that reflected the opinions of experts.
The survey was open and closed to accommodate the opinions of the expert panel. The first survey consisted of 72 closed-ended and four open-ended questions. Each participant was encouraged to submit their revised recommendations and additional comments regarding the questionnaire. Round 1 of the survey took approximately 30 minutes to complete.
The Round 2 survey was revised to reflect the responses of the expert panels in Round 1. It consisted of 72 closed-ended questions in four domains (physical activity, nutrition, social relationships, and social participation). Each participant was informed via e-mail and asked to answer each question using a 4-point Likert scale (strongly irrelevant, not relevant, relevant, or strongly relevant). Neutral (ordinary) scores were not included ( ) to obtain clear opinions from the expert panel. In the second round, four items from the first round were modified.
To derive opinions, the collected response data were analyzed for means, standard deviations, medians, content validity, convergence, and stability. For general information from the expert panel, the mean, standard deviation, and median were obtained using descriptive statistics. The convergence, stability, and consensus for the collected values were calculated to confirm the validity of the item content. The degree of convergence has a value of zero when all opinions converge at a point and increases when opinion deviations are large. This means that closer to zero, the more valid the item [23–8]. The calculation formula for the CVR value is as follows ( ): During repeated survey, the smaller the response difference of the panel, the higher the stability if the response is constant, the higher the stability if the response is 0.5 ~ 0.8, and the additional investigation is required if it is 0.8 or more , The degree of agreement has a value of 1 when the opinion is congruent, and the closer to 1, the more valid the item . For the fitness value of each item, the content validity ratio (CVR) value was analyzed based on the criteria of Lawshe (1975) , and the data were analyzed based on 0.51, the minimum value according to the 14 expert panels participating in this study.
Demographics of the expert panels Sixteen Delphi expert invitations were sent to potential participants. Of these, 14 participants responded by expressing their consent to participate. A panel of 14 experts who participated in this study completed the first and second surveys. The expert panel that participated in this study had a 100% response rate and no participants dropped out. The demographic characteristics of the participants are shown in . Seven (50%) participants were female and seven (50%) were male. They had an average of 8.86 years of experience in their major fields of expertise. The expert panel comprised various health experts in nursing, nutrition, occupational therapy, and physical therapy. Results of round 1 The categories and items of the detailed results of the Delphi Round 1 are presented in . Fourteen experts responded to all questions in the first-round survey, and the first round consisted of 76 questions, including 72 closed-ended questions and 4 open-ended questions in four categories. In the physical activity category, the range of the CVR ratio was 0.57 ~ 1.00, and content validity was verified for all items. In the nutrition category, among the imbalanced items, the CVR ratio of two items asking about eating speed, drinking, and smoking was 0.43, falling short of the standard value. In the case of the social relations and social participation categories, the CVR ratio of each item was 0.43, and consensus was not reached. Four questions for which a consensus was not reached in the first Delphi survey were modified according to the opinions of experts written in open-ended questions, and then a second Delphi was conducted. Results of round 2 The detailed results of the Delphi Round 2 are presented in . Fourteen experts responded to all the questions in the second survey. In the first round, the questionnaire was administered by modifying four items that did not reach consensus because they did not reach the standard CVR ratio. In the physical activity category, content validity was verified for all items, with CVR ratios ranging from 0.57 to 1.00. In the case of nutrition, the two items asking about eating speed, township, and smoking, which were not agreed upon, were corrected. The CVR ratios were 0.86 and 0.57, respectively, confirming their validity. However, there was no consensus on nut consumption, with a score of 0.43. In the social relationship category, the question of whether to live alone was modified to reflect the first open-ended opinion, but the CVR ratio was 0.29; thus, a consensus was not reached. In the social relations category, the CVR ratio of the question asking about regular activity participation was 0.43, which did not reach a consensus, and the question of whether there was help for community participation reached agreement with a CVR ratio of 0.57 through modification reflecting the first open-ended opinion was reached. The second Delphi results showed a relatively low standard deviation and relatively high content validity compared with the first ( ).
Sixteen Delphi expert invitations were sent to potential participants. Of these, 14 participants responded by expressing their consent to participate. A panel of 14 experts who participated in this study completed the first and second surveys. The expert panel that participated in this study had a 100% response rate and no participants dropped out. The demographic characteristics of the participants are shown in . Seven (50%) participants were female and seven (50%) were male. They had an average of 8.86 years of experience in their major fields of expertise. The expert panel comprised various health experts in nursing, nutrition, occupational therapy, and physical therapy.
The categories and items of the detailed results of the Delphi Round 1 are presented in . Fourteen experts responded to all questions in the first-round survey, and the first round consisted of 76 questions, including 72 closed-ended questions and 4 open-ended questions in four categories. In the physical activity category, the range of the CVR ratio was 0.57 ~ 1.00, and content validity was verified for all items. In the nutrition category, among the imbalanced items, the CVR ratio of two items asking about eating speed, drinking, and smoking was 0.43, falling short of the standard value. In the case of the social relations and social participation categories, the CVR ratio of each item was 0.43, and consensus was not reached. Four questions for which a consensus was not reached in the first Delphi survey were modified according to the opinions of experts written in open-ended questions, and then a second Delphi was conducted.
The detailed results of the Delphi Round 2 are presented in . Fourteen experts responded to all the questions in the second survey. In the first round, the questionnaire was administered by modifying four items that did not reach consensus because they did not reach the standard CVR ratio. In the physical activity category, content validity was verified for all items, with CVR ratios ranging from 0.57 to 1.00. In the case of nutrition, the two items asking about eating speed, township, and smoking, which were not agreed upon, were corrected. The CVR ratios were 0.86 and 0.57, respectively, confirming their validity. However, there was no consensus on nut consumption, with a score of 0.43. In the social relationship category, the question of whether to live alone was modified to reflect the first open-ended opinion, but the CVR ratio was 0.29; thus, a consensus was not reached. In the social relations category, the CVR ratio of the question asking about regular activity participation was 0.43, which did not reach a consensus, and the question of whether there was help for community participation reached agreement with a CVR ratio of 0.57 through modification reflecting the first open-ended opinion was reached. The second Delphi results showed a relatively low standard deviation and relatively high content validity compared with the first ( ).
The necessity of developing an assessment tool that can identify lifestyle tendencies reflecting the diversity of activities and the intensity of health-related routines and habits repeated in daily life was confirmed. The development of these assessment tools will support the multi-faceted and systematic measurement of an individual’s physical, social, and mental health. The field of health promotion is increasingly emphasizing the importance of evaluating a multi-dimensional and dynamic lifestyle as an approach to shaping and improving a healthy lifestyle . In addition, Studies and reinforce our understanding that healthy lifestyles support physical and mental well-being, with specifically linking lifestyle diversity to successful aging.However, in previous studies, quantifying and analyzing the frequency and exposure of behaviors such as physical activity, smoking, and drinking had limitations in measuring aspects of habit, behavior, management, and diversity . Accordingly, we decided to develop a comprehensive evaluation tool that could evaluate an individual’s lifestyle in a multi-faceted and systematic manner by applying the Delphi technique. In this study, we discuss our proposal based on the consensus of an expert panel. Previous studies have shown that physical activity can be considered a lifestyle factor . Most national and international physical activity guidelines and statements recommend regular physical activity to prevent diseases, promote health, and delay loss . However, according to the expert panel, it was necessary to consider aspects of habits, behaviors, management, and variety and to identify individual preferences or tendencies, including quantified measures such as high-intensity, moderate-intensity, and low-intensity activities. Previous studies have reported that maintaining a healthy lifestyle increases a chronic disease-free lifespan and improves overall physical and emotional well-being . One study highlights the role of exercise in reducing chronic disease risks, while another links lifestyle adherence with longer disease-free years. Conversely, being overweight or obese is associated with chronic illness and reduced quality of life . This suggests that regular physical activity is necessary to improve health and quality of life. Therefore, when evaluating an individual’s multi-faceted lifestyle, it is necessary to assess various types of physical activities and habits. Nutrition is described as a healthy lifestyle factor along with eating habits . This study found that it is necessary to identify unbalanced eating habits, including balanced eating habits. Despite the CVR value of 0.43, nut intake was retained due to its established importance in studies on cardiovascular and immune health [ – ]. Adequate nutrition is critical for the optimal functioning of the body and for strengthening immune function, and nutritional imbalances can adversely affect the immune system . Additionally, adequate nutrition, physical activity, and proper eating habits can reduce the risk of chronic diseases and improve the immune system . A balanced diet that includes fruits, vegetables, whole grains, plant and animal proteins, and healthy fat is a good way to promote better health and proper functioning of the immune system . A balanced diet cannot control infections but can improve the immune system and mental health Therefore, when evaluating a multi-faceted lifestyle, it is necessary to evaluate balanced nutritional status and eating habits. Social relationships are important factors in physical and psychological health . Individual social relationships are a major factor in mental health, including depression and anxiety. Previous studies have reported that positive interactions with others and maintaining supportive relationships have a positive effect on physical and psychological health . This study confirms that it is necessary to ask questions about the physical and human resources that have positive and negative effects on social relationships. However, the question of whether to live alone was 0.29 based on the second round CVR ratio, and a standard value was not reached. It was difficult to determine whether a person lived alone because the type of residence had changed owing to the rapid increase in single-person households in Korea. However, this question was not addressed in this study. Previous studies have reported that social relationships are divided into intra-family and non-family relationships and that objective characteristics of situations, such as the number of social contacts or proximity, are the characteristics of social relationships . In addition, social relationships should consider the volume of relationships (how many social ties are nearby, available, and dependable) and the emotional volume (the degree of intimacy, understanding, and interest shared with others) . Therefore, the question was not deleted, but it is necessary to conduct additional research on the factors of social relationships according to the presence or absence of single-person households in the future by reflecting the additional opinions of experts. Social participation plays an important role in the quality of life as we age . Social participation is defined as engaging in community activities that provide social interaction . Social participation is an important factor in determining healthy aging . Age, gender, health status, neighborhood, and social environment have been reported as additional factors [ – ]. that have complex relationships with social participation . Similarly, the results of this study confirmed that questions on social participation require consideration of various factors. However, for the presence or absence of participation in religious activities, the second-round CVR was 0.43, and a reference value was not reached. This is because they do not consider non-religious people. However, in previous studies, religious participation has been reported to have potential effects on mental health as an important factor in social participation and to play a positive role in the daily lives of the elderly . Religious participation is closely related to the psychological, physical, and social dimensions of successful aging . Religious participation should be considered when evaluating social participation, but a more subjective understanding of religiosity and participation in religious activities may be questioned. Therefore, the results need to be interpreted cautiously. The limitations of this study were as follows. The first is the number of Delphi experts. The expert panel comprised a relatively small group of 14 members. Although useful results can be obtained with a panel of 4–11 people , there is a limit to generalization, considering that reliability increases as the number of panels increases. Nevertheless, unlike previous studies, this study is meaningful in that a panel of experts in various health-related fields was formed, and conclusions were drawn by collecting various opinions. Second, existing studies were reviewed to develop Delphi questionnaires. It cannot be ruled out that researchers may have missed various theories, backgrounds, and studies in this process. Third, the reliability and validity of the results were analyzed using the opinions of an expert panel. In future, additional studies on the validity and reliability of this questionnaire should be conducted with the public. Finally, the expert panel consisted of only Korean health experts. Therefore, this study reflects only a limited perspective on the Republic of Korea. Therefore, for an international version of this tool, expanding the panel to include experts from multiple countries would provide additional perspectives.
This study developed a multi-faceted lifestyle assessment tool using a modified Delphi technique. As a result, the 72 items were classified into eight subcategories and four categories. This assessment tool can systematically evaluate lifestyles to achieve health and physical, mental, and social well-being and can be used to identify individual lifestyles necessary for interventions in more depth. However, because the tool was developed in Korea, the possibility that it can be affected by various environmental factors, such as geographical and cultural factors, cannot be ruled out. Therefore, there is a need to conduct research on people living in various communities. Additional research should be conducted to confirm the reliability and validity of the evaluation tools developed.
|
The Chinese Version of the DigiHealthCom (Digital Health Competence) Instrument for Assessing Digital Health Competence of Health Care Professionals: Translation, Adaptation, and Validation Study | bc1665ed-72a7-4970-85fb-a002ce887554 | 11951814 | Medicine[mh] | The World Health Organization (WHO) defines digital health as the field that involves the development and use of digital technologies to improve health outcomes . In 2019, the WHO released the world’s first guidelines for digital health interventions, outlining 10 ways to use digital health technologies to enhance health and primary services . This concept extends beyond eHealth to include a wide array of smart devices and connected equipment, encompassing digital technologies such as the Internet of Things, big data, artificial intelligence, and robotics . The release of these guidelines was followed by the Global Strategy for Digital Health (2020‐2025), which emphasizes prioritized digital health strategies for global health care development . China has incorporated measures for the development of a digitally healthy nation in the 14th Five-Year Plan . With the global market projected to grow from US $211 billion in 2022 to US $809.2 billion by 2030 , digital health care is recognized as a rapidly expanding sector. Digital health has emerged as a significant trend in the evolution of global health care services. The digital transformation that the health care sector is currently undergoing is redefining the roles and responsibilities of health care professionals , creating an urgent need for digital health competence among them. Given increasing prominence of digital health in global health care landscape, it is critical for health care professionals to possess sufficient digital health competence. Digital health competence is increasingly recognized as one of the core competencies for health care professionals, which would enable them to design and evaluate the impact of digital solutions on patient care and determine the best way to implement digital solutions in their work . Although patients are becoming more accepting of and motivated to use digital health care services and tools, health care professionals face a digital skills shortage that impedes the adoption of digital solutions . Inadequate digital health competence may lead to negative experiences and frustration with technology adoption among these professionals . There is a strong association between health care professionals’ digital health competence and their willingness to use such tools . Studies have highlighted that the acceptance of digital health technologies by health care professionals significantly influences the adoption of digital solutions and emphasizes the critical role of digital health competence in ensuring patient safety . Therefore, assessing digital health competence is essential to effectively providing digital health care solutions to the public. Previous studies have attempted to explore digital health competence among health care professionals but have struggled to define it comprehensively due to the evolving nature of digital technologies . Existing assessment tools primarily focus on informatics competence, digital health literacy, or skills related to the application of digital technologies. Examples include the Digital Health Literacy Instrument (DHLI, 2017) , the eHealth literacy questionnaire (eHLQ, 2018) , and the Nursing Digital Application Skill Scale (NDASS, 2024) . The eHLQ is designed for eHealth user, especially individuals with low digital health literacy and those with chronic conditions . The NDASS targets nurses’ digital application skills in clinical settings . Digital health literacy reflects users’ knowledge and skills within their cultural, social, and institutional context . Competence entails an integrative understanding of the knowledge, skills, performance, values, and attitudes essential for the effective execution of a given task . Therefore, it is crucial to comprehensively evaluate the knowledge, skills, performance, and attitudes required for various health care professionals to adapt to the evolving digital health technologies. Developed and validated among Finnish health care professionals in 2022, the DigiHealthCom (Digital Health Competence) instrument offers a more comprehensive scope than existing tools and is applicable to a wide range of health care professionals. In addition to assessing competence in using digital solutions and information and communication technology (ICT), it also explores previously unaddressed domains that reflect future requirements, such as acceptance of digital solutions, human-centered remote counseling, and ethical competence concerning digital solutions . Furthermore, the instrument has been used to explore digital health competence profiles and associated factors in 817 health care professionals from 9 organizations in Finland . It has been translated into 15 languages, and a large-scale international cross-sectional study on the digital health competence of health care professionals is currently in progress. Our team is part of this collaborative research effort. This study aimed to culturally adapt and validate the Chinese version of DigiHealthCom for Chinese health care professionals.
Study Design A cross-sectional study was conducted. Participants Participants were recruited between May 2023 and April 2024 via convenience sampling. Recruitment posters with QR codes were disseminated on social networks specific to health care professionals. In addition, health care professionals attending local academic conferences were invited to participate. The inclusion criteria for participants were (1) employment within a health care organization and (2) consent to participate. The exclusion criteria were (1) individuals who were retired or had less than 1 year of work experience; (2) health care students; (3) individuals who completed the questionnaire in less than 150 seconds, as this indicated random clicking; or (4) individuals who displayed erratic response patterns. According to the thumb rule, the sample size should be 5 to 10 times the number of items. With an anticipated 10% rate of invalid responses, a minimum of 231 samples was necessary. Finally, a total of 398 cases were included in the study. To assess the test-retest reliability of the instrument, 43 participants who provided contact information completed the questionnaire again after a 2-week interval. Instrument The DigiHealthCom instrument comprises 5 domains, comprising a total of 42 items—competence in human-centered remote counseling (16 items), digital solutions as part of work (9 items), competence in ICT (5 items), competence in using and evaluating digital solutions (8 items), and ethical competence related to digital solutions (4 items). Each item was rated on a 4-point Likert scale (1=completely disagree, 2=partially disagree, 3=partially agree, and 4=completely agree). For each domain, a mean value of ≤2.49 indicated low competence, 2.50‐3.49 indicated intermediate competence, and ≥3.50 indicated high competence . DigiHealthCom has been validated among health care professionals across tertiary, primary, and private health care settings (n=817), demonstrating satisfactory internal consistency (Cronbach α=0.80‐0.97) and content validity (item content validity index [I-CVI]: 0.77‐1.00; S-CVI/Ave: 0.94) . Translation and Cross-Cultural Adaptation To ensure a high-quality Chinese translation of DigiHealthCom, a rigorous translation and cross-cultural adaptation process was followed . illustrates the translation and cross-cultural adaptation process. Forward and Backward Translation In total, 2 bilingual nursing professionals, who are native Chinese speakers (a nursing expert and a nursing graduate student), independently translated the English version of DigiHealthCom into Chinese, resulting in 2 forward translations (T1 and T2). A consensus panel reviewed these translations for conceptual equivalence, clarity, and comprehensibility, producing a reconciled version (T3). Subsequently, the T3 version was back-translated into English by a professional translator and a PhD student unfamiliar with the original version, both knowledgeable about Chinese and English-speaking cultures. The consensus panel reviewed and resolved discrepancies between the back-translations and the original version, finalizing the initial Chinese version. The consensus panel comprised 2 nursing researchers experienced in scale development and a linguistic expert. Expert Review To assess the content validity of the Chinese version of DigiHealthCom, 11 experts were invited to evaluate the relevance of its dimensions and items using a 4-point ordinal scale (1 [not relevant], 2 [weakly relevant], 3 [strongly relevant], and 4 [very relevant]). The expert panel consisted of 9 health care specialists and 2 IT experts. In addition, expert evaluated the comprehensibility of the items. The consensus panel, initially involved in the translation process, conducted a detailed discussion to resolve any potential discrepancies. Cognitive Interviewing Cognitive interviewing was used to evaluate the clarity and cultural suitability of the initial Chinese version. A total of 7 native Chinese speakers, including 2 doctors, 4 nurses, and 1 IT technician, were recruited. Participants were briefed on the study’s objectives and methods before interviews, and their consent was obtained. The first author, trained in qualitative research methods, conducted the interviews in a meeting room. Participants completed the Chinese version of DigiHealthCom independently and engaged in cognitive interviews. Interviewers evaluated whether participants found the items relevant to their condition and if they encountered any understanding difficulties. Field notes were taken. Modifications were deemed necessary if at least 1 participant (1) found an item difficult to understand, (2) demonstrated mostly or completely inaccurate comprehension of an item, or (3) provided feedback indicating the need for improvements, especially regarding cultural relevance. The consensus panel determined whether to retain or alter item wording following expert review and cognitive interviewing. Discrepancies affecting item clarity were resolved to create the pretest version of the Chinese DigiHealthCom version. Specific item modifications were detailed in Table S1 in . Pilot Study A pilot study was conducted with 39 health care professionals from a general hospital in Guangzhou, China. The pilot study confirmed that no further modifications were required. The final Chinese version of DigiHealthCom consists of 5 dimensions and 42 items. Data Collection Data were collected using the Wen Juan Xing (a Chinese web questionnaire platform) . Respondents accessed the questionnaire by scanning a QR code or clicking a link, with 1 response allowed per IP address to prevent duplicates. The questionnaire platform performed completeness checks before submission. The Chinese version of DigiHealthCom was presented alongside a sociodemographic questionnaire. The questionnaire consisted of 2 pages and 73 items. Sex, age, education, service area, professional license, work experience, type of organization, and frequency of patient work were collected. Participants who voluntarily chose to provide contact information participated in a retest after a 2-week interval. An introduction outlined the study’s purpose and provided questionnaire instructions. Statistical Analysis Item analysis was used to evaluate item discrimination, item correlation, and item homogeneity. For item discrimination, respondents were classified into high-score (top 27%) and low-score (bottom 27%) groups based on their total scores. An independent t test determined whether each item could significantly distinguish between these groups. Items with a critical ratio (| t |) less than 3.0 were considered for exclusion . In addition, item homogeneity and item correlation were tested using Cronbach α if the item was deleted and corrected item-total correlation coefficient. Items with a corrected item-total correlation coefficient below 0.40 were considered for exclusion . Cronbach α and Intraclass correlation coefficient (ICC) were used to assess the internal consistency and test-retest reliability of the instrument. A Cronbach α≥0.70 indicated good internal consistency, while ICC>0.70 indicated good time stability . The I-CVI and the Scale CVI/Average (S-CVI/Ave) were used to evaluate content validity of the instrument. I-CVI≥0.78 and S-CVI/Ave≥0.90 indicate satisfactory content validity . Construct validity was evaluated using confirmatory factor analysis (CFA) with a robust chi-square to df ratio ( χ 2 /df ) less than 3, Tucker-Lewis Index (TLI), Incremental Fit Index (IFI), and Comparative Fit Index (CFI) greater than 0.90, and root-mean-square error of approximation (RMSEA) and standardized root-mean-square residual (SRMR) less than 0.08, indicating an acceptable data-model fit . The average variance extracted (AVE) and composite reliability (CR) were used to assess convergent validity, with AVE greater than 0.50 and CR greater than 0.70 indicating good convergent validity . Discriminant validity was performed using the heterotrait-monotrait ratio (HTMT), with a correlation matrix value <0.90 considered good . Furthermore, participants were divided into 2 groups based on geographic location for the sensitivity analysis. The DigiHealthCom scores were compared to evaluate potential selection bias. The absence of a significant difference in digital health competence between the groups indicates that selection bias is unlikely in the study sample. Statistical analysis was conducted using IBM SPSS (version 27.0), IBM AMOS (version 29.0), and Smart PLS (Version 4.1.0.0; GmbH Corp). A P value of less than .05 was considered statistically significant. Ethical Considerations Ethical approval was obtained from the Ethical Committee at Nanfang Hospital, Southern Medical University, China (NFEC-2023‐165). This research adhered to the principals of the Declaration of Helsinki. All participants provided informed consent and voluntarily completed the web questionnaire. Participants had the option to skip questions, review, and delete their responses. Participants had the right to withdraw from the survey at any time. Those who completed it received a random monetary reward ranging from CNY 2 to 5 (US $0.28 to 0.69). Participants’ rights and researcher’s contact information were provided on the first page of the web survey. Minimal sociodemographic was collected to maintain ethical standards. Participant information was kept confidential and anonymous. The CHERRIES (Checklist for Reporting Results of Internet E-Surveys; ) was used to enhance the transparency of the study .
A cross-sectional study was conducted.
Participants were recruited between May 2023 and April 2024 via convenience sampling. Recruitment posters with QR codes were disseminated on social networks specific to health care professionals. In addition, health care professionals attending local academic conferences were invited to participate. The inclusion criteria for participants were (1) employment within a health care organization and (2) consent to participate. The exclusion criteria were (1) individuals who were retired or had less than 1 year of work experience; (2) health care students; (3) individuals who completed the questionnaire in less than 150 seconds, as this indicated random clicking; or (4) individuals who displayed erratic response patterns. According to the thumb rule, the sample size should be 5 to 10 times the number of items. With an anticipated 10% rate of invalid responses, a minimum of 231 samples was necessary. Finally, a total of 398 cases were included in the study. To assess the test-retest reliability of the instrument, 43 participants who provided contact information completed the questionnaire again after a 2-week interval.
The DigiHealthCom instrument comprises 5 domains, comprising a total of 42 items—competence in human-centered remote counseling (16 items), digital solutions as part of work (9 items), competence in ICT (5 items), competence in using and evaluating digital solutions (8 items), and ethical competence related to digital solutions (4 items). Each item was rated on a 4-point Likert scale (1=completely disagree, 2=partially disagree, 3=partially agree, and 4=completely agree). For each domain, a mean value of ≤2.49 indicated low competence, 2.50‐3.49 indicated intermediate competence, and ≥3.50 indicated high competence . DigiHealthCom has been validated among health care professionals across tertiary, primary, and private health care settings (n=817), demonstrating satisfactory internal consistency (Cronbach α=0.80‐0.97) and content validity (item content validity index [I-CVI]: 0.77‐1.00; S-CVI/Ave: 0.94) . Translation and Cross-Cultural Adaptation To ensure a high-quality Chinese translation of DigiHealthCom, a rigorous translation and cross-cultural adaptation process was followed . illustrates the translation and cross-cultural adaptation process. Forward and Backward Translation In total, 2 bilingual nursing professionals, who are native Chinese speakers (a nursing expert and a nursing graduate student), independently translated the English version of DigiHealthCom into Chinese, resulting in 2 forward translations (T1 and T2). A consensus panel reviewed these translations for conceptual equivalence, clarity, and comprehensibility, producing a reconciled version (T3). Subsequently, the T3 version was back-translated into English by a professional translator and a PhD student unfamiliar with the original version, both knowledgeable about Chinese and English-speaking cultures. The consensus panel reviewed and resolved discrepancies between the back-translations and the original version, finalizing the initial Chinese version. The consensus panel comprised 2 nursing researchers experienced in scale development and a linguistic expert. Expert Review To assess the content validity of the Chinese version of DigiHealthCom, 11 experts were invited to evaluate the relevance of its dimensions and items using a 4-point ordinal scale (1 [not relevant], 2 [weakly relevant], 3 [strongly relevant], and 4 [very relevant]). The expert panel consisted of 9 health care specialists and 2 IT experts. In addition, expert evaluated the comprehensibility of the items. The consensus panel, initially involved in the translation process, conducted a detailed discussion to resolve any potential discrepancies. Cognitive Interviewing Cognitive interviewing was used to evaluate the clarity and cultural suitability of the initial Chinese version. A total of 7 native Chinese speakers, including 2 doctors, 4 nurses, and 1 IT technician, were recruited. Participants were briefed on the study’s objectives and methods before interviews, and their consent was obtained. The first author, trained in qualitative research methods, conducted the interviews in a meeting room. Participants completed the Chinese version of DigiHealthCom independently and engaged in cognitive interviews. Interviewers evaluated whether participants found the items relevant to their condition and if they encountered any understanding difficulties. Field notes were taken. Modifications were deemed necessary if at least 1 participant (1) found an item difficult to understand, (2) demonstrated mostly or completely inaccurate comprehension of an item, or (3) provided feedback indicating the need for improvements, especially regarding cultural relevance. The consensus panel determined whether to retain or alter item wording following expert review and cognitive interviewing. Discrepancies affecting item clarity were resolved to create the pretest version of the Chinese DigiHealthCom version. Specific item modifications were detailed in Table S1 in . Pilot Study A pilot study was conducted with 39 health care professionals from a general hospital in Guangzhou, China. The pilot study confirmed that no further modifications were required. The final Chinese version of DigiHealthCom consists of 5 dimensions and 42 items.
To ensure a high-quality Chinese translation of DigiHealthCom, a rigorous translation and cross-cultural adaptation process was followed . illustrates the translation and cross-cultural adaptation process.
In total, 2 bilingual nursing professionals, who are native Chinese speakers (a nursing expert and a nursing graduate student), independently translated the English version of DigiHealthCom into Chinese, resulting in 2 forward translations (T1 and T2). A consensus panel reviewed these translations for conceptual equivalence, clarity, and comprehensibility, producing a reconciled version (T3). Subsequently, the T3 version was back-translated into English by a professional translator and a PhD student unfamiliar with the original version, both knowledgeable about Chinese and English-speaking cultures. The consensus panel reviewed and resolved discrepancies between the back-translations and the original version, finalizing the initial Chinese version. The consensus panel comprised 2 nursing researchers experienced in scale development and a linguistic expert.
To assess the content validity of the Chinese version of DigiHealthCom, 11 experts were invited to evaluate the relevance of its dimensions and items using a 4-point ordinal scale (1 [not relevant], 2 [weakly relevant], 3 [strongly relevant], and 4 [very relevant]). The expert panel consisted of 9 health care specialists and 2 IT experts. In addition, expert evaluated the comprehensibility of the items. The consensus panel, initially involved in the translation process, conducted a detailed discussion to resolve any potential discrepancies.
Cognitive interviewing was used to evaluate the clarity and cultural suitability of the initial Chinese version. A total of 7 native Chinese speakers, including 2 doctors, 4 nurses, and 1 IT technician, were recruited. Participants were briefed on the study’s objectives and methods before interviews, and their consent was obtained. The first author, trained in qualitative research methods, conducted the interviews in a meeting room. Participants completed the Chinese version of DigiHealthCom independently and engaged in cognitive interviews. Interviewers evaluated whether participants found the items relevant to their condition and if they encountered any understanding difficulties. Field notes were taken. Modifications were deemed necessary if at least 1 participant (1) found an item difficult to understand, (2) demonstrated mostly or completely inaccurate comprehension of an item, or (3) provided feedback indicating the need for improvements, especially regarding cultural relevance. The consensus panel determined whether to retain or alter item wording following expert review and cognitive interviewing. Discrepancies affecting item clarity were resolved to create the pretest version of the Chinese DigiHealthCom version. Specific item modifications were detailed in Table S1 in .
A pilot study was conducted with 39 health care professionals from a general hospital in Guangzhou, China. The pilot study confirmed that no further modifications were required. The final Chinese version of DigiHealthCom consists of 5 dimensions and 42 items.
Data were collected using the Wen Juan Xing (a Chinese web questionnaire platform) . Respondents accessed the questionnaire by scanning a QR code or clicking a link, with 1 response allowed per IP address to prevent duplicates. The questionnaire platform performed completeness checks before submission. The Chinese version of DigiHealthCom was presented alongside a sociodemographic questionnaire. The questionnaire consisted of 2 pages and 73 items. Sex, age, education, service area, professional license, work experience, type of organization, and frequency of patient work were collected. Participants who voluntarily chose to provide contact information participated in a retest after a 2-week interval. An introduction outlined the study’s purpose and provided questionnaire instructions.
Item analysis was used to evaluate item discrimination, item correlation, and item homogeneity. For item discrimination, respondents were classified into high-score (top 27%) and low-score (bottom 27%) groups based on their total scores. An independent t test determined whether each item could significantly distinguish between these groups. Items with a critical ratio (| t |) less than 3.0 were considered for exclusion . In addition, item homogeneity and item correlation were tested using Cronbach α if the item was deleted and corrected item-total correlation coefficient. Items with a corrected item-total correlation coefficient below 0.40 were considered for exclusion . Cronbach α and Intraclass correlation coefficient (ICC) were used to assess the internal consistency and test-retest reliability of the instrument. A Cronbach α≥0.70 indicated good internal consistency, while ICC>0.70 indicated good time stability . The I-CVI and the Scale CVI/Average (S-CVI/Ave) were used to evaluate content validity of the instrument. I-CVI≥0.78 and S-CVI/Ave≥0.90 indicate satisfactory content validity . Construct validity was evaluated using confirmatory factor analysis (CFA) with a robust chi-square to df ratio ( χ 2 /df ) less than 3, Tucker-Lewis Index (TLI), Incremental Fit Index (IFI), and Comparative Fit Index (CFI) greater than 0.90, and root-mean-square error of approximation (RMSEA) and standardized root-mean-square residual (SRMR) less than 0.08, indicating an acceptable data-model fit . The average variance extracted (AVE) and composite reliability (CR) were used to assess convergent validity, with AVE greater than 0.50 and CR greater than 0.70 indicating good convergent validity . Discriminant validity was performed using the heterotrait-monotrait ratio (HTMT), with a correlation matrix value <0.90 considered good . Furthermore, participants were divided into 2 groups based on geographic location for the sensitivity analysis. The DigiHealthCom scores were compared to evaluate potential selection bias. The absence of a significant difference in digital health competence between the groups indicates that selection bias is unlikely in the study sample. Statistical analysis was conducted using IBM SPSS (version 27.0), IBM AMOS (version 29.0), and Smart PLS (Version 4.1.0.0; GmbH Corp). A P value of less than .05 was considered statistically significant.
Ethical approval was obtained from the Ethical Committee at Nanfang Hospital, Southern Medical University, China (NFEC-2023‐165). This research adhered to the principals of the Declaration of Helsinki. All participants provided informed consent and voluntarily completed the web questionnaire. Participants had the option to skip questions, review, and delete their responses. Participants had the right to withdraw from the survey at any time. Those who completed it received a random monetary reward ranging from CNY 2 to 5 (US $0.28 to 0.69). Participants’ rights and researcher’s contact information were provided on the first page of the web survey. Minimal sociodemographic was collected to maintain ethical standards. Participant information was kept confidential and anonymous. The CHERRIES (Checklist for Reporting Results of Internet E-Surveys; ) was used to enhance the transparency of the study .
Characteristics of the Participants Initially, 431 participants were recruited for this study. Furthermore, 33 participants were subsequently excluded due to not meeting the inclusion criteria (n=7), having less than 1 year work experience (n=4), completing the questionnaire in under 150 seconds (n=2), and exhibiting erratic response patterns (n=20). Finally, 398 eligible health care professionals were included. illustrates the participant enrollment flowchart. The participants enrolled from 36 cities across 15 provinces, as identified by IP addresses. Among the participants, 249 (62.6%) were recruited from Guangzhou, a megacity in Guangdong province that contains 6125 registered medical facilities. As shown in , 364 (91.5%) participants were female, 386 (97%) worked in health care services, and 357 (89.7%) were nurses. The participants had an average of 13.7 (SD 9.4) years of work experience, with 281 (70.6%) working directly with patients for at least 5 days per week. Participant characteristics are summarized in . Results of Item Analysis Item analysis showed a significant difference between high-score and low-score groups. The critical ratio values for all items were above 3.0 (range 16.05‐23.77, P <.001; ), indicating excellent item discrimination. All corrected item-total correlation coefficients exceeded 0.4 ( ). The Cronbach α if the item deleted (range 0.91‐0.96) were acceptable ( ). Internal Consistency and Test-Retest Reliability As shown in , internal consistency (Cronbach α=0.94‐0.98) and test-retest reliability were good (ICC=0.90; 95% CI 0.81‐0.95). Content Validity To test content validity, 11 experts evaluated the relevance of the items. The S-CVI/Ave for the instrument was 0.97, and the I-CVI ranged from 0.82 to 1.00 ( ), indicating satisfactory content validity. Construct Validity As illustrated in , the results of CFA supported the 5-factors structure (digital solutions as part of work; ethical competence related to digital solutions; ICT competence; human-centered remote counselling competence; competence in using and evaluating digital solutions). The indices, including χ 2 /df (3.10), CFI (0.91), TLI (0.90), IFI (0.91), RMSEA (0.07), and SRMR (0.05), indicated an acceptable model fit. The factor loadings of items ranged from 0.69 to 0.93 ( ). The AVE values for each dimension exceeded 0.50 (range: 0.60‐0.79), with the CR values >0.70 (range: 0.94‐0.95), indicating excellent convergent validity. The HTMT values within the matrix were less than 0.90 (range: 0.72‐0.89), indicating good divergent validity. These results collectively demonstrated satisfactory construct validity of the instrument. Sensitivity Analysis Participants were divided into 2 groups based on geographic location: Southern China and Northern and Western China. The DigiHealthCom scores were compared to evaluate potential selection bias. As illustrated in Table S3 in , no significant difference was found in DigiHealthCom scores between the 2 groups ( P >.05), indicating that selection bias is unlikely in the study sample.
Initially, 431 participants were recruited for this study. Furthermore, 33 participants were subsequently excluded due to not meeting the inclusion criteria (n=7), having less than 1 year work experience (n=4), completing the questionnaire in under 150 seconds (n=2), and exhibiting erratic response patterns (n=20). Finally, 398 eligible health care professionals were included. illustrates the participant enrollment flowchart. The participants enrolled from 36 cities across 15 provinces, as identified by IP addresses. Among the participants, 249 (62.6%) were recruited from Guangzhou, a megacity in Guangdong province that contains 6125 registered medical facilities. As shown in , 364 (91.5%) participants were female, 386 (97%) worked in health care services, and 357 (89.7%) were nurses. The participants had an average of 13.7 (SD 9.4) years of work experience, with 281 (70.6%) working directly with patients for at least 5 days per week. Participant characteristics are summarized in .
Item analysis showed a significant difference between high-score and low-score groups. The critical ratio values for all items were above 3.0 (range 16.05‐23.77, P <.001; ), indicating excellent item discrimination. All corrected item-total correlation coefficients exceeded 0.4 ( ). The Cronbach α if the item deleted (range 0.91‐0.96) were acceptable ( ).
As shown in , internal consistency (Cronbach α=0.94‐0.98) and test-retest reliability were good (ICC=0.90; 95% CI 0.81‐0.95).
To test content validity, 11 experts evaluated the relevance of the items. The S-CVI/Ave for the instrument was 0.97, and the I-CVI ranged from 0.82 to 1.00 ( ), indicating satisfactory content validity.
As illustrated in , the results of CFA supported the 5-factors structure (digital solutions as part of work; ethical competence related to digital solutions; ICT competence; human-centered remote counselling competence; competence in using and evaluating digital solutions). The indices, including χ 2 /df (3.10), CFI (0.91), TLI (0.90), IFI (0.91), RMSEA (0.07), and SRMR (0.05), indicated an acceptable model fit. The factor loadings of items ranged from 0.69 to 0.93 ( ). The AVE values for each dimension exceeded 0.50 (range: 0.60‐0.79), with the CR values >0.70 (range: 0.94‐0.95), indicating excellent convergent validity. The HTMT values within the matrix were less than 0.90 (range: 0.72‐0.89), indicating good divergent validity. These results collectively demonstrated satisfactory construct validity of the instrument.
Participants were divided into 2 groups based on geographic location: Southern China and Northern and Western China. The DigiHealthCom scores were compared to evaluate potential selection bias. As illustrated in Table S3 in , no significant difference was found in DigiHealthCom scores between the 2 groups ( P >.05), indicating that selection bias is unlikely in the study sample.
Principal Findings In this study, we translated and culturally adapted the DigiHealthCom instrument into Chinese and assessed its reliability and validity among Chinese health care professionals. The Chinese version of DigiHealthCom demonstrated satisfactory internal consistency, test-retest reliability, content validity, and construct validity. To our knowledge, this study represents the first effort to validate a comprehensive Chinese tool for measuring digital health competence among health care professionals. In this study, each item of the Chinese version of DigiHealthCom was translated and back-translated strictly following the dual direct-to-back translation model to ensure alignment in semantic, conceptual, and content with the original English version. And then, we conducted expert review involving 11 experts in nursing and IT to examine content validity of the Chinese version. High content validity indices imply that the instrument provides a broad enough range of content to allow conclusions about the targeted construct. We also conducted cognitive interviewing and a pilot study to ensure its clarity and comprehensibility . Issues related to semantic validation and understanding identified in the initial Chinese version were addressed after receiving feedback from the expert review and cognitive interviewing, leading to improvements such as clarifying complex terms and vague definitions. Item analysis revealed good differentiation and high correlations between the items in this study. Furthermore, this instrument underwent rigorous testing for internal consistency, test-retest reliability, and construct validity, exhibiting excellent internal consistency, time stability, and construct validity, resonating with the original study . The types of reliability and validity assessed in this study represent essential psychometric properties for a measurement instrument. Discrepancies between existing instruments largely stem from differing conceptual frameworks. The DHLI measures 7 individual skills: operation, navigation, information searching, evaluating reliability, determining relevance, adding self-generated content, and protecting privacy . The eHLQ comprises 7 dimensions, addressing users’ attributes, their interaction with technologies, and their experience with systems . Our CFA findings confirm a 5-factor structure, consistent with the exploratory factor analysis findings of the initial study . Beyond competence in using and evaluating digital solutions and ICTs, the DigiHealthCom instrument addresses additional essential domains that were previously unaddressed, that is acceptance of digital solutions, human-centered remote counseling, digital interaction skills with patients and interprofessional teams, and ethical competence related to digital solutions . Notably, acceptance of digital solutions significantly influences health care professionals’ adoption of digital solutions . Competence in person-centered remote consultations is crucial for fostering patient engagement and improving accessibility to equitable digital health services . Furthermore, proficient digital interaction skills facilitate effective coordination in digital settings, thereby supporting decision-making and optimizing treatment plans . As digital solutions become more prevalent, attention to data privacy and information security has intensified . Ethical competence related to digital solutions ensures appropriate management of patient information, fostering trust in digital health technologies and enhancing the safety of digital health services . These domains provide a unified theoretical framework for comprehensively measuring digital health competencies among various professionals , which are essential for delivering high-quality digital health services to meet future demands. The Finnish health care system is renowned for its universality, robust public funding, and centralized digital infrastructure . Regional variations exist in Finland; for example, northern districts have fewer professionals specializing in remote counseling and the integration of digital solutions compared with southern areas . Strategies have been implemented to enhance health care professionals’ digital health competence and improve education related to health care digitalization, including technology to support client engagement, digital services integrated into nursing work, and considerations of safety and ethics in the digital environment . In contrast, China experiences regional disparities in digital health infrastructure and access to digital health services, especially in rural areas, despite the rapid growth of digital health services . In addition, system compatibility across hospitals remains an unsolved issue . These factors may influence the development of digital health competence among health care professionals. A cross-sectional study conducted in a central Chinese province reported that 49.9 % (1690/3386) of clinical nurses demonstrate low telehealth readiness . Consequently, evaluating, addressing, and reducing regional disparities in health care professionals’ digital health competence is vital for promoting equity in health care service provision. Validating a comprehensive digital health competence measurement instrument for health care professionals is imperative to address challenges posed by diverse health care environments, such as China’s. The Chinese version of DigiHealthCom is now ready for application in a wide range of settings and various health care professionals. Researchers, educators, health care providers, and policy makers can use it to evaluate digital health competence among diverse health care professionals, and develop digital health training curricula and policies based on the assessment. Limitations This study acknowledges several limitations. First, the primary participants were nurses. Future research should involve a broader spectrum of health care professionals. Second, although recruited from diverse health care institutions, the participants mainly came from southern China, which may introduce selection bias due to regional disparities in digital health infrastructure. To address this, participants were divided into 2 groups (Southern China versus Northern and Western China) for sensitivity analysis, and their DigiHealthCom scores were compared. We did not find significant differences in digital health competence between the 2 groups, indicating that selection bias is unlikely in the study sample. Future studies could benefit from employing a stratified sampling method. Third, the absence of a standardized method for assessing digital health competence among health care professionals hindered the evaluation of criterion validity. Assessments of digital health competence predominantly rely on subjective self-reports, which might be susceptible to response bias. However, relying exclusively on performance-based assessments or other objective measures to evaluate competence also presents challenges. Therefore, using a more holistic combination of methods could offer a viable solution. Future research should consider larger sample sizes and multicenter external evaluations to address these limitations. Conclusions In conclusion, we have successfully translated, culturally adapted, and validated DigiHealthCom for Chinese health care professionals. Our findings demonstrate that the Chinese version of DigiHealthCom is a reliable and valid instrument for assessing digital health competence among these health care professionals.
In this study, we translated and culturally adapted the DigiHealthCom instrument into Chinese and assessed its reliability and validity among Chinese health care professionals. The Chinese version of DigiHealthCom demonstrated satisfactory internal consistency, test-retest reliability, content validity, and construct validity. To our knowledge, this study represents the first effort to validate a comprehensive Chinese tool for measuring digital health competence among health care professionals. In this study, each item of the Chinese version of DigiHealthCom was translated and back-translated strictly following the dual direct-to-back translation model to ensure alignment in semantic, conceptual, and content with the original English version. And then, we conducted expert review involving 11 experts in nursing and IT to examine content validity of the Chinese version. High content validity indices imply that the instrument provides a broad enough range of content to allow conclusions about the targeted construct. We also conducted cognitive interviewing and a pilot study to ensure its clarity and comprehensibility . Issues related to semantic validation and understanding identified in the initial Chinese version were addressed after receiving feedback from the expert review and cognitive interviewing, leading to improvements such as clarifying complex terms and vague definitions. Item analysis revealed good differentiation and high correlations between the items in this study. Furthermore, this instrument underwent rigorous testing for internal consistency, test-retest reliability, and construct validity, exhibiting excellent internal consistency, time stability, and construct validity, resonating with the original study . The types of reliability and validity assessed in this study represent essential psychometric properties for a measurement instrument. Discrepancies between existing instruments largely stem from differing conceptual frameworks. The DHLI measures 7 individual skills: operation, navigation, information searching, evaluating reliability, determining relevance, adding self-generated content, and protecting privacy . The eHLQ comprises 7 dimensions, addressing users’ attributes, their interaction with technologies, and their experience with systems . Our CFA findings confirm a 5-factor structure, consistent with the exploratory factor analysis findings of the initial study . Beyond competence in using and evaluating digital solutions and ICTs, the DigiHealthCom instrument addresses additional essential domains that were previously unaddressed, that is acceptance of digital solutions, human-centered remote counseling, digital interaction skills with patients and interprofessional teams, and ethical competence related to digital solutions . Notably, acceptance of digital solutions significantly influences health care professionals’ adoption of digital solutions . Competence in person-centered remote consultations is crucial for fostering patient engagement and improving accessibility to equitable digital health services . Furthermore, proficient digital interaction skills facilitate effective coordination in digital settings, thereby supporting decision-making and optimizing treatment plans . As digital solutions become more prevalent, attention to data privacy and information security has intensified . Ethical competence related to digital solutions ensures appropriate management of patient information, fostering trust in digital health technologies and enhancing the safety of digital health services . These domains provide a unified theoretical framework for comprehensively measuring digital health competencies among various professionals , which are essential for delivering high-quality digital health services to meet future demands. The Finnish health care system is renowned for its universality, robust public funding, and centralized digital infrastructure . Regional variations exist in Finland; for example, northern districts have fewer professionals specializing in remote counseling and the integration of digital solutions compared with southern areas . Strategies have been implemented to enhance health care professionals’ digital health competence and improve education related to health care digitalization, including technology to support client engagement, digital services integrated into nursing work, and considerations of safety and ethics in the digital environment . In contrast, China experiences regional disparities in digital health infrastructure and access to digital health services, especially in rural areas, despite the rapid growth of digital health services . In addition, system compatibility across hospitals remains an unsolved issue . These factors may influence the development of digital health competence among health care professionals. A cross-sectional study conducted in a central Chinese province reported that 49.9 % (1690/3386) of clinical nurses demonstrate low telehealth readiness . Consequently, evaluating, addressing, and reducing regional disparities in health care professionals’ digital health competence is vital for promoting equity in health care service provision. Validating a comprehensive digital health competence measurement instrument for health care professionals is imperative to address challenges posed by diverse health care environments, such as China’s. The Chinese version of DigiHealthCom is now ready for application in a wide range of settings and various health care professionals. Researchers, educators, health care providers, and policy makers can use it to evaluate digital health competence among diverse health care professionals, and develop digital health training curricula and policies based on the assessment.
This study acknowledges several limitations. First, the primary participants were nurses. Future research should involve a broader spectrum of health care professionals. Second, although recruited from diverse health care institutions, the participants mainly came from southern China, which may introduce selection bias due to regional disparities in digital health infrastructure. To address this, participants were divided into 2 groups (Southern China versus Northern and Western China) for sensitivity analysis, and their DigiHealthCom scores were compared. We did not find significant differences in digital health competence between the 2 groups, indicating that selection bias is unlikely in the study sample. Future studies could benefit from employing a stratified sampling method. Third, the absence of a standardized method for assessing digital health competence among health care professionals hindered the evaluation of criterion validity. Assessments of digital health competence predominantly rely on subjective self-reports, which might be susceptible to response bias. However, relying exclusively on performance-based assessments or other objective measures to evaluate competence also presents challenges. Therefore, using a more holistic combination of methods could offer a viable solution. Future research should consider larger sample sizes and multicenter external evaluations to address these limitations.
In conclusion, we have successfully translated, culturally adapted, and validated DigiHealthCom for Chinese health care professionals. Our findings demonstrate that the Chinese version of DigiHealthCom is a reliable and valid instrument for assessing digital health competence among these health care professionals.
10.2196/65373 Multimedia Appendix 1 Modified items of the Chinese version of the DigiHealthCom (Digital Health Competence) instrument after cultural adaption. 10.2196/65373 Multimedia Appendix 2 Comparison of DigiHealthCom (Digital Health Competence) scores between health care professionals in the Southern China group and in the Northern and Western China group. 10.2196/65373 Checklist 1 Checklist for Reporting Results of Internet E-Surveys (CHERRIES).
|
Maturation of the Visceral (Gut-Adipose-Liver) Network in Response to the Weaning Reaction versus Adult Age and Impact of Maternal High-Fat Diet | e1a7ae3a-6ab5-42a2-92b2-cd8e14a0001b | 8541006 | Anatomy[mh] | Due to a pandemic spread of obesity, metabolic-associated fatty liver disease and hepatic inflammation have become the most frequent causes of liver-related and metabolic co-morbidity in adults and children . Primary prevention is the only way to circumvent the current lack of effective therapy and of screening tools able to capture individuals at risk of progression. Among early, modifiable, and most common risk factors, maternal obesity has a life-long impact on body weight and gut microbiota composition, programming metabolic disorders in the offspring ; however, its direct contribution to liver disease remains controversial. Some studies indicate that a maternal high-fat diet (HFD) is sufficient , whereas others indicate that post-natal HFD in the offspring is necessary to determine severe liver damage . In addition, one study indicates that a mismatch of normal diet (ND) in mothers and HFD in offspring is the only triggering combination . The liver is located in a strategic position to handle substrates, metabolites, and pro-inflammatory factors deriving from the gut and visceral fat, and regulates their access to the systemic circulation. In fact, these three organs receive blood directly from mesenteric or portal routes, thereby establishing an internal (pre-systemic) anatomical and functional network before connecting with the rest of the body. They share a primary role in triggering/blunting inflammatory states [ , , , , ], and they determine the fate of endogenous and exogenous metabolic substrates. Their composition and function undergo important changes during early growth, including the hematopoietic-to-metabolic transition of the liver , the thickening and replication of intestinal glands, with immune cell and microbiota colonization , and the proliferation and differentiation of (pro)adipocytes via hyperplasia , which typically occurs in early life. In particular, the weaning period represents a critical time window for the programming of a life-long cooperation between the gut, visceral fat, and liver, as the dramatic and rapid bacterial changes caused by weaning provoke a benign inflammatory reaction, imprinting the immune system, and reducing the likelihood of chronic and inflammatory diseases in adult life [ , , , ]. During the life course, these organs continue to co-modulate the risk of systemic and liver dysmetabolism. Each of them has been implicated in the pathogenesis of chronic metabolic diseases, but their cross-talk along maturation windows remains to be elucidated, together with the impact of maternal obesity. Understanding these relationships will translate into diversified, mechanism-targeted prevention strategies and time windows. Multimodal imaging of abdominal organs by, e.g., positron emission tomography (PET) and computerized tomography (CT) attenuation measures (reflecting organ structure, fibrosis, lipid content) offers the unique possibility to address relationships between gut, liver, and visceral fat in vivo. This has already revealed that tissue glucose uptake plays a role in triglyceride accumulation and release (16), that intestinal glucose uptake is impaired in metabolic diseases , and that visceral fat and liver glucose metabolism are correlated in adult humans , in whom adipose tissue may serve as a sink against liver substrate overload , at the expense of adipocyte enlargement and low-grade systemic inflammation. In the current study, we hypothesized that the simultaneous evaluation of glucose handling and tissue radiodensity in interactive abdominal organs in the context of the weaning reaction and in adulthood, as well as in high- vs. low-risk mice, would disclose new insights in the pathogenesis and indicators of metabolic and liver damage, its potential modulators or primary prevention targets, and time windows. We conjectured that the pattern of microbiota maturation and its location (caecum versus colon) could participate in the regulation of the visceral cross-talk. To explore these hypotheses, in vivo imaging of pre-systemic and post-systemic glucose extraction and glucose uptake (by PET and 18 F-labelled fluorodeoxyglucose: 18 FDG) and tissue radiodensity (by CT) in the liver, visceral fat, and intestine in combination with liver histology and microbiota and a metabolic pathway KEGG analysis (caecum and colon) were carried out in the offspring of HFD and ND dams at the ages of weaning or 6 months.
2.1. Study Design B6129SF2/J (stock no: 101045, The Jackson Laboratory, Bar Harbor, Maine) female mice underwent ND (11% kcals from fat, Mucedola s.r.l., Milan, Italy, n = 5), or HFD (58% kcals from fat, Mucedola, n = 4) for 3 months before mating, and during gestation and lactation. After weaning, offspring (total n = 38) were fed with a standard diet. Animals were housed under standard conditions (22 °C, 12-hour light/dark cycles), with ad libitum access to food and water. Food intake, body weight, and glycemia have been already reported together with cognitive data . At weaning ( n = 19, NDoff n = 11, HFDoff n = 8) or at 6 months of age (adulthood, n = 19, NDoff n = 10, HFDoff n = 9), liver, visceral fat, and intestinal glucose uptake and density were measured by PET-CT imaging. Then, animals were euthanized by anesthetic overdose, and liver biopsies, blood, and fecal samples (colon, caecum) were collected. 2.2. Tissue Metabolism Imaging of 18 FDG was performed under fasting conditions (PET-CT IRIS, Inviscan SAS, Strasbourg, France), under isofluorane anesthesia (IsoFlo ® , Abbott Laboratories, Chicago, IL, USA). Breath frequency and temperature were monitored during the study and a heated pad was used to prevent hypothermia. 18 FDG was administered by intraperitoneal (i.p.) injection to enable first-pass glucose uptake via mesenteric and portal vein vessels in visceral organs (pre-systemic), before entry into/delivery from the general circulation (post-systemic glucose uptake). A 60-minute whole-body dynamic PET scan was performed, and glycemia was measured in tail blood by a glucometer (OneTouch, Johnson&Johnson Medical SpA, Pomezia, Italy). PET data were corrected for dead time, random coincidences, and radioactive decay, and reconstructed by a 3D-Ordered Subset Expectation Maximization (3D-OSEM) algorithm. CT images where corrected for beam hardening and ring artifacts, reconstructed with cone-beam filtered backprojection (FBP), and calibrated in Hounsfield units (HU). All PET and CT images were exported to DICOM format after reconstruction, and fused within the AMIDE Medical Image Data Examiner 1.0.4 ( http://amide.sourceforge.net/ , accessed on 2020). Regions of interest were manually drawn on PET-CT images in correspondence with the right liver lobe, the large intestine, and lower abdominal visceral fat. Tissue radiodensity in HU (from now on defined as density) was extracted from the CT images in corresponding areas. Tissue time activity curves were normalized to the injected dose per gram of body weight (%ID/g), representing the glucose fractional extraction ( A–C), and multiplied by glycaemia to estimate glucose uptake . Glucose extraction and uptake rates were integrated over the first 10 min to primarily reflect organ pre-systemic glucose entry, and in the subsequent 10–60 min interval to reflect post-systemic glucose extraction. This was supported by the observation ( ) that in the first 10 min, left ventricular blood had received 4% of the total amount of tracer entering the general circulation in the entire imaging period, with 96% of 18 FDG reaching the general circulation in the following 10–60 min. Instead, 15–20% of the tracer reached the gut and visceral fat very rapidly, followed by the liver, within the pre-systemic 10 min time window. The systemic clearance of glucose was computed as a ratio of injected to integrated blood 18 FDG activity, and multiplied by glucose levels to reflect the endogenous glucose production (EGP) , as expressed per gram of body weight. 2.3. Biochemical Analyses Blood samples were collected at the end of the imaging procedures for a plasma or serum biochemical analysis. Triglycerides and liver enzymes (aspartate aminotransferase, AST, alanine aminotransferase, ALT) were determined by a bench clinical chemistry analyzer (Reflovet ® Plus, scil animal care company S.r.l., Treviglio, Italy) in 28–30 high quality samples. Triglyceride results falling below the measurable range were equaled to the lowest value of the accessible range (70 mg/dl). Inflammatory markers were measured by Luminex ® xMAP ® technology (Merck-Millipore Corp., Boston, MA, USA); they have been previously reported , and were only used here to confirm the occurrence of the weaning reaction and examine associations with the metabolism of visceral organs. 2.4. Liver Histology Tissue samples for histology were collected and processed in half of the cases ( n = 19). They were fixed in 10% formalin for 24 h, dehydrated, and included in paraffin using the Donatello Diapath automatic tissue processor (Martinengo, Bergamo, Italy), sliced (HistoCore Autocut, Leica BioSystems microtome) with 2 μm thickness, and stained with hematoxylin and eosin using the automated Dako CoverStainer (Santa Clara, CA, USA). Each section was documented at 20× and 40× magnification, by using the Olympus BX51 microscope and connected with an Olympus DP70 digital camera and AnalySIS 5.0 imaging system software (Olympus, Tokyo, Japan). Analyses were adapted from the method of Kleiner et al. to evaluate and score (yes/no, 1/0) vessel dilatation, fibrosis, portal and lobular inflammation (also graded by foci number at 20× magnification, 1 = one focus, 2 = two-four foci, 3 = >four foci), or ballooning degeneration, and micro/macrovesicular steatosis (also graded as the percentage of the affected cells). The diagnosis of non-alcoholic steatohepatitis was defined as the sum of grades ≥4 (sum score: range 4–7, including steatosis, lobular inflammation, and ballooning), and categorically (non-alcoholic steatohepatitis, NASH yes/no). 2.5. Gut Bacteria 16S rRNA Gene Sequencing Total DNA was isolated from the caecum and colon fecal pellets by the MasterPure Complete DNA&RNA Purification Kit (Epicentre, Illumina, San Diego, WI, USA), as previously described . DNA was measured using a Qubit ® 2.0 Fluorometer (Life Technology, Carlsbad, CA, USA) and normalized to 10 ng/μL. The V3-V4 region of the 16S rRNA gene was amplified by PCR using Illumina adapter overhang nucleotide sequences according to Illumina protocols. The multiplexing step was performed using a Nextera XT Index Kit (Illumina, San Diego, CA, USA). A Bioanalyzer DNA 1000 chip (Agilent Technologies, Santa Clara, CA, USA) was used to check the PCR product, and libraries were sequenced using a 2 × 300 bp paired-end run (MiSeq Reagent kit v3) on a MiSeq-Illumina platform (FISABIO sequencing service, Valencia, Spain) according to the manufacturer’s instructions (Illumina). For quality control, reagents employed for DNA extraction and PCR amplification were also sequenced. A quality assessment was performed by the prinseq-lite program (min_length: 50; trim_qual_right: 20; trim_qual_type: mean; trim_qual_window: 20 ). R1 and R2 from sequencing were joined using fastq-join from ea-tools suite ( http://code.google.com/p/ea-utils , sequences from 2016–2017). Data were obtained using an ad-hoc pipeline written in R Statistics environment and data processing was performed by a QIIME pipeline (version 1.9.0) . Chimeric sequences and sequences that could not be aligned were removed. The clustered sequences were utilized to construct OTUs tables (97% identity), then classified into phylum, family, and genus taxonomic levels based on the Greengenes database v13.8. Sequences not taxonomically classified or belonging to cyanobacteria and chloroplasts (representing ingested plant material) were removed. 2.6. Statistical Analysis The results were first analyzed by age groups, and then stratified by maternal habitus*age, under the rationale given in the results section. The metabolic profile, imaging results, and circulating markers were compared between groups with the analysis of variance or the Student’s t-test, as appropriate. Standard regression analyses were used to evaluate associations. These results are presented as mean ± sem, and p values ≤0.05 were regarded as statistically significant. Relative microbial abundances were obtained with the Calypso pipeline tool, version 8.84, by using total sum normalization. A redundancy analysis (RDA) was used to assess the complex associations of gut bacteria community composition in the offspring. A linear discriminant analysis (LDA) and effect size (LEfSe) analyses were used to identify unique biomarkers (LDA score > 3.0) in relative abundance of bacterial taxonomy . Spearman’s correlation analysis was used to explore univariate associations between bacteria taxa relative abundances and imaging parameters. A predictive inferred functional analysis was performed using PICRUSt with the Kyoto Encyclopedia of Genes and Genomes (KEGG) . Then, the LEfSe analysis and Wilcoxon-rank test were used to explore associations between KEGG pathways and groups. A false discovery rate (FDR) correction was applied in multiple comparisons/associations analyses.
B6129SF2/J (stock no: 101045, The Jackson Laboratory, Bar Harbor, Maine) female mice underwent ND (11% kcals from fat, Mucedola s.r.l., Milan, Italy, n = 5), or HFD (58% kcals from fat, Mucedola, n = 4) for 3 months before mating, and during gestation and lactation. After weaning, offspring (total n = 38) were fed with a standard diet. Animals were housed under standard conditions (22 °C, 12-hour light/dark cycles), with ad libitum access to food and water. Food intake, body weight, and glycemia have been already reported together with cognitive data . At weaning ( n = 19, NDoff n = 11, HFDoff n = 8) or at 6 months of age (adulthood, n = 19, NDoff n = 10, HFDoff n = 9), liver, visceral fat, and intestinal glucose uptake and density were measured by PET-CT imaging. Then, animals were euthanized by anesthetic overdose, and liver biopsies, blood, and fecal samples (colon, caecum) were collected.
Imaging of 18 FDG was performed under fasting conditions (PET-CT IRIS, Inviscan SAS, Strasbourg, France), under isofluorane anesthesia (IsoFlo ® , Abbott Laboratories, Chicago, IL, USA). Breath frequency and temperature were monitored during the study and a heated pad was used to prevent hypothermia. 18 FDG was administered by intraperitoneal (i.p.) injection to enable first-pass glucose uptake via mesenteric and portal vein vessels in visceral organs (pre-systemic), before entry into/delivery from the general circulation (post-systemic glucose uptake). A 60-minute whole-body dynamic PET scan was performed, and glycemia was measured in tail blood by a glucometer (OneTouch, Johnson&Johnson Medical SpA, Pomezia, Italy). PET data were corrected for dead time, random coincidences, and radioactive decay, and reconstructed by a 3D-Ordered Subset Expectation Maximization (3D-OSEM) algorithm. CT images where corrected for beam hardening and ring artifacts, reconstructed with cone-beam filtered backprojection (FBP), and calibrated in Hounsfield units (HU). All PET and CT images were exported to DICOM format after reconstruction, and fused within the AMIDE Medical Image Data Examiner 1.0.4 ( http://amide.sourceforge.net/ , accessed on 2020). Regions of interest were manually drawn on PET-CT images in correspondence with the right liver lobe, the large intestine, and lower abdominal visceral fat. Tissue radiodensity in HU (from now on defined as density) was extracted from the CT images in corresponding areas. Tissue time activity curves were normalized to the injected dose per gram of body weight (%ID/g), representing the glucose fractional extraction ( A–C), and multiplied by glycaemia to estimate glucose uptake . Glucose extraction and uptake rates were integrated over the first 10 min to primarily reflect organ pre-systemic glucose entry, and in the subsequent 10–60 min interval to reflect post-systemic glucose extraction. This was supported by the observation ( ) that in the first 10 min, left ventricular blood had received 4% of the total amount of tracer entering the general circulation in the entire imaging period, with 96% of 18 FDG reaching the general circulation in the following 10–60 min. Instead, 15–20% of the tracer reached the gut and visceral fat very rapidly, followed by the liver, within the pre-systemic 10 min time window. The systemic clearance of glucose was computed as a ratio of injected to integrated blood 18 FDG activity, and multiplied by glucose levels to reflect the endogenous glucose production (EGP) , as expressed per gram of body weight.
Blood samples were collected at the end of the imaging procedures for a plasma or serum biochemical analysis. Triglycerides and liver enzymes (aspartate aminotransferase, AST, alanine aminotransferase, ALT) were determined by a bench clinical chemistry analyzer (Reflovet ® Plus, scil animal care company S.r.l., Treviglio, Italy) in 28–30 high quality samples. Triglyceride results falling below the measurable range were equaled to the lowest value of the accessible range (70 mg/dl). Inflammatory markers were measured by Luminex ® xMAP ® technology (Merck-Millipore Corp., Boston, MA, USA); they have been previously reported , and were only used here to confirm the occurrence of the weaning reaction and examine associations with the metabolism of visceral organs.
Tissue samples for histology were collected and processed in half of the cases ( n = 19). They were fixed in 10% formalin for 24 h, dehydrated, and included in paraffin using the Donatello Diapath automatic tissue processor (Martinengo, Bergamo, Italy), sliced (HistoCore Autocut, Leica BioSystems microtome) with 2 μm thickness, and stained with hematoxylin and eosin using the automated Dako CoverStainer (Santa Clara, CA, USA). Each section was documented at 20× and 40× magnification, by using the Olympus BX51 microscope and connected with an Olympus DP70 digital camera and AnalySIS 5.0 imaging system software (Olympus, Tokyo, Japan). Analyses were adapted from the method of Kleiner et al. to evaluate and score (yes/no, 1/0) vessel dilatation, fibrosis, portal and lobular inflammation (also graded by foci number at 20× magnification, 1 = one focus, 2 = two-four foci, 3 = >four foci), or ballooning degeneration, and micro/macrovesicular steatosis (also graded as the percentage of the affected cells). The diagnosis of non-alcoholic steatohepatitis was defined as the sum of grades ≥4 (sum score: range 4–7, including steatosis, lobular inflammation, and ballooning), and categorically (non-alcoholic steatohepatitis, NASH yes/no).
Total DNA was isolated from the caecum and colon fecal pellets by the MasterPure Complete DNA&RNA Purification Kit (Epicentre, Illumina, San Diego, WI, USA), as previously described . DNA was measured using a Qubit ® 2.0 Fluorometer (Life Technology, Carlsbad, CA, USA) and normalized to 10 ng/μL. The V3-V4 region of the 16S rRNA gene was amplified by PCR using Illumina adapter overhang nucleotide sequences according to Illumina protocols. The multiplexing step was performed using a Nextera XT Index Kit (Illumina, San Diego, CA, USA). A Bioanalyzer DNA 1000 chip (Agilent Technologies, Santa Clara, CA, USA) was used to check the PCR product, and libraries were sequenced using a 2 × 300 bp paired-end run (MiSeq Reagent kit v3) on a MiSeq-Illumina platform (FISABIO sequencing service, Valencia, Spain) according to the manufacturer’s instructions (Illumina). For quality control, reagents employed for DNA extraction and PCR amplification were also sequenced. A quality assessment was performed by the prinseq-lite program (min_length: 50; trim_qual_right: 20; trim_qual_type: mean; trim_qual_window: 20 ). R1 and R2 from sequencing were joined using fastq-join from ea-tools suite ( http://code.google.com/p/ea-utils , sequences from 2016–2017). Data were obtained using an ad-hoc pipeline written in R Statistics environment and data processing was performed by a QIIME pipeline (version 1.9.0) . Chimeric sequences and sequences that could not be aligned were removed. The clustered sequences were utilized to construct OTUs tables (97% identity), then classified into phylum, family, and genus taxonomic levels based on the Greengenes database v13.8. Sequences not taxonomically classified or belonging to cyanobacteria and chloroplasts (representing ingested plant material) were removed.
The results were first analyzed by age groups, and then stratified by maternal habitus*age, under the rationale given in the results section. The metabolic profile, imaging results, and circulating markers were compared between groups with the analysis of variance or the Student’s t-test, as appropriate. Standard regression analyses were used to evaluate associations. These results are presented as mean ± sem, and p values ≤0.05 were regarded as statistically significant. Relative microbial abundances were obtained with the Calypso pipeline tool, version 8.84, by using total sum normalization. A redundancy analysis (RDA) was used to assess the complex associations of gut bacteria community composition in the offspring. A linear discriminant analysis (LDA) and effect size (LEfSe) analyses were used to identify unique biomarkers (LDA score > 3.0) in relative abundance of bacterial taxonomy . Spearman’s correlation analysis was used to explore univariate associations between bacteria taxa relative abundances and imaging parameters. A predictive inferred functional analysis was performed using PICRUSt with the Kyoto Encyclopedia of Genes and Genomes (KEGG) . Then, the LEfSe analysis and Wilcoxon-rank test were used to explore associations between KEGG pathways and groups. A false discovery rate (FDR) correction was applied in multiple comparisons/associations analyses.
Results are first presented by age periods (pooling HFDoff and NDoff at each age), to reflect the maturation of the gut–visceral fat–liver axis in a general population, with the expected prevalence of being overweight and the frequency of individuals born to overweight mothers seen in a general population. Results are then stratified by maternal group at each age to dissect the impact of early-life dietary exposures on the development of the visceral network and metabolic disorders. 3.1. Effect of Age on Systemic and Tissue Metabolism Maternal body weight was similar between the pooled age groups. Adult offspring were expectedly heavier than weaning mice (31 ± 2 vs. 17 ± 1 g, p < 0.0001), whereas circulating glucose and triglyceride (TG) levels did not differ significantly (n.s.). Weaning was characterized by a pronounced proinflammatory reaction at the systemic level, as suggested by high circulating concentrations of TNFα, IL6, and PAI-1, which were markedly reduced in adults ( A). Adults were characterized by a greater CT density in the liver and gut wall ( B), higher frequency of hepatic lobular inflammation, higher steatohepatitis scores in biopsies ( C), and greater systemic glucose clearance and EGP ( D). Comparing the kinetics of the glucose tracer among the three tissues ( ), visceral fat and gut were more glucose avid at weaning, whereas the liver became more glucose avid in adult life. This was due to a marked reduction in pre-systemic and post-systemic glucose extraction and uptake in the intestine and visceral fat, as opposed to a mild decline only involving pre-systemic glucose extraction and uptake in the liver ( E, ). As a result, the visceral fat/liver and gut/liver glucose uptake ratios (0–60 imaging time-range) were three times higher in weaning than adult mice, and conversely the relative amount of glucose going into the liver (liver-to-visceral fat and liver-to-gut partitioning) was more than doubled in adults than weaning mice ( F). 3.2. Effect of Maternal Diet on Systemic and Tissue Metabolism In comparing the weaning reaction in mice born to HFD vs. ND mothers ( H–L, ), the increase in TNFα, PAI-1, and intestinal glucose extraction and uptake was present in both groups (compared to adult age), but HFDoff tended towards higher levels of IL6, together with a two-to-six-fold elevation in glucose extraction and uptake in visceral fat and (less markedly) in the gut than NDoff. These differences were not seen in adults. Instead, adult HFDoff, were characterized by a 40% elevation in post-systemic hepatic glucose uptake compared to NDoff ( L). Consequently, in HFDoff the visceral fat/liver (glucose uptake) ratio was very high at weaning (+300%) compared to NDoff ( M). These patterns suggest that visceral fat and the gut received and sequestered more glucose in the weaning reaction of HFDoff than NDoff, reducing the within-network (i.e., pre-systemic) flux of glucose to the liver, preserving fasting EGP (n.s., not shown) and glucose clearance ( K). This pre-systemic glucose-sinking effect was lost in adult HFDoff, resulting in a high post-systemic liver glucose uptake and a lack in the physiological increase in systemic glucose clearance observed in NDoff. In addition, the CT density ( I) in the visceral fat and liver was high in weaning HFDoff, thereby failing to undergo the physiological increase from weaning-to-adulthood, as seen in NDoff. Biopsy data in the available samples demonstrated a significant trend of steatohepatitis grades, worsening across the four age*maternal-diet subgroups ( J), supporting a summative contribution of age and maternal HFD, in which adult HFDoff had the most severe liver condition with a 40% elevation in circulating triglyceride levels (1.15 ± 0.10 vs. 0.79 ± 0.00 mmol/L, p < 0.003 vs. adult NDoff). This group also lacked the physiological elevation in MCP1 levels observed in age-matched NDoff, thereby showing a four-fold cytokine deficiency ( H), as previously reported . 3.3. Tissue Metabolism Associates with Systemic Inflammation, Liver Steatosis, and Steatohepatitis Correlations between imaging markers and inflammatory cytokines or histological liver results were analyzed over the full range of pooled groups to interpret the CT and PET findings ( G). Results of these analyses documented that the visceral fat metabolism was the strongest indicator of systemic inflammation. Pre-systemic liver glucose extraction and uptake, and pre-/post-systemic intestinal and visceral fat glucose extraction and uptake were positively associated with TNFα and IL6, and (the former two) with PAI-1. Higher pre-systemic hepatic glucose extraction ( r = −0.35, p = 0.03) and pre-/post-systemic gut glucose extraction ( r = −0.33, p = 0.05, r = −0.38, p = 0.02) were also predictors of the lowering of hepatic CT density, which was a valid in vivo indicator of histology proven macrovescicular steatosis ( r = −0.50, p = 0.03). The visceral fat CT density was also strongly and negatively related to liver macrosteatosis ( r = −0.61, p = 0.009), whereas increasing liver CT density was proportional to biopsy-proven lobular inflammation ( r = 0.42, p = 0.07). Higher pre-systemic liver glucose extraction and uptake were positive predictors of portal inflammation ( r = 0.48, p = 0.036, r = 0.46, p = 0.05), whereas higher post-systemic visceral fat glucose uptake ( r = −0.57, p = 0.017) and colon glucose uptake ( r = −0.52, p = 0.03) were negative predictors of lobular inflammation. Portal inflammation ( r = −0.49, p = 0.03) was negatively related to liver steatosis. The intestinal wall CT density was inversely related to post-systemic gut glucose extraction and uptake ( r = −0.37, p = 0.025, r = −0.36, p = 0.031). The visceral fat CT density was positively related to post-systemic visceral fat glucose extraction ( r = 0.35, p = 0.039). The above findings suggest that liver glucose overexposure, especially via the portal route (pre-systemic) may subtend a risk of inflammation and steatohepatitis, which can be partly counterbalanced by the ability of visceral fat and gut to capture glucose. 3.4. Microbiota and Metabolic Pathway Analyses Among the three visceral tissues, microbiota composition and metabolic pathways were more frequently and more strongly related to imaging parameters describing the gut, followed by the liver, and then visceral fat. Caecum microbiota and metabolic pathways seemed more influential than their respective colon counterparts. In the current KEGG evaluation, we focused on functions related to substrate metabolism and inflammation, excluding pathways more strictly related to bacterial constituents (e.g., RNA, ribosome, DNA), cofactors-vitamins, or drugs and toxicants. 3.4.1. Effect of Age The RDA analyses showed highly significant differences in the microbiota profile between weaning and adult age in both the caecum and colon ( A). Explaining this difference, LDA analyses and LEfSE scores identified a panel of unique biomarkers ( B). The weaning microbiota was typified by high abundance in taxa belonging to the Bacteroidia (Bacteroidetes phylum) and the Clostridia classes (Firmicutes phylum), and the Biophila genus (Proteobacteria phylum) in the caecum, whereas only the S24-7 family/genus (also Muribaculaceae/Muribaculum ) were enriched in the weaning colon. In adults, taxonomic biomarkers belonged to the Actinobacteria (only caecum), Proteobacteria, Tenericutes, Verrucomicrobia phyla (both caecum and colon), Porphyromonadaceae , and Rikinellaceae families (Bacteroidetes phylum, only colon), several taxa belonging to Bacilli order (including Bacillus , Streptococcus , Unclassified Planococcaceae in both caecum and colon, or Lactobacillus genera in caecum), and Blautia and Coprobacillus genera (only caecum) (Firmicutes phylum). Considering the significant increase in NASH susceptibility observed in the adult age, we conducted targeted multivariant RDA and LDA-LEfSE analyses ( D–E) to identify unique biomarkers of NASH, and found that NASH-positive mice were discriminated by high abundance in Desulfovibrio genus in the colon, and Aerococcus genus and Aerococcaceae family (belonging to Bacilli class, Firmicutes phylum) and Unclassified RF39 genera, Tenericutes phylum, Mollicutes class, RF39 order, Unclassified RF39 families in the caecum. Most tissue parameters measured in vivo and ex vivo in the liver, gut, and visceral fat showed opposite correlations with weaning- vs. adulthood-dominating bacteria, suggesting that the changing microbiota composition is strictly aligned to, and may explain the changes occurring in the metabolism and structure of these tissues during maturation. In A,B and , we restricted the selection only to the tissue parameters that were significantly different between ages, and the bacteria (genus level) and metabolic pathways showing significant correlations with these parameters. Among them, the increase in Adelcreutzia , Lactobacillus , RF39 , Desulfovibrio , Bacillus , and Akkermansia genera, and the decrease in S24-7 , Oscillospira , Ruminococcus , and RC44 genera, especially in the caecum, from weaning to adulthood were associated with the corresponding increase in tissue density and decrease in glucose uptake observed between ages. In addition, the increase in Desulfovibrio , RF39 , and Coprobacillus genus abundance were predictive of greater liver inflammation, which is consistent with the results of the LDA/LEfSE analysis in E. Metabolic pathways involved in proinflammatory imprinting (lipopolysaccharide-LPS, sphyngolipids, MAPK signaling) and in organ proliferation and maturation (energy and lipid metabolism, glycan production, bile acids, and several amino-acids biosyntheses) were high at weaning ( C and B), whereas glycolysis-gluconeogenesis, the phosphotransferase system, amino-acid (tyrosine, alanine), and sulfur related pathways were hallmarks of adult age. Correlations between the expression level of these pathways and organ density or metabolism were coherent with the changes occurring in these imaging biomarkers from weaning to adulthood. The bile acids, LPS, sphingolipid, glycosaminoglycan degradation, and amino-acid metabolic pathways were positively correlated with liver inflammation or were unique identifiers of NASH, whereas the caecum glutamine-glutamate pathway seemed protective ( F and B). 3.4.2. Effect of Maternal Diet Microbiota richness decreased significantly in the caecum from 1 to 6 months of age only in adult HFDoff. We have previously reported a description of differently abundant taxa in HFDoff and NDoff, together with KEGG pathways in the colon . In the current study, we extended KEGG analyses to the caecum, and we presented differences in the pathways related to substrate metabolism and inflammation in both the caecum and colon ( ). In C–F and , we selected the tissue parameters that were significantly different between age-matched HFDoff and NDoff, and the bacteria (family and genus level) and metabolic pathways showing significant correlations with these parameters. These involved an enrichment in Clostridia and Bacteroidia classes, and the Tenericutes- RF39 genus, and a depletion in the Anaerotruncus genus in weaning HFDoff vs. NDoff, correlating with the observed metabolic features characterizing HFDoff at this age (increased visceral fat and intestinal glucose uptake). In adult HFDoff, showing high liver and intestinal glucose uptake (vs. NDoff), the associated biomarkers were Desulfovibrio and Allobaculum genera, Clostridiales order ( Unclassified Clostridiales genera, Peptococcaceae family), and Coriobacteriaceae family. In the KEGG analysis, only three amino-acid related pathways (lysine, phenylalanine, tryptophan) were simultaneously associated with the maternal diet and with the tissue differences observed at weaning. In adults, the upregulation of insulin signaling and downregulation of nitrogen metabolism and butyrate (short-chain fatty acid) pathways detected in HFDoff were predictors of histological liver inflammation, whereas pathways regulating arachidonic acid, glycan, sulfur, and amino-acid functions were associated to the imaging outcomes.
Maternal body weight was similar between the pooled age groups. Adult offspring were expectedly heavier than weaning mice (31 ± 2 vs. 17 ± 1 g, p < 0.0001), whereas circulating glucose and triglyceride (TG) levels did not differ significantly (n.s.). Weaning was characterized by a pronounced proinflammatory reaction at the systemic level, as suggested by high circulating concentrations of TNFα, IL6, and PAI-1, which were markedly reduced in adults ( A). Adults were characterized by a greater CT density in the liver and gut wall ( B), higher frequency of hepatic lobular inflammation, higher steatohepatitis scores in biopsies ( C), and greater systemic glucose clearance and EGP ( D). Comparing the kinetics of the glucose tracer among the three tissues ( ), visceral fat and gut were more glucose avid at weaning, whereas the liver became more glucose avid in adult life. This was due to a marked reduction in pre-systemic and post-systemic glucose extraction and uptake in the intestine and visceral fat, as opposed to a mild decline only involving pre-systemic glucose extraction and uptake in the liver ( E, ). As a result, the visceral fat/liver and gut/liver glucose uptake ratios (0–60 imaging time-range) were three times higher in weaning than adult mice, and conversely the relative amount of glucose going into the liver (liver-to-visceral fat and liver-to-gut partitioning) was more than doubled in adults than weaning mice ( F).
In comparing the weaning reaction in mice born to HFD vs. ND mothers ( H–L, ), the increase in TNFα, PAI-1, and intestinal glucose extraction and uptake was present in both groups (compared to adult age), but HFDoff tended towards higher levels of IL6, together with a two-to-six-fold elevation in glucose extraction and uptake in visceral fat and (less markedly) in the gut than NDoff. These differences were not seen in adults. Instead, adult HFDoff, were characterized by a 40% elevation in post-systemic hepatic glucose uptake compared to NDoff ( L). Consequently, in HFDoff the visceral fat/liver (glucose uptake) ratio was very high at weaning (+300%) compared to NDoff ( M). These patterns suggest that visceral fat and the gut received and sequestered more glucose in the weaning reaction of HFDoff than NDoff, reducing the within-network (i.e., pre-systemic) flux of glucose to the liver, preserving fasting EGP (n.s., not shown) and glucose clearance ( K). This pre-systemic glucose-sinking effect was lost in adult HFDoff, resulting in a high post-systemic liver glucose uptake and a lack in the physiological increase in systemic glucose clearance observed in NDoff. In addition, the CT density ( I) in the visceral fat and liver was high in weaning HFDoff, thereby failing to undergo the physiological increase from weaning-to-adulthood, as seen in NDoff. Biopsy data in the available samples demonstrated a significant trend of steatohepatitis grades, worsening across the four age*maternal-diet subgroups ( J), supporting a summative contribution of age and maternal HFD, in which adult HFDoff had the most severe liver condition with a 40% elevation in circulating triglyceride levels (1.15 ± 0.10 vs. 0.79 ± 0.00 mmol/L, p < 0.003 vs. adult NDoff). This group also lacked the physiological elevation in MCP1 levels observed in age-matched NDoff, thereby showing a four-fold cytokine deficiency ( H), as previously reported .
Correlations between imaging markers and inflammatory cytokines or histological liver results were analyzed over the full range of pooled groups to interpret the CT and PET findings ( G). Results of these analyses documented that the visceral fat metabolism was the strongest indicator of systemic inflammation. Pre-systemic liver glucose extraction and uptake, and pre-/post-systemic intestinal and visceral fat glucose extraction and uptake were positively associated with TNFα and IL6, and (the former two) with PAI-1. Higher pre-systemic hepatic glucose extraction ( r = −0.35, p = 0.03) and pre-/post-systemic gut glucose extraction ( r = −0.33, p = 0.05, r = −0.38, p = 0.02) were also predictors of the lowering of hepatic CT density, which was a valid in vivo indicator of histology proven macrovescicular steatosis ( r = −0.50, p = 0.03). The visceral fat CT density was also strongly and negatively related to liver macrosteatosis ( r = −0.61, p = 0.009), whereas increasing liver CT density was proportional to biopsy-proven lobular inflammation ( r = 0.42, p = 0.07). Higher pre-systemic liver glucose extraction and uptake were positive predictors of portal inflammation ( r = 0.48, p = 0.036, r = 0.46, p = 0.05), whereas higher post-systemic visceral fat glucose uptake ( r = −0.57, p = 0.017) and colon glucose uptake ( r = −0.52, p = 0.03) were negative predictors of lobular inflammation. Portal inflammation ( r = −0.49, p = 0.03) was negatively related to liver steatosis. The intestinal wall CT density was inversely related to post-systemic gut glucose extraction and uptake ( r = −0.37, p = 0.025, r = −0.36, p = 0.031). The visceral fat CT density was positively related to post-systemic visceral fat glucose extraction ( r = 0.35, p = 0.039). The above findings suggest that liver glucose overexposure, especially via the portal route (pre-systemic) may subtend a risk of inflammation and steatohepatitis, which can be partly counterbalanced by the ability of visceral fat and gut to capture glucose.
Among the three visceral tissues, microbiota composition and metabolic pathways were more frequently and more strongly related to imaging parameters describing the gut, followed by the liver, and then visceral fat. Caecum microbiota and metabolic pathways seemed more influential than their respective colon counterparts. In the current KEGG evaluation, we focused on functions related to substrate metabolism and inflammation, excluding pathways more strictly related to bacterial constituents (e.g., RNA, ribosome, DNA), cofactors-vitamins, or drugs and toxicants. 3.4.1. Effect of Age The RDA analyses showed highly significant differences in the microbiota profile between weaning and adult age in both the caecum and colon ( A). Explaining this difference, LDA analyses and LEfSE scores identified a panel of unique biomarkers ( B). The weaning microbiota was typified by high abundance in taxa belonging to the Bacteroidia (Bacteroidetes phylum) and the Clostridia classes (Firmicutes phylum), and the Biophila genus (Proteobacteria phylum) in the caecum, whereas only the S24-7 family/genus (also Muribaculaceae/Muribaculum ) were enriched in the weaning colon. In adults, taxonomic biomarkers belonged to the Actinobacteria (only caecum), Proteobacteria, Tenericutes, Verrucomicrobia phyla (both caecum and colon), Porphyromonadaceae , and Rikinellaceae families (Bacteroidetes phylum, only colon), several taxa belonging to Bacilli order (including Bacillus , Streptococcus , Unclassified Planococcaceae in both caecum and colon, or Lactobacillus genera in caecum), and Blautia and Coprobacillus genera (only caecum) (Firmicutes phylum). Considering the significant increase in NASH susceptibility observed in the adult age, we conducted targeted multivariant RDA and LDA-LEfSE analyses ( D–E) to identify unique biomarkers of NASH, and found that NASH-positive mice were discriminated by high abundance in Desulfovibrio genus in the colon, and Aerococcus genus and Aerococcaceae family (belonging to Bacilli class, Firmicutes phylum) and Unclassified RF39 genera, Tenericutes phylum, Mollicutes class, RF39 order, Unclassified RF39 families in the caecum. Most tissue parameters measured in vivo and ex vivo in the liver, gut, and visceral fat showed opposite correlations with weaning- vs. adulthood-dominating bacteria, suggesting that the changing microbiota composition is strictly aligned to, and may explain the changes occurring in the metabolism and structure of these tissues during maturation. In A,B and , we restricted the selection only to the tissue parameters that were significantly different between ages, and the bacteria (genus level) and metabolic pathways showing significant correlations with these parameters. Among them, the increase in Adelcreutzia , Lactobacillus , RF39 , Desulfovibrio , Bacillus , and Akkermansia genera, and the decrease in S24-7 , Oscillospira , Ruminococcus , and RC44 genera, especially in the caecum, from weaning to adulthood were associated with the corresponding increase in tissue density and decrease in glucose uptake observed between ages. In addition, the increase in Desulfovibrio , RF39 , and Coprobacillus genus abundance were predictive of greater liver inflammation, which is consistent with the results of the LDA/LEfSE analysis in E. Metabolic pathways involved in proinflammatory imprinting (lipopolysaccharide-LPS, sphyngolipids, MAPK signaling) and in organ proliferation and maturation (energy and lipid metabolism, glycan production, bile acids, and several amino-acids biosyntheses) were high at weaning ( C and B), whereas glycolysis-gluconeogenesis, the phosphotransferase system, amino-acid (tyrosine, alanine), and sulfur related pathways were hallmarks of adult age. Correlations between the expression level of these pathways and organ density or metabolism were coherent with the changes occurring in these imaging biomarkers from weaning to adulthood. The bile acids, LPS, sphingolipid, glycosaminoglycan degradation, and amino-acid metabolic pathways were positively correlated with liver inflammation or were unique identifiers of NASH, whereas the caecum glutamine-glutamate pathway seemed protective ( F and B). 3.4.2. Effect of Maternal Diet Microbiota richness decreased significantly in the caecum from 1 to 6 months of age only in adult HFDoff. We have previously reported a description of differently abundant taxa in HFDoff and NDoff, together with KEGG pathways in the colon . In the current study, we extended KEGG analyses to the caecum, and we presented differences in the pathways related to substrate metabolism and inflammation in both the caecum and colon ( ). In C–F and , we selected the tissue parameters that were significantly different between age-matched HFDoff and NDoff, and the bacteria (family and genus level) and metabolic pathways showing significant correlations with these parameters. These involved an enrichment in Clostridia and Bacteroidia classes, and the Tenericutes- RF39 genus, and a depletion in the Anaerotruncus genus in weaning HFDoff vs. NDoff, correlating with the observed metabolic features characterizing HFDoff at this age (increased visceral fat and intestinal glucose uptake). In adult HFDoff, showing high liver and intestinal glucose uptake (vs. NDoff), the associated biomarkers were Desulfovibrio and Allobaculum genera, Clostridiales order ( Unclassified Clostridiales genera, Peptococcaceae family), and Coriobacteriaceae family. In the KEGG analysis, only three amino-acid related pathways (lysine, phenylalanine, tryptophan) were simultaneously associated with the maternal diet and with the tissue differences observed at weaning. In adults, the upregulation of insulin signaling and downregulation of nitrogen metabolism and butyrate (short-chain fatty acid) pathways detected in HFDoff were predictors of histological liver inflammation, whereas pathways regulating arachidonic acid, glycan, sulfur, and amino-acid functions were associated to the imaging outcomes.
The RDA analyses showed highly significant differences in the microbiota profile between weaning and adult age in both the caecum and colon ( A). Explaining this difference, LDA analyses and LEfSE scores identified a panel of unique biomarkers ( B). The weaning microbiota was typified by high abundance in taxa belonging to the Bacteroidia (Bacteroidetes phylum) and the Clostridia classes (Firmicutes phylum), and the Biophila genus (Proteobacteria phylum) in the caecum, whereas only the S24-7 family/genus (also Muribaculaceae/Muribaculum ) were enriched in the weaning colon. In adults, taxonomic biomarkers belonged to the Actinobacteria (only caecum), Proteobacteria, Tenericutes, Verrucomicrobia phyla (both caecum and colon), Porphyromonadaceae , and Rikinellaceae families (Bacteroidetes phylum, only colon), several taxa belonging to Bacilli order (including Bacillus , Streptococcus , Unclassified Planococcaceae in both caecum and colon, or Lactobacillus genera in caecum), and Blautia and Coprobacillus genera (only caecum) (Firmicutes phylum). Considering the significant increase in NASH susceptibility observed in the adult age, we conducted targeted multivariant RDA and LDA-LEfSE analyses ( D–E) to identify unique biomarkers of NASH, and found that NASH-positive mice were discriminated by high abundance in Desulfovibrio genus in the colon, and Aerococcus genus and Aerococcaceae family (belonging to Bacilli class, Firmicutes phylum) and Unclassified RF39 genera, Tenericutes phylum, Mollicutes class, RF39 order, Unclassified RF39 families in the caecum. Most tissue parameters measured in vivo and ex vivo in the liver, gut, and visceral fat showed opposite correlations with weaning- vs. adulthood-dominating bacteria, suggesting that the changing microbiota composition is strictly aligned to, and may explain the changes occurring in the metabolism and structure of these tissues during maturation. In A,B and , we restricted the selection only to the tissue parameters that were significantly different between ages, and the bacteria (genus level) and metabolic pathways showing significant correlations with these parameters. Among them, the increase in Adelcreutzia , Lactobacillus , RF39 , Desulfovibrio , Bacillus , and Akkermansia genera, and the decrease in S24-7 , Oscillospira , Ruminococcus , and RC44 genera, especially in the caecum, from weaning to adulthood were associated with the corresponding increase in tissue density and decrease in glucose uptake observed between ages. In addition, the increase in Desulfovibrio , RF39 , and Coprobacillus genus abundance were predictive of greater liver inflammation, which is consistent with the results of the LDA/LEfSE analysis in E. Metabolic pathways involved in proinflammatory imprinting (lipopolysaccharide-LPS, sphyngolipids, MAPK signaling) and in organ proliferation and maturation (energy and lipid metabolism, glycan production, bile acids, and several amino-acids biosyntheses) were high at weaning ( C and B), whereas glycolysis-gluconeogenesis, the phosphotransferase system, amino-acid (tyrosine, alanine), and sulfur related pathways were hallmarks of adult age. Correlations between the expression level of these pathways and organ density or metabolism were coherent with the changes occurring in these imaging biomarkers from weaning to adulthood. The bile acids, LPS, sphingolipid, glycosaminoglycan degradation, and amino-acid metabolic pathways were positively correlated with liver inflammation or were unique identifiers of NASH, whereas the caecum glutamine-glutamate pathway seemed protective ( F and B).
Microbiota richness decreased significantly in the caecum from 1 to 6 months of age only in adult HFDoff. We have previously reported a description of differently abundant taxa in HFDoff and NDoff, together with KEGG pathways in the colon . In the current study, we extended KEGG analyses to the caecum, and we presented differences in the pathways related to substrate metabolism and inflammation in both the caecum and colon ( ). In C–F and , we selected the tissue parameters that were significantly different between age-matched HFDoff and NDoff, and the bacteria (family and genus level) and metabolic pathways showing significant correlations with these parameters. These involved an enrichment in Clostridia and Bacteroidia classes, and the Tenericutes- RF39 genus, and a depletion in the Anaerotruncus genus in weaning HFDoff vs. NDoff, correlating with the observed metabolic features characterizing HFDoff at this age (increased visceral fat and intestinal glucose uptake). In adult HFDoff, showing high liver and intestinal glucose uptake (vs. NDoff), the associated biomarkers were Desulfovibrio and Allobaculum genera, Clostridiales order ( Unclassified Clostridiales genera, Peptococcaceae family), and Coriobacteriaceae family. In the KEGG analysis, only three amino-acid related pathways (lysine, phenylalanine, tryptophan) were simultaneously associated with the maternal diet and with the tissue differences observed at weaning. In adults, the upregulation of insulin signaling and downregulation of nitrogen metabolism and butyrate (short-chain fatty acid) pathways detected in HFDoff were predictors of histological liver inflammation, whereas pathways regulating arachidonic acid, glycan, sulfur, and amino-acid functions were associated to the imaging outcomes.
The current study documents that maturation from weaning to adulthood leads to profound changes in the metabolism and structure of visceral tissues and in microbiota composition and function, with an increased susceptibility to liver inflammation. Our results identify clear in vivo imaging indicators of organ maturation, namely tissue glucose uptake and organ density, showing strict associations with changes in microbiota composition and metabolic pathway expression, modulating plasma levels of inflammatory markers. We provide evidence that the weaning reaction is amplified, and organ metabolism is affected in mice born to HFD compared to ND mothers, in line with microbiota characteristics, which may contribute to the metabolic abnormalities observed especially in adult age, i.e., being overweight, hyperglycemia, hypertriglyceridemia, and liver inflammation. The suggested sequence of events is exemplified in . Our data were first examined across age periods to reflect the maturation of the gut–visceral fat–liver axis in a general population, and then addressed in relation to maternal diet. As a general finding, caecum compared to colon microbiota showed stronger connections with our imaging and histology measures, underlining its role in substrate exchange and immunity development. 4.1. Effects of Age: Maturation of the Visceral Network Maturation of the intestinal wall is concentrated in early life and implies crypt fission, goblet cell proliferation, mucus thickening, and impermeability, while adipose tissue undergoes differentiation and hyperplasia, and the liver undergoes the transition from being a predominantly hematopoietic to the most important metabolic organ, with a growing ability to produce glucose, replacing maternal sources [ , , , , , , ]. In coherence with these changes, we found a significant increase in hepatic density from weaning to adult age, and the liver became the major (in adulthood), from being the minor (at weaning) glucose consumer per unit of tissue volume compared to gut and visceral fat, also increasing its release of glucose (EGP) in proportion to body weight and systemic glucose clearance. Notably, bacteria (especially S24-7 ) and metabolites (i.e., LPS, bile acids, amino-acids, energy metabolism) that were dominant at weaning in our mice have been implicated in liver regeneration processes after liver resection, co-regulating the proliferative and metabolic roles of the liver . Similar to the liver, the immature gut was characterized by low wall density and high glucose requirements, signaling an active proliferation. Microbiota-induced metabolic pathways (energy metabolism, oxidative phosphorylation, glycan, glycosamine, sphingolipid, glycosphingolipid metabolism, adipocytokine signalling, peroxisomes, branched-chain, and other amino acids) related to energy-fueling, protein-building, and mucin-promoting functions prevailed in this period, and were strongly predictive of the intestinal imaging hallmarks, establishing a coherent mechanistic scenario. Finally, visceral fat showed a high glucose uptake at weaning, consistent with the hyperplastic phase, and a (negative) CT density (Hounsfield unit) range indicating a prevailing lipid occupancy. The visceral fat glucose uptake was associated with the abundance in Oscillospira ( Clostridiales order) and RC44 genera, and with metabolic pathways involving peroxisome, oxidative phosphorylation, and energy metabolism, beyond ammino-acid biosynthesis. One important phenomenon occurring during our early-life observation period is the weaning reaction, i.e., a benign proinflammatory reaction, consisting of microbiota induced elevations in TNFα levels in the circulation, and an immature impermeability of the developing intestinal barrier, playing a permissive role on microbiota interactions with visceral organs [ , , ]. The reaction lasts two weeks in mice and impacts a number of disease conditions during life. Consistent with this, we observed a several-fold elevation in TNFα, IL6, and PAI-1 levels in our weaning mice, compared to adults. Their microbiota was typified by a high abundance in LPS producing Bacteroidia ( Rikenellaceae in caecum, S24-7 in caecum and colon) and Clostridiales genera, and an upregulation in LPS biosynthetic and MAPK signaling pathways in the KEGG analysis, all of which have been associated with the weaning reaction, including hepatic oxidative stress . Older studies have shown that the infusion of LPS results in a pronounced increment in tissue glucose uptake, due to the build-up of glucose-avid immune cells, with the greatest response occurring in the liver . On the same principle, 18 FDG-PET imaging has been used to capture inflammation in adipose tissue , and permeability in the gut wall . Notably, visceral fat plays an important role in generating cytokines, and visceral fat glucose uptake showed the strongest correlation with the levels of circulating TNFα and IL6 in this study, followed by gut and liver glucose uptake and CT densities, whereas PAI-1 was preferentially related to hepatic pre-systemic glucose uptake. In fact, pre-systemic liver glucose uptake was high in our weaning mice, correlating with the degree of biopsy-proven portal inflammation, which is typically seen in pediatric liver dysmetabolism . Interestingly, pre-systemic liver glucose uptake was also the only metabolic process associating with the abundance of all bacteria that were dominant at weaning, and showing a significant association with the LPS biosynthetic pathway. In synthesis, weaning mice displayed a very high tissue glucose uptake resulting from rapid proliferation and a marked proinflammatory reaction; the structural immaturity of the gut and liver tissues translated in low CT density. Both imaging hallmarks are closely associated with the microbiota and its metabolic pathways. The pre-systemic liver glucose uptake was reflective of biopsy-proven portal inflammation and associated to the microbiota induced LPS pathway. The three folds higher visceral fat/liver and gut/liver glucose uptake ratios seen in this life-phase are consistent with the role of the gut and visceral fat in generating and buffering the proinflammatory reaction, and may protect the liver from more severe substrate overload and damage, as suggested by the inverse correlations observed between visceral fat or gut glucose uptake and biopsy-measured liver inflammation. Maturation towards adulthood involved an increase in tissue density and a major decrease in pre-/post-systemic glucose uptake especially in visceral fat and the gut, with higher EGP and whole-body glucose clearance. Bacterial composition was significantly shifted away from the dominance of Bacteroidia and Clostridia classes during weaning, in favor of other phyla or classes, namely Actinobacteria, Proteobacteria ( Desulfovibrio genus), Tenericutes ( RF39 genus), Verrucomicrobia ( Akkermansia ), Bacteroidetes ( Porphyromonadaceae , Rikinellaceae genera), and Firmicutes ( Bacilli , Blautia genera). In line with the transient nature of the weaning reaction, systemic inflammatory markers were expectedly declined, and the decrease was correlated with the changes in organ metabolism and structure. The relative abundance of Verrucomicrobia phylum characterized by the presence of Akkermansia genus was associated with the lowering of pre-systemic tissue glucose uptake, most strongly in visceral fat, which was dependent on the reduction of systemic inflammation. The Akkermansia genus has been proven to reduce systemic inflammation, by suppressing LPS, TNFα, and other cytokine levels and increasing fatty acid oxidation in adipose tissue, as reviewed in . This previous knowledge provides mechanistic support to the observations of the current study, also suggesting that Akkermansia should be investigated as a potential factor terminating the weaning reaction. As opposed to a decline in systemic inflammatory markers, we observed an increased susceptibility of the liver to develop steatohepatitis in adult mice. The inverse association between post-systemic glucose uptake in visceral fat and gut (reduced in adults) and lobular liver inflammation (increased in adults) suggests that a diminished sequestration of glucose in these extra-hepatic tissues may detract protection against oxidative stress and liver inflammation. In fact, the sum of glucose release and uptake by the liver indicates that the inflow of glucose and gluconeogenic substrates into the liver was increased in adult mice. It is of note that the reduction in visceral fat and gut glucose uptake in adults was proportional to the abundance of Desulfovibrio and RF39 genera (i.e., the two main predictors of NASH) and to the upregulation in the sulfur relay system and glycolysis-gluconeogenesis pathways in the KEGG analysis. In fact, the emerging roles of hydrogen sulfide point towards a negative effect on glucose uptake in adipocytes and a stimulatory effect on hepatic pyruvate carboxylase sulfhydration, promoting gluconeogenesis in liver cells . Hence, the upregulation of the sulfur relay system and glycolysis-gluconeogenesis pathways might underlie the reduction in visceral fat and gut glucose uptake, as well as the increase in EGP observed in our adult mice. In synthesis, the above findings suggest that the increase in Akkermansia genus abundance may be involved in the downregulation of systemic inflammation and the associated pre-systemic glucose uptake, especially in visceral fat, whereas the upregulation of Desulfovibrio and RF39 bacteria may activate metabolic pathways linked with liver inflammation susceptibility. Among these, our data point towards the (in)ability of visceral fat and gut to extract glucose from the systemic circulation (post-systemic glucose uptake), leading to liver glucose overexposure and an increase in EGP. The immune modulatory impact of the gut microbiota can significantly affect intrahepatic and intestinal inflammation, and hepatocarcinogenesis. Our findings are consistent with the evidence that patients with cirrhosis and steatohepatitis who developed hepatocellular carcinoma (HCC) lacked similar protective bacteria profiles and had enhanced liver and intestinal inflammation as compared to patients with cirrhosis, but without HCC . Cirrhosis patients had a lower abundance of Akkermansia genus abundance than controls in association with an inflammatory intestinal environment and higher fecal calprotectin levels, and Bifidobacterium was also depleted in HCC patients . Thus, the combined deficiency of these beneficial bacteria might enhance intestinal and liver inflammation, influencing liver disease progression as well as the initiation and/or progression of hepatocarcinogenesis, and the administration of probiotics was shown to reduce the incidence and growth of HCC lesions in mice, by modulating gut microbiota composition and restoring intestinal permeability [ , , ]. 4.2. Effects of Maternal Diet*Age on the Visceral Network We asked to which extent does an early exposure to ND or HFD contribute to, or deviate from, the results obtained in a more general, i.e., admixed population. The microbiota of weaning HFDoff showed an amplified abundance of Bacterioidia and Clostridia class bacteria, namely Rikenellaceae , Parabacteriodes , and Peptostreptococcaceae genera, and in RF39 . In line with this, visceral fat glucose uptake was several folds higher in HFDoff than NDoff at weaning; the gut was affected to a lower extent, and a trend towards higher IL6 levels in HFDoff was seen. The liver showed a greater CT density in this group. Interestingly, both RF39 abundance and lobular inflammation were positive predictors of hepatic CT density, and an LDA analysis indicated that RF39 was a significant discriminator of biopsy-proven NASH. We screened for functional pathways that could mediate the interactions between microbiota and tissue outcomes, and we identified three (upregulated) metabolic pathways that were associated (positively) with liver CT density and/or visceral fat glucose uptake, namely tryptophan metabolism, lysine degradation, and phenylalanine metabolism. Lysine degradation was the most upregulated (+15–30%) pathway in both the colon and caecum. Lysine can be processed by Clostridia to generate short-chain fatty acids (SCFAs) , and we correspondingly observed an upregulation of fatty acid and butanoate (butyric acid) metabolic pathways in the colon and caecum. SCFAs also affect tryptophan metabolism, which has major effects on intestinal barrier integrity and transit, and fatty liver disease and metabolism . In states of obesity and inflammatory bowel diseases, the metabolism of tryptophan is enhanced in the adipose tissue, correlating with IL6 levels and leading to increased intestinal permeability and LPS translocation in the systemic circulation, resulting in peripheral inflammation, altered glucose metabolism, and impaired hepatic protection . These aminoacidic actions recapitulate very well the findings of the current study, namely the potentiation of cytokine IL6 levels and of visceral fat and gut glucose uptake observed in weaning HFDoff, as well as the tendency towards liver inflammation and the occurrence of hyperglycemia in the fed state, as previously shown in this mice group . We have previously reported that maternal HFD led to adulthood obesity, hyperglycemia, hypoinsulinemia, hyperresistinemia, and impaired MCP1 levels in these mice . In this study, we demonstrated that maternal HFD contributes to increasing the susceptibility towards liver inflammation in the offspring from early to adult life, and provokes a 40% elevation in plasma triglyceride levels in adult HFDoff, limiting the accumulation of tissue lipids. The main difference between adult HFDoff and NDoff at the tissue metabolic level was the upregulation in post-systemic glucose uptake in the liver and (less) in the gut in HFDoff. It is of note that the glucose uptake in these two tissues was related to fasting hyperglycemia, suggesting that visceral fat was unable to act as the substrate sink. As previously shown , the microbiota was significantly different in relation to the maternal diet. Among bacteria that were different in adult NDoff and HFDoff, the abundance of Coriobacteriaceae , Desulfovibrio , and Allobaculum were positively associated with liver glucose uptake, and the deficiency of Clostridiales and Peptococcaceae families in HFDoff predicted higher liver and gut glucose uptake. The abundance of Desulfovibrio , Allobaculum , and Actinobacteria have been previously shown to reflect the progression of liver disease in Western-diet-fed or HFD-fed diabetic mice , and the abundance of Desulfovibrio genus emerged as a significant LDA-discriminator of biopsy-proven NASH in this study. Consistent with our findings, steatohepatitis in HFD and in overweight (but not in lean) subjects, and in patients with type 2 diabetes was associated with an increase in gut Desulfovibrio and sulfate-lowering bacteria . We found that arachidonic acid metabolism was augmented by 3.5 folds in adult HFDoff and positively related to gut glucose uptake, together with sulfur and phenylalanine metabolism, and both sulfur reduction and the arachidonic acid cascade have been implicated in gut inflammation, steatohepatitis, and an unhealthy gut–adipose–liver relationship [ , , ]. We also observed that metabolic pathways of methionine-cysteine (i.e., the only sulfur-containing amino acids) were associated with liver and gut post-systemic glucose uptake. These amino acids have been related to an unhealthy gut–liver–adipose metabolism and to obesity, and their dietary restriction has many benefits, as referenced in . In addition, we noted that the phosphatydilinositol and insulin signaling pathways, and glycan biosynthesis and metabolism were related to liver outcomes in our adult HFDoff mice, and it is of interest that the insulin-phosphatydilinositol kinase pathway has been recently involved in the progression of non-alcoholic liver disease . Our results provide a cause–effect demonstration linking weaning, adult age, and maternal diet with respective metabolic, inflammatory, and microbiota outcomes; however, the associations between imaging and microbiota biomarkers support, but do not prove causality. However, a causal relationship was documented in studies addressing microbiota transplantation from 2-week-old infants born to obese (compared to lean) mothers into adult germ-free mice. This led to alterations in intestinal permeability, increased hepatic endoplasmic reticulum stress, and signs of periportal inflammation , together with impaired macrophage phagocytic function and cytokine production, the latter promoting weight gain, beyond hepatic macrophage accumulation. These findings are consistent with the overweight, associating with hepatic inflammation susceptibility, and the four-fold MCP1 defect observed in our adult HFDoff.
Maturation of the intestinal wall is concentrated in early life and implies crypt fission, goblet cell proliferation, mucus thickening, and impermeability, while adipose tissue undergoes differentiation and hyperplasia, and the liver undergoes the transition from being a predominantly hematopoietic to the most important metabolic organ, with a growing ability to produce glucose, replacing maternal sources [ , , , , , , ]. In coherence with these changes, we found a significant increase in hepatic density from weaning to adult age, and the liver became the major (in adulthood), from being the minor (at weaning) glucose consumer per unit of tissue volume compared to gut and visceral fat, also increasing its release of glucose (EGP) in proportion to body weight and systemic glucose clearance. Notably, bacteria (especially S24-7 ) and metabolites (i.e., LPS, bile acids, amino-acids, energy metabolism) that were dominant at weaning in our mice have been implicated in liver regeneration processes after liver resection, co-regulating the proliferative and metabolic roles of the liver . Similar to the liver, the immature gut was characterized by low wall density and high glucose requirements, signaling an active proliferation. Microbiota-induced metabolic pathways (energy metabolism, oxidative phosphorylation, glycan, glycosamine, sphingolipid, glycosphingolipid metabolism, adipocytokine signalling, peroxisomes, branched-chain, and other amino acids) related to energy-fueling, protein-building, and mucin-promoting functions prevailed in this period, and were strongly predictive of the intestinal imaging hallmarks, establishing a coherent mechanistic scenario. Finally, visceral fat showed a high glucose uptake at weaning, consistent with the hyperplastic phase, and a (negative) CT density (Hounsfield unit) range indicating a prevailing lipid occupancy. The visceral fat glucose uptake was associated with the abundance in Oscillospira ( Clostridiales order) and RC44 genera, and with metabolic pathways involving peroxisome, oxidative phosphorylation, and energy metabolism, beyond ammino-acid biosynthesis. One important phenomenon occurring during our early-life observation period is the weaning reaction, i.e., a benign proinflammatory reaction, consisting of microbiota induced elevations in TNFα levels in the circulation, and an immature impermeability of the developing intestinal barrier, playing a permissive role on microbiota interactions with visceral organs [ , , ]. The reaction lasts two weeks in mice and impacts a number of disease conditions during life. Consistent with this, we observed a several-fold elevation in TNFα, IL6, and PAI-1 levels in our weaning mice, compared to adults. Their microbiota was typified by a high abundance in LPS producing Bacteroidia ( Rikenellaceae in caecum, S24-7 in caecum and colon) and Clostridiales genera, and an upregulation in LPS biosynthetic and MAPK signaling pathways in the KEGG analysis, all of which have been associated with the weaning reaction, including hepatic oxidative stress . Older studies have shown that the infusion of LPS results in a pronounced increment in tissue glucose uptake, due to the build-up of glucose-avid immune cells, with the greatest response occurring in the liver . On the same principle, 18 FDG-PET imaging has been used to capture inflammation in adipose tissue , and permeability in the gut wall . Notably, visceral fat plays an important role in generating cytokines, and visceral fat glucose uptake showed the strongest correlation with the levels of circulating TNFα and IL6 in this study, followed by gut and liver glucose uptake and CT densities, whereas PAI-1 was preferentially related to hepatic pre-systemic glucose uptake. In fact, pre-systemic liver glucose uptake was high in our weaning mice, correlating with the degree of biopsy-proven portal inflammation, which is typically seen in pediatric liver dysmetabolism . Interestingly, pre-systemic liver glucose uptake was also the only metabolic process associating with the abundance of all bacteria that were dominant at weaning, and showing a significant association with the LPS biosynthetic pathway. In synthesis, weaning mice displayed a very high tissue glucose uptake resulting from rapid proliferation and a marked proinflammatory reaction; the structural immaturity of the gut and liver tissues translated in low CT density. Both imaging hallmarks are closely associated with the microbiota and its metabolic pathways. The pre-systemic liver glucose uptake was reflective of biopsy-proven portal inflammation and associated to the microbiota induced LPS pathway. The three folds higher visceral fat/liver and gut/liver glucose uptake ratios seen in this life-phase are consistent with the role of the gut and visceral fat in generating and buffering the proinflammatory reaction, and may protect the liver from more severe substrate overload and damage, as suggested by the inverse correlations observed between visceral fat or gut glucose uptake and biopsy-measured liver inflammation. Maturation towards adulthood involved an increase in tissue density and a major decrease in pre-/post-systemic glucose uptake especially in visceral fat and the gut, with higher EGP and whole-body glucose clearance. Bacterial composition was significantly shifted away from the dominance of Bacteroidia and Clostridia classes during weaning, in favor of other phyla or classes, namely Actinobacteria, Proteobacteria ( Desulfovibrio genus), Tenericutes ( RF39 genus), Verrucomicrobia ( Akkermansia ), Bacteroidetes ( Porphyromonadaceae , Rikinellaceae genera), and Firmicutes ( Bacilli , Blautia genera). In line with the transient nature of the weaning reaction, systemic inflammatory markers were expectedly declined, and the decrease was correlated with the changes in organ metabolism and structure. The relative abundance of Verrucomicrobia phylum characterized by the presence of Akkermansia genus was associated with the lowering of pre-systemic tissue glucose uptake, most strongly in visceral fat, which was dependent on the reduction of systemic inflammation. The Akkermansia genus has been proven to reduce systemic inflammation, by suppressing LPS, TNFα, and other cytokine levels and increasing fatty acid oxidation in adipose tissue, as reviewed in . This previous knowledge provides mechanistic support to the observations of the current study, also suggesting that Akkermansia should be investigated as a potential factor terminating the weaning reaction. As opposed to a decline in systemic inflammatory markers, we observed an increased susceptibility of the liver to develop steatohepatitis in adult mice. The inverse association between post-systemic glucose uptake in visceral fat and gut (reduced in adults) and lobular liver inflammation (increased in adults) suggests that a diminished sequestration of glucose in these extra-hepatic tissues may detract protection against oxidative stress and liver inflammation. In fact, the sum of glucose release and uptake by the liver indicates that the inflow of glucose and gluconeogenic substrates into the liver was increased in adult mice. It is of note that the reduction in visceral fat and gut glucose uptake in adults was proportional to the abundance of Desulfovibrio and RF39 genera (i.e., the two main predictors of NASH) and to the upregulation in the sulfur relay system and glycolysis-gluconeogenesis pathways in the KEGG analysis. In fact, the emerging roles of hydrogen sulfide point towards a negative effect on glucose uptake in adipocytes and a stimulatory effect on hepatic pyruvate carboxylase sulfhydration, promoting gluconeogenesis in liver cells . Hence, the upregulation of the sulfur relay system and glycolysis-gluconeogenesis pathways might underlie the reduction in visceral fat and gut glucose uptake, as well as the increase in EGP observed in our adult mice. In synthesis, the above findings suggest that the increase in Akkermansia genus abundance may be involved in the downregulation of systemic inflammation and the associated pre-systemic glucose uptake, especially in visceral fat, whereas the upregulation of Desulfovibrio and RF39 bacteria may activate metabolic pathways linked with liver inflammation susceptibility. Among these, our data point towards the (in)ability of visceral fat and gut to extract glucose from the systemic circulation (post-systemic glucose uptake), leading to liver glucose overexposure and an increase in EGP. The immune modulatory impact of the gut microbiota can significantly affect intrahepatic and intestinal inflammation, and hepatocarcinogenesis. Our findings are consistent with the evidence that patients with cirrhosis and steatohepatitis who developed hepatocellular carcinoma (HCC) lacked similar protective bacteria profiles and had enhanced liver and intestinal inflammation as compared to patients with cirrhosis, but without HCC . Cirrhosis patients had a lower abundance of Akkermansia genus abundance than controls in association with an inflammatory intestinal environment and higher fecal calprotectin levels, and Bifidobacterium was also depleted in HCC patients . Thus, the combined deficiency of these beneficial bacteria might enhance intestinal and liver inflammation, influencing liver disease progression as well as the initiation and/or progression of hepatocarcinogenesis, and the administration of probiotics was shown to reduce the incidence and growth of HCC lesions in mice, by modulating gut microbiota composition and restoring intestinal permeability [ , , ].
We asked to which extent does an early exposure to ND or HFD contribute to, or deviate from, the results obtained in a more general, i.e., admixed population. The microbiota of weaning HFDoff showed an amplified abundance of Bacterioidia and Clostridia class bacteria, namely Rikenellaceae , Parabacteriodes , and Peptostreptococcaceae genera, and in RF39 . In line with this, visceral fat glucose uptake was several folds higher in HFDoff than NDoff at weaning; the gut was affected to a lower extent, and a trend towards higher IL6 levels in HFDoff was seen. The liver showed a greater CT density in this group. Interestingly, both RF39 abundance and lobular inflammation were positive predictors of hepatic CT density, and an LDA analysis indicated that RF39 was a significant discriminator of biopsy-proven NASH. We screened for functional pathways that could mediate the interactions between microbiota and tissue outcomes, and we identified three (upregulated) metabolic pathways that were associated (positively) with liver CT density and/or visceral fat glucose uptake, namely tryptophan metabolism, lysine degradation, and phenylalanine metabolism. Lysine degradation was the most upregulated (+15–30%) pathway in both the colon and caecum. Lysine can be processed by Clostridia to generate short-chain fatty acids (SCFAs) , and we correspondingly observed an upregulation of fatty acid and butanoate (butyric acid) metabolic pathways in the colon and caecum. SCFAs also affect tryptophan metabolism, which has major effects on intestinal barrier integrity and transit, and fatty liver disease and metabolism . In states of obesity and inflammatory bowel diseases, the metabolism of tryptophan is enhanced in the adipose tissue, correlating with IL6 levels and leading to increased intestinal permeability and LPS translocation in the systemic circulation, resulting in peripheral inflammation, altered glucose metabolism, and impaired hepatic protection . These aminoacidic actions recapitulate very well the findings of the current study, namely the potentiation of cytokine IL6 levels and of visceral fat and gut glucose uptake observed in weaning HFDoff, as well as the tendency towards liver inflammation and the occurrence of hyperglycemia in the fed state, as previously shown in this mice group . We have previously reported that maternal HFD led to adulthood obesity, hyperglycemia, hypoinsulinemia, hyperresistinemia, and impaired MCP1 levels in these mice . In this study, we demonstrated that maternal HFD contributes to increasing the susceptibility towards liver inflammation in the offspring from early to adult life, and provokes a 40% elevation in plasma triglyceride levels in adult HFDoff, limiting the accumulation of tissue lipids. The main difference between adult HFDoff and NDoff at the tissue metabolic level was the upregulation in post-systemic glucose uptake in the liver and (less) in the gut in HFDoff. It is of note that the glucose uptake in these two tissues was related to fasting hyperglycemia, suggesting that visceral fat was unable to act as the substrate sink. As previously shown , the microbiota was significantly different in relation to the maternal diet. Among bacteria that were different in adult NDoff and HFDoff, the abundance of Coriobacteriaceae , Desulfovibrio , and Allobaculum were positively associated with liver glucose uptake, and the deficiency of Clostridiales and Peptococcaceae families in HFDoff predicted higher liver and gut glucose uptake. The abundance of Desulfovibrio , Allobaculum , and Actinobacteria have been previously shown to reflect the progression of liver disease in Western-diet-fed or HFD-fed diabetic mice , and the abundance of Desulfovibrio genus emerged as a significant LDA-discriminator of biopsy-proven NASH in this study. Consistent with our findings, steatohepatitis in HFD and in overweight (but not in lean) subjects, and in patients with type 2 diabetes was associated with an increase in gut Desulfovibrio and sulfate-lowering bacteria . We found that arachidonic acid metabolism was augmented by 3.5 folds in adult HFDoff and positively related to gut glucose uptake, together with sulfur and phenylalanine metabolism, and both sulfur reduction and the arachidonic acid cascade have been implicated in gut inflammation, steatohepatitis, and an unhealthy gut–adipose–liver relationship [ , , ]. We also observed that metabolic pathways of methionine-cysteine (i.e., the only sulfur-containing amino acids) were associated with liver and gut post-systemic glucose uptake. These amino acids have been related to an unhealthy gut–liver–adipose metabolism and to obesity, and their dietary restriction has many benefits, as referenced in . In addition, we noted that the phosphatydilinositol and insulin signaling pathways, and glycan biosynthesis and metabolism were related to liver outcomes in our adult HFDoff mice, and it is of interest that the insulin-phosphatydilinositol kinase pathway has been recently involved in the progression of non-alcoholic liver disease . Our results provide a cause–effect demonstration linking weaning, adult age, and maternal diet with respective metabolic, inflammatory, and microbiota outcomes; however, the associations between imaging and microbiota biomarkers support, but do not prove causality. However, a causal relationship was documented in studies addressing microbiota transplantation from 2-week-old infants born to obese (compared to lean) mothers into adult germ-free mice. This led to alterations in intestinal permeability, increased hepatic endoplasmic reticulum stress, and signs of periportal inflammation , together with impaired macrophage phagocytic function and cytokine production, the latter promoting weight gain, beyond hepatic macrophage accumulation. These findings are consistent with the overweight, associating with hepatic inflammation susceptibility, and the four-fold MCP1 defect observed in our adult HFDoff.
Our study is the first to investigate tissue specific CT radiodensity and glucose uptake in vivo, in relation to microbiota composition and function in the context of the weaning reaction, tissue maturation, and maternal diet. The results suggest that pre-systemic hypermetabolism and high CT density reflect inflammatory processes, whereas post-systemic glucose uptake (gut, visceral fat) confers protection against hepatic metabolic stress. The tissue glucose uptake was high in the weaning period and related to the dominance of Clostridia and Bacteroidia classes ( Oscillospira , Coprococcus , RC44 , Clostridiales , Ruminococcaceae , Rikenellaceae , S24-7 genera) and of metabolic pathways involved in energy-fueling, constitutive growth, and endotoxemia-inflammation-immunity. Maturation affected microbiota composition, reducing systemic inflammation and tissue metabolism, but increasing the susceptibility for liver inflammation, whose microbiota discriminators were RF39 and Desulfovibrio genera. Maternal HFD promoted a proinflammatory microbiota and metabolic pathway profile at both ages, amplifying the weaning reaction (high visceral fat glucose uptake, IL6, liver CT density) and causing adult dysmetabolism (high post-systemic gut and liver glucose uptake in relation to hyperglycemia, with visceral fat being unable to compensate), and monocyte-macrophage dysfunction (low MCP1), increasing proneness for liver injury. We suggest that the visceral network establishes an optimal balance between metabolism and inflammation, and the weaning time window is crucial for its modulation. Our results point to visceral fat as an organ undergoing metabolic stress in early life in the offspring of HFD dams, which may result in a dysfunctional storage capacity in adulthood, and we identify clear imaging biomarkers and bacteria to test the reprogramming of the weaning reaction as a necessary future line of investigation.
|
Simulated data for census-scale entity resolution research without privacy restrictions: a large-scale dataset generated by individual-based modeling | dbb915cf-20fd-4d62-bf9f-f0e7d66b89d0 | 11518969 | Psychiatry[mh] | Entity resolution (ER) is a foundational element of data science and has emerged as a crucial research task in a variety of disciplines, from the social sciences to epidemiology to forensics . Put simply, ER is the process of linking the records corresponding to a single “entity” (e.g., an individual person) from one or multiple data sources when there is not a unique key on which to join them. In this context, an entity may be anything a row of data corresponds to, for example a person, household, business, or establishment. Record linkage of administrative data can enable the analysis of events across government services and systems – . For researchers who work with large-scale, individual-level data, such as those working with the US Census Bureau, ER typically uses personally identifiable information (PII) such as name, address, date of birth, or government-issued identification numbers. Protecting PII is crucial to safeguarding individuals’ privacy, security, and personal well-being in an increasingly interconnected and data-driven world. As such, restrictions on access to these data have presented a barrier to developing and testing new methods and software for ER . Although it is possible for some research to proceed using perturbed or synthetic data or for researchers to work with confidential data in a secure data enclave, it seems that technical barriers inherent in these approaches have prevented them from overcoming these barriers at present. In 2021, the US Census Bureau (USCB) awarded a cooperative agreement to the University of Washington’s Institute for Health Metrics and Evaluation (IHME) Simulation Science team to expand and improve ER methodological research and technology . As part of this work, we have used simulation to address the research barrier caused by PII. Through the development of a simulated version of various administrative datasets, including a simulation of the confidential data gathered by the USCB, we hope to help researchers develop new techniques for linking datasets together that are compatible with the privacy protections necessary for such sensitive and consequential information – and to do so without needing access to the real data. The goal of the generation of these data is to use them for ER methods research, and the datasets themselves are not intended to replicate or reconstruct protected data for social scientific research. It should be noted that our team is not the first to attempt such a data synthesis project. Prior approaches include the Australian National University’s “Freely extensible biomedical record linkage” and Data Generator and Corruptor projects ; and from the University of Arkansas Little Rock, the synthetic occupancy generator approach . There is also relevant work from the University of Edinburgh, which developed an R package for producing synthetic data called synthpop ; and from the United Kingdom Ministry of Justice, which developed synthetic data for testing the Python package Splink . Recently another group has use simulation specifically to generate data for record linkage, which might be considered the first paper to use microsimulation for generating data for record linkage . There are certain features of our project, however, that differentiate it from other efforts, most notably the scale of our simulated data: We have simulated a 100% sample of the USA over 20 years. We have recently released pseudopeople, a Python software package that allows users to generate realistic simulated data about a fictional United States population over multiple decades. Both the generation and distribution of the dataset are governed by a system of “relational governance” , in which data subjects play a central role; a paper on our governance approach is currently under preparation . The package produces simulated datasets similar to census, survey, and administrative datasets routinely used by USCB in their data linkage practice, and it allows the user to configure the levels of noise in each dataset. This includes noise that leaves fields blank, chooses wrong options, replaces names with nicknames and fake names, swaps months and days in dates, misreports ages, and writes wrong digits in numbers and zip codes, as well as adding phonetic, optical character recognition, and typographical errors, following the approach pioneered by Christen and colleagues , , . Readers interested in more details on how our pseudopeople software adds noise to the data generated by this simulation are referred to the pseudopeople documentation website, pseudopeople.readthedocs.io , which includes implementations and extensions of many of the data corruption approaches listed in the previous paragraph. The simulated datasets generated by pseudopeople are based on the results of an individual-based microsimulation built with our Vivarium simulation framework. This simulation is calibrated with real, publicly accessible data about the United States population, including realistic household and family structures, at a large scale. The purpose of this data note is to describe this simulation, which we hope will aid researchers in using our pseudopeople package to develop new algorithms and software.
Using Vivarium Vivarium is a mature, open-source simulation framework that uses standard scientific Python tools, such as NumPy and pandas , . A simulation in Vivarium consists of user-written components that encapsulate the simulation logic, a machine-readable model specification that describes what components are in the model and how they are configured, and a data file containing all data used in the simulation. The framework provides a set of services to assist users in writing their model components, an engine for executing the simulations from both an interactive Python session and from the command line, and abstractions to help manage and format model input data. Models built in Vivarium are typically individual-based, representing people in a population as agents or “simulants,” each with their own age, sex, and other characteristics relevant to the specific model. They typically use discrete “time steps” at which events may occur. In this work, each Vivarium simulant represented a person living in the United States, and each discrete time step included changes relevant to data used in record linkage, such as births, deaths, moving to another address, and changing jobs. We use a pre-established workflow when developing a Vivarium simulation, with roles for researchers and software engineers. The researchers lead the model development process through background research, conceptualizing modeling strategies, validating strategies with domain experts, guiding the conceptual development of the modeling software, and generating analytics for simulation inputs and outputs. The software engineers lead the development of simulation code, including model components and outputs, and tools supporting model and input data analytics. In the following sections, we will cover the different input data sources and data processing strategies used to inform our simulation of the US population. We describe the Vivarium model components for simulant characteristics including basic demographics, household structure, mortality, fertility, migration, and employment dynamics. We also describe the addition of simulant names, physical addresses, employer names, and other attributes which we implemented as a post-processing step, rather than during the simulation itself. Simulation time We initialized our simulation to begin on January 1, 2019, and step forward in time with 28-day time steps until the simulation clock exceeded May 1, 2041. We chose this time step duration to balance the complexity of changes in demographics, housing, employment, etc. with the computational demand of running a simulation with over 300 million simulants. Concept model The concept model diagrams ( and ) provide a visualization of the logical dynamics underlying this simulation and indicate how the various components of the simulation relate. The simulation components can be divided into three overarching categories: i) simulation events (i.e., birth, death, migration, and employment change), ii) simulant attributes (i.e., demographics, household structure, location, and government-issued identification numbers such as SSNs), and iii) simulated dataset observers (i.e., how the simulants are observed over the course of the simulation, through routinely collected surveys and administrative datasets, such as the Decennial Census, tax forms, household surveys, and government-related social safety programs). shows what influences the occurrence of each simulation event and how these events are captured in our data collection, while shows how the simulant components interact with one another at the individual level. When a simulant undergoes an event (e.g., gives birth, changes jobs, changes address), the simulant’s attributes change accordingly. Those attributes are then captured by the observers. Input data. We informed the simulated datasets we developed for pseudopeople using open-source input data, including data released publicly by the Social Security Administration (SSA) and the USCB. We informed physical addresses from the training data of the Python package libpostal, as repackaged by the deepparse project . In the sections that follow, we elaborate on how we used these data sources, and how our simulation could be extended to be even more realistic in future work. Basic demographics. We initialized the simulated population’s demographic characteristics, including age and date of birth, race/ethnicity, sex, nativity (i.e., whether a simulant was born within the US), geographic area, and household structure by sampling from the 2016–2020 ACS Public Use Microdata Sample (PUMS) . By sampling from PUMS, we were able to match the univariate distribution of each attribute as well as joint distributions of arbitrary complexity between the attributes at the Public Use Microdata Area level, while also preserving structure within sampled household units. For instance, the PUMS data capture the age distribution of people in America, where more people were born from 1945 to 1965 than from 1965 to 1985. The PUMS data are not without limitations, however. For example, the granularity of the PUMS data is limited by privacy considerations, and specific details that might be crucial for a detailed analysis are sometimes obscured, which could affect the precision of simulations based on these data, especially in socioeconomic and health-related contexts. Age is reported in PUMS in floored integer years, but our simulation uses precise ages in fractional years. We assigned simulants a uniformly random precise age consistent with their nominal age as sampled from PUMS. For ER research and development, it was particularly important that we did not generate simulants who are much more similar to one another than would be expected in a real population, which would make linkage unrealistically difficult. Our simulated population is the size of the US population, but every simulant is initialized from a person in PUMS, which is a 5% sample of the US. Therefore, many simulants are created from the same person in PUMS, which could create unrealistic clustering. To decrease similarity without assuming total independence between attributes, we perturbed age values at sampling time. In different components of the simulation, we sampled different entity types from the PUMS: entire households, individuals living in group quarters (GQs), or individuals living in households (non-GQs). For each entity sampled, we added a random age shift taken from a standard normal distribution to that entity’s age value(s). When perturbation led to a negative age value, we flipped the negative age value’s sign. We then defined each simulant’s date of birth to be consistent with their precise age. Sex is reported in PUMS as binary (male or female), so we initialized a sex attribute this way as well for each simulant. We mapped separate PUMS indicators of race and ethnicity to a single composite “race/ethnicity” indicator, with the following exhaustive and mutually exclusive categories: “White,” “Black,” “Latino,” “American Indian and Alaskan Native,” “Asian,” “Native Hawaiian and Other Pacific Islander,” and “Multiracial or Some Other Race.” We defined these categories in accordance with the guidelines provided by the US Office of Management and Budget (OMB) . Nativity describes whether a simulant was born in the United States or elsewhere, and we modeled this as a binary variable in our simulation. We used this nativity attribute to inform the likelihood that the simulant had a Social Security Number (SSN). provides a sample of the basic demographics present in the simulated population used in pseudopeople (note that these are entirely simulated data and therefore do not constitute Confidential Unclassified Information under US Code). Household structure. Our simulants lived in either residential households or group quarters (GQ). We used the ACS PUMS data to inform the residential household structure regarding how each simulant is related to a reference person in their household. Simulants living in GQ do not have such a relationship and GQs do not have a reference person. Residential households and GQs have geographic locations as well as physical and mailing street addresses, which may be different, because some residential households receive mail at a PO box (we do not simulate other kinds of mailing-only addresses, such as rural route addresses). PUMS data were not sufficient to identify precisely which type of GQ each simulant resided in; they only provided information on whether it was an institutional or non-institutional GQ. We subdivided institutional GQ into three mutually exclusive and collectively exhaustive categories of carceral, nursing homes, and other institutional. We also subdivided non-institutional GQ into college, military, and other non-institutional. We chose a GQ type uniformly at random for each simulant out of the three types consistent with their institutional/non-institutional status. For simulants living in residential households, we modeled a relationship to the reference person of their household based on the relationship values in the PUMS . Possible relationship values were reference person, biological child, adopted child, stepchild, sibling, parent, grandchild, parent-in-law, child-in-law, other relative, roommate, foster child, other non-relative. Mortality. To model mortality, we used our standard Vivarium approach, informed by data from the age- and sex-specific estimate of all-cause mortality for the US in 2019 as produced by the IHME Global Burden of Disease Study . When a simulant who was the reference person in a non-GQ household died, we made the oldest remaining simulant in that household the new reference person and updated all other relationships (this produces some households with an unrealistically young simulant as the reference person). Unlike many of our past Vivarium simulations, we did not model the underlying cause for any simulant’s death. However, we could extend this simulation to model specific causes of death in future iterations of the simulation, such as to facilitate research and development in cancer registry linkage applications. Fertility. We used our standard Vivarium approach to an age-specific fertility model in which each female simulant has a probability of having a birth event at each time step, derived from the age-specific fertility rate for the USA. In the current version of our model, only one female parent is identified, representing the simulant who gave birth. The birth event is considered to occur at a randomly chosen time during the 28-day time step, which informs the date of birth and age of the simulants born. We select a random 4% of birth events to be the birth of twins (two newborn simulants), and for the other birth events we add a single newborn simulant. We expect that the inclusion of twins will create some particularly challenging ER data, where simulants have the same last name, address, and date of birth. We do not include adoption or any other complexities of family structure. The newborn simulant inherits certain attributes from their mother simulant, including household, race/ethnicity, and last name (recall that the simulation associates a newborn with only a single parent, so these attributes are inherited from this individual unambiguously). These simplifying assumptions allowed us to avoid modeling the complex dynamics of relationships but precluded us from following the dominant patriarchal naming pattern present in the US. The nativity of children born in the simulation is set to reflect that they were born in the US; therefore, all children born in the simulation are assigned an SSN. Additionally, we assigned newborns a relationship to the reference person in their household (which is also their parent’s household) based on the relationship between their parent and the household reference person, using a set of logical business rules. Migration. We attempted to include accurate patterns of migration in our simulation, as migration leads to changing addresses, which constitutes an important challenge in ER. As with basic demographics, all data informing migration in our simulation come from ACS PUMS. We used PUMS to calculate migration by demographics. There are a huge number of attributes that could explain moving behavior, and they may interact in complex ways in the real world. We modeled only some of this complexity and captured three types of household and individual migration events: migration within the simulation (domestic migration), migration into the simulation (in-migration), and migration out of the simulation (out-migration). Domestic migration We modeled domestic migration events as happening at a rate determined by age, sex, and race/ethnicity; we held these rates constant across time in the simulation. Individual domestic migration caused a single simulant to move and might reflect an individual moving out of their current living situation (i.e., GQ or residential household) and establishing a new one-person household, moving into GQ, or joining an existing residential household as a non-reference person. For individual migrations in which a simulant establishes a new household, we always classified the simulant as the reference person. For individual migrations in which a simulant joins an existing household, the simulant is always classified as an “Other nonrelative.” We assumed that simulants have at most a single individual migration event per time step. When a simulant who was the reference person in a non-GQ household moved, we assigned the oldest remaining simulant in their household to be the reference person and updated all other relationships in the household according to logical business rules. Household domestic migration caused an entire household of more than one simulant to move as a unit. As with individual migration, we calculated the rate of household migration per household-year and stratified by demographics. Because households do not have overall demographic characteristics, we used the demographics of the reference person for this stratification. Unlike individual migration, we did not change any relationships in household migrations. We used a simplifying assumption that all simulants who moved and were of working age (which we define as age 18 and older) changed employment. International immigration We modeled immigration by adding new simulants to our simulation to represent individuals moving into the US from other countries. We sampled simulants immigrating to the US from the subset of the 2016–2020 ACS PUMS who had immigrated to the US in the year before they were surveyed and did not perturb their age. This approach relies on our assumption that the number-per-year and demographic characteristics of recent immigrants in the 2016–2020 ACS PUMS will not change substantially for all future years of the simulation. We modeled three kinds of immigration events in our simulation: household moves, GQ person moves, and non-reference-person moves. As with domestic migration, a household move is when an entire non-GQ household enters from outside the country as a unit, preserving relationships within the unit. A GQ person move is when a simulant enters from outside the country and joins group quarters. Because simulants who reside in GQs do not have tracked relationships in PUMS or our simulation, these moves have no relationship structure. Lastly, a non-reference-person move is when an individual simulant enters from outside the country and joins an existing non-GQ household with some relationship other than “reference person.” We used the weighted number of last-year immigration events of each type from the ACS PUMS to inform the yearly rate at which immigration events of each move type occurred in our simulation. We simulated constant rates over time and did not model seasonal or temporal fluctuation in immigration. International emigration Emigration occurs when a simulant leaves the US to live in another country. We used the Net International Migration (NIM) estimates from the Census Bureau’s Population Housing Unit (PopEst) program to determine the number of emigrants per year, by subtracting immigration numbers from ACS to isolate emigration. The NIM estimates are made by the PopEst team by combining information about immigration from ACS with information about emigration from demographic analysis (for those born outside the US) and analysis of foreign censuses (for those born in the US) . There are three types of emigration events that can occur in our simulation: household moves, GQ-person moves, and non-reference-person moves. These cause an entire household, a GQ person, or a household member who is not a reference person to leave the US, respectively. We stratified emigration rates by age group, sex, race/ethnicity, nativity, and US state of residence, and we assumed that these stratified rates were constant over time, without a long-term trend or seasonal variation. We stopped tracking households and individuals after an emigration event and assumed that they would not return to the US or appear in any pseudopeople data after they had emigrated. Employment dynamics. We consider all simulants aged 18 years or older to be working age; all such simulants either have an employer or are considered unemployed. We only allow a single employer at a time for each simulant. We initialized the working-age simulants to be unemployed, employed in the military, or employed otherwise, and we considered the military to be a single employer. To employ the rest of the simulants (those with non-military jobs), we generated employers with an initial size attribute chosen from a skewed distribution to ensure that there are a few large employers and many small employers. In order to assign individual simulants to employers such that the size attribute is (roughly) accurate at the population level, we selected each simulant’s employer from the categorical distribution where the probability of each employer is proportional to its initial size attribute. Working-age simulants (including those who are unemployed) change employment randomly at a rate of 50 changes per 100 person-years (a rate we selected subjectively to provide an appropriate challenge in record linkage). When a simulant changes employment, we sample a new employer with the same procedure used at initialization. This approach to selecting a new employer ensures that at the population level, the number of simulants employed by any given employer will remain roughly proportional to the initial size attribute sampled for that employer. We also simulate income, which affected which datasets a simulant appeared in. For instance, the WIC dataset only recorded simulants with household income below a certain threshold. We approximated the income distribution with a log-normal distribution for each age group, sex, and race/ethnicity combination, fit to the ACS PUMS. See Appendix 1 for more detail on the distribution parameters used for each demographic group. We simulated only income earned through wages; unemployed simulants had no income. To simplify our model, we assumed statistical independence between wages and employer for employed simulants. Post-processing We added some elements to the simulated data after the simulation ran. This includes features that would require additional computing resources to track for the simulation’s duration, such as simulant names (first and last), employer names, and government-issued identification numbers (i.e., SSNs and ITINs). We developed simulant first and last names based on two distinct data sources: first and middle names are sourced from SSA data, which allowed use to match name frequencies to age and sex; while last names are generated from Census data with hyphens and spaces added to make linking tasks realistically more challenging, which allowed use to match the frequencies by race/ethnicity , . We generated SSNs in accordance with the current algorithm used to issue unique SSNs. We selected the first three digits uniformly at random from 001 to 899, excluding 666; the next two digits from 01 to 99; and the final four digits from 0001 to 9999 . We generated Individual Taxpayer Identification Numbers (ITINs) for simulated 1040 filings by simulants without an SSN using a similar process . We based our simulated employer names on a database of 5,321,506 “location names” from the SafeGraph “Core Places of Interest USA” dataset released in June 2020 . To create a representation of bigrams from this dataset, we constructed a directed multigraph. Each word in a location name was treated as a node, and we included special <start> and <end> nodes. We included a directed multi-edge for each occurrence of a word pair in sequence in each location name. To generate simulated employer names, we performed a random walk through the bigram graph. Starting from the <start> node, we traversed directed edges selected uniformly at random until we reached the <end> node or exceeded a predetermined maximum path length. We then combined the words associated with each node that was encountered along the path to form the simulated employer name. This approach resulted in a diverse range of names that maintained a realistic quality. In the sample data included with the pseudopeople package, the W2 and 1099 employer names in 2020 include 212 distinct names and the three most commons are “San Benito Martinez Landscape Supply”, “Tony's Family Practice Inc”, and “Pikes Creek Campground”.
Vivarium is a mature, open-source simulation framework that uses standard scientific Python tools, such as NumPy and pandas , . A simulation in Vivarium consists of user-written components that encapsulate the simulation logic, a machine-readable model specification that describes what components are in the model and how they are configured, and a data file containing all data used in the simulation. The framework provides a set of services to assist users in writing their model components, an engine for executing the simulations from both an interactive Python session and from the command line, and abstractions to help manage and format model input data. Models built in Vivarium are typically individual-based, representing people in a population as agents or “simulants,” each with their own age, sex, and other characteristics relevant to the specific model. They typically use discrete “time steps” at which events may occur. In this work, each Vivarium simulant represented a person living in the United States, and each discrete time step included changes relevant to data used in record linkage, such as births, deaths, moving to another address, and changing jobs. We use a pre-established workflow when developing a Vivarium simulation, with roles for researchers and software engineers. The researchers lead the model development process through background research, conceptualizing modeling strategies, validating strategies with domain experts, guiding the conceptual development of the modeling software, and generating analytics for simulation inputs and outputs. The software engineers lead the development of simulation code, including model components and outputs, and tools supporting model and input data analytics. In the following sections, we will cover the different input data sources and data processing strategies used to inform our simulation of the US population. We describe the Vivarium model components for simulant characteristics including basic demographics, household structure, mortality, fertility, migration, and employment dynamics. We also describe the addition of simulant names, physical addresses, employer names, and other attributes which we implemented as a post-processing step, rather than during the simulation itself.
We initialized our simulation to begin on January 1, 2019, and step forward in time with 28-day time steps until the simulation clock exceeded May 1, 2041. We chose this time step duration to balance the complexity of changes in demographics, housing, employment, etc. with the computational demand of running a simulation with over 300 million simulants.
The concept model diagrams ( and ) provide a visualization of the logical dynamics underlying this simulation and indicate how the various components of the simulation relate. The simulation components can be divided into three overarching categories: i) simulation events (i.e., birth, death, migration, and employment change), ii) simulant attributes (i.e., demographics, household structure, location, and government-issued identification numbers such as SSNs), and iii) simulated dataset observers (i.e., how the simulants are observed over the course of the simulation, through routinely collected surveys and administrative datasets, such as the Decennial Census, tax forms, household surveys, and government-related social safety programs). shows what influences the occurrence of each simulation event and how these events are captured in our data collection, while shows how the simulant components interact with one another at the individual level. When a simulant undergoes an event (e.g., gives birth, changes jobs, changes address), the simulant’s attributes change accordingly. Those attributes are then captured by the observers. Input data. We informed the simulated datasets we developed for pseudopeople using open-source input data, including data released publicly by the Social Security Administration (SSA) and the USCB. We informed physical addresses from the training data of the Python package libpostal, as repackaged by the deepparse project . In the sections that follow, we elaborate on how we used these data sources, and how our simulation could be extended to be even more realistic in future work. Basic demographics. We initialized the simulated population’s demographic characteristics, including age and date of birth, race/ethnicity, sex, nativity (i.e., whether a simulant was born within the US), geographic area, and household structure by sampling from the 2016–2020 ACS Public Use Microdata Sample (PUMS) . By sampling from PUMS, we were able to match the univariate distribution of each attribute as well as joint distributions of arbitrary complexity between the attributes at the Public Use Microdata Area level, while also preserving structure within sampled household units. For instance, the PUMS data capture the age distribution of people in America, where more people were born from 1945 to 1965 than from 1965 to 1985. The PUMS data are not without limitations, however. For example, the granularity of the PUMS data is limited by privacy considerations, and specific details that might be crucial for a detailed analysis are sometimes obscured, which could affect the precision of simulations based on these data, especially in socioeconomic and health-related contexts. Age is reported in PUMS in floored integer years, but our simulation uses precise ages in fractional years. We assigned simulants a uniformly random precise age consistent with their nominal age as sampled from PUMS. For ER research and development, it was particularly important that we did not generate simulants who are much more similar to one another than would be expected in a real population, which would make linkage unrealistically difficult. Our simulated population is the size of the US population, but every simulant is initialized from a person in PUMS, which is a 5% sample of the US. Therefore, many simulants are created from the same person in PUMS, which could create unrealistic clustering. To decrease similarity without assuming total independence between attributes, we perturbed age values at sampling time. In different components of the simulation, we sampled different entity types from the PUMS: entire households, individuals living in group quarters (GQs), or individuals living in households (non-GQs). For each entity sampled, we added a random age shift taken from a standard normal distribution to that entity’s age value(s). When perturbation led to a negative age value, we flipped the negative age value’s sign. We then defined each simulant’s date of birth to be consistent with their precise age. Sex is reported in PUMS as binary (male or female), so we initialized a sex attribute this way as well for each simulant. We mapped separate PUMS indicators of race and ethnicity to a single composite “race/ethnicity” indicator, with the following exhaustive and mutually exclusive categories: “White,” “Black,” “Latino,” “American Indian and Alaskan Native,” “Asian,” “Native Hawaiian and Other Pacific Islander,” and “Multiracial or Some Other Race.” We defined these categories in accordance with the guidelines provided by the US Office of Management and Budget (OMB) . Nativity describes whether a simulant was born in the United States or elsewhere, and we modeled this as a binary variable in our simulation. We used this nativity attribute to inform the likelihood that the simulant had a Social Security Number (SSN). provides a sample of the basic demographics present in the simulated population used in pseudopeople (note that these are entirely simulated data and therefore do not constitute Confidential Unclassified Information under US Code). Household structure. Our simulants lived in either residential households or group quarters (GQ). We used the ACS PUMS data to inform the residential household structure regarding how each simulant is related to a reference person in their household. Simulants living in GQ do not have such a relationship and GQs do not have a reference person. Residential households and GQs have geographic locations as well as physical and mailing street addresses, which may be different, because some residential households receive mail at a PO box (we do not simulate other kinds of mailing-only addresses, such as rural route addresses). PUMS data were not sufficient to identify precisely which type of GQ each simulant resided in; they only provided information on whether it was an institutional or non-institutional GQ. We subdivided institutional GQ into three mutually exclusive and collectively exhaustive categories of carceral, nursing homes, and other institutional. We also subdivided non-institutional GQ into college, military, and other non-institutional. We chose a GQ type uniformly at random for each simulant out of the three types consistent with their institutional/non-institutional status. For simulants living in residential households, we modeled a relationship to the reference person of their household based on the relationship values in the PUMS . Possible relationship values were reference person, biological child, adopted child, stepchild, sibling, parent, grandchild, parent-in-law, child-in-law, other relative, roommate, foster child, other non-relative. Mortality. To model mortality, we used our standard Vivarium approach, informed by data from the age- and sex-specific estimate of all-cause mortality for the US in 2019 as produced by the IHME Global Burden of Disease Study . When a simulant who was the reference person in a non-GQ household died, we made the oldest remaining simulant in that household the new reference person and updated all other relationships (this produces some households with an unrealistically young simulant as the reference person). Unlike many of our past Vivarium simulations, we did not model the underlying cause for any simulant’s death. However, we could extend this simulation to model specific causes of death in future iterations of the simulation, such as to facilitate research and development in cancer registry linkage applications. Fertility. We used our standard Vivarium approach to an age-specific fertility model in which each female simulant has a probability of having a birth event at each time step, derived from the age-specific fertility rate for the USA. In the current version of our model, only one female parent is identified, representing the simulant who gave birth. The birth event is considered to occur at a randomly chosen time during the 28-day time step, which informs the date of birth and age of the simulants born. We select a random 4% of birth events to be the birth of twins (two newborn simulants), and for the other birth events we add a single newborn simulant. We expect that the inclusion of twins will create some particularly challenging ER data, where simulants have the same last name, address, and date of birth. We do not include adoption or any other complexities of family structure. The newborn simulant inherits certain attributes from their mother simulant, including household, race/ethnicity, and last name (recall that the simulation associates a newborn with only a single parent, so these attributes are inherited from this individual unambiguously). These simplifying assumptions allowed us to avoid modeling the complex dynamics of relationships but precluded us from following the dominant patriarchal naming pattern present in the US. The nativity of children born in the simulation is set to reflect that they were born in the US; therefore, all children born in the simulation are assigned an SSN. Additionally, we assigned newborns a relationship to the reference person in their household (which is also their parent’s household) based on the relationship between their parent and the household reference person, using a set of logical business rules. Migration. We attempted to include accurate patterns of migration in our simulation, as migration leads to changing addresses, which constitutes an important challenge in ER. As with basic demographics, all data informing migration in our simulation come from ACS PUMS. We used PUMS to calculate migration by demographics. There are a huge number of attributes that could explain moving behavior, and they may interact in complex ways in the real world. We modeled only some of this complexity and captured three types of household and individual migration events: migration within the simulation (domestic migration), migration into the simulation (in-migration), and migration out of the simulation (out-migration). Domestic migration We modeled domestic migration events as happening at a rate determined by age, sex, and race/ethnicity; we held these rates constant across time in the simulation. Individual domestic migration caused a single simulant to move and might reflect an individual moving out of their current living situation (i.e., GQ or residential household) and establishing a new one-person household, moving into GQ, or joining an existing residential household as a non-reference person. For individual migrations in which a simulant establishes a new household, we always classified the simulant as the reference person. For individual migrations in which a simulant joins an existing household, the simulant is always classified as an “Other nonrelative.” We assumed that simulants have at most a single individual migration event per time step. When a simulant who was the reference person in a non-GQ household moved, we assigned the oldest remaining simulant in their household to be the reference person and updated all other relationships in the household according to logical business rules. Household domestic migration caused an entire household of more than one simulant to move as a unit. As with individual migration, we calculated the rate of household migration per household-year and stratified by demographics. Because households do not have overall demographic characteristics, we used the demographics of the reference person for this stratification. Unlike individual migration, we did not change any relationships in household migrations. We used a simplifying assumption that all simulants who moved and were of working age (which we define as age 18 and older) changed employment. International immigration We modeled immigration by adding new simulants to our simulation to represent individuals moving into the US from other countries. We sampled simulants immigrating to the US from the subset of the 2016–2020 ACS PUMS who had immigrated to the US in the year before they were surveyed and did not perturb their age. This approach relies on our assumption that the number-per-year and demographic characteristics of recent immigrants in the 2016–2020 ACS PUMS will not change substantially for all future years of the simulation. We modeled three kinds of immigration events in our simulation: household moves, GQ person moves, and non-reference-person moves. As with domestic migration, a household move is when an entire non-GQ household enters from outside the country as a unit, preserving relationships within the unit. A GQ person move is when a simulant enters from outside the country and joins group quarters. Because simulants who reside in GQs do not have tracked relationships in PUMS or our simulation, these moves have no relationship structure. Lastly, a non-reference-person move is when an individual simulant enters from outside the country and joins an existing non-GQ household with some relationship other than “reference person.” We used the weighted number of last-year immigration events of each type from the ACS PUMS to inform the yearly rate at which immigration events of each move type occurred in our simulation. We simulated constant rates over time and did not model seasonal or temporal fluctuation in immigration. International emigration Emigration occurs when a simulant leaves the US to live in another country. We used the Net International Migration (NIM) estimates from the Census Bureau’s Population Housing Unit (PopEst) program to determine the number of emigrants per year, by subtracting immigration numbers from ACS to isolate emigration. The NIM estimates are made by the PopEst team by combining information about immigration from ACS with information about emigration from demographic analysis (for those born outside the US) and analysis of foreign censuses (for those born in the US) . There are three types of emigration events that can occur in our simulation: household moves, GQ-person moves, and non-reference-person moves. These cause an entire household, a GQ person, or a household member who is not a reference person to leave the US, respectively. We stratified emigration rates by age group, sex, race/ethnicity, nativity, and US state of residence, and we assumed that these stratified rates were constant over time, without a long-term trend or seasonal variation. We stopped tracking households and individuals after an emigration event and assumed that they would not return to the US or appear in any pseudopeople data after they had emigrated. Employment dynamics. We consider all simulants aged 18 years or older to be working age; all such simulants either have an employer or are considered unemployed. We only allow a single employer at a time for each simulant. We initialized the working-age simulants to be unemployed, employed in the military, or employed otherwise, and we considered the military to be a single employer. To employ the rest of the simulants (those with non-military jobs), we generated employers with an initial size attribute chosen from a skewed distribution to ensure that there are a few large employers and many small employers. In order to assign individual simulants to employers such that the size attribute is (roughly) accurate at the population level, we selected each simulant’s employer from the categorical distribution where the probability of each employer is proportional to its initial size attribute. Working-age simulants (including those who are unemployed) change employment randomly at a rate of 50 changes per 100 person-years (a rate we selected subjectively to provide an appropriate challenge in record linkage). When a simulant changes employment, we sample a new employer with the same procedure used at initialization. This approach to selecting a new employer ensures that at the population level, the number of simulants employed by any given employer will remain roughly proportional to the initial size attribute sampled for that employer. We also simulate income, which affected which datasets a simulant appeared in. For instance, the WIC dataset only recorded simulants with household income below a certain threshold. We approximated the income distribution with a log-normal distribution for each age group, sex, and race/ethnicity combination, fit to the ACS PUMS. See Appendix 1 for more detail on the distribution parameters used for each demographic group. We simulated only income earned through wages; unemployed simulants had no income. To simplify our model, we assumed statistical independence between wages and employer for employed simulants.
We added some elements to the simulated data after the simulation ran. This includes features that would require additional computing resources to track for the simulation’s duration, such as simulant names (first and last), employer names, and government-issued identification numbers (i.e., SSNs and ITINs). We developed simulant first and last names based on two distinct data sources: first and middle names are sourced from SSA data, which allowed use to match name frequencies to age and sex; while last names are generated from Census data with hyphens and spaces added to make linking tasks realistically more challenging, which allowed use to match the frequencies by race/ethnicity , . We generated SSNs in accordance with the current algorithm used to issue unique SSNs. We selected the first three digits uniformly at random from 001 to 899, excluding 666; the next two digits from 01 to 99; and the final four digits from 0001 to 9999 . We generated Individual Taxpayer Identification Numbers (ITINs) for simulated 1040 filings by simulants without an SSN using a similar process . We based our simulated employer names on a database of 5,321,506 “location names” from the SafeGraph “Core Places of Interest USA” dataset released in June 2020 . To create a representation of bigrams from this dataset, we constructed a directed multigraph. Each word in a location name was treated as a node, and we included special <start> and <end> nodes. We included a directed multi-edge for each occurrence of a word pair in sequence in each location name. To generate simulated employer names, we performed a random walk through the bigram graph. Starting from the <start> node, we traversed directed edges selected uniformly at random until we reached the <end> node or exceeded a predetermined maximum path length. We then combined the words associated with each node that was encountered along the path to form the simulated employer name. This approach resulted in a diverse range of names that maintained a realistic quality. In the sample data included with the pseudopeople package, the W2 and 1099 employer names in 2020 include 212 distinct names and the three most commons are “San Benito Martinez Landscape Supply”, “Tony's Family Practice Inc”, and “Pikes Creek Campground”.
For dataset validation, we followed the standard workflow used across all of our Vivarium models, using a process often referred to as verification and validation (V&V) . In this process, model results are verified by the research team by checking that a given model approximately replicates target values it was explicitly designed to replicate (e.g., verifying that the proportion of simulants living in group quarters as opposed to individual households matched the value specified by the research team). Results are also validated by the research team, ensuring that model results are logically viable or sensible (e.g., checking that the US population size and structure does not change drastically over the time period modeled). In the event that model results did not meet verification and validation criteria, model implementation and/or design were iteratively adjusted appropriately until criteria were satisfied. We validated pseudopeople datasets through automated testing conducted on the engineering side as well as manual, systematic testing of the simulated population and post-processing data on the research side. In an effort to be as systematic as possible with our user-led data testing processes, we aspired to specify our verification and validation strategies before the synthetic population model was developed by our engineers. For example, we used an interactive simulation in a Jupyter Notebook to verify that simulants were dying at the age- and sex-specific all-cause mortality rates estimated for the USA by the IHME Global Burden of Disease Study. Dataset limitations There are a variety of limitations to our simulation strategy which may affect its ability to reflect real-world dynamics, including but not limited to those regarding migration, employment, physical or mailing address, guardianship, household structure and simulant relationships, and simulant identification. For instance, to make possible the simulation of complex migration dynamics, there are a series of assumptions we made regarding how simulants and households move around, into, and out of the simulation. We assumed that domestic and international migration do not change over time, but rather remain at the average rate from 2016 to 2020 in each future year of the simulation. When a simulant moves, we assume that their mailing address, physical address, and employer all change. In addition, complexities particular to household sub-structure interacting with migration are largely not captured in our simulation. For instance, a child can move out of their household without a parent, or a simulant could move without their spouse into a different household. Additionally, we assume that relationship does not affect emigration rates and that all household types are equally likely to have a simulant move out of or into them. Furthermore, for any individual migration of a simulant from one household into another, we assign the relationship “other nonrelative” in their new household. Thus, as time passes within the simulation, the proportion of households with irregular relationship structures grows. In the sample data included with the pseudopeople package, the 2020 decennial census has 4% of rows have “Other nonrelative” as relationship to the reference person, and in 2030 this rises to 16%. Even in the early years of our simulation, it is possible that there are rare, but challenging, household structures which are not sampled in ACS and therefore not represented in our data either, for example a very small fraction of very large households might present a problem in real-work linkage work that would not be identified when testing with pseudopeople data. Similarly, there are several assumptions that we made to simplify our model of employment dynamics in our simulation. We do not model retirement, and each simulant can only have one employer at a time. There is a myriad of business dynamics that we currently do not model, including new businesses opening, existing businesses closing, business name changes, or business mergers and acquisitions. As with household physical and mailing addresses, when a business address is vacated, it is not reused. In effect, this likely makes business record linkage with these data easier than it will be in practice. Our age-specific fertility and mortality models do not account for variations related to income or race/ethnicity, and in future iterations of this work, we wish to address more complicated dynamics between the various elements of our simulation. There are also limitations in simulant identification because of privacy protections in the name data we have used. The data on first names excludes names with fewer than five occurrences while the data on last names included only names with at least 100 occurrences. Furthermore, we did not model the correlation between first and last names explicitly. We hope to address these limitations in future refinements of our model.
There are a variety of limitations to our simulation strategy which may affect its ability to reflect real-world dynamics, including but not limited to those regarding migration, employment, physical or mailing address, guardianship, household structure and simulant relationships, and simulant identification. For instance, to make possible the simulation of complex migration dynamics, there are a series of assumptions we made regarding how simulants and households move around, into, and out of the simulation. We assumed that domestic and international migration do not change over time, but rather remain at the average rate from 2016 to 2020 in each future year of the simulation. When a simulant moves, we assume that their mailing address, physical address, and employer all change. In addition, complexities particular to household sub-structure interacting with migration are largely not captured in our simulation. For instance, a child can move out of their household without a parent, or a simulant could move without their spouse into a different household. Additionally, we assume that relationship does not affect emigration rates and that all household types are equally likely to have a simulant move out of or into them. Furthermore, for any individual migration of a simulant from one household into another, we assign the relationship “other nonrelative” in their new household. Thus, as time passes within the simulation, the proportion of households with irregular relationship structures grows. In the sample data included with the pseudopeople package, the 2020 decennial census has 4% of rows have “Other nonrelative” as relationship to the reference person, and in 2030 this rises to 16%. Even in the early years of our simulation, it is possible that there are rare, but challenging, household structures which are not sampled in ACS and therefore not represented in our data either, for example a very small fraction of very large households might present a problem in real-work linkage work that would not be identified when testing with pseudopeople data. Similarly, there are several assumptions that we made to simplify our model of employment dynamics in our simulation. We do not model retirement, and each simulant can only have one employer at a time. There is a myriad of business dynamics that we currently do not model, including new businesses opening, existing businesses closing, business name changes, or business mergers and acquisitions. As with household physical and mailing addresses, when a business address is vacated, it is not reused. In effect, this likely makes business record linkage with these data easier than it will be in practice. Our age-specific fertility and mortality models do not account for variations related to income or race/ethnicity, and in future iterations of this work, we wish to address more complicated dynamics between the various elements of our simulation. There are also limitations in simulant identification because of privacy protections in the name data we have used. The data on first names excludes names with fewer than five occurrences while the data on last names included only names with at least 100 occurrences. Furthermore, we did not model the correlation between first and last names explicitly. We hope to address these limitations in future refinements of our model.
Our simulation process produced over 900 gigabytes of simulated censuses, surveys, and administrative data for pseudopeople, representing hundreds of millions of simulants. A sample simulated population of thousands of simulants is now openly available to all users of the pseudopeople package, and large-scale simulated populations of millions and hundreds of millions of simulants are also available by online request through GitHub ( github.com/ihmeuw/pseudopeople/issues ). These simulated population data are structured for use by the pseudopeople package, which includes additional affordances to add various kinds of noise to the data to provide realistic, sharable challenges for ER researchers. shows a sample of the simulated data that might be found in administrative sources on income tax (note that these are entirely simulated data and therefore do not constitute Confidential Unclassified Information under US Code).
By generating population data with complexity and scale comparable to that of large organizations and federal agencies, like the US Census Bureau, we hope to circumvent the common data privacy– and access-related barriers to ER research and development. We intend for this data note to serve as a comprehensive guide for researchers contemplating the use of pseudopeople to develop and test their fresh theories, algorithms, and software systems.
|
A case of solitary plasmacytoma of bone showing co-expression of both immunoglobulin light chains | fb15b5c1-356e-4d54-9663-0df243f3989b | 8686560 | Anatomy[mh] | Solitary plasmacytoma of bone (SPB) is a rare subtype of plasma cell neoplasm, accounting for less than 5% of plasma cell neoplasms overall. SPB may develop at any bone site, but is especially associated with active bone marrow hematopoiesis. Reported sites of origin have included the vertebrae, ribs, and pelvis . Tumor cells resemble plasma cells which have abundant basophilic cytoplasm, often with pale paranuclear area. The nuclei are located eccentrically and contain condensed chromatin. Tumor cells are positive for cytoplasmic immunoglobulin (Ig), surface Ig is usually undetectable, and M-proteins are often present, with IgG being the most common type (about 60%), IgA (15%), and IgE and IgD (about 1%) . Tumor cells are usually negative for CD45 and the pan-B-cell markers CD19 and CD20, and positive for CD38, CD79a, and CD138. Cytologically and immunophenotypically, SPB is similar to plasma cell myeloma (PCM), although it presents as a single localized neoplasm, shows no evidence of bone marrow involvement, and no clinical or laboratory evidence of PCM is evident in affected patients. The median age of patients with SPB is 55 years, i.e., about 10 years younger than those with PCM, and males are affected twice as frequently as females. SPB is treated mainly with radiation and surgery, with additional chemotherapy in cases that have progressed to PCM. The prognosis of SPB is better than that of PCM, and 50% of SPB patients survive longer than 10 years. More than 50% of SPB cases progress to PCM within 2 years or less, suggesting that some cases of SPB represent an early manifestation of PCM [ , – ]. A number of cytokines have been shown to be involved in the pathogenesis of plasma cell neoplasm . These factors, secreted by tumor cells or the surrounding microenvironment, have been implicated in tumor growth and apoptosis, and may also result in the secretion of pathological Ig. It has also been suggested that these cytokines may affect the immune status of plasma cell neoplasm patients . Normal B cells or plasma cells express a single Ig kappa or lambda light chain, and the ratio of B cells or plasma cells expressing kappa relative to those expressing lambda ranges from 0.5 to 3.0 . B cell and plasma cell neoplasms are believed to arise from a single cell after initiation of immunoglobulin heavy chain (IgH) and immunoglobulin light chain (IgL) rearrangement, and the neoplastic plasma cells essentially express a single type of IgL. Thus, restriction of IgL expression can be detected by flow cytometry, in situ hybridization and immunohistochemistry. In neoplastic plasma cells, co-expression of both IgL types is extremely rare. Here, we report a very rare case of SPB showing dual expression of IgL.
A 36-year-old woman presented to a local hospital with a history of neck pain. Computed tomography (CT) and magnetic resonance imaging (MRI) demonstrated a tumor arising from the anterior elements of the C1 and C2 vertebrae. Cervical spine fusion and mass reduction surgery were performed. The resected specimen showed diffuse proliferation of plasma cells and was diagnosed as “plasmacytoma” at that time. Additional radiotherapy was performed, but the patient later dropped out from the treatment course. Sixteen years later, at the age of 52 years, the patient returned complaining of dysarthria. CT and MRI showed a similar but much larger mass at the same location, and recurrence of the tumor was diagnosed (Fig. ). The mass compressed the spinal cord and was thought to be responsible for the dysarthria. The patient was referred to our hospital for further examination and treatment. Quantitative serum Ig analysis showed an increased level of IgG (2096 mg/dl; reference range 870–1700 mg/dl) and normal levels of IgM (203 mg/dl; reference range 46–260 mg/dl) and IgA (293 mg/dl; reference range 110–410 mg/dl). A serum Ig-free light chain study revealed increased levels of both free kappa light chain (61.5 mg/l; reference range 2.42–18.92 mg/l) and free lambda light chain (88.1 mg/l; reference range 4.44–26.18 mg/l) (kappa/lambda: 0.70). Excisional biopsy of the tumor was performed. Flow cytometry analysis demonstrated a distinct population of abnormal plasma cells which were positive for CD56 (96%), CD38 (70%), CD45 (9%), and CD19 (1%). Approximately 96% of these tumor cells co-expressed cytoplasmic kappa and lambda light chain based on a CD38-positive gate strategy (Fig. ). Histologically, the specimen showed diffuse proliferation of plasmacytoid tumor cells with pale paranuclear area and dense chromatin, and immunohistochemistry showed that these tumor cells were strongly and diffusely positive for CD138 (Fig. a, b) and MUM1, and negative for CD3, CD20, and CD56. Immunohistochemistry for IgL demonstrated co-expression of kappa and lambda light chain (Fig. c, d). Additionally, we demonstrated in situ hybridization (ISH) targeting IgL mRNA using FFPE specimen, and ISH also revealed that tumor cells co-expressed kappa and lambda light chain (Fig. e, f). Bone marrow biopsy showed no evidence of plasmacytes showing dual expression or deviation of kappa and lambda light chain, and additional CT and MRI revealed no skeletal abnormality except for the primary lesion, and absence of end-organ damage attributable to a proliferative plasma cell disorder. We reviewed the previous specimen and performed additional IgL immunohistochemistry. The sample revealed a histological pattern similar to that of the later sample as well as co-expression of kappa and lambda light chain (Additional file : Fig. S1). Taking these findings together, the tumor was diagnosed as SPB with dual expression of the kappa and lambda light chains. After diagnosis, the patient underwent nine courses of VRD (bortezomib, lenalidomide, and dexamethasone) therapy, which reduced the size of the lesion. Two years after diagnosis, the lesions have not increased in size, and bone marrow examinations have shown no progression to PCM.
To our knowledge, this case of SPB showing dual expression of the kappa and lambda light chains is the first of its kind to have been reported. Normal plasma cells differentiate from B cells and produce Ig. B cells can rearrange their Ig genes to recognize a huge variety of different antigens. Ig is normally composed of one heavy and one light chain. There are five types of IgH—IgM, IgG, IgA, IgD, and IgE produced by a single gene—and two types of IgL—kappa and lambda produced by two distinct genes. The IgH , IgL kappa , and IgL lambda gene loci are located on chromosomes 14, 2, and 22, respectively. If the IgH and one of the IgL genes are rearranged successfully, the other allele of the IgH and IgL gene is excluded to maintain the antigen specificity of the B cell, a phenomenon known as “allelic exclusion.” In this way, a single B cell expresses the rearranged IgH and the either IgL genes transcribed from only one allele each, with the other 4 alleles remaining in the germline configuration. Normal B cells first undergo IgH rearrangement, followed by IgL kappa rearrangement. If a productive IgL kappa rearrangement occurs, the IgL lambda gene never rearranges. If IgL kappa rearrangement is non-productive for both alleles, the IgL kappa locus is inactivated by deletion and IgL lambda rearrangement occurs . This mechanism of expressing either IgL kappa or lambda genes is called “isotypic exclusion,” and thus B cells (or plasma cells) express one type of light chain, but not both . “Isotypic exclusion” of IgL is not fully understood, although some hypotheses have been proposed on the basis of experimental studies. Inactivation of IgL kappa is accomplished by recombination activating gene (RAG)-mediated joining of the non-coding recombining sequence (RS). The RS is the IgL kappa -deleting element in humans, located ~ 25 kilobases downstream of the constant region of IgL kappa (C-kappa) . RS recombination leads to deletion of the C-kappa exon and silencing of the IgL kappa allele, and it is also known that RS recombination promotes the formation of B cells which express IgL lambda chain [ – ]. However, Diaw et al. reported a mouse-origin plasmacytoma that produced both IgL kappa and lambda, and using micro-manipulation and reverse transcription polymerase chain reaction (RT-PCR) confirmed that both were expressed simultaneously in a single cell . This suggests that in some plasma cell neoplasms both kappa and lambda light chains may exist in a single plasma cell. Furthermore, few cases of PCM showing dual expression of IgL have been reported. The majority of such cases have tended to show a high incidence of complex cytogenetic or fluorescence in situ hybridization (FISH) abnormalities, suggesting involvement of the light chain genes, subsequent isotypic exclusion error, and dual light chain expression [ – ]. Unfortunately, as no investigation of chromosome abnormalities was undertaken in the present case, any genetic dysfunction related to isotypic exclusion remained unclear. On the other hand, Shi et al. subjected human peripheral B cells to single-cell sequencing and found that more than one antibody was produced in some individual B cells, albeit accounting for a small proportion of the total (about 10%) . This unprecedented finding appears to cast doubt on the traditional “one cell – one antibody” paradigm and suggested that dual expression of kappa and lambda light chain can occur in normal conditions and might be the origin of the dual IgL expression neoplasm. However, due to the limited number of cases, further investigation is needed to elucidate the mechanism of tumor co-expression of kappa and lambda light chains. The findings in the present case invite speculation as to how both the kappa and lambda light chains can exist in a single neoplastic plasma cell. Two hypotheses to explain this have been suggested: (1) two types of light chain are expressed in the same antibody; (2) two types of antibodies constructed by each of the kappa and lambda light chains are present in the same plasma cell. In the present case, we detected co-expression of both IgLs in the neoplastic cells by flow cytometry, immunohistochemistry and ISH targeting IgL mRNA, but none of the results offered any suggestion about either hypothesis. There is no way to test these hypotheses at this stage, and further analysis is needed.
To our knowledge, the present case of SPB showing dual IgL expression is the first of its kind to have been reported. IgL rearrangement is under strict genetic control, although the mechanism involved is still unclear. The present appears to represent an exceptional event that deviates from the traditional “isotypic exclusion” mechanism.
Additional file 1: Figure S1. Tumor biopsy of the neck (sampled at 2003). ( a ) H&E staining (×400) shows marked plasma cells infiltration. ( b ) CD138 staining (×200) shows diffuse positivity for these plasma cells. ( c , d ) Immunohistochemistry of kappa and lambda light chain (×200).
|
Clinical outcome assessment trends in clinical trials—Contrasting oncology and non‐oncology trials | 488bd13a-74b2-464c-89de-f72b036828c2 | 10501237 | Internal Medicine[mh] | INTRODUCTION Clinical outcome assessments (COAs) are measures of how patients feel, function, and survive. COAs play a crucial role in holistic evaluation of patient care and can be particularly informative in oncology where the multitude of symptoms, treatment‐related toxicities, and often poor prognosis of patients necessitate a focus on symptom management and quality of life. Landmark clinical trials have demonstrated that incorporation of COAs in trial outcomes can inform whether the intervention benefit is meaningful, that COAs can capture within‐patient effects, and that patient reporting of symptoms can even lead to prolonged survival. Therefore, designing oncology trials to incorporate COAs in conjunction with survival, tumor response, and safety outcomes is key to patient‐centered assessment of net clinical benefit of novel interventions and supportive care. COAs include four categories: patient‐reported outcomes (PRO), observer‐reported outcomes (ObsRO), clinician‐reported outcomes (ClinRO), and performance outcomes (PerfO) measures, which serve complementary roles in clinical trials. PROs and PerfOs enable multifaceted assessments of treatment toxicity and efficacy whereas ClinROs serve in determining trial eligibility and treatment course, and as correlates to objective outcomes and disease progression. COAs presently encompass over 2500 individual instruments and 300 composite measures which range in scope and applicability, with indications covering from specific conditions and neoplasms to all areas of clinical care. This diversity has invariably presented its own challenges, including the difficulty in selecting valid, reliable, and clinically useful instruments, since quality of life and other constructs of interest are often multidimensional, dynamic, and unique to an individual or situation. Regulatory agencies have recognized the value and challenges of incorporating COAs in drug development, , and multiple initiatives and guidelines for trial design, analysis, and reporting advocate for the use of COAs across clinical research. , , , PRO use has been evaluated in general oncology trials and focused cancer trial subsets , , , , , but the comprehensive use of COAs in oncology trials across time and how it compares to the use of COAs in non‐oncology clinical trials has not been evaluated. Moreover, trends in relation to landmark regulatory and advocacy events have not been assessed, and with few exceptions , previous evaluations relied on manual curation of the literature which is not feasible for periodic assessments. In this study, we extensively evaluated COA use across oncology trials and how it compares to the COA use in the rest of clinical trials in the ClinicalTrials.gov registry. We leveraged the registry data through programmatic semi‐automated means and assessed the impact of key regulatory and advocacy events.
METHODS 2.1 Selecting and obtaining study data ClinicalTrials.gov registry was accessed programmatically on August 1, 2021 using R ( RPostgreSQL library) through the Aggregate Analysis of ClinicalTrials.gov database. , Interventional trials with start date 1985–2020 were retrieved from the registry. Oncology trials were determined by searching for Medical Subject Headings terms under “Neoplasms by Site” (Table ) in study title, description, or design. The European Union (EU) or The United States (US)‐based trials were determined by site country location (Table ). Trial data obtained included initiation year, phase, primary purpose, intervention model, randomization, blinding, data monitoring committee (DMC) use, regulatory status of drug or device, and primary sponsors. 2.2 Obtaining clinical outcome assessment information COA instrument details including name, acronyms, COA type (PRO, ClinRO, PerfO, ObsRO, or composite), and therapeutic indications, were extracted from PROQOLID using Python's BeautifulSoup library on October 26, 2020. Several COAs with ambiguous acronyms (such as QLQ or QOL) or ambiguous meanings in trials (e.g., PFS referring either to the Parkinson Fatigue Scale or Progression Free Survival) were excluded from search terms to improve specificity. COA use in trials was identified by searching for COA instrument names and acronyms among trial entries relevant to outcome measures. COA use was further categorized into primary, secondary, or other outcome by matching COA instrument names with trial outcome design descriptions. Similarly, trials assessing overall survival, progression free survival, and tumor response endpoints were identified by searching for relevant terms among trial entries relevant to outcome measures. 2.3 Trends of COA use over time Trends in COA use over time were assessed using Pearson's correlation and linear regression analysis between trial initiation year and the proportion of trials using COAs. When stratified by trial primary purpose or COA category, the proportion was calculated with respect to the baseline count of all (oncology or non‐oncology) trials and all (oncology or non‐oncology) trials using COAs, respectively. Years with statistically outlying (beyond three times the interquartile range) proportion of trials using COAs were excluded from regression. We removed outlier years with regression standardized residuals larger than three and representing years with fewer than 1% of trials to avoid skew in COA use rates from years with low trial counts and confirmed that regression residuals were normally distributed (Shapiro–Wilk deviation from normality: W = 0.96, p = 0.204) and not dependent on year. 2.4 COA use and landmark events Chi‐square tests with continuity correction were used to compare COA use in trials initiated before versus after five landmark events driven by regulatory bodies and initiatives including the European Medicines Agency (EMA), Food and Drug Administration (FDA), Consolidated Standards of Reporting Trials (CONSORT), International Society for Quality of Life Research (ISOQOL), and Standard Protocol Items: Recommendations for Interventional Trials (SPIRIT). Logistic regression was used to assess linear trends between COA use with binary indicator variables for events (whether the trial took place before or after the event) and trial initiation year as a continuous variable and an interaction term between the two covariates. False discovery rate multiple hypothesis correction was performed across models and covariates. 2.5 Trial characteristics associated with COA use Logistic regression was performed to investigate associations between COA use and trial characteristics: phase (as ordinal variable), intervention FDA regulatory status, lead sponsor and collaborator class (National Institutes of Health (NIH) or other US federal agencies, industry, or other, which includes trial networks, academic institutions, and individual investigators), primary purpose (treatment, supportive care, diagnostic, or other, which includes basic science, health services research, prevention, screening, device feasibility, educational/counseling/training), blinding, number of trial arms (single‐ or multi‐arm), data monitoring committee (DMC) use, and randomization. Analyses were conducted using R version 3.6.0.
Selecting and obtaining study data ClinicalTrials.gov registry was accessed programmatically on August 1, 2021 using R ( RPostgreSQL library) through the Aggregate Analysis of ClinicalTrials.gov database. , Interventional trials with start date 1985–2020 were retrieved from the registry. Oncology trials were determined by searching for Medical Subject Headings terms under “Neoplasms by Site” (Table ) in study title, description, or design. The European Union (EU) or The United States (US)‐based trials were determined by site country location (Table ). Trial data obtained included initiation year, phase, primary purpose, intervention model, randomization, blinding, data monitoring committee (DMC) use, regulatory status of drug or device, and primary sponsors.
Obtaining clinical outcome assessment information COA instrument details including name, acronyms, COA type (PRO, ClinRO, PerfO, ObsRO, or composite), and therapeutic indications, were extracted from PROQOLID using Python's BeautifulSoup library on October 26, 2020. Several COAs with ambiguous acronyms (such as QLQ or QOL) or ambiguous meanings in trials (e.g., PFS referring either to the Parkinson Fatigue Scale or Progression Free Survival) were excluded from search terms to improve specificity. COA use in trials was identified by searching for COA instrument names and acronyms among trial entries relevant to outcome measures. COA use was further categorized into primary, secondary, or other outcome by matching COA instrument names with trial outcome design descriptions. Similarly, trials assessing overall survival, progression free survival, and tumor response endpoints were identified by searching for relevant terms among trial entries relevant to outcome measures.
Trends of COA use over time Trends in COA use over time were assessed using Pearson's correlation and linear regression analysis between trial initiation year and the proportion of trials using COAs. When stratified by trial primary purpose or COA category, the proportion was calculated with respect to the baseline count of all (oncology or non‐oncology) trials and all (oncology or non‐oncology) trials using COAs, respectively. Years with statistically outlying (beyond three times the interquartile range) proportion of trials using COAs were excluded from regression. We removed outlier years with regression standardized residuals larger than three and representing years with fewer than 1% of trials to avoid skew in COA use rates from years with low trial counts and confirmed that regression residuals were normally distributed (Shapiro–Wilk deviation from normality: W = 0.96, p = 0.204) and not dependent on year.
COA use and landmark events Chi‐square tests with continuity correction were used to compare COA use in trials initiated before versus after five landmark events driven by regulatory bodies and initiatives including the European Medicines Agency (EMA), Food and Drug Administration (FDA), Consolidated Standards of Reporting Trials (CONSORT), International Society for Quality of Life Research (ISOQOL), and Standard Protocol Items: Recommendations for Interventional Trials (SPIRIT). Logistic regression was used to assess linear trends between COA use with binary indicator variables for events (whether the trial took place before or after the event) and trial initiation year as a continuous variable and an interaction term between the two covariates. False discovery rate multiple hypothesis correction was performed across models and covariates.
Trial characteristics associated with COA use Logistic regression was performed to investigate associations between COA use and trial characteristics: phase (as ordinal variable), intervention FDA regulatory status, lead sponsor and collaborator class (National Institutes of Health (NIH) or other US federal agencies, industry, or other, which includes trial networks, academic institutions, and individual investigators), primary purpose (treatment, supportive care, diagnostic, or other, which includes basic science, health services research, prevention, screening, device feasibility, educational/counseling/training), blinding, number of trial arms (single‐ or multi‐arm), data monitoring committee (DMC) use, and randomization. Analyses were conducted using R version 3.6.0.
RESULTS 3.1 Trial characteristics and associations with COA use A total of 279,855 interventional trials with start dates between 1985 and 2020 (a date range chosen to encompass 95% of all interventional trials in the registry) were retrieved from the registry: 35,415 were determined to be oncology trials studying one or more neoplasms, and the remaining 244,440 were considered non‐oncology trials. The design features and the number of trials reporting them are summarized in Figure and Table . Briefly, the oncology trials reporting relevant information most commonly had a primary purpose of treatment (77%), studied a non‐FDA‐regulated intervention (59%), were completed (44%), phase 2 (48%), non‐blinded (84%), multi‐arm (54%), and non‐randomized (57%). Similarly, non‐oncology trials predominantly had a primary purpose of treatment (64%) and studied a non‐FDA‐regulated intervention (72%), but they had more balanced phase distribution (with phase 1, phase 2, and phase 3 at 22%, 25%, and 21%, respectively), and a higher degree of blinding (52% non‐blinded), multi‐arm design (77%), and randomization (70%) (Figure ; Table ). Eighteen percent of oncology trials reported COA use as determined by the search for individual instruments among trial design entries related to outcomes. Early phase trials had relatively lower rates of COA use (less than 15% each) compared to phase 2/phase 3 (25%), phase 3 (30%), and phase 4 (19%) trials. Nineteen percent of trials studying an FDA‐regulated intervention used a COA, versus 27% among trials studying non‐FDA‐regulated interventions. Eighteen percent of industry‐sponsored trials used COAs, versus 13% of NIH and US federal government‐sponsored, and 18% of all other sponsors. Trials focused on supportive care had the greatest rate of COA use (49%), followed by trials focused on treatment (17%), diagnostic (6%), with cumulative other trial categories having 15% COA use. Trials with blinding had greater rates of COA use (29%) than open‐label trials (16%). Multi‐arm trials had greater rates of COA use (25%) compared to single‐arm trials (12%). Twenty‐seven percent of randomized trials used COAs compared to 12% in non‐randomized trials (Table ). We next evaluated the interplay of the trial design elements in the implementation of COAs using multi‐variate logistic regression across trials reporting sufficient information ( N = 7339; Table ). The likelihood of COA use increased significantly with progressing clinical trial phases (OR = 1.30, p < 0.001). Trials studying FDA‐regulated interventions were significantly less likely to use COAs than those studying non‐FDA‐regulated interventions (OR = 0.81, p = 0.001) whereas trials using DMCs were more likely to use COAs than those without DMCs (OR = 1.26, p < 0.001). Compared to the baseline rate of COA use among supportive care trials, trials focused on treatment (OR = 0.34, p < 0.001), diagnostic (OR = 0.12, p < 0.001), and all other purposes (OR = 0.26, p < 0.001) were significantly less likely to use COAs. Randomized trials were significantly more likely to use COAs than non‐randomized studies (OR = 2.32, p < 0.001; Figure ). Primary sponsor class, blinding, and single‐ versus multi‐arm design were not significantly associated with COA use. Non‐oncology trials in the registry had higher rates of COA use overall (26%) and similar trends in COA use by phase, DMC use, and randomization, whereas unlike in oncology trials, treatment‐focused trials had similar rates of COA use compared to supportive care studies (OR = 1.16, p = 0.100). Additionally, compared to industry‐sponsored trials, trials with other non‐classified sponsors were less likely to use COAs than industry‐sponsored trials (OR = 0.89, p < 0.001), while trials with sponsorship from NIH or US federal government (versus industry; OR = 1.39, p < 0.001), blinding schemes (OR = 1.82, p < 0.001), and single‐arm designs (OR = 1.26, p < 0.001) were significantly more likely to use COAs. The FDA regulatory status of the trial intervention was not significantly associated with COA use among non‐oncology trials (Table ). 3.2 Characteristics of COA instruments used We identified 655 unique instruments used across the oncology trials, representing a diverse set of COAs and comprising of 24% of all instruments obtained from PROQOLID ( N = 2695). The top instruments in order of frequency were the European Organization for Research and Treatment of Cancer Quality of Life Questionnaire‐Core Questionnaire (EORTC QLQ‐C30) used by 29%, Eastern Cooperative Oncology Group (ECOG) Performance Status used by 14%, Euroqol EQ‐5D used by 10%, Hospital Anxiety and Depression Scale (HADS) used by 5%, and Brief Pain Inventory (BPI) used by 4% of trials using COAs (Table ). Fifty‐six percent of trials using COAs reported using one instrument (range: 1–13 instruments per trial). Of trials reporting details of trial outcomes ( N = 5795), COAs were used predominantly as a secondary trial outcome (84%), followed by primary trial outcome (23%), and other outcome (7%). Non‐oncology trials reported a broader range of instruments encompassing 1709 unique COAs; most frequently used instruments were the SF‐36 Health Survey (8%), EQ‐5D (6%), Patient Health Questionnaire (4%), and HADS (3%) (Table ). Similarly to oncology, 50% of non‐oncology trials using COAs reported using one instrument (range: 1–19). Compared to oncology, a greater proportion of non‐oncology trials reported COAs as a primary outcome (49%), with 75% and 7% of trials reporting COAs as secondary and other outcome. Among the COA categories, PROs were used in 84% of oncology trials reporting COA use ( N = 6331), followed by ClinROs in 27%, ObsROs in 6%, PerfOs in 4%, while 4% used composite measures. Oncology trials most often utilized COAs indicated as generic for neoplasms (63%), all conditions (32%), chronic diseases (10%), depression (8%), anxiety disorders (8%), pain (7%), breast neoplasms (6%), and lung neoplasms (5%); all other COA indication groups were represented in less than 5% of trials using COAs (Figures and ). Among non‐oncology trials reporting COA use ( N = 62,544), 78% used PROs, 43% used ClinROs, 13% used ObsROs, 11% used PerfOs, and 16% used composite measures. These instruments were indicated for all conditions (47%), depression (16%), anxiety disorders (13%), schizophrenia (9%), generic for mental disorders (9%), generic for neoplasms (6%), Parkinson's disease (6%), pain (6%), and generic for psychiatry/psychology (5%), with all other indications representing less than 5% of trials using COAs (Figures and ). Stratifying oncology trials by primary purpose, treatment‐focused trials ( N = 4648 reporting COAs) most commonly used oncology‐specific quality of life measures (EORTC QLQ‐C30 in 33%), measures of overall health status (ECOG Performance Status in 18%), and other quality of life measures (EQ‐5D in 11%); all other instruments were reported in fewer than 5% of trials (Table ). Among trials with a primary purpose of supportive care ( N = 946 reporting COAs), the EORTC QLQ‐C30 (20%) and EQ‐5D (6%) were also among the most commonly used; additionally, a broader range of COAs were more prevalent, including HADS (16%), Functional Assessment of Chronic Illness Therapy (8%), Functional Assessment of Cancer Therapy‐General (8%), Short Form Health Survey (8%), BPI (7%), Functional Assessment of Cancer Therapy‐Breast Cancer (6%), Center for Epidemiologic Studies Depression Scale (5%), and Pittsburgh Sleep Quality Index (5%) (see Tables and for top COAs used for trials of other purpose and non‐oncology trials). 3.3 Trends of COA use over time We assessed the landscape of COA use over time using linear regression, revealing that both the number and proportion of oncology trials using COAs increased significantly over time ( R = 0.91, p < 0.001 and R = 0.98, p < 0.001, respectively), at rates of an additional 19 trials per year (95% Confidence Interval (CI): 16–22) and 0.81% of trials per year (95% CI: 0.75%–0.88%/year), respectively (Figure ). The increase was observed for all trial subsets stratified by primary purpose (treatment: +0.80% trials/year, 95% CI: 0.74–0.87%/year; diagnostic: +0.28% trials/year, 95% CI: 0.18–0.37%/year, supportive care: +2.40% trials/year, 95% CI: 1.95–2.84%/year; and cumulative all other: +0.75% trials/year, 95% CI: 0.61–0.89%/year). Moreover, the use of each of the five COA categories increased significantly over time. PROs had the greatest rate of increase with an additional 0.72% of trials using PROs per year (95% CI: 0.64–0.81%/year), followed by ClinROs (+0.20% trials/year, 95% CI: 0.18–0.22%/year), ObsROs (+0.05% trials/year, 95% CI: 0.04–0.06%/year), PerfOs (+0.04% trials/year, 95% CI: 0.03–0.05%/year), and composite measures (+0.04% trials/year, 95% CI: 0.03–0.05%/year) (Table and Figure ). The proportion of non‐oncology trials using COAs also increased significantly over time ( R = 0.98, p < 0.001), at a rate of 0.94% of trials per year (95% CI: 0.87%–1.02%/year). Stratification by trial purpose and COA type revealed similar significant increases in COA use compared to baseline across all levels (Figures and ; Table ). 3.4 COA use across landmark events We focused on five regulatory guidelines and interest group recommendations relevant to the use of COAs in clinical trials (Figure ) and assessed changes in COA use before versus after each event for the subset of applicable trials. We noted a significant increase in COA use around every event considered (Table ). To decouple the contribution of the individual events from the linear increase in COA use over the years, we considered regression models that included separate covariates for a given event, the initiation year, and the interaction between the two (Table ; Figure ). In all these models the association between year and increase in likelihood of COA use across the relevant oncology study subset remained significant. Moreover, there were additional significant increases in the likelihood of PRO use among trials with a site in the EU before versus after 2005 (OR > 10.00, p = 0.007) corresponding to the EMA guidance for HRQoL. A similar change was noted for industry‐sponsored trials with a site in the US before versus after 2009 (OR > 10.00, p = 0.003) corresponding to the FDA Guidance for Industry PRO Use, and all interventional oncology trials before versus after 2013 (OR > 10.00, p < 0.001) corresponding to the ISOQOL PRO Recommendations, when accounting for the underlying increase in COA use over time and its interaction with the respective event years. As in oncology trials, the increase in COA use rates among non‐oncology trials across each event considered, as well as the association between year and COA use among every trial subset considered were significant. Compared to trends among oncology trials, PRO use rates among non‐oncology trials in the EU did not increase after 2005 (OR > 10.00, p = 0.187). However, PRO use as a primary or secondary outcome increased significantly after 2018 (OR > 10.00, p < 0.001) corresponding to SPIRIT‐PRO extension (Table ; Figure ).
Trial characteristics and associations with COA use A total of 279,855 interventional trials with start dates between 1985 and 2020 (a date range chosen to encompass 95% of all interventional trials in the registry) were retrieved from the registry: 35,415 were determined to be oncology trials studying one or more neoplasms, and the remaining 244,440 were considered non‐oncology trials. The design features and the number of trials reporting them are summarized in Figure and Table . Briefly, the oncology trials reporting relevant information most commonly had a primary purpose of treatment (77%), studied a non‐FDA‐regulated intervention (59%), were completed (44%), phase 2 (48%), non‐blinded (84%), multi‐arm (54%), and non‐randomized (57%). Similarly, non‐oncology trials predominantly had a primary purpose of treatment (64%) and studied a non‐FDA‐regulated intervention (72%), but they had more balanced phase distribution (with phase 1, phase 2, and phase 3 at 22%, 25%, and 21%, respectively), and a higher degree of blinding (52% non‐blinded), multi‐arm design (77%), and randomization (70%) (Figure ; Table ). Eighteen percent of oncology trials reported COA use as determined by the search for individual instruments among trial design entries related to outcomes. Early phase trials had relatively lower rates of COA use (less than 15% each) compared to phase 2/phase 3 (25%), phase 3 (30%), and phase 4 (19%) trials. Nineteen percent of trials studying an FDA‐regulated intervention used a COA, versus 27% among trials studying non‐FDA‐regulated interventions. Eighteen percent of industry‐sponsored trials used COAs, versus 13% of NIH and US federal government‐sponsored, and 18% of all other sponsors. Trials focused on supportive care had the greatest rate of COA use (49%), followed by trials focused on treatment (17%), diagnostic (6%), with cumulative other trial categories having 15% COA use. Trials with blinding had greater rates of COA use (29%) than open‐label trials (16%). Multi‐arm trials had greater rates of COA use (25%) compared to single‐arm trials (12%). Twenty‐seven percent of randomized trials used COAs compared to 12% in non‐randomized trials (Table ). We next evaluated the interplay of the trial design elements in the implementation of COAs using multi‐variate logistic regression across trials reporting sufficient information ( N = 7339; Table ). The likelihood of COA use increased significantly with progressing clinical trial phases (OR = 1.30, p < 0.001). Trials studying FDA‐regulated interventions were significantly less likely to use COAs than those studying non‐FDA‐regulated interventions (OR = 0.81, p = 0.001) whereas trials using DMCs were more likely to use COAs than those without DMCs (OR = 1.26, p < 0.001). Compared to the baseline rate of COA use among supportive care trials, trials focused on treatment (OR = 0.34, p < 0.001), diagnostic (OR = 0.12, p < 0.001), and all other purposes (OR = 0.26, p < 0.001) were significantly less likely to use COAs. Randomized trials were significantly more likely to use COAs than non‐randomized studies (OR = 2.32, p < 0.001; Figure ). Primary sponsor class, blinding, and single‐ versus multi‐arm design were not significantly associated with COA use. Non‐oncology trials in the registry had higher rates of COA use overall (26%) and similar trends in COA use by phase, DMC use, and randomization, whereas unlike in oncology trials, treatment‐focused trials had similar rates of COA use compared to supportive care studies (OR = 1.16, p = 0.100). Additionally, compared to industry‐sponsored trials, trials with other non‐classified sponsors were less likely to use COAs than industry‐sponsored trials (OR = 0.89, p < 0.001), while trials with sponsorship from NIH or US federal government (versus industry; OR = 1.39, p < 0.001), blinding schemes (OR = 1.82, p < 0.001), and single‐arm designs (OR = 1.26, p < 0.001) were significantly more likely to use COAs. The FDA regulatory status of the trial intervention was not significantly associated with COA use among non‐oncology trials (Table ).
Characteristics of COA instruments used We identified 655 unique instruments used across the oncology trials, representing a diverse set of COAs and comprising of 24% of all instruments obtained from PROQOLID ( N = 2695). The top instruments in order of frequency were the European Organization for Research and Treatment of Cancer Quality of Life Questionnaire‐Core Questionnaire (EORTC QLQ‐C30) used by 29%, Eastern Cooperative Oncology Group (ECOG) Performance Status used by 14%, Euroqol EQ‐5D used by 10%, Hospital Anxiety and Depression Scale (HADS) used by 5%, and Brief Pain Inventory (BPI) used by 4% of trials using COAs (Table ). Fifty‐six percent of trials using COAs reported using one instrument (range: 1–13 instruments per trial). Of trials reporting details of trial outcomes ( N = 5795), COAs were used predominantly as a secondary trial outcome (84%), followed by primary trial outcome (23%), and other outcome (7%). Non‐oncology trials reported a broader range of instruments encompassing 1709 unique COAs; most frequently used instruments were the SF‐36 Health Survey (8%), EQ‐5D (6%), Patient Health Questionnaire (4%), and HADS (3%) (Table ). Similarly to oncology, 50% of non‐oncology trials using COAs reported using one instrument (range: 1–19). Compared to oncology, a greater proportion of non‐oncology trials reported COAs as a primary outcome (49%), with 75% and 7% of trials reporting COAs as secondary and other outcome. Among the COA categories, PROs were used in 84% of oncology trials reporting COA use ( N = 6331), followed by ClinROs in 27%, ObsROs in 6%, PerfOs in 4%, while 4% used composite measures. Oncology trials most often utilized COAs indicated as generic for neoplasms (63%), all conditions (32%), chronic diseases (10%), depression (8%), anxiety disorders (8%), pain (7%), breast neoplasms (6%), and lung neoplasms (5%); all other COA indication groups were represented in less than 5% of trials using COAs (Figures and ). Among non‐oncology trials reporting COA use ( N = 62,544), 78% used PROs, 43% used ClinROs, 13% used ObsROs, 11% used PerfOs, and 16% used composite measures. These instruments were indicated for all conditions (47%), depression (16%), anxiety disorders (13%), schizophrenia (9%), generic for mental disorders (9%), generic for neoplasms (6%), Parkinson's disease (6%), pain (6%), and generic for psychiatry/psychology (5%), with all other indications representing less than 5% of trials using COAs (Figures and ). Stratifying oncology trials by primary purpose, treatment‐focused trials ( N = 4648 reporting COAs) most commonly used oncology‐specific quality of life measures (EORTC QLQ‐C30 in 33%), measures of overall health status (ECOG Performance Status in 18%), and other quality of life measures (EQ‐5D in 11%); all other instruments were reported in fewer than 5% of trials (Table ). Among trials with a primary purpose of supportive care ( N = 946 reporting COAs), the EORTC QLQ‐C30 (20%) and EQ‐5D (6%) were also among the most commonly used; additionally, a broader range of COAs were more prevalent, including HADS (16%), Functional Assessment of Chronic Illness Therapy (8%), Functional Assessment of Cancer Therapy‐General (8%), Short Form Health Survey (8%), BPI (7%), Functional Assessment of Cancer Therapy‐Breast Cancer (6%), Center for Epidemiologic Studies Depression Scale (5%), and Pittsburgh Sleep Quality Index (5%) (see Tables and for top COAs used for trials of other purpose and non‐oncology trials).
Trends of COA use over time We assessed the landscape of COA use over time using linear regression, revealing that both the number and proportion of oncology trials using COAs increased significantly over time ( R = 0.91, p < 0.001 and R = 0.98, p < 0.001, respectively), at rates of an additional 19 trials per year (95% Confidence Interval (CI): 16–22) and 0.81% of trials per year (95% CI: 0.75%–0.88%/year), respectively (Figure ). The increase was observed for all trial subsets stratified by primary purpose (treatment: +0.80% trials/year, 95% CI: 0.74–0.87%/year; diagnostic: +0.28% trials/year, 95% CI: 0.18–0.37%/year, supportive care: +2.40% trials/year, 95% CI: 1.95–2.84%/year; and cumulative all other: +0.75% trials/year, 95% CI: 0.61–0.89%/year). Moreover, the use of each of the five COA categories increased significantly over time. PROs had the greatest rate of increase with an additional 0.72% of trials using PROs per year (95% CI: 0.64–0.81%/year), followed by ClinROs (+0.20% trials/year, 95% CI: 0.18–0.22%/year), ObsROs (+0.05% trials/year, 95% CI: 0.04–0.06%/year), PerfOs (+0.04% trials/year, 95% CI: 0.03–0.05%/year), and composite measures (+0.04% trials/year, 95% CI: 0.03–0.05%/year) (Table and Figure ). The proportion of non‐oncology trials using COAs also increased significantly over time ( R = 0.98, p < 0.001), at a rate of 0.94% of trials per year (95% CI: 0.87%–1.02%/year). Stratification by trial purpose and COA type revealed similar significant increases in COA use compared to baseline across all levels (Figures and ; Table ).
COA use across landmark events We focused on five regulatory guidelines and interest group recommendations relevant to the use of COAs in clinical trials (Figure ) and assessed changes in COA use before versus after each event for the subset of applicable trials. We noted a significant increase in COA use around every event considered (Table ). To decouple the contribution of the individual events from the linear increase in COA use over the years, we considered regression models that included separate covariates for a given event, the initiation year, and the interaction between the two (Table ; Figure ). In all these models the association between year and increase in likelihood of COA use across the relevant oncology study subset remained significant. Moreover, there were additional significant increases in the likelihood of PRO use among trials with a site in the EU before versus after 2005 (OR > 10.00, p = 0.007) corresponding to the EMA guidance for HRQoL. A similar change was noted for industry‐sponsored trials with a site in the US before versus after 2009 (OR > 10.00, p = 0.003) corresponding to the FDA Guidance for Industry PRO Use, and all interventional oncology trials before versus after 2013 (OR > 10.00, p < 0.001) corresponding to the ISOQOL PRO Recommendations, when accounting for the underlying increase in COA use over time and its interaction with the respective event years. As in oncology trials, the increase in COA use rates among non‐oncology trials across each event considered, as well as the association between year and COA use among every trial subset considered were significant. Compared to trends among oncology trials, PRO use rates among non‐oncology trials in the EU did not increase after 2005 (OR > 10.00, p = 0.187). However, PRO use as a primary or secondary outcome increased significantly after 2018 (OR > 10.00, p < 0.001) corresponding to SPIRIT‐PRO extension (Table ; Figure ).
DISCUSSION We assessed the landscape of clinical trials with a focus on understanding differences in COA incorporation distinct from traditional survival and response endpoints in oncology versus non‐oncology trials and the impact of regulatory and advocacy efforts to promote COA use in clinical research. COAs were used by 18% of oncology trials, lagging non‐oncology trials where use was reported in 26% of studies, albeit the latter rate represents use across diverse disease areas with previously reported heterogeneous landscape of COA use. Use of COAs by sponsor class was found to be similar for oncology trials, whereas for non‐oncology trials sponsorship of trials by NIH or other US federal agencies was associated with higher rates of COA use. Similar trends were found in associations between progressing phases, DMC use, randomization, and higher rates of COA use in both categories of trials. However, in contrast to non‐oncology trials, oncology trials focused on treatment had significantly lower rates of COA use compared to those focused on supportive care. The overall use of COAs increased over time across both oncology and non‐oncology trials, likely reflecting the cumulative impact of regulatory and advocacy efforts. Our assessment also found significant increases in COA use coinciding with focused guidelines, including the EMA guidance for HRQoL, the FDA Guidance for Industry PRO Use, and the ISOQOL PRO Recommendations but the magnitude of the impact could not be determined reliably, pointing to the need to assess additional factors underlying COA use trends. Lower COA use rates in early phase clinical trials reiterate previously reported concerns about phase 1 cancer trials, , and likely reflect challenges of incorporating COA instruments into small, non‐randomized trials often assessing heterogeneous patient populations or multiple tumor types. , The importance of COAs in patient‐focused toxicity assessment and the drive to include medical benefit outcomes in early phase cancer trials call for incorporation of meaningful COA outcomes to supplement or strengthen early phase studies as advocated by groups including the NIH Office of Patient‐Centered Outcomes Research. , Guidelines targeting early‐phase COA use or adaptation of existing strategies for randomized trials may be beneficial. , , The growing role of COAs to establish net clinical benefit of novel therapies and impact regulatory decision making , , underscores the need to address the low COA use rates in treatment‐focused oncology trials. However, an important caveat when interpreting these rates is that they exclude the use of traditional survival and tumor response endpoints in oncology trials, therefore underestimating the use of ClinROs as they pertain to those endpoints. In fact, we estimate that 54% of oncology trials use these traditional endpoints (66% of treatment‐focused trials and 9% of supportive care trials), versus only 8% of non‐oncology trials (Table ). The need to augment these traditional assessments for treatment‐focused trials is driven by the challenges of objectively assessing tumor response, resulting, for example, from inter‐tumor heterogeniety, clinical consequences of immune and targeted therapies that are distinct from traditional chemotherapy on which traditional radiographic criteria were developed, , and the need to establish that the assessed response translates into future benefit. Even among supportive care oncology trials, 51% did not use COAs despite emphasis on using COAs to justify the clinical benefit of non‐curative therapies. Moreover, as a recent assessment of the use of COAs in regulatory decision‐making discovered, patient experience as captured by COAs played a central role in FDA decisions when COAs were used as primary endpoints with uses in other endpoints primarily providing supporting information to contextualize rare or poorly‐characterized conditions. Our assessment found that among trials that reported COA use, only 23% of oncology trials versus 49% of non‐oncology trials used COAs as primary or co‐primary endpoints. Our results and prior reports , highlight the unmet recommendations for COA use in trials, which may contribute to the underrealized impact of COAs in drug development. PROs were found to be the most frequently used category of COAs in both categories of trials (84% for oncology and 78% non‐oncology trials reporting COA use) tracking Eastern Research Group's recent assessment that 84% of FDA reviews for approved applications (submitted between 2017 and 2020) that mention patient experience data, mention PROs. The most frequently reported COA instruments in oncology trials reflected recommended instruments including the EORTC QLQ‐C30, reported by nearly one third of all oncology trials using COAs. However, our analysis also revealed a broad, heterogeneous group of COA instruments used in oncology compared to non‐oncology, with a third of the trials using generic instruments not specific to a disease group. Generic measures of quality of life that are more translatable across disease groups may be paired with measures of cancer site‐ and therapy‐specific efficacy and toxicity profiles, given the growing development of cancer‐specific therapies. Standardization of these efforts should reflect the FDA Patient‐Focused Drug Development guidelines and may be further facilitated by core outcome sets with specific COA recommendations within cancer types, cancer‐site‐specific extensions to validated COAs as in the EORTC QLQ, and instrument libraries such as the Patient‐Reported Outcomes Measurement Information System from which targeted items can be selected from a validated pool of questionnaires. Our study was limited to the search and analysis of data from the ClinicalTrials.gov registry, which was chosen for being the largest and among the more user‐friendly clinical trial registries. However, ClinicalTrials.gov , similarly to other registries, is marked by low quality of data reporting, , often lacking structured or standardized information, and often missing information on exploratory endpoints whose reporting in the registry remains optional. Similarly, use of the free‐access version of PROQOLID limited our search to a large but non‐exhaustive subset of curated COAs and which, for example, excludes global assessment scales and omits custom‐made instruments from industry or individual investigators. While these limitations have likely led to COA use estimates that are too conservative, it is reasonable to posit that oncology and non‐oncology trials are similarly affected, and that the differences between the two classes of trials are not artifacts of the search methodology. Among the more complex limitations, we recognize that while the cataloged instruments were curated to be valid and reliable, we did not have information on how reliably and appropriately they were used in the trials or whether they captured the experience of diverse participants. Event‐oriented analyses were limited by subjective selection of events, challenges in disambiguating the effect of individual events, and potential bias from increasing reporting quality over time in the registry. However, our programmatic approach enabled a comprehensive, replicable, and efficient review of the data from the registry and utilization of information from additional databases. Future analyses of this kind may benefit from a greater push for COA reporting standardization not only at the existing level of protocol‐building and publications, , but also for trial registries. Recommendations may include reporting of outcome placement of COAs, reporting specific instrument names, use of common terminologies within ClinicalTrials.gov , which, to the best of our knowledge, do not exist for COAs yet, and reporting of post‐hoc analyses where COA data from trials are published, if at all. The resulting data transparency may also incentivize more intentional, a priori planning of COA outcomes in trial design, and guide future iterations of similar evaluations of COA use trends to gauge the field's progress. These steps may be necessary in the ongoing efforts to establish COAs as integral parts of clinical trial outcomes and advance the state of the practice.
Yeonju Kim: Conceptualization (equal); data curation (lead); formal analysis (lead); methodology (equal); visualization (lead); writing – original draft (equal); writing – review and editing (equal). Mark R Gilbert: Conceptualization (supporting); funding acquisition (equal); writing – review and editing (equal). Terri S. Armstrong: Conceptualization (supporting); funding acquisition (equal); writing – review and editing (equal). Orieta Celiku: Conceptualization (equal); data curation (supporting); formal analysis (supporting); methodology (equal); supervision (lead); visualization (supporting); writing – original draft (equal); writing – review and editing (equal).
Intramural Research Program of the National Cancer Institute, National Institutes of Health.
The authors declare no conflicts of interest.
Data S1. Supporting Information Click here for additional data file.
|
A new guiding strategy for treating the free margin of the leaflet in an aortic root remodelling procedure | 04183287-544b-4e2e-bef6-835b50d4dc00 | 11929233 | Cardiovascular System[mh] | Yacoub et al. were the first to propose the valve-sparing aortic root remodelling procedure to improve the postoperative quality of life in patients with aortic root aneurysms. Furthermore, Lansac et al. proposed the modified aortic root remodelling procedure involving an external aortic annular ring , which improved long-term efficacy, aiming at the defect that classical aortic root remodelling failed to stabilizing the basal ring. However, during the valve-sparing aortic root replacement procedure, the normalisation of the aortic root dimension, especially that of the sinotubular junction (STJ), may cause the prolapse of one or more aortic valve leaflets. This prolapse should be corrected by reducing the length of the free margin (FML) using techniques such as central plication of the leaflet. Therefore, the extent to which the length of the free margin of the leaflet should be reduced to restore the cusp height to normal is a critical challenge. Dr Schäfers guided the procedure using a self-designed leaflet-height-measuring calliper . In the valve-sparing reimplantation procedure, Dr Yu Cuntao proposed the following empirical formula to guide FML treatment: the target length of FML = the diameter of selected artificial graft for root reconstruction + 3 ~ 5 mm. In this study, we hypothesised that a method that manages the length of the free margin of the leaflet based on the diameter of the selected graft could be effectively used in a modified aortic root remodelling procedure with an external sub-valvular ring. We reviewed the results of patients with aortic root aneurysms treated at our institution to confirm the effectiveness of this strategy.
This study was a retrospective study. We analyzed data from January 2021 to May 2024 on consecutive patients with aortic root aneurysms who received modified aortic root remodelling with external sub-valvular ring. A total of 14 patients were admitted during this period, 2 patients with bicuspid aortic valves were excluded, and 12 patients with tri-leaflet configuration were included in the analysis. The mean age of the 12 patients (10 males and two females) was 42 ± 14 years. Marfan syndrome was diagnosed in three patients. Notably, all the patients had aortic valve insufficiency preoperatively. Table shows the degree of insufficiency and the status of aortic root dilation. Notably, all patients underwent full-phase aortic root scans with Siemens dual-source computed tomography (CT) (Siemens Healthcare GmbH, Forchheim, Germany) preoperatively, and the data were analysed using 3mensio software (Dutch PIE Medical Imaging Company, Maastricht, Netherlands) to calculate the diameter of each patient’s aortic annulus (basal ring) based on the area and circumference. Furthermore, based on the calculated aortic annulus size, the Lansac group’s experience was recommended for deciding the diameter of the Dacron graft to be reconstructed for the aortic root (26 mm, 28 mm, and 30 mm grafts for an annulus diameter of 25–27 mm, 28–30 mm, and 31–35 mm, respectively). The surgery was performed under conventional cardiac surgery anaesthesia and cardiopulmonary bypass support with femoral artery perfusion. Furthermore, the quality of the valve leaflets was assessed after dissecting the ascending aorta, and the geometric height (gH) of each leaflet was measured. Annulus size was confirmed using a St Jude mechanical valve size plug gauge (St. Jude Medical, Minneapolis, MN, USA) that approximated the preoperative CT estimate value, and the diameter of the selected Dacron graft was finally determined. A 7/0 prolene suture (PROLENE™ Polypropylene Suture, Ethicon LLC, Johnson & Johnson, New Brunswick, NJ, USA) was applied to the nodule of Arantius in the centre of each of the three leaflets, and the leaflets were gently pulled toward the centre of the aortic orifice without excessive tension. After straightening the free margin of the leaflet on both sides of the 7/0 prolene traction suture, the length of the free margin of each leaflet was measured in two sections using medical calliper forceps (Jinzhong Surgical Instrument Factory, Shanghai, China) with the suture point of 7/0 prolene traction suture as the boundary (Figs. and ). The sum of these two sections was known as the FML. The standard for evaluation was as follows: target length of the FML = the diameter of the selected artificial graft for root reconstruction + 3 ~ 5 mm. Leaflets with the FML lengths within this range were not processed. In cases exceeding this range, the free margin of the leaflet was centrally plicated with a 6/0 prolene suture (Fig. ) to reduce the FML to the target length of approximately ‘the diameter of selected artificial graft + 4 mm’. The left and right coronary artery openings were freed as a button, and the aortic root was dissected along the lateral aortic wall to the sub-valvular level of the basal ring. Furthermore, at the nadir of each sinus and the same level as the bottom of the interleaflet triangle, a 2/0 polyester braided suture (Ethicon 3/8 25-mm needle; Ethicon LLC, Johnson & Johnson, New Brunswick, NJ, USA) with pericardial patch was sewn, with a total of 5–6 stitches. The suture was perforated from the arterial lumen to the lateral side of the aortic wall and was used as the external ring fixation suture. Therefore, to protect the membranous septum, no suture was placed at the interleaflet triangular area of the Right-Non commissure junction. However, a 5/0 prolene suture with a patch was sewn onto the roof of the right atrium just against the outer wall of the aorta as an external ring fixation suture. The aortic sinus wall was trimmed, leaving a tissue margin of 3–5 mm from the fibrous annulus as the suture margin. One end of the selected Dacron graft was cut along three longitudinal incisions into three equal parts, and each part was then trimmed to a scalloped shape. The height of the longitudinal incision (the position of the apex of the new leaflet commissure junction after root reconstruction) was equal to the diameter of the selected Dacron graft. A 5/0 prolene suture was used to anastomose the three scalloped part to the remaining tissue margins of the three sinus walls. A circular ring approximately 4 mm in height, corresponding to three to four folds on the graft, was cut from the selected Dacron graft as the external sub-valvular ring. The external ring was then pushed into a sub-valvular position along the sutured graft. Pre-set fixation sutures were used to pass around the Dacron ring and knots. Two holes were made using an electric cautery in the suitable position on the Dacron graft and got trimmed. Then, they were anastomosed to the left and right coronary artery openings, respectively. Finally, the graft was anastomosed with the distal incision of the ascending aorta. The surgery was completed only when transoesophaegeal echocardiography confirmed that the aortic valve had no more than mild regurgitation or remnants of mild regurgitation of central reflux after the cardiopulmonary bypass support stopped. Otherwise, cardiopulmonary bypass should be performed again and the aortic valve should be further treated. By using echocardiography, aortic valve function was evaluated, the diameter of the newly formed aortic sinus was measured, and the effective height (eH) and coaptation length (CL) of the leaflets were also measured. The medical records of the patients were collected during hospitalisation, and echocardiographic data were collected at 3-, 6-month and 1 year after discharge. Furthermore, continuous variables were described in terms of mean ± standard deviation.
All 12 patients with aortic root aneurysms and valve insufficiency successfully underwent valve-sparing aortic root replacement without in-hospital mortality or complications. Furthermore, all patients underwent simultaneous ascending aortic replacement, one underwent partial aortic arch replacement, and another underwent tricuspid valve repair. Table shows the intraoperative measurements of the gH and FML valves. The aortic cross-clamping and cardiopulmonary bypass time were 222.1 ± 32.2 min and 262.3 ± 34.1 min, respectively. However, the average length of hospital stay postoperatively was 9 ± 2 days. For aortic root reconstruction, a straight tubular graft (Hemashield Platinum; Maquet, Rastatt, Germany) was used in nine patients. In the other three patients, a Valsalva graft (Terumo Aortic, Renfrewshire, United Kingdom) was used. The Valsalva graft was preferred, and straight tubular graft was used when a Valsalva graft not available. A 28-mm diameter graft was used in seven patients (one with a Valsalva graft); a 26-mm diameter graft, in four patients (two with a Valsalva graft); and a 30-mm tubular graft, in one patient. Furthermore, based on the standard for determining the length of the free margin of the leaflet in this study, intraoperatively, all three free margins of the leaflets were treated with central plication in four patients, the free margins of two leaflets needed to be treated in two patients, only one free margin of the leaflets needed to be treated in three patients, and no free margin needed to be treated in three patients. In 5 patients, 6 leaflets with FML equal to “the diameter of the selected graft + 6 mm” were also not treated. Because it has technically difficult to shorten the length of free margin by 2 mm through central plication (Table ). After aortic root reconstruction, the eH measured through intraoperative transoesophaegeal echocardiography was 8.9 ± 1.3 mm, and the CL of the leaflets was 5.3 ± 0.9 mm. However, regardless of whether Valsalva or tubular grafts were used, transoesophaegeal echocardiography revealed an obvious sinus morphology after aortic root remodelling. The diameter of the new sinus with the 28-mm Valsalva graft reconstruction was 35 mm (the maximum value in the whole group). However, the average diameter of the new sinus was 33.5 mm after reconstructing the root with six tubular grafts with diameters of 28 mm. In two other patients, the new sinuses reconstructed with a 26-mm Valsalva graft were 32 mm and 30 mm in diameter, respectively. Furthermore, in two patients, the diameter of the new sinus after root reconstruction with the 26-mm tubular graft was 29 mm. The diameter of the new sinus after root reconstruction with the 30-mm tubular graft was 33 mm. Postoperatively, aortic valve closure function improved in all patients (Table ). Notably, all patients completed a 6-month postoperative follow-up, and echocardiography showed that the degree of aortic regurgitation was not more than mild; the left ventricular dilatation at the end of diastole was reduced postoperatively compared with preoperative conditions (preoperatively: 58.1 ± 6.9 mm vs. 6 months postoperatively: 51.1 ± 4.1 mm). Six patients were followed up for > 1 year; the degree of aortic regurgitation was not more than mild.
Valve-sparing aortic root replacement for treating aortic root aneurysms comprises two main categories of techniques: remodelling and reimplantation. In the early days of valve-sparing aortic root replacement, it was thought that restoring normal configuration at the STJ would improve aortic valve closure in patients with aortic root aneurysms. However, in 2002, the Hannover group in Germany reported the treatment experience of reimplantation in the largest group of cases at that time . They found that if the aortic valve leaflets were not specifically treated intraoperatively, postoperative aortic valve coaptation would take on three forms: Type A coaptation point ≥ 2 mm above the lower prosthetic rim; Type B, coaptation close to the lower border of the woven Dacron graft; and Type C, coaptation ≥ 2 mm below the prosthesis. Notably, all patients with type C presented with an AI ≥ grade 2 at 1 year postoperatively . Furthermore, Schäfers et al. proposed the concept of “eH” , which suitably explained the clinical findings of the Hannover group. He found that the reduction of STJ diminishes eH during the remodelling procedure. A reduction in the STJ diameter by 3–4 mm consistently led to a height decrease of 2–3 mm . Therefore, to maintain proper closure of the aortic valve, eH must reach 8–9 mm or > 45% of a leaflet gH. To correct the prolapse caused by the root reconstruction procedure, it is necessary to use a technique of free margin shortening, such as leaflet central plication, to increase eH to the normal level. In a large series of valve-sparing root replacements, these techniques were reported to be necessary in 60–90% of patients . Thus, to accurately guide the management of the free margin, Dr Schäfers used a specially designed intraoperative eH measuring calliper. Through direct measurement with callipers, a leaflet with eH < 8–9 mm was centrally plicated, the length of the free margin was shortened, and the leaflet was raised to achieve the ideal eH target value. However, other researchers have used the proportional relationship between critical anatomical structures of the aortic root to guide the treatment of the free margin length of the leaflet. In 2021, Dr Yu Cuntao of Fuwai Hospital proposed an empirical formula to guide reimplantation procedure. His empirical formula was as follows: ‘the diameter of artificial graft selected for root reconstruction = (gH of the smallest leaflet of the three leaflets– 2 ~ 3 mm) *2’ and ‘the target length of FML = the diameter of selected artificial graft + 3 ~ 5 mm’. The patients in this study received modified aortic root remodelling; however, the guidance method of ‘the target length of FML = the diameter of selected artificial graft for root reconstruction + 3 ~ 5 mm’ was also directly used in the treatment of free margin length of the leaflet. Aortic valve closure function improved postoperatively, and eH and CL measured using oesophageal ultrasound also reached a satisfactory level. Anatomical studies on the aortic root have been conducted for many years [ – ]. Owing to different research times, measurement methods, specimen types, and simulated states of the aortic root, the proportional relationship between critical anatomical structures of the aortic root obtained by various studies is not completely consistent. According to De Kerchove et al. , the ratio of the FML to STJ diameter is approximately 1.29:1. Based on this relationship, FML is approximately equal to ‘the diameter of selected artificial graft + 8 mm’. However, De Kerchove et al. studied normal homogenous aortic root specimens. In this batch of specimens, the height of the leaflet commissure junction was 70% of the STJ diameter. When modified root remodelling was performed, the fixed height of the new leaflet commissure junction was equal to the diameter of the STJ (diameter of the artificial graft). In these two states, the angles formed between the free margins of the leaflets and the horizontal planes were different, which may account for the difference. Morishita et al. performed root remodelling based on the fact that the diameter of the STJ was approximately 90% of the FML of the leaflet. The size of the artificial graft used in the procedure was selected based on the FML of the leaflet . Based on the proportional relationship they provided, the FML was approximately equal to “the diameter of the selected artificial graft + 3 mm”, which is similar to our empirical value. Morishita et al. laid sutures along the free margins of the casts of the aortic leaflets and measured the lengths of the sutures . We sewed 7/0 prolene traction sutures on the nodule of Arantius, and the free margin of the leaflet on both sides of the traction suture was straightened without excessive tension. The two straightened free margins were measured directly using a calliper, and the sum of the two sections was the FML. During the measurement, the outer edges of the two measuring arms of the calliper were aligned with the traction line and inner wall of the leaflet commissure junction to ensure that the length of each section was the distance between the outer edges of the two measuring arms. The choice of graft size for aortic root reconstruction was based on the diameter of the aortic annulus (basal ring), according to the empirical Table provided by Lansac et al. In one patient, the aortic annulus diameter was 28 mm, and a 28-mm diameter artificial graft should be selected. However, the geometric height of the smallest leaflet measured intraoperatively was only 15 mm in this patient. Therefore, due to the concern regarding small leaflet areas, an artificial graft with a diameter of 26-mm was used based on the selection principle of graft calibre proposed by Dr Yu Cuntao. In the present study, the selection of the external sub-valvular aortic annuloplasty ring was inconsistent with that of Lansac et al., and a circular Dacron ring with the same calibre as the selected artificial graft was used as the external ring. According to Ismail et al. , the diameter of the aortic basal ring is reduced to 4 ~ 5 mm smaller than that of the external Dacron ring after extra-aortic annuloplasty ring implantation in the root remodelling procedure. Therefore, the ratio of STJ diameter to basal ring diameter after root reconstruction can reach the level of 1.1–1.2:1 required by De Paulis et al. and De Kerchove et al. , using a Dacron ring with the same diameter as the selected graft. Of course, this study is only a single-center pilot study, with a limited number of cases and a short follow-up period. A larger number of cases will need to be accumulated to further confirm the effectiveness of this guidance strategy.
The clinical experience of this group of patients showed that the method based on the diameter of the selected graft to manage the length of the free margin of the leaflet can be effectively used in the modified aortic root remodelling procedure with an external sub-valvular ring.
|
Surgeon experience, robotic perioperative outcomes, and complications in gynecology | 821999b7-a105-4ff9-ad3b-b1a00c8c96ba | 9720764 | Gynaecology[mh] | Robotic surgery is currently on the rise and has been widely applied all over the world . At present, there are more than 50,000 certified surgeons and more than 6,730 robots distributed across 69 countries , . Gynecology offers many opportunities for the development of innovative techniques due to the magnitude of benign and malignant diagnoses with surgical needs , . To date, 84 hospitals in Brazil have the robotic daVinci system and these institutions, as well as medical associations and regulatory entities, have been searching and working for the safety in the use of this technology and its association with good results for the patient . On this topic, we recently published an article on the importance of proctors in robotic gynecological procedures when surgeons are performing their first procedures, providing a learning curve without increasing risks to the patient . But the literature is scarce, and there is no global credentialing rules to assure the best and safest outcomes to patients , , . Barbash and Glied have recommended a minimum of 150–200 procedures for the surgeon to become skilled in the use of robotics, and Lenihan Jr described a shorter surgical time after 50 hysterectomies , . In contrast, Geller et al. showed consistent improvement in surgical time and robotic suture time after 20 procedures, including hysterectomies and sacrocolpopexies . To seek for possible differences after some experience with the technology, we arbitrarily divided surgeons into two groups: those with more than 20 robotic procedures (qualified surgeons) and those with 20 or less procedures (in-training surgeons). Our goal was to correlate perioperative outcomes such as surgical time, complications, and length of hospital stay, with the surgical diagnosis, procedure performed, and surgeon experience. This was a cross-sectional, retrospective survey, including an analysis of 632 medical records from patients who underwent robotic gynecological surgeries from January 2008 to December 2017. All procedures were performed at the Hospital Israelita Albert Einstein (HIAE), Brazil. Cases in which the main procedure was not gynecological were not considered in our analysis. This study was approved by the Institutional Research Ethics Committee (no. 38045414.7.0000.007). We collected information on the number of surgeries per year, patient’s age, body mass index (BMI), diagnoses, procedures performed, surgical time, length of hospital stay, and complications, such as intraoperative injuries, transfusions, and conversions. Comorbidities and previous surgeries were not evaluated. Surgeries included in the study were composed of those for benign and malignant pathologies. Surgical time was determined as the time for the whole surgical act, including the docking and undocking of the robot, but not the time for the anesthetic induction. Perioperative complications were identified, evaluated one by one, and divided into three major groups: perioperative injuries, transfusions, and conversions. Perioperative injuries that need extra care or treatment for correction were considered. The prescription of blood cell transfusion was individualized on a case-by-case basis, based on patient’s background, bleeding amount, and decision of the surgical team. Conversions of robot-assisted laparoscopy, when necessary, took place for conventional laparoscopy or open surgery. All the participating gynecological surgeons were board certified by the National Obstetrics and Gynecology Association (FEBRASGO) and the first six robotic procedures of each surgeon were always leaded by a proctor, with recognized experience in robotic surgery, according to the local protocol. We analyzed the results of 32 surgeons who were divided into two groups as follows: qualified (>20 previous robotic surgeries) and in-training surgeons (≤20 previous robotic surgeries). With these criteria, 9 qualified surgeons were responsible for 475 surgeries in the period (75% of the total), while the remaining 157 surgeries were performed by 23 in-training surgeons (25% of the total). We first considered those surgeons qualified in robotics who had done at least 20 procedures before the study period, considering that when reaching the preestablished threshold (20 cases) during the time of the study, the new surgical data had been changed to the qualified group from the next case on. Therefore, the initial 20 records were counted as in-training and from case 21 onward as qualified group. Sample size was defined as the total number of surgeries performed in an established period of time, i.e., 10 years. Comparisons among surgeon experience, length of hospital stay, and duration of the surgery were performed using linear mixed models applied to the log-transformed data. The results are presented as mean ratios and p-values. We used mixed logistic regression models, accounting for the dependence between different surgeries performed by the same surgeon. Results were expressed as odds ratios with 95% confidence intervals. All analyses were performed considering the level of significance as 5% (p=0.05). The analyzed sample consisted of 632 robot-assisted gynecological surgeries performed at HIAE from January 2008 to December 2017. In the study period, the number of surgeries consistently increased. The number of surgeries performed in 2017 was 50 times higher than that in 2008, which confirmed a significant growth in the adoption of this technology by the gynecological surgeons at the institution over the years. Patients’ age ranged from 19 to 84, with a mean of 39 years old, and an interquartile range (IQR) of 34–46 years. BMI ranged from 16 to 49, with a median of 23 kg/m . A total of 632 patients underwent robotic surgery, with 756 main diagnoses. In total, 1,929 procedures were performed, since several patients underwent two or more procedures at the same time. Of the 756 indications for surgery, we observed 381 cases of endometriosis, which corresponded to 50.4% of the total, being the most common diagnosis in our sample. The same predominance was observed in the frequency of procedures, with endometriosis treatment representing 19.7% of the total, as other procedures often compose those patients’ surgeries. Surgical time varied from 30 to 600 min, with a median of 205 min (IQR: 135–270 min), and the length of hospital stay ranged from <1 to 13 days, with a median of 2 days (IQR: 2–4 days). Patients who underwent surgery with qualified surgeons had a shorter surgical time (p<0.001) and shorter hospitalization (p=0.005), compared with patients operated on by in-training surgeons . In the present study, we aimed to measure the incidence of perioperative complications, such as perioperative injuries, transfusions, and conversions. In 632 patients, we had 20 complications, corresponding to 3.2% of the total, composed by 1.6% incidence of blood cell transfusions, 1.1% of perioperative injuries, and 0.5% of surgical conversion. We tested the correlation of adverse events with the most relevant clinical categories such as indication of surgery, type of procedure, and surgeon qualifying status. In the set of cases, it should be noted that the 3 of 632 conversions were necessary either because of heavy bleeding that made it difficult to continue the procedure or due to technical limitations to handle big fibroids robotically. The need for transfusion could be associated with patient’s previous hemoglobin level. Therefore, we considered perioperative complication as only the cases where excess of bleeding in the procedure resulted in the need for blood transfusion (10 cases). It is also worth highlighting the perioperative injuries found in this 10-year period: one ureter injury, one intraperitoneal hematoma, one vaginal cuff dehiscence, one rectovaginal fistula, one abdominal bleeding, one hemorrhagic shock, and one post-spinal anesthesia headache (7/632). presents the analysis, in a multiple model, for these categories: qualified surgeon (yes vs. no), indication of surgery (endometriosis or myoma vs. other), procedure (hysterectomy vs. myomectomy vs. other), and endometriosis treatment (yes vs. no). We observed that qualified surgeons had 64% less adverse outcomes than in-training surgeons (p=0.038). Regarding the indication of surgery, no relation with perioperative complications was noted (p=0.654). When myomectomy was compared to hysterectomy, we found a 4.78 times higher probability of perioperative complications in patients who underwent myomectomy than in those who underwent hysterectomy (p=0.029). Regarding the endometriosis treatment, no difference was noted (p=0.480). Considering perioperative complications, only blood cell transfusion had enough events to allow individualized statistical analysis. We used the simple mixed model and observed that in the qualified surgeons’ procedures the percentage of blood transfusion was 79% lower (p=0.018), as can be seen in . Concerning the indication of surgery and procedures performed, no difference in the frequency of blood transfusion was noted in our set of cases. Robotic surgery is a substantial breakthrough in the field of minimally invasive surgery, and this type of intervention has shown a large growth in the past decade , , . There are more than 6,730 robots over 69 countries , . In the United States, there has been a 23% increase rate per year in the number of procedures using robotics . In our hospital, in Sao Paulo / Brazil, this percentage growth was even higher in the past few years, with the number of robotic procedures increasing around 42% every year from 2008 to 2017. In this 10-year period, the most frequent diagnosis for surgery was endometriosis, representing 381 (50.4%) cases. Previous studies have shown benefits in the use of robotics for the endometriosis treatment, given its extension, depth, and complexity , . Advantages go beyond ergonomics, with more precise identification of lesions, as confirmed by Mosbrucker et al. showing a significantly higher rate of confirmed diagnosis in patients treated with robotic technology . The robotic-assisted approach seems to better preserve ovarian function in patients with bilateral ovarian endometrioma when compared to traditional laparoscopy, based on AMH levels before and after ovarian cystectomy . However, this growth in the adoption of the technology raises concerns regarding safety and surgical training , , . There is no consensus on the minimum number of procedures to achieve good surgical performance , . Barbash and Glied, who evaluated robotic surgery in different specialties, concluded that 150–200 procedures were the requirements for a surgeon to become proficient . Woelk et al. claimed that 91 cases are needed to reach feasibility in robotic hysterectomy, while Lenihan mentioned 50 as the number of procedures required to develop enough competence in the same procedure, according to suture and surgical time , . Geller et al. demonstrated improvements in surgical time for robotic hysterectomy after 20 cases, which is in accordance with the data presented on different procedures comparing surgeons with more than 20 robotic surgeries to those with 20 or less . Qualified surgeons (>20 robotic surgeries) had a 20% decreasing in surgical time and 15% in length of hospital stay. Regarding the occurrence of unfavorable outcomes, these qualified surgeons had 64% less complications than in-training ones (≤20 robotic surgeries). Comparison between robotic myomectomy and hysterectomy showed that patients who underwent myomectomy had risk of complications four times higher, therefore reflecting the complexity of myomectomy and enforcing the need of this procedure to be performed by a qualified surgeon with the most appropriate therapeutic tools. The number of perioperative complications was low, represented by 20 in total, making individual parameters difficult to be analyzed. No statistical significance was noted when comparing perioperative injuries and conversions due to the small number of occurrences, and a larger set of cases would be necessary to compare those numbers. Transfusion rate was the only adverse event with sufficient numbers to be separately analyzed and patients operated on by qualified surgeons had 79% less transfusions than those by in-training surgeons. Taking into account patient safety priorities, although considering that our study has limitations of a retrospective descriptive noncontrolled trial, we present relevant data for establishing policies regarding the minimum number of procedures that a surgeon should perform under supervision of a proctor to assure low risks of perioperative complications to the gynecological patient. By adding suitable training to safety protocols and development of technology, robotics can be a very useful tool in the operating room. It is worth to highlight that this article has limitations, as only one medical institution was evaluated. In addition, there was no standardization on surgical technique and extension of procedures when gynecological surgeries were compared. There was a wide range of surgical complexity as endometriosis treatment, myomectomy, and oncological procedures are usually more complicated than benign hysterectomies and adnexal procedures. Qualified surgeons with more than 20 robotic surgeries may have better perioperative results and less complications than in-training surgeons during their 20 initial robotic procedures. |
The evolving role of family physicians during the coronavirus disease 2019 crisis: An appreciative reflection | dc723290-8f70-4ff4-bbf6-1617155ec001 | 7343922 | Family Medicine[mh] | Ten family physicians and family medicine registrars are in their first quarterly training complex meeting for 2020. The setting is the Garden Route district in South Africa, a semi-rural area with a population of 620 000 people, of which 80% are dependent on six government hospitals for their health needs. The mood in the room is circumspect. Many questions hang in cyberspace. We can all see and hear each other, but there is a difference. For the first time since 2007 we are meeting via videoconferencing, all sitting in front of computers, spaced out over 120 km. Since our last physical meeting in 2019, the world has changed.
An invisible enemy, the novel coronavirus, has infected more than 1.9 million people globally, with a case fatality rate of 6.4%. The world is locked down; most businesses are closed; and people are keeping a social distance from each other, isolating themselves, washing hands for 20 s, not touching their faces, wearing masks and avoiding hospitals. The elderly of the society and those with chronic diseases are at the highest risk of dying. Five weeks of lockdown with almost no cars on the roads and no selling of alcohol have dramatically reduced mortality and morbidity from road accidents, murders and assaults. All routine outpatient clinics, elective surgery, outreach support and continuous medical education meetings have stopped. Hospitals and clinics are preparing for the coronavirus disease 2019 (COVID-19) crises, expecting to be overwhelmed. Traditional roles have changed; for example, orthopaedic medical officers and ophthalmology specialists are screening people for acute respiratory symptoms, some joining nurses and community health workers. Clinical rotations have changed; for example, ‘1 month in theatre’ has become ‘1 week in emergency centre, followed by 1 week in theatre’. Student assignments and examinations have been deferred or cancelled. There is a complete relook at the way we teach and learn. The healthcare system is being redesigned.
While ‘disruption’ is a negative term, it creates an environment for innovation, which is a positive term, well-described in technological industries. A disruption can be defined as an event in which a substantial share of agents belonging to a system is disrupted, typically requiring new skills and creating a pressure to change the value generation models of an organisation. The questions in the ‘room’ reflected a cautious concern for an uncertain future: How will this crisis affect the registrar training programme, including my learning plans? How will I complete my research, particularly the fieldwork? How will the exams work? How will we do outreach and support? How will I finalise my quality improvement cycle implemented at the clinic in order to complete my Leadership and Management module? Can this crisis be the opportunity to relook at how family physicians live out our expected roles? Particularly, moving from a hospital-centred to a community-centred environment, working in multi-professional teams, focusing on preventative health and lifestyle versus curative, technocentric medicine? While we need intensive care units and ventilators, do we not need to understand self-care, community health and public health in a better way? The clinical expert role has been well developed, while the roles of community advocate and collaborator have been less well developed.
The roles of the family physician in South Africa and the contribution to district health services have been well described. , Apart from clinical competence, the family physician needs to be a consultant, capacity-builder, clinical governance leader, champion of community-oriented primary care (COPC) and clinical trainer. , Researchers in a recent Harvard Business Review highlighted four traps that may prevent leaders in a crisis from balancing the management of the present with leading beyond the crisis. Family physicians, having a diverse and adaptive skill set, are regarded as expert generalists. Our roles make us leaders in our clinical and community contexts and, therefore, vulnerable to falling into these traps (see ). For example, over-centralising the response, in an attempt to control the crisis, can reduce capacity-building, missing an ideal opportunity to build capacity in others. In this example, our challenge is not to over-centralise, but remain ‘centred’ to our purpose and leadership role as family physicians. The solution is to adopt the cognitive aids depicted in (lower half). Despite the roles of the family physician being broad, inclusive and adaptive, the COVID-19 crisis imposes challenges to exercise these roles. There are restrictions on movement, social distancing, de-escalation of elective procedures and minimisation of patient–clinician contact. Time for regular activities is reduced as we are called upon to assist in many contexts and capacities, from training cleaners to advising joint operations committees (JOC). We therefore need to use every opportunity to exercise our roles. (left side) presents the ‘traditional’ model of how a family physician might exercise his or her roles during the COVID-19 crisis. In contrast, the right side presents an example of an activity that the family physician is called upon to assist with, in this case, optimising personal protective equipment (PPE) use. This shows how the various ‘roles’ can be exercised from this one activity. The idea is to facilitate family physicians in exercising roles through activity rather than through ourselves . This enables agility and responsiveness during the crisis but still enables the expert generalist to use this crisis as an opportunity to strengthen the health system.
Coronavirus disease 2019 has made us realise how important understanding human behaviour is, something we still have not figured out with the tuberculosis epidemic. It has created a pressure to change our focus from routine activities to thinking critically about real health needs of individuals in communities, as well as staff health and safety. We need to innovate to prevent infection, promote health (including mental health), work closer with many more sectors of society, including social services, community volunteers, business sector, manage resources better and cooperate with the private health sector. A ‘new normal’ is setting in. A shift is happening from reactive curative care towards a more promotive, preventative community-based approach. Perhaps this is the time to shift from just measuring the leading causes of death to also include the leading causes of life. Health workers as well as people in communities need a sense of agency (power to do), to have hope (believe in solutions), to connect more (with self, others and environment) and to appreciate intergenerativity (realising that everyone is part of a continuum of those that went before and those that are coming after). Family physicians, with their skills set of clinical competence across all clinical disciplines, leadership skills of communication and collaboration within teams and viewing patients as part of families in communities, are ideally placed to respond appropriately to this crisis.
The roles of the family physician have been disrupted by an invisible enemy. Preventative care in a patient-centred, community-oriented approach is of critical importance, now more than ever. The governance of resources, such as masks, gloves and medicines, has global implications. The isolation, uncertainty and fears of patients and colleagues necessitate diligent self-care, mental healthcare and a renewed focus on caring for each other. The skills to reflect, communicate and share within our teams are being sharpened like never before.
|
Metagenomic Characterization of the | a4154ec4-7c6c-4c47-848e-92913640135c | 11942469 | Microbiology[mh] | Maerua crassifolia , commonly referred to as Sarh, is a perennial, drought-resistant plant belonging to the Capparaceae family . Its leaves are rich in minerals such as calcium and iron, and amino acids such as phenylalanine, tyrosine, linoleic and alpha-linolenic acids, which serve as a valuable food source in low-income areas . Traditionally, this plant has been used to treat various ailments due to its medicinal properties [ , , , , , , ]. M. crassifolia is now considered endangered, prompting conservation efforts. A detailed micropropagation protocol has been developed for the in vitro culture of the species , but further strategies are still needed to support sustainable agricultural practices within traditional communities. Drought, one of the most important environmental challenges, significantly limits plant growth and poses a major obstacle to achieving sustainability . Plant resistance to drought and growth can be enhanced by the rhizobiome . Among these rhizosphere-dwelling microorganisms are plant growth-promoting microorganisms (PGPMs), which have been demonstrated to improve plant stress tolerance and facilitate growth . The rhizosphere serves as a notable interface facilitating the exchange of resources between plants and microorganisms [ , , ]. This interaction is primarily initiated by root exudation, the main communication pathway between plants and soil microorganisms. After this initial chemical signaling, a complex sequence of exchanges involving metabolites, molecular signals and nutrients takes place. The symbiotic associations observed in the rhizosphere stem from this intricate network of below-ground interactions . Using bacterial and fungal inocula in combination with organic amendments is a viable way to incorporate nutrient management methods into soils that have deteriorated . These inoculums may improve soil fertility by moving, relocating, mineralizing and using important nutrients including phosphorus (P), potassium (K) and iron (Fe). They also help with the buildup of organic matter and the fixation of nitrogen (N) from the atmosphere [ , , ]. According to previous research, arbuscular mycorrhizal fungi (AMF) and nitrogen-fixing bacteria are responsible for 5–20% of the total annual nitrogen requirement in grasslands and savannahs. In temperate and boreal forests, AMF give up to 80% of the needed N, whereas bacteria and fungi combined assist 75% of total P uptake by plants . The key methods by which bacteria and fungi promote nutrient bioavailability are N fixation and the mobilization of P, K and Fe via the formation of organic acids and siderophores. In addition, these microorganisms release organo-polysaccharides and proteins, including glomalin, mucilages and hydrophobins. These substances are important for enhancing the stability of soil aggregates [ , , ]. The metagenome, the genetic material within an environment , is currently being discovered using metagenomics and metataxonomics as primary methodologies. Metataxonomics, or amplicon sequencing (AS), focuses on specific genomic regions such as 16S and 18S ribosomal DNA (rDNA) to create taxonomic profiles limited to a single domain of microorganisms. AS can only characterize microbes down to the genus level, whereas metagenomics, or whole genome shotgun sequencing (WGS), provides a broader approach to characterize functional and taxonomic profiles across all microbial domains, down to the species or strain level. WGS can excel in detecting lower abundance microbes compared to AS but requires a higher cost and more complex bioinformatic analysis [ , , ]. Furthermore, the fact that 98% of bacteria are unculturable suggests that these methodologies are powerful for uncovering previously hidden microbial diversity, despite their limitations and resource demands . The current study aims to utilize WGS to uncover distinctive features of bacteria and fungi in the rhizosphere microbiomes associated with the plant M. crassifolia , as well as those present in neighboring bulk soil microbiomes.
2.1. Samples Collection The soil was sampled in the morning during October 2023 from the M. crassifolia plant located in the Makkah district of southeastern Saudi Arabia, at coordinates 21°30′23.8″ N latitude and 40°01′35.1″ E longitude at an altitude of 439 m above sea level ( ). Eight samples, each weighing 5 g, were collected, four of them targeting the upper 0–10 cm layer of bulk soil and four from the rhizosphere at depths between 18 and 25 cm . Immediately after collection, the soil samples were rapidly frozen in liquid nitrogen and subsequently stored on dry ice, then transferred at −20 °C until later use. A soil–water slurry was prepared at a 1:1 dry-weight ratio to measure the soil’s pH. 2.2. DNA Extraction DNA extraction was conducted using the PowerMax ® Soil DNA Isolation Kit (MO BIO Laboratories, Carlsbad, CA, USA) following the guidelines provided by the manufacturer. Next, the DNA integrity was assessed through electrophoresis on a 0.8% agarose gel, while the purity of the DNA was determined using a NanoDrop Spectrophotometer (Thermo Fisher Scientific, Waltham, MA, USA) as a pre-initial step in the Quality Control (QC) procedures. 2.3. Library Preparation and DNA Sequencing Library preparation and DNA sequencing tasks were conducted at Novogene Co., Ltd. in Singapore. All procedures utilize one microgram of sample DNA as input. The initial QC measures included quantification, integrity assessment and purity analysis of the DNA samples using the Agilent 5400 system (Santa Clara, CA, USA). Following this step, DNA fragmentation was achieved via sonication. DNA fragments then underwent end-polished, phosphorylation and A-tailing reactions. Subsequently, the fragments were ligated with Illumina adapters and amplified using polymerase chain reaction (PCR). Library preparation was conducted using the Illumina Ultra DNA Library Prep kit (NEB, Ipswich, MA, USA), followed by PCR primer removal. Size selection was performed to isolate DNA fragments of 350 base pairs (bp). Library quantification was conducted using Qubit and real-time PCR. Library size distribution was assessed using a bioanalyzer as a secondary QC measure. Finally, the pooled libraries were sequenced on the Illumina Novaseq TM platform 6000 (San Diego, CA, USA). 2.4. Bioinformatics Analysis Clean data reads extraction was performed using Readfq software ( https://github.com/cjfields/readfq , 20 January 2025). Reads were subjected to a stringent QC process involving three stages. First, reads exceeding a 10 bp threshold with N bases were removed. Second, reads that had adapter overlaps greater than 15 bp were discarded. Third, reads that contained a proportion of 40 bp of low-quality bases (Phred score ≤ 38) were filtered out. Then, the reads were aligned to the plant reference genome of M. crassifolia using Bowtie2 software (version 2.2.4) to identify and eliminate any residual plant-derived sequences [ , , ]. Genome assembly was conducted using MEGAHIT software (version 1.0.4) [ , , ]. To capture genomic information from low-abundance microbial taxa potentially underrepresented in the initial assembly, all unused scaffolds were combined into a composite sample designated NOVO_MIX. This composite sample was then subjected to an independent de novo assembly. All scaffolds containing “N” base regions were excluded from the final scaffold following assembly. This fourth QC step involved fragmentation of the scaffolds at the site of each “N” base. Finally, the clean sequencing data were aligned to the reference genome using Soap software version 2.21 to generate usable scaftig sequences . Open reading frames (ORFs) indicative of potential genes were identified within scaftig sequences ≥ 500 bp in length using MetaGeneMark software version 2.10 [ , , , , ]. Identified ORFs shorter than 100 nucleotides were excluded [ , , , , ]. To establish a non-redundant catalog of identified genes from the assembled metagenome, redundant sequences with coverage of ≥90% and a sequence identity of ≥95% were eliminated using the Cluster Database at High Identity with Tolerance (CD-HIT) software (version 4.5.8) [ , , , , ]. The abundance of genes in the non-redundant genes catalog (nrGC) was then determined. A computational pipeline developed by Novogene Corporation facilitated the categorization of these data at the gene level. As a fifth QC measure, longer metagenomic segments, termed “contigs”, were subjected to taxonomic classification that depended on consistent annotation. Contigs were classified as putatively hybrid, and thus erroneous or chimeric, if less than 95% of their constituent reads could be assigned to a single reported species. This singularity threshold allowed for identifying and excluding contigs with ambiguous taxonomic classification. The clean sequencing data were then aligned to the nrGC using Bowtie2 software. This alignment enabled the quantification of mapped reads for each unique gene, providing a measure of the relative abundance (RA) of each gene within the samples. Genes with ≤2 mapped reads, indicating very low abundance, were removed . The remaining gene set constituted the final unigene catalog. Gene abundance within this final catalog was quantified based on the length of the corresponding gene sequence and the number of mapped reads [ , , ]. Several downstream bioinformatic analyses were performed. These analyses included the generation of basic descriptive statistics, core–pan gene analyses and Venn diagram visualizations. Unigene sequences were compared to the National Center for Biotechnology Information (NCBI) non-redundant (NR) protein database using DIAMOND software (version 2.1.8), which implements a BLASTP (NCBI) search algorithm . The taxonomic annotation was determined using the Lowest Common Ancestor (LCA) method to account for sequences with multiple alignments. The LCA algorithm traverses the NCBI taxonomic tree for each alignment to identify the most specific common ancestor taxon. This annotation process was performed using MEGAN software (version 4). Gene abundance tables were generated by combining sequence read counts with LCA annotation data at different taxonomic levels. Taxon abundance within the sample was determined based on the cumulative read counts for all genes associated with that taxon [ , , , ]. The visualization of data was conducted utilizing the Matplotlib package (version 3.10.0) within a Python environment (version 3.11.11.). 2.5. Statistical Analysis Beta diversity was calculated to examine taxonomic composition based on RA at different taxonomic levels. General trends in Beta diversity and clustering of samples within the studied groups were assessed by principal component analysis (PCA) of the taxonomic abundance data using the ade4 package in the R statistical computing environment. To statistically evaluate significant differences in microbial community profiles between groups relative to within groups, an analysis of similarities (ANOSIM) was performed using the vegan package in R statistical computing environment. In addition, to identify specific taxa that showed significant differential abundance between the designated sample groups, a differential abundance analysis was conducted using the Linear Discriminant Analysis (LDA) Effect Size (LEfSe) with a default LDA threshold of 4 .
The soil was sampled in the morning during October 2023 from the M. crassifolia plant located in the Makkah district of southeastern Saudi Arabia, at coordinates 21°30′23.8″ N latitude and 40°01′35.1″ E longitude at an altitude of 439 m above sea level ( ). Eight samples, each weighing 5 g, were collected, four of them targeting the upper 0–10 cm layer of bulk soil and four from the rhizosphere at depths between 18 and 25 cm . Immediately after collection, the soil samples were rapidly frozen in liquid nitrogen and subsequently stored on dry ice, then transferred at −20 °C until later use. A soil–water slurry was prepared at a 1:1 dry-weight ratio to measure the soil’s pH.
DNA extraction was conducted using the PowerMax ® Soil DNA Isolation Kit (MO BIO Laboratories, Carlsbad, CA, USA) following the guidelines provided by the manufacturer. Next, the DNA integrity was assessed through electrophoresis on a 0.8% agarose gel, while the purity of the DNA was determined using a NanoDrop Spectrophotometer (Thermo Fisher Scientific, Waltham, MA, USA) as a pre-initial step in the Quality Control (QC) procedures.
Library preparation and DNA sequencing tasks were conducted at Novogene Co., Ltd. in Singapore. All procedures utilize one microgram of sample DNA as input. The initial QC measures included quantification, integrity assessment and purity analysis of the DNA samples using the Agilent 5400 system (Santa Clara, CA, USA). Following this step, DNA fragmentation was achieved via sonication. DNA fragments then underwent end-polished, phosphorylation and A-tailing reactions. Subsequently, the fragments were ligated with Illumina adapters and amplified using polymerase chain reaction (PCR). Library preparation was conducted using the Illumina Ultra DNA Library Prep kit (NEB, Ipswich, MA, USA), followed by PCR primer removal. Size selection was performed to isolate DNA fragments of 350 base pairs (bp). Library quantification was conducted using Qubit and real-time PCR. Library size distribution was assessed using a bioanalyzer as a secondary QC measure. Finally, the pooled libraries were sequenced on the Illumina Novaseq TM platform 6000 (San Diego, CA, USA).
Clean data reads extraction was performed using Readfq software ( https://github.com/cjfields/readfq , 20 January 2025). Reads were subjected to a stringent QC process involving three stages. First, reads exceeding a 10 bp threshold with N bases were removed. Second, reads that had adapter overlaps greater than 15 bp were discarded. Third, reads that contained a proportion of 40 bp of low-quality bases (Phred score ≤ 38) were filtered out. Then, the reads were aligned to the plant reference genome of M. crassifolia using Bowtie2 software (version 2.2.4) to identify and eliminate any residual plant-derived sequences [ , , ]. Genome assembly was conducted using MEGAHIT software (version 1.0.4) [ , , ]. To capture genomic information from low-abundance microbial taxa potentially underrepresented in the initial assembly, all unused scaffolds were combined into a composite sample designated NOVO_MIX. This composite sample was then subjected to an independent de novo assembly. All scaffolds containing “N” base regions were excluded from the final scaffold following assembly. This fourth QC step involved fragmentation of the scaffolds at the site of each “N” base. Finally, the clean sequencing data were aligned to the reference genome using Soap software version 2.21 to generate usable scaftig sequences . Open reading frames (ORFs) indicative of potential genes were identified within scaftig sequences ≥ 500 bp in length using MetaGeneMark software version 2.10 [ , , , , ]. Identified ORFs shorter than 100 nucleotides were excluded [ , , , , ]. To establish a non-redundant catalog of identified genes from the assembled metagenome, redundant sequences with coverage of ≥90% and a sequence identity of ≥95% were eliminated using the Cluster Database at High Identity with Tolerance (CD-HIT) software (version 4.5.8) [ , , , , ]. The abundance of genes in the non-redundant genes catalog (nrGC) was then determined. A computational pipeline developed by Novogene Corporation facilitated the categorization of these data at the gene level. As a fifth QC measure, longer metagenomic segments, termed “contigs”, were subjected to taxonomic classification that depended on consistent annotation. Contigs were classified as putatively hybrid, and thus erroneous or chimeric, if less than 95% of their constituent reads could be assigned to a single reported species. This singularity threshold allowed for identifying and excluding contigs with ambiguous taxonomic classification. The clean sequencing data were then aligned to the nrGC using Bowtie2 software. This alignment enabled the quantification of mapped reads for each unique gene, providing a measure of the relative abundance (RA) of each gene within the samples. Genes with ≤2 mapped reads, indicating very low abundance, were removed . The remaining gene set constituted the final unigene catalog. Gene abundance within this final catalog was quantified based on the length of the corresponding gene sequence and the number of mapped reads [ , , ]. Several downstream bioinformatic analyses were performed. These analyses included the generation of basic descriptive statistics, core–pan gene analyses and Venn diagram visualizations. Unigene sequences were compared to the National Center for Biotechnology Information (NCBI) non-redundant (NR) protein database using DIAMOND software (version 2.1.8), which implements a BLASTP (NCBI) search algorithm . The taxonomic annotation was determined using the Lowest Common Ancestor (LCA) method to account for sequences with multiple alignments. The LCA algorithm traverses the NCBI taxonomic tree for each alignment to identify the most specific common ancestor taxon. This annotation process was performed using MEGAN software (version 4). Gene abundance tables were generated by combining sequence read counts with LCA annotation data at different taxonomic levels. Taxon abundance within the sample was determined based on the cumulative read counts for all genes associated with that taxon [ , , , ]. The visualization of data was conducted utilizing the Matplotlib package (version 3.10.0) within a Python environment (version 3.11.11.).
Beta diversity was calculated to examine taxonomic composition based on RA at different taxonomic levels. General trends in Beta diversity and clustering of samples within the studied groups were assessed by principal component analysis (PCA) of the taxonomic abundance data using the ade4 package in the R statistical computing environment. To statistically evaluate significant differences in microbial community profiles between groups relative to within groups, an analysis of similarities (ANOSIM) was performed using the vegan package in R statistical computing environment. In addition, to identify specific taxa that showed significant differential abundance between the designated sample groups, a differential abundance analysis was conducted using the Linear Discriminant Analysis (LDA) Effect Size (LEfSe) with a default LDA threshold of 4 .
WGS was utilized to analyze the microbiome of the rhizosphere and bulk soil of M. crassifolia plants, allowing the detection of bacterial and fungal structures in each soil type. The soil samples were collected at a temperature of 34 °C and a pH of 7.6. According to climatological data for the Makkah region from the Climate Change Knowledge Portal, the average annual rainfall is 95.71 mm. A previous study in this region reported soil organic matter (SOM) concentrations ranging from 0.01 to 4.5 mg kg −1 , with higher levels observed in the southern areas, including the present study area . The microbiome groups were categorized according to soil type: group ’A’ represents the rhizosphere microbiome, while group ’B’ represents the bulk soil microbiome. In addition, ’R’ stands for individual samples from the rhizosphere, and ’S’ for individual samples from the bulk soil. Detailed statistics on raw metagenomic data can be found in and clean data in , while a Venn diagram is shown in ; both are available in the . The distribution of ORF lengths (measured in nucleotides (nt)) and scaffold lengths (for sequences of 500 bp or longer, also known as scaftigs) are displayed in . Core–pan rarefaction curves, which depict the identification of shared and distinct ORFs in the metagenomic datasets as more samples are included, are provided in . 3.1. Analysis of Similarities (ANOSIM) ANOSIM was employed to assess compositional differences in microbial communities between groups A and B. While the ANOSIM analysis revealed slight differences between groups A and B at the phylum level (R = −0.146), this difference was not statistically significant ( p = 0.853). A similar result was observed at the genus level. At the species level, the ANOSIM test indicated no significant difference in community composition between the two groups (R = 0, p = 0.447) ( A–C). 3.2. Principal Component Analyses (PCA) PCA revealed differences in microbiome composition between rhizosphere and bulk soil samples at the phylum, genus and species levels. At the phylum level, rhizosphere microbiomes clustered primarily along the positive side of the PC2 axis, while bulk soil microbiomes were positioned primarily along the positive side of the PC1 axis. At the genus level, rhizosphere microbiomes clustered toward the negative side of the PC1 axis, while bulk soil microbiomes clustered toward the positive side of the PC2 axis. A similar trend was observed at the species level, with rhizosphere microbiomes clustering toward the negative side of the PC1 axis, and bulk soil microbiomes exhibiting the opposite pattern. The distinct clustering of rhizosphere and bulk soil samples across the phylum, genus and species levels indicates significant differences in microbial composition between soil types ( A–C). 3.3. Microbial Annotation As mentioned above, gene predictions and mapping were performed for the assembled scaftigs ( ) and ORFs were identified ( ). Non-redundant genes (nRGs) were then clustered, leading to the development of catalogs of the original nrGCs. Gene abundances were then classified at three different taxonomic levels, including phylum, genus and species. This was followed by an analysis of RA. Archaea and viruses were excluded and are shown as ’other’ with an unidentified taxon. However, the annotation indicates that detailed information is not available for most taxa in the metagenomes, as evidenced by the high gene abundance of the unknown taxa. Furthermore, much of this unidentified genetic information corresponds to the high proportion of low-abundance, unculturable microorganisms. The analysis revealed a significantly higher abundance of bacteria compared to eukaryotes (domain that encompasses fungi) ( ). The top ten most predominant taxa across soil types (e.g., rhizosphere and bulk soil) were selected for further analysis at the three taxonomic levels (phylum, genus and species). Based on the nRG, the phyla Actinobacteria, Pseudomonadota and Acidobacteria were identified as the most prevalent among bacteria, while Mucoromycota, Ascomycota and Basidiomycota were the most prevalent among fungi ( ). At the genus level Rubrobacter , Solirubrobacter and Streptomyces were revealed as the dominant bacterial taxa, while Rhizophagus , Funneliformis and Hyaloraphidium were the most abundant fungal genera ( ). At the species level, unclassified members of the Actinobacteria and Acidobacteria, along with an uncultured Rubrobacteraceae bacterium, were the most abundant bacterial taxa. Within the fungi, Rhizophagus irregularis , Rhizophagus clarus and Hyaloraphidium curvatum were the dominant species ( ). shows a comparative analysis of the samples without NOVO_MIX. To investigate the RA of microbial taxa in different soil types (e.g., rhizosphere and bulk soil), the taxonomic assignment of nRGs in metagenomes was analyzed. This analysis was performed at the phylum, genus and species levels for bacteria ( , and and ) and Eukaryota ( , and and ). As some of the taxa were not detected in the soil type-specific analyses, their occurrence is attributed to the NOVO_MIX group. To allow for contrasts in the RA of fungi and bacteria between the rhizosphere and bulk soil, an initial analysis was performed to assess the overall abundance of microorganisms in the combined dataset that includes the NOVO_MIX group. The NOVO_MIX category consists of short gene sequences collected from all samples; therefore, its abundance cannot be assigned to a specific soil type. The present approach establishes a basic understanding of the microbial community composition before investigating specific variations between the rhizosphere and bulk soil. Although slight differences in RA between soil types were detected, the next discussion will focus on relative rather than absolute abundance. 3.4. Linear Discriminant Analysis (LDA) Effect Size (LEfSe) LEfSe software (version 1.1.2.) was employed to identify biomarkers exhibiting significant variations in abundance between A and B groups, enabling the detection of differential species abundance between groups A and B. Within the bulk soil microbiome, one taxonomic biomarker exhibited statistically significant enrichment ( p ≤ 0.05, LDA score (log<sub>10</sub>) > 4.0). This biomarker, identified at family and species levels, corresponds to an uncultured bacterium within the order Acidimicrobiales (phylum Acidobacteria) ( A,B).
ANOSIM was employed to assess compositional differences in microbial communities between groups A and B. While the ANOSIM analysis revealed slight differences between groups A and B at the phylum level (R = −0.146), this difference was not statistically significant ( p = 0.853). A similar result was observed at the genus level. At the species level, the ANOSIM test indicated no significant difference in community composition between the two groups (R = 0, p = 0.447) ( A–C).
PCA revealed differences in microbiome composition between rhizosphere and bulk soil samples at the phylum, genus and species levels. At the phylum level, rhizosphere microbiomes clustered primarily along the positive side of the PC2 axis, while bulk soil microbiomes were positioned primarily along the positive side of the PC1 axis. At the genus level, rhizosphere microbiomes clustered toward the negative side of the PC1 axis, while bulk soil microbiomes clustered toward the positive side of the PC2 axis. A similar trend was observed at the species level, with rhizosphere microbiomes clustering toward the negative side of the PC1 axis, and bulk soil microbiomes exhibiting the opposite pattern. The distinct clustering of rhizosphere and bulk soil samples across the phylum, genus and species levels indicates significant differences in microbial composition between soil types ( A–C).
As mentioned above, gene predictions and mapping were performed for the assembled scaftigs ( ) and ORFs were identified ( ). Non-redundant genes (nRGs) were then clustered, leading to the development of catalogs of the original nrGCs. Gene abundances were then classified at three different taxonomic levels, including phylum, genus and species. This was followed by an analysis of RA. Archaea and viruses were excluded and are shown as ’other’ with an unidentified taxon. However, the annotation indicates that detailed information is not available for most taxa in the metagenomes, as evidenced by the high gene abundance of the unknown taxa. Furthermore, much of this unidentified genetic information corresponds to the high proportion of low-abundance, unculturable microorganisms. The analysis revealed a significantly higher abundance of bacteria compared to eukaryotes (domain that encompasses fungi) ( ). The top ten most predominant taxa across soil types (e.g., rhizosphere and bulk soil) were selected for further analysis at the three taxonomic levels (phylum, genus and species). Based on the nRG, the phyla Actinobacteria, Pseudomonadota and Acidobacteria were identified as the most prevalent among bacteria, while Mucoromycota, Ascomycota and Basidiomycota were the most prevalent among fungi ( ). At the genus level Rubrobacter , Solirubrobacter and Streptomyces were revealed as the dominant bacterial taxa, while Rhizophagus , Funneliformis and Hyaloraphidium were the most abundant fungal genera ( ). At the species level, unclassified members of the Actinobacteria and Acidobacteria, along with an uncultured Rubrobacteraceae bacterium, were the most abundant bacterial taxa. Within the fungi, Rhizophagus irregularis , Rhizophagus clarus and Hyaloraphidium curvatum were the dominant species ( ). shows a comparative analysis of the samples without NOVO_MIX. To investigate the RA of microbial taxa in different soil types (e.g., rhizosphere and bulk soil), the taxonomic assignment of nRGs in metagenomes was analyzed. This analysis was performed at the phylum, genus and species levels for bacteria ( , and and ) and Eukaryota ( , and and ). As some of the taxa were not detected in the soil type-specific analyses, their occurrence is attributed to the NOVO_MIX group. To allow for contrasts in the RA of fungi and bacteria between the rhizosphere and bulk soil, an initial analysis was performed to assess the overall abundance of microorganisms in the combined dataset that includes the NOVO_MIX group. The NOVO_MIX category consists of short gene sequences collected from all samples; therefore, its abundance cannot be assigned to a specific soil type. The present approach establishes a basic understanding of the microbial community composition before investigating specific variations between the rhizosphere and bulk soil. Although slight differences in RA between soil types were detected, the next discussion will focus on relative rather than absolute abundance.
LEfSe software (version 1.1.2.) was employed to identify biomarkers exhibiting significant variations in abundance between A and B groups, enabling the detection of differential species abundance between groups A and B. Within the bulk soil microbiome, one taxonomic biomarker exhibited statistically significant enrichment ( p ≤ 0.05, LDA score (log<sub>10</sub>) > 4.0). This biomarker, identified at family and species levels, corresponds to an uncultured bacterium within the order Acidimicrobiales (phylum Acidobacteria) ( A,B).
Fungi are ecologically important, but their diversity, especially in the dry and semi-arid Arabian Peninsula, has been little investigated compared to bacteria . This work uses metagenomics WGS to define the top 10 bacterial and fungal structures of M. crassifolia’s rhizosphere and bulk soil at the phylum, genus and species levels. Due to its superior taxonomic resolution, WGS may find bacterial and fungal species SA missed . This precision is significant because soil microbial populations are susceptible to complex biotic and abiotic variables like drought . AS is useful for processing large datasets cheaply and efficiently, but it fails to capture microbial community diversity. Soil environments’ great microbial diversity and intricate interactions show these limits. Since most 16S rDNA primers do not bind to Planctomycetota’s sequences, PCR bias makes identification problematic . AS-based research is sometimes called “metagenomics” instead of “metataxonomics.” This distinction is important for scientific clarity since metagenomics analyzes genetic material from whole microbial communities, whereas metataxonomics explore taxonomic diversity, reflecting AS-based research priorities. Mislabeling “metagenomics” confounds academics and concepts. Clear communication that achieves research aims requires technique-appropriate terminology. Microorganisms are considered PGPMs based on a set of specific criteria. It must fulfill at least two of the following three requirements: enhancement of plant development, successful colonization and mitigation of biotic and abiotic stressors . It is important to note that the efficiency of plant growth promotion is not solely determined by the quantity of plant growth-promoting (PGP) activities. Rather, each strain’s unique combination and mechanisms play an essential role in its effectiveness. Therefore, to optimize plant development, it is necessary to identify and analyze the most suitable PGP properties for each specific application . The current investigation found the ten most prevalent bacterial phyla in M. crassifolia rhizosphere and bulk soil. The core phylum in the rhizosphere and bulk soils was actinobacteria. Significant compositional variations were found across soil types. Pseudomonadota and Bacteroidota predominated in the rhizosphere. Acidobacteria was more abundant in bulk soil. Gemmatimonadota, Chloroflexota, Candidatus Rokubacteria, Myxococcota, Planctomycetota and Cyanobacteria were consistent throughout both soil types ( ). Pseudonocaradia , Sphingomonas , Steroidobacter and Nonomuraea had increased rhizosphere abundances. In contrast, bulk soil Rubrobacter , Solirubrobacter and Geodermatophil were abundant. The two soil types had similar abundances of Streptomyces , Nocardioides and Longimicrobium ( ). At the species level, unclassified Actinobacteria bacterium had a higher RA in the rhizosphere than in the bulk soil, whereas unclassified Acidobacteria had a higher prevalence in the bulk soil. Both environments had comparable RA for the remaining species ( ). Current findings match Actinobacteria abundance in drought-affected soils . Actinobacteria regulate ethylene levels, lowering plant stress . During drought, plants generate ethylene, but too much might limit growth . Streptomyces spp., for example, produces antibiotics and other secondary metabolites [ , , , , , ]. Rhizosphere and bulk soil have greater Streptomyces spp. RA. Nocardioides spp. dominated both soil types, showing their widespread prevalence. The genus produces phytohormones including salicylic acid (SA) and indole-3-acetic acid. SA helps plants develop and IAA helps them deal with salt . Pseudonocardia and Nonomuraea predominated in the rhizosphere, unlike the bulk soil. IAA production makes Nonomuraea and Pseudonocardia spp. suitable PGPMs. Nonomuraea is antifungal and Pseudonocardia species is antimalarial [ , , , ]. Due to intense colonization by several taxa, Geodermatophilus spp. is abundant in bulk soil. Acidobacteria were more abundant in bulk soil, while Pseudomonadota and Bacteroidota dominated the rhizosphere, due to their differing nutritional needs. The oligotrophic nature of Acidobacteria, favoring low-nutrient environments, aligns with their prevalence in the nutrient-poor bulk soil. In contrast, copiotrophic Pseudomonadota and Bacteroidota thrive in carbon-rich conditions, consistent with their dominance in the rhizosphere, enriched by root exudates. Root exudates, through chemotaxis, may actively recruit PGPMs from the bulk soil, functioning as a reservoir of potential symbionts . On the one hand, Pseudomonadota acts as a PGPM by fixing nitrogen, producing IAA and solubilizing phosphorus, which is especially beneficial during droughts [ , , , ]. Bacteroidota is known for its ability to produce the enzyme 1-aminocyclopropane-1-carboxylate (ACC) deaminase . Studies have associated soil degradation and reduced fertility with a decline in Bacteroidota abundance, making it a potential indicator of soil quality . On the other hand, Acidobacteria members, though less prevalent in the rhizosphere than in bulk soil, exhibit root colonization, potentially facilitated by their production of exopolysaccharide (EPS). The presence of Acidobacteria species suggests a symbiotic relationship with plants, which could precede their function as PGPMs. This symbiotic interaction contributes to plant growth promotion by regulating key metabolic processes (sulfur, nitrogen and carbon cycling) and improving soil health via EPS-mediated soil structure enhancement. It also enhances plant growth directly through IAA production and indirectly by increasing iron availability via siderophore production . Higher soil macronutrients, plant species richness and Gemmatimonadota abundance are positively correlated . This investigation found a hitherto uncharacterized microbial species from this phylum, but its restricted culturability prevents a full elucidation of its functional significance in plant–microbe interactions . It is premature to categorize Gemmatimonadota as a PGPM. Candidatus Rokubacteria is uncultivable, preventing functional characterization. The phyla Planctomycetota and Myxococcota cause Panax notoginseng root rot , although Myxococcota also respires, denitrifies and ammoniifies . Some Myxococcota species modulate rhizosphere microbiome structure as Fusarium wilt biocontrol agents . This apparent contrast reflects the complex dynamics of plant–microbe systems, where plant species and root exudate components affect how particular bacterial taxa affect plant health . Some reports indicate a strong link between Chloroflexi and the rhizosphere of medicinal plants, as in the present study . The current results also confirm the presence of Cyanobacteria in the rhizobiome of arid regions in western Saudi Arabia, aligning with previous studies . Cyanobacteria, known for their potential as biofertilizers [ , , , , , ], enhance plant growth through nitrogen fixation, phytohormone production, phosphate solubilization and pathogen control [ , , , ]. Their production of auxins, cytokinin, SA and gibberellic acid is critical in alleviating abiotic stress and promoting plant development, especially during drought [ , , , , , , , ]. Previous studies have shown that Sphingomonas spp. act as PGPMs by mitigating abiotic stress and improving nutrient uptake [ , , , ]. Solirubrobacter and Geodermatophilus spp. contribute to nutrient cycling, with the former exhibiting resistance to UV radiation, drought and carbon limitation [ , , , , , , ]. Rubrobacter spp. possess photosynthetic proteins and are involved in the carbon, nitrogen and sulfur cycles [ , , ]. Steroidobacter spp. confer cadmium resistance , while Betaproteobacteria species play a role in nitrogen fixation, degradation of aromatic compounds, inhibition of pathogens and arsenic resistance [ , , , ]. Longimicrobium spp. improve heavy metal resistance and plant biomass . The present study also revealed a significant prevalence of unidentified bacterial species in all soil types, especially in the phyla Actinobacteria, Gemmatimonadota, Chloroflexota and Betaproteobacteria, suggesting that they could serve as new candidates for PGPMs. In the bulk soil, an unclassified Acidobacteria was identified as the most dominant species after Actinobacteria. These unclassified bacteria may indirectly have the potential to promote plant growth via soil as novel taxa, emphasizing the need for further investigation. The top ten phyla observed in the rhizosphere and bulk soil microbiomes at the Eukaryota domain shared dominance, although their relative abundances varied. Notably, Mucoromycota was present in both soil types but showed higher levels in the bulk soil. On the other hand, Ascomycota exhibited greater abundance in the rhizosphere while Chytridiomycota was more prevalent in the bulk soil. Basidiomycota and Zoopagomycota displayed similar abundance levels in both regions, whereas Blastocladiomycota were only found in the rhizosphere. The RA of genera and species was similar across both regions, with a few notable exceptions. Podosphaera aphanis was exclusively detected in the rhizosphere, and Morchella spp. and R. clarus exhibited higher RA in the rhizosphere. In contrast, H. curvatum and Rhizopus arrhizus demonstrated greater RA in the bulk soil ( , and ). Zygomycetous fungi (Mucoromycota and Zoopagomycota), which lack zoospores, have important ecological roles. Their influence is attributed to plankton and pathogen population management, nutrient cycling within food webs and heavy metal bioremediation . Despite resource limitations, Ascomycota significantly contribute to carbon and nitrogen cycling through decomposition, symbiotic relationships and pathogenic interactions. Their enzymatic arsenal (cellulases, hemicellulases and ligninases) and effector proteins mediate these interactions with plants . Basidiomycetes, prevalent in warm and humid climates, are noteworthy for producing bioactive secondary metabolites with antioxidant and anti-pathogenic properties . Zoosporic fungi, including Chytridiomycota, demonstrate ecological versatility as saprotrophs and parasites, employing diverse enzymatic mechanisms . Likewise, the lately delineated Blastocladiomycota phylum includes notable plant pathogens such as Physoderma spp. . Microsporidia, found in diverse soil environments, including tomato rhizosphere, often serve as biocontrol agents or act as pathogens targeting insect species . Symbiotic root colonization and nutrient exchange by AMFs including R. clarus , R. irregularis and Funneliformis geosporum increase host plant drought tolerance . However, dryness significantly limits AMF development and activity, reducing the efficacy of this symbiotic interaction in relieving water stress . The lack of phytase-encoding genes in AMF species limits their ability to metabolize organic phosphorus, requiring cooperative interactions with phosphate-solubilizing bacteria (PSB). Fungal exudates induce PSB migration toward AMF hyphae, which activates PSB chemotaxis genes. The chemotactic recruitment of PSB is necessary for this symbiotic phosphorus acquisition approach to work . Several fungal species improve plant stress tolerance. R. clarus improves soybean and tomato drought tolerance, boosting growth and recovery [ , , ]. R. irregularis improves root architecture, glomalin production and microbial communities in different plants to reduce water stress. These effects improve nutrient absorption, secondary metabolites and antioxidants [ , , , , , , ]. F. geosporum , found in Capsicum annuum , Celtis caucasica and Zingiber spp. rhizospheres, promotes Solanum tuberosum , C. annuum and C. caucasica growth and reduces legume salinity stress [ , , , , ]. By enhancing phosphorus uptake, Entrophospora candida boosts Allium vineale development . On the contrary, the presence of the strawberry powdery mildew pathogen P. aphanis in the rhizosphere is likely non-specific, resulting from wind or sand dispersal . The plant pathogenic fungi, such as R. arrhizus , Aspergillus spp. and Fusarium spp., detected in bulk soil, may be passively displaced from rhizospheres via chemotaxis mediated by bacteria or root exudates [ , , ] ( ). Intriguingly, some of these pathogens may indirectly benefit plants by catabolizing complex organic molecules into simpler, more readily utilizable forms . The presence of Actinobacteria, particularly Streptomyces , in soil has been correlated with the occurrence of Ascomycota, Basidiomycota, Mucoromycota, Pseudomonadota and Bacteroidota in the rhizosphere of Ziziphus jujuba . This suggests a positive interaction between Actinobacteria and these fungal and bacterial phyla, in agreement with present findings. However, it is essential to recognize that multiple factors beyond microbial interactions influence such associations in the rhizosphere. Plant species, root exudates, soil pH, temperature, moisture levels and organic matter content shape these relationships . Thus, it cannot be definitively concluded that these associations are for the rhizosphere of all plant species. Furthermore, the present study observed the presence of Chytridiomycota and Cyanobacteria, which has been corroborated by prior research . Plant–microbe signaling networks mediate the structuring of rhizosphere bacterial communities, resulting in differential distribution of certain taxa compared to bulk soil. However, drought exerts a stronger influence on bacterial community composition than plant species identity. On the other hand, fungal community composition is primarily driven by soil properties, reflecting the saprophytic nature of many fungi and their reliance on SOM decomposition rather than plant-derived carbon only [ , , , ].
The present study illustrates the bacterial and fungal communities inhabiting the M. crassifolia rhizosphere and surrounding bulk soil using WGS. Our results indicate intricate symbiotic relationships within the rhizosphere microbiome, including associations between AMF and PSB. Additionally, the analysis of the results suggests that specific microbial taxa may enhance plant resilience to biotic and abiotic stresses. Significantly, this study also identified novel microbial taxa not currently represented in existing databases, highlighting the unexplored diversity within the M. crassifolia rhizosphere. These unclassified microbes may represent a promising avenue for future research, with the potential to uncover novel PGPMs. To advance research on M. crassifolia and its soil rhizosphere microbiome, future studies should focus on isolating and characterizing microbial taxa within the Actinobacteria, Betaproteobacteria, Gemmatimonadota and Chloroflexota phyla to uncover novel plant growth-promoting microbes and their roles in nutrient cycling and plant resilience in arid conditions. Additionally, the molecular mechanisms underlying the tripartite interactions between AMF, PSB and M. crassifolia plant should be explored, with an emphasis on nutrient uptake and stress tolerance pathways. Advanced omics tools can also be employed to profile microbial functions in the soil rhizosphere, shedding light on active metabolic pathways that support plant health. The development of microbial inoculants, particularly those based on PSB and AMF, should also be pursued to improve plant growth and soil fertility under water scarcity. Additionally, analysis via the Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway and enzyme can further be employed to detect and map the functional roles of microbes in the rhizosphere, enhancing our understanding of their contributions to plant health. Finally, integrating microbial-based strategies into land rehabilitation and sustainable agriculture efforts in arid regions could enhance soil health and plant nutrition, ultimately supporting the conservation and sustainable use of M. crassifolia in these ecosystems.
|
Bridging the gap: evaluation of preoperative patients’ education by comparing expectations and real-perioperative surgical experiences: a mixed-methods descriptive cross-sectional study | 8a59bfbb-b870-4d04-9520-9e4e17e974eb | 11337772 | Patient Education as Topic[mh] | Patient education is essential for providing optimal patient care in all healthcare sectors. It leads to changes in knowledge, attitudes, and behavior, with the ultimate goal of improving therapeutic outcomes . Effective patient education recognizes patients’ needs and goes beyond knowledge delivery to instill major behavioral changes in patients . Education during the preoperative phase helps to prepare patients for coming surgical experience by offering access to health information, postoperative psychosocial support, and the ability to relieve surgical anxiety . Preoperative education included information about the surgical process, preoperative preparation, postoperative expectations, anesthesia, and prospective results . Preoperative instruction has been shown to reduce postoperative problems, shorten hospital stay, and increase patient and family satisfaction . The significance of preoperative education is consistent with broader healthcare programs such as shared decision making (SDM) and enhanced recovery after surgery (ERAS) protocols. SDM involves patients in their healthcare decisions, promoting a collaborative approach that improves patient engagement and satisfaction . ERAS protocols use evidence-based perioperative care to reduce physiological stress during surgery, improve recovery periods, and shorten hospital stays . Both SDM and ERAS stress the significance of preoperative education in preparing patients for surgery and improving outcomes . Preoperative education is a critical component of these patient-centered care paradigms. Healthcare practitioners can minimize fear, increase knowledge, and improve overall patient experiences by informing patients about what to expect before, during, and after surgery . Despite its acknowledged benefits, many institutions continue to use obsolete preoperative techniques that may not properly serve patients’ demands . Traditional surgical care components, such as overnight fasting, can aggravate the stress response and impede postoperative recovery. Modern perioperative therapies seek to control the surgical stress response and mitigate its deleterious consequences, accentuating the significance of upgrading preoperative patient education strategies to match current evidence-based standards . Healthcare practitioners can play an important role in reinforcing patient education on treatment, which can boost patient confidence, their active involvement, and adherence with treatment plans . Effective two-way communication between patients and physicians lowers knowledge asymmetry and promotes collaborative decision-making, boosting patients’ confidence and trust in their doctors . This trust is essential for successful preoperative education because patients who trust their healthcare professionals are more likely to participate and adhere to their treatment plans . Our research focused on the interaction between several modifying factors, as stated in the Health Belief Model (HBM), and preoperative anxiety. Psychosocial and psychological factors play an important role in preparing patients for surgery, influencing their emotional well-being and physiological reactions. As previously hypothesized and verified by research, anxiety can modify the body’s stress response, influencing immune function and postoperative recovery outcomes . Preoperative patient education is a critical component of surgical care. It has the potential to greatly improve clinical outcomes and patient satisfaction. This investigation aimed to assess the effectiveness of preoperative education provided to patients. Areas with higher levels of dissatisfaction should be identified, and recommendations for future studies should be given. This study is unique because of its primary focus on the perceptions of patients after a thorough preoperative education session and the identification of which additional preoperative information the patient expects, be it a practical demonstration or a different choice of words for communication. Thus, the boundaries of patient satisfaction should be stretched in light of the guidance provided by the patients.
Study design and sample This descriptive cross-sectional mixed-method study aimed to assess patients’ experiences with preoperative education and postsurgery satisfaction levels using both the qualitative and quantitative elements. Consecutive sampling was employed, and 65 patients were selected from the ENT department of Khyber Teaching Hospital, Peshawar, Pakistan from October 9th to November 2nd, 2023. The criteria included competent patients who consented to undergo surgery, while nonconsenting individuals were excluded. Data collection The data were collected using the “Patient Data Collection Form” (PDCF), which was developed by researchers based on a thorough literature review . The form comprised 25 questions focusing on preoperative education, patient demographics, and previous surgical experiences. During data collection, patients who met the inclusion criteria were informed about the purpose and procedure of the study. The researchers educated and interviewed the patients face-to-face. Comments provided by the patients were transcribed verbatim. The interview duration was approximately 10–15 min preoperatively and 3–5 min postoperatively for each patient. In addition to interviews, group discussions were held to facilitate a deeper understanding of patients’ perspectives. Analysis The transcribed data from interviews and group discussions were analyzed using thematic analysis techniques manually. Comments were judged and categorized based on the patients’ responses into three levels of satisfaction: satisfied, partially satisfied, and not satisfied. The thematic data derived from the qualitative analysis were then quantified as frequencies and percentages for each satisfaction category in each point of PDCF using SPSS software. Additionally, SPSS was used to analyze the descriptive data of the sample, providing an overview of demographic characteristics and other relevant variables. To evaluate the impact of the preoperative education on perioperative fear and anxiety, a chi-square test was applied.
This descriptive cross-sectional mixed-method study aimed to assess patients’ experiences with preoperative education and postsurgery satisfaction levels using both the qualitative and quantitative elements. Consecutive sampling was employed, and 65 patients were selected from the ENT department of Khyber Teaching Hospital, Peshawar, Pakistan from October 9th to November 2nd, 2023. The criteria included competent patients who consented to undergo surgery, while nonconsenting individuals were excluded.
The data were collected using the “Patient Data Collection Form” (PDCF), which was developed by researchers based on a thorough literature review . The form comprised 25 questions focusing on preoperative education, patient demographics, and previous surgical experiences. During data collection, patients who met the inclusion criteria were informed about the purpose and procedure of the study. The researchers educated and interviewed the patients face-to-face. Comments provided by the patients were transcribed verbatim. The interview duration was approximately 10–15 min preoperatively and 3–5 min postoperatively for each patient. In addition to interviews, group discussions were held to facilitate a deeper understanding of patients’ perspectives.
The transcribed data from interviews and group discussions were analyzed using thematic analysis techniques manually. Comments were judged and categorized based on the patients’ responses into three levels of satisfaction: satisfied, partially satisfied, and not satisfied. The thematic data derived from the qualitative analysis were then quantified as frequencies and percentages for each satisfaction category in each point of PDCF using SPSS software. Additionally, SPSS was used to analyze the descriptive data of the sample, providing an overview of demographic characteristics and other relevant variables. To evaluate the impact of the preoperative education on perioperative fear and anxiety, a chi-square test was applied.
A cohort of 65 patients was included in this study, the descriptive and clinical details of which are provided in Table while the various causes of anxiety in the preoperative and perioperative periods are described in Table . Table depicts a gender-based analysis regarding the impact of patient education on anxiety, showing a significant reduction in females as compared to males. Satisfaction status in all the different domains of preoperative education is provided in Table . Figure serves as a visual summary of the satisfaction status. Some of the remarks given by patients are provided below; a curious remark was from a patient who said there were no instructions regarding the usage of Naswar/Snuff. Bringing to our notice to incorporate instructions regarding the use of narcotics/other drugs in future preoperative education protocols. Patient no. 17- Regarding ROM and deep breathing exercises , ‘I had pain and cough. I was informed about exercises , but a practical demonstration at that time would have been better.’ Patient no. 37- Regarding the induction of anesthesia , “everything went blurred. My heart beat raised. I was told that it will be like sleeping , but it was not’ . Patient no. 38- Post-Tonsillectomy ‘The surgery took longer than I was told. I felt pain while talking , which was not mentioned.’ Patient no. 39- Due to time mismanagement , ‘I had a fast of almost 10 hours. I was told to stand a few hours after surgery , but nobody told me that I would feel dizzy while doing so.’ Patient no. 42- Due to mismanagement regarding timing , ‘I had a fast of 11 hours , which was way too long than I was told. The whole surgery took more time than previously mentioned. Nothing was mentioned regarding change in voice which I am feeling.’ Patient no. 45- ‘No one told me when I could resume smoking and take Naswar [Snuff] after surgery.’ Patient no. 46- ‘When I woke up , I felt feverish with a burning sensation all over my face.’
Informed consent includes information disclosure along with the recommendation of a plan of care, patient education, voluntary decision-making, and authorization to proceed with the plan of care . This research investigated the effect of structured preoperative education on patient anxiety levels, satisfaction with the information supplied, and perceived knowledge gaps among people undergoing surgery. The findings provide important insights into the efficacy of current preoperative education programs and suggestions for improvement. Preoperative anxiety is a typical concern for surgical patients and has a substantial impact on their perioperative experience. Our findings revealed that a significant majority of patients (60%) experienced preoperative worry, which was mostly caused by a lack of information (10.25%), a lack of prior experience (41.02%), or both (48.71%). These findings highlight the value of personalized preoperative education in reducing anxiety by addressing individual patient concerns. These results align with earlier research and support the idea that well-informed patients are better prepared to deal with postoperative stressors. Effective patient education has also been found to reduce readmission and complication rates . The study revealed a considerable gender gap in anxiety levels, with females being more likely to have preoperative anxiety than males (83.8% vs. 28.6%). This gender-specific variance in anxiety responses during surgical trips emphasizes the significance of personalizing educational efforts to meet individual psychological requirements. This conclusion is consistent with the findings of Kennedy and Parish (2021), who reported sex-related differences in postoperative stress responses . However, this study revealed that although preoperative anxiety levels were greater in females, perioperative anxiety was significantly lower in females than in males [Table ] ( p = 0.001). This clearly demonstrates the importance of adequate patient education in nullifying sex differences. Although these results support combat anxiety, there is no evidence to suggest that preoperative education has any effect on decreasing depression levels. There is a need to employ additional strategies to ensure emotional wellbeing . While overall satisfaction with preoperative information was high (100%) for most factors, such as surgery type, treatment alternatives, success rates, and postoperative care instructions, other regions indicated lower levels of satisfaction. Dissatisfaction was particularly high with the time and duration of surgery (33.8%) and prior fasting instructions (12.3%). These findings are consistent with those of Forner et al. (2020), who emphasized the crucial link between information adequacy and patient satisfaction . Educational interventions have also been noted to improve satisfaction levels among family members . The findings of this study have multiple ramifications for clinical practice. First, improving the clarity and comprehensiveness of preoperative teaching materials can greatly reduce anxiety and increase patient satisfaction. Second, a patient-centered approach that takes into account demographic aspects such as gender and educational background is critical for effectively tailoring educational interventions . Patients commonly express concerns related to surgical procedures, including fears about equipment, anesthesia, surgical complications, and postsurgical outcomes . Third, incorporating practical demonstrations and interactive sessions into preoperative education may address patient discontent with specific areas of care, such as posture exercises and postoperative protocols. The diverse ways patients approach their healthcare presents difficulties for public engagement efforts and healthcare providers, underscoring the importance of personalized and nuanced approaches to patient treatment . Furthermore, continuing the review and revision of preoperative education methods are critical for meeting changing patient needs and expectations. Collaboration among healthcare practitioners, patients, and caregivers is critical to ensuring that instructional content remains current and accessible. Trust and confidence in healthcare professionals are important for patients, and miscommunication or a perceived lack of care can impact patient-professional relationships . Understanding patients’ expectations across three domains, namely, health outcomes, individual clinicians, and the healthcare system, is essential for the delivery of effective patient-centered care . Patients often face challenges in adjusting to new hospital environments, leading to increased anxiety and worry, which underscores the importance of thorough preoperative education . Research has shown that patients’ understanding of provided information is often limited, hence making it more desirable to include both written and oral forms of informed consent . Last, it is not just about healthcare providers’ priorities because the work environment and institutional focus on task-centered care, rather than emphasis on patient-centered care, are significant barriers to addressing patients’ needs . Tran recently proposed the person-centered communication model of care (PCCMC), which involves establishing rapport, followed by information exchange, shared decision making and eventually discussing care outcomes . Such models, though very effective in some settings, remain ineffective in settings where staff shortages, increased workloads, burnout and limited time hamper patient-centered communication . However, communication skills training has been found to improve patient-centered communication without increasing consultation duration . All these findings, while highlighting the need for further studies, emphasize the need for efforts at both the institutional and individual levels. Limitations While the study offers useful insights, it is not without limits. The sample size was small and restricted to a single population, perhaps restricting generalizability. Furthermore, relying on self-reported measures and intervieweres interpretation may result in response bias and collector bias, respectively. Future studies should include larger, more diverse patient populations to validate these findings in a variety of healthcare settings. Longitudinal studies could also examine the long-term effects of preoperative education on patient outcomes after the procedure.
While the study offers useful insights, it is not without limits. The sample size was small and restricted to a single population, perhaps restricting generalizability. Furthermore, relying on self-reported measures and intervieweres interpretation may result in response bias and collector bias, respectively. Future studies should include larger, more diverse patient populations to validate these findings in a variety of healthcare settings. Longitudinal studies could also examine the long-term effects of preoperative education on patient outcomes after the procedure.
This study emphasizes the importance of systematic preoperative education for reducing anxiety, increasing patient satisfaction, and improving surgical results. Healthcare professionals can foster a supportive environment that allows patients to make informed decisions and actively engage in their treatment journey by addressing particular patient concerns and using evidence-based communication skills. Continued research and innovation in preoperative education are critical for enhancing patient-centered care and surgical outcomes.
|
null | 95b3abf2-a9f9-4c77-8b27-ead145d3ad01 | 10423414 | Debridement[mh] | Necrotizing fasciitis (NF) is an aggressive skin and soft tissue infection (SSTI) that causes necrosis in muscular fascia and subcutaneous tissues. Infection travels along the fascial plane, which has a low blood supply. Hence, the overlying tissues are initially spared, while the fascia and preifascial planes are necrotizing, which might delay diagnosis. That is followed by an extension of infection to nearby soft tissues and muscles . NF is a life-threatening condition that requires prompt surgical intervention . Causative microbiologic pathogens are either polymicrobial or monomicrobial. Most cases are attributed to polymicrobial involvement and usually occur in immunocompromised patients, while monomicrobial cases are rare, seen mostly after a penetrating trauma injury. The usual pathologic isolates are Staphylococcus aureus , group A streptococci, anaerobic organisms (peptostreptococci, Bacteroides ), and Clostridium perfringes . NF is rarely attributed to other pathogens . Only a few cases of non-typhoid Salmonella NF have been reported to date. Herein, we report a case of NF caused by Salmonella typhimurium followed by a superimposed infection with Fusarium SSTI.
Background and presentation A 20-year-old Caucasian male patient presented to the emergency department (ED) with 2-day history of watery, nonbloody diarrhea, high-grade fever, and 1-day history of painful left lower extremity violaceous discoloration (Fig. ). The patient’s past medical history was pertinent for hypoplastic right heart syndrome (HRHS) with atretic tricuspid and pulmonic valves, large atrial septal defect, and hypoplastic right ventricle status post multiple interventions, namely right modified Blalock–Taussig shunt, Glenn procedure, failed fenestrated Fontan procedure with an obligatory right to left shunt, and dual-chamber pacemaker insertion. He suffers from severe left-sided heart failure (estimated left ventricular ejection fraction of 20–24%), for which he is maintained on daily bisoprolol, enalapril, furosemide, hydrochlorothiazide, spironolactone, and warfarin with recurrent hospitalizations for heart failure exacerbation. He had been taking oral prednisone 20 mg two times daily for the past 4 months for a suspected diagnosis of protein-losing enteropathy in the setting of persistent hypoproteinemia and hypoalbuminemia. The patient is maintained on prednisone owing to the unavailability of oral budesonide in Lebanon because of the major economic crisis the country is facing and the tremendous burden on the healthcare sector and medication shortage. Other medical comorbidities included congestive hepatopathy and hypothyroidism, for which he is treated with levothyroxine 50 mcg daily. Of note, the patient is a nonsmoker, does not consume alcohol, and denied the use of any recreational drugs. On the day of his presentation, the patient noticed a small area of violaceous discoloration on his left lower extremity (LE) shin that progressed rapidly with worsening tenderness and severe pain requiring opioids for pain relief. He denied recent trauma or penetrating injury at the site of the left LE. No animal contact was reported. He recalled consuming poultry-containing meals with mayonnaise sauce 4 days before the diarrhea onset. On review of systems, the patient noted dark-colored urine for the past 2 days and decreased appetite. Upon presentation to the ED, the blood pressure was 70/30 mmHg, the heart rate was 95, the respiratory rate was 18, and the oxygen saturation was 81% on room air (his baseline oxygen saturation was around 82% due to right-to-left shunt and Fontan anatomy). Patient was conscious but ill-appearing with no focal neurological deficit. Examination of the left lower extremity revealed severe tenderness, extensive erythema, bluish-violaceous discoloration more accentuated on the lateral aspect of the leg reaching below the knee, weak peripheral pulses, and normal foot dorsal and plantar flexion (Fig. ). Chest and abdominal examination was relevant for bibasilar crackles on auscultation and increased abdominal girth along with hepatomegaly; otherwise the abdomen was soft and nontender. Two sets of peripheral blood cultures were taken, and a stool sample was sent for culture. Meanwhile, the patient received an intravenous (IV) fluid bolus of 250 mL normal saline 0.9% and was started on a norepinephrine drip along with broad-spectrum antibiotics in view of his immunosuppression and shock status. IV meropenem 1 g every 8 h, clindamycin 900 mg every 8 h, and vancomycin 1 g every 8 h for possible NF were initiated. Oral azithromycin 1 g every 24 h was started for adequate salmonellosis coverage as it was suspected given exposure history. A stress dose of intravenous hydrocortisone 100 mg every 8 h was also given. Relevant laboratory work-up on presentation is summarized in Table . Investigations and management NF was high on the differential diagnosis, which also included gas gangrene, necrotizing myositis, cellulitis, and toxic shock syndrome. The plastic and reconstructive surgical team was consulted, and a computed tomography (CT) angiography of the left LE was obtained. Imaging showed cutaneous and subcutaneous soft tissue thickening involving the left leg, suggestive of cellulitis and diffuse body wall edema with no radiological evidence of necrotizing fasciitis and no vascular filling defect. Lower extremities venous duplex showed no evidence of deep or superficial venous thromboembolism. The patient was transferred to the medical intensive care unit (ICU) as he was deemed at high risk for general anesthesia given his cardiac condition. Serial follow-up examinations showed no further progression of the lower extremity discoloration compared with his initial presentation; however, the patient required the use of two vasopressors (norepinephrine and epinephrine) delivered through an urgently inserted jugular central venous catheter during the first 24 h of his hospital stay for septic and possible cardiogenic shock. Moreover, he had increased oxygen requirements reaching 10 L/min delivered through a face mask to maintain his baseline oxygen saturation. On day 1 of hospitalization, Gram-negative rods were recovered from the two sets of blood cultures taken at presentation. Broad-spectrum antibiotics regimen was maintained in the setting of ongoing septic shock and possible polymicrobial SSTI. The patient started to show hemodynamics improvement within 48 h from admission, and pressors were completely stopped on day 6 of hospitalization. The Gram-negative rods in the blood were identified as Salmonella typhimurium . Antibiotic susceptibility was tested by standard disk diffusion method as per Clinical Laboratory Standards Institute guidelines. The isolate was susceptible to ampicillin, ceftriaxone, cefotaxime, ceftazidime, ciprofloxacin, and trimethoprim/sulfamethoxazole. The antimicrobial regimen was modified to ceftriaxone 2 g IV every 24 h and oral azithromycin 1 g every 24 h as a dual treatment for severe Salmonella bacteremia with deep-seated infection . Although the patient showed significant clinical and hemodynamic improvement, the lower extremity discoloration showed no improvement, and multiple serous fluid-filled bullae developed. A fluid sample was taken under sterile conditions from one of the LE bullae and was incubated in aerobic and anaerobic blood culture bottles that later grew Salmonella typhimurium , hence confirming the diagnosis of Salmonella non-typhi bacteremia with associated cellulitis. No Salmonella species were recovered from the stool culture taken on the day of presentation. Repeated central and peripheral blood cultures were negative at day 4 of presentation. After the vasopressors were stopped, oral azithromycin was stopped, and IV ceftriaxone was kept, treating Salmonella bacteremia in an immunocompromised patient with a life-threatening presentation. Transthoracic cardiac ultrasound showed no evidence of valvular vegetation. Transesophageal cardiac ultrasound was not possible, given his cardiac and respiratory status. Direct hyperbilirubinemia and slightly elevated liver enzymes were attributed to the shock state and improved on follow-up laboratory workup. Hypervolemic hyponatremia improved with a low-dose continuous infusion of IV furosemide (5 mg/h) for 2 days. Thrombocytopenia on presentation was likely sepsis related and improved subsequently. The patient was then transferred to the regular medical ward. On days 11 and 15, he underwent wide surgical debridement, excision of necrotic tissues, and thorough irrigation under regional anesthesia, as conservative management alone had failed to improve his LE condition. Intraoperative tissue cultures taken during the two surgeries showed Gram-negative rods on Gram staining and grew Salmonella typhimurium with similar susceptibility patterns to the one isolated in blood. Surgical pathology confirmed the diagnosis of necrotizing fasciitis. The patient was discharged on day 16, and he was prescribed oral ciprofloxacin 500 mg two times daily for an additional month to complete a total duration of 6 weeks and an oral prednisone taper regimen over 2 weeks with regular follow-up. Figure summarizes the timeline of the events. Follow-up Twenty-three days later, the patient underwent a third surgical debridement for necrotic tissues because of poor wound healing with the application of a dermal regeneration template. Intraoperative bacterial tissue culture showed no evidence of Salmonella typhimurium ; however, the pathologic specimen showed invasive fungal organisms with a superficial fungal ball. Then, intraoperative tissue fungal culture grew Fusarium species. Blood cultures were taken subsequently and showed no evidence of bacteremia or fungemia. The patient was started on oral voriconazole 200 mg two times daily to cover superimposed Fusarium SSTI. The patient completed the ciprofloxacin course and was kept on oral voriconazole until undergoing skin grafting. A follow-up transthoracic cardiac ultrasound showed no evidence of endocardial involvement. In the setting of invasive Salmonella , fungal SSTI, and poor wound healing, the patient was investigated for the presence of immunodeficiency. He was found to have low serum immunoglobulin G (IgG) (3.3 g/L; reference range 7–16 g/L), and blood immune profile by flow cytometry showed a decrease in T cytotoxic lymphocytes, T helper lymphocytes, natural killer cells, and B lymphocytes. He was referred to an immunology clinic for management and follow-up, and he received a single dose of 1 g/kg of intravenous immunoglobulin. Six weeks following his last surgical debridement, the wound showed adequate healing, and the patient underwent skin grafting, after which oral voriconazole was stopped. No treatment-related side effects were identified during hospital stay and frequent clinical follow-ups. The patient was followed up every 2 weeks in plastic surgery clinics for the first 3 months after the last surgical debridement, then monthly. Six months follow-up showed adequate wound healing.
A 20-year-old Caucasian male patient presented to the emergency department (ED) with 2-day history of watery, nonbloody diarrhea, high-grade fever, and 1-day history of painful left lower extremity violaceous discoloration (Fig. ). The patient’s past medical history was pertinent for hypoplastic right heart syndrome (HRHS) with atretic tricuspid and pulmonic valves, large atrial septal defect, and hypoplastic right ventricle status post multiple interventions, namely right modified Blalock–Taussig shunt, Glenn procedure, failed fenestrated Fontan procedure with an obligatory right to left shunt, and dual-chamber pacemaker insertion. He suffers from severe left-sided heart failure (estimated left ventricular ejection fraction of 20–24%), for which he is maintained on daily bisoprolol, enalapril, furosemide, hydrochlorothiazide, spironolactone, and warfarin with recurrent hospitalizations for heart failure exacerbation. He had been taking oral prednisone 20 mg two times daily for the past 4 months for a suspected diagnosis of protein-losing enteropathy in the setting of persistent hypoproteinemia and hypoalbuminemia. The patient is maintained on prednisone owing to the unavailability of oral budesonide in Lebanon because of the major economic crisis the country is facing and the tremendous burden on the healthcare sector and medication shortage. Other medical comorbidities included congestive hepatopathy and hypothyroidism, for which he is treated with levothyroxine 50 mcg daily. Of note, the patient is a nonsmoker, does not consume alcohol, and denied the use of any recreational drugs. On the day of his presentation, the patient noticed a small area of violaceous discoloration on his left lower extremity (LE) shin that progressed rapidly with worsening tenderness and severe pain requiring opioids for pain relief. He denied recent trauma or penetrating injury at the site of the left LE. No animal contact was reported. He recalled consuming poultry-containing meals with mayonnaise sauce 4 days before the diarrhea onset. On review of systems, the patient noted dark-colored urine for the past 2 days and decreased appetite. Upon presentation to the ED, the blood pressure was 70/30 mmHg, the heart rate was 95, the respiratory rate was 18, and the oxygen saturation was 81% on room air (his baseline oxygen saturation was around 82% due to right-to-left shunt and Fontan anatomy). Patient was conscious but ill-appearing with no focal neurological deficit. Examination of the left lower extremity revealed severe tenderness, extensive erythema, bluish-violaceous discoloration more accentuated on the lateral aspect of the leg reaching below the knee, weak peripheral pulses, and normal foot dorsal and plantar flexion (Fig. ). Chest and abdominal examination was relevant for bibasilar crackles on auscultation and increased abdominal girth along with hepatomegaly; otherwise the abdomen was soft and nontender. Two sets of peripheral blood cultures were taken, and a stool sample was sent for culture. Meanwhile, the patient received an intravenous (IV) fluid bolus of 250 mL normal saline 0.9% and was started on a norepinephrine drip along with broad-spectrum antibiotics in view of his immunosuppression and shock status. IV meropenem 1 g every 8 h, clindamycin 900 mg every 8 h, and vancomycin 1 g every 8 h for possible NF were initiated. Oral azithromycin 1 g every 24 h was started for adequate salmonellosis coverage as it was suspected given exposure history. A stress dose of intravenous hydrocortisone 100 mg every 8 h was also given. Relevant laboratory work-up on presentation is summarized in Table .
NF was high on the differential diagnosis, which also included gas gangrene, necrotizing myositis, cellulitis, and toxic shock syndrome. The plastic and reconstructive surgical team was consulted, and a computed tomography (CT) angiography of the left LE was obtained. Imaging showed cutaneous and subcutaneous soft tissue thickening involving the left leg, suggestive of cellulitis and diffuse body wall edema with no radiological evidence of necrotizing fasciitis and no vascular filling defect. Lower extremities venous duplex showed no evidence of deep or superficial venous thromboembolism. The patient was transferred to the medical intensive care unit (ICU) as he was deemed at high risk for general anesthesia given his cardiac condition. Serial follow-up examinations showed no further progression of the lower extremity discoloration compared with his initial presentation; however, the patient required the use of two vasopressors (norepinephrine and epinephrine) delivered through an urgently inserted jugular central venous catheter during the first 24 h of his hospital stay for septic and possible cardiogenic shock. Moreover, he had increased oxygen requirements reaching 10 L/min delivered through a face mask to maintain his baseline oxygen saturation. On day 1 of hospitalization, Gram-negative rods were recovered from the two sets of blood cultures taken at presentation. Broad-spectrum antibiotics regimen was maintained in the setting of ongoing septic shock and possible polymicrobial SSTI. The patient started to show hemodynamics improvement within 48 h from admission, and pressors were completely stopped on day 6 of hospitalization. The Gram-negative rods in the blood were identified as Salmonella typhimurium . Antibiotic susceptibility was tested by standard disk diffusion method as per Clinical Laboratory Standards Institute guidelines. The isolate was susceptible to ampicillin, ceftriaxone, cefotaxime, ceftazidime, ciprofloxacin, and trimethoprim/sulfamethoxazole. The antimicrobial regimen was modified to ceftriaxone 2 g IV every 24 h and oral azithromycin 1 g every 24 h as a dual treatment for severe Salmonella bacteremia with deep-seated infection . Although the patient showed significant clinical and hemodynamic improvement, the lower extremity discoloration showed no improvement, and multiple serous fluid-filled bullae developed. A fluid sample was taken under sterile conditions from one of the LE bullae and was incubated in aerobic and anaerobic blood culture bottles that later grew Salmonella typhimurium , hence confirming the diagnosis of Salmonella non-typhi bacteremia with associated cellulitis. No Salmonella species were recovered from the stool culture taken on the day of presentation. Repeated central and peripheral blood cultures were negative at day 4 of presentation. After the vasopressors were stopped, oral azithromycin was stopped, and IV ceftriaxone was kept, treating Salmonella bacteremia in an immunocompromised patient with a life-threatening presentation. Transthoracic cardiac ultrasound showed no evidence of valvular vegetation. Transesophageal cardiac ultrasound was not possible, given his cardiac and respiratory status. Direct hyperbilirubinemia and slightly elevated liver enzymes were attributed to the shock state and improved on follow-up laboratory workup. Hypervolemic hyponatremia improved with a low-dose continuous infusion of IV furosemide (5 mg/h) for 2 days. Thrombocytopenia on presentation was likely sepsis related and improved subsequently. The patient was then transferred to the regular medical ward. On days 11 and 15, he underwent wide surgical debridement, excision of necrotic tissues, and thorough irrigation under regional anesthesia, as conservative management alone had failed to improve his LE condition. Intraoperative tissue cultures taken during the two surgeries showed Gram-negative rods on Gram staining and grew Salmonella typhimurium with similar susceptibility patterns to the one isolated in blood. Surgical pathology confirmed the diagnosis of necrotizing fasciitis. The patient was discharged on day 16, and he was prescribed oral ciprofloxacin 500 mg two times daily for an additional month to complete a total duration of 6 weeks and an oral prednisone taper regimen over 2 weeks with regular follow-up. Figure summarizes the timeline of the events.
Twenty-three days later, the patient underwent a third surgical debridement for necrotic tissues because of poor wound healing with the application of a dermal regeneration template. Intraoperative bacterial tissue culture showed no evidence of Salmonella typhimurium ; however, the pathologic specimen showed invasive fungal organisms with a superficial fungal ball. Then, intraoperative tissue fungal culture grew Fusarium species. Blood cultures were taken subsequently and showed no evidence of bacteremia or fungemia. The patient was started on oral voriconazole 200 mg two times daily to cover superimposed Fusarium SSTI. The patient completed the ciprofloxacin course and was kept on oral voriconazole until undergoing skin grafting. A follow-up transthoracic cardiac ultrasound showed no evidence of endocardial involvement. In the setting of invasive Salmonella , fungal SSTI, and poor wound healing, the patient was investigated for the presence of immunodeficiency. He was found to have low serum immunoglobulin G (IgG) (3.3 g/L; reference range 7–16 g/L), and blood immune profile by flow cytometry showed a decrease in T cytotoxic lymphocytes, T helper lymphocytes, natural killer cells, and B lymphocytes. He was referred to an immunology clinic for management and follow-up, and he received a single dose of 1 g/kg of intravenous immunoglobulin. Six weeks following his last surgical debridement, the wound showed adequate healing, and the patient underwent skin grafting, after which oral voriconazole was stopped. No treatment-related side effects were identified during hospital stay and frequent clinical follow-ups. The patient was followed up every 2 weeks in plastic surgery clinics for the first 3 months after the last surgical debridement, then monthly. Six months follow-up showed adequate wound healing.
Salmonella species usually cause gastroenteritis, especially in developing countries, where 5% of the cases are usually complicated by secondary bacteremia, especially among immunocompromised patients . Focal nontyphoidal Salmonella infections are relatively rare, representing around 6% of Salmonella infections, and may affect any organ . Despite its unusualness, its prevalence varies according to host factors. In fact, it is three to four times more prevalent among immunocompromised patients in comparison with the general population . The most common sites of focal Salmonella infection are bones, joints, urinary tract, intraabdominal cavity, and rarely skin and soft tissues . In fact, SSTI secondary to Salmonella non-typhi is seen in around 1.5% of Salmonella non-typhi cases with a wide spectrum of clinical manifestations ranging from pustular lesions to subcutaneous abscesses and very rarely NF [ , – ]. Necrotizing fasciitis is a serious and potentially life-threatening soft tissue infection that should be promptly handled. A multidisciplinary team should be involved in managing this aggressive condition while continuously monitoring patient hemodynamics in a controlled setting . NF is a surgical emergency requiring wide and extensive debridement and reconstructive surgery . Multiple surgical debridements are usually warranted until tissue necrosis ceases and the growth of viable tissue is observed. The amputation of an involved limb should also be considered in the setting of irreversible necrosis and widespread gangrene with hemodynamic compromise . After the initial debridement, thorough wound follow-up is warranted with daily antibiotic dressing change, as hemodynamic instability may lead to progressive skin necrosis. The patient may return as often as necessary for further surgical debridement. NF is usually a polymicrobial entity; hence, broad-spectrum antimicrobial agents covering aerobic Gram-positive, Gram-negative organisms and anaerobes should be started with no delay. A reasonable regimen includes a combination of penicillin G, aminoglycoside, and clindamycin, a potent suppressor of bacterial toxin synthesis. The maximum doses of the antibiotics should be used and adjusted according to the patient’s weight, kidney and liver functions . In the present case report, we describe a rare incidence of monomicrobial nontraumatic Salmonella typhimurium NF of the lower extremity in a patient with multiple comorbid cardiac anomalies, prolonged steroid use, and immunodeficiency who presented with Salmonella bacteremia and improved on pathogen-directed antibiotics along with multiple surgical debridements. Only a few cases in the literature discuss complicated SSTI secondary to nontyphoidal Salmonella infections that commonly present in patients with predisposing immunocompromising conditions such as systemic lupus erythematosus, chronic steroids use, or immunomodulatory medications . Despite the high mortality associated with NF, a case series by Khawcharoenporn et al. in 2006 found that five out of the six reported cases of Salmonella NF survived after a combination of prompt medical and surgical management, such as our case ( ). Moreover, the presented case was complicated by a superimposed fungal infection with Fusarium species that was successfully managed with surgical debridement and oral voriconazole. Hence, a nonhealing wound in a similar setting should prompt consideration of superimposed opportunistic infections such as fungus. Despite its novelty in describing a rare entity with an unorthodox management plan dictated by the multiple comorbid conditions in our patient and his clinical evolution, this manuscript remains a case report with limited possibility of validity generalization. However, it may serve as a potential management plan in similar rare cases in light of the scarcity of evidence. Further larger-scale observational studies are required to establish a consensus on the management of this rare entity in frail patients. NF is a serious infection of soft tissues associated with high morbidity and mortality rates if not managed in a timely manner. The initial management should include a broad-spectrum antibiotic regimen, which would be ultimately adjusted according to culture results. Even though NF is most commonly related to polymicrobial infection, it is important to maintain a high index of suspicion for rare causative entities such as Salmonella non-typhi SSTI, in particular in immunosuppressed patients in whom a delay in diagnosis and management may cause significant morbidity and mortality.
|
Microvolt T-wave alternans complemented with electrophysiologic study for prediction of ventricular tachyarrhythmias in patients with arrhythmogenic right ventricular cardiomyopathy: a long-term follow-up study | a6c04b0f-91d0-4443-98e1-79cd6f088dab | 6629327 | Physiology[mh] | Arrhythmogenic right ventricular cardiomyopathy (ARVC) is an inherited cardiomyopathy characterized by right ventricular (RV) dysfunction and ventricular arrhythmias. Studies have shown that ARVC is present in up to 20% of individuals who experience sudden cardiac death (SCD) and is even more common among athletes who die suddenly. Prevention of SCD is the most important goal in the clinical management of patients with ARVC. According to the consensus guidelines, currently, the implantable cardioverter defibrillator (ICD) is the only effective strategy for the prevention of SCD. More effective markers are needed to determine the long-term prognosis of patients with ARVC. Microvolt T-wave alternans (MTWA) is a non-invasive test of arrhythmia vulnerability. Previous studies have shown that MTWA can predict ICD shocks and ventricular tachyarrhythmic events in diverse patient populations, including those with heart failure and ischemic cardiomyopathy. Ambulatory electrocardiograph-based TWA was reported to exhibit prognostic ability in patients with ARVC. Currently, electrophysiologic studies (EPSs) are also used to predict the prognosis in patients with ARVC. However, studies have shown mixed results. [ – ] Saguner et al reported that inducible sustained monomorphic ventricular tachycardia (VT) might predict long-term adverse outcomes in patients with ARVC. Therefore, the objective of this study was to explore the prognostic value of MTWA using modified moving average (MMA) analysis complemented by an EPS in patients with ARVC with preserved left ventricular (LV) function during long-term follow-up.
Ethical approval The study protocol was approved by the institutional ethics committee of the First Affiliated Hospital of Nanjing Medical University. All participants provided written informed consent. Patient recruitment Between January 2007 and November 2008, all in-hospital patients with ARVC were consecutively recruited. The diagnosis of ARVC was established on the basis of the criteria set by the Task Force of the Working Group of Myocardial and Pericardial Disease of the European Society of Cardiology and the Scientific Council on Cardiomyopathies of the International Society and Federation of Cardiology. All patients underwent MTWA tests and EPSs. Patients with acute infectious disease, acute myocardial ischemia, or uncontrolled heart failure or who were unable to complete the treadmill exercise test were excluded. Treadmill exercise test and measurement of microvolt T-wave alternans The exercise test was performed using a treadmill ergometer after enrollment. Continuous electrocardiograms (ECGs) were digitally recorded at 500 Hz with a CardioSoft exercise ECG system (version 4.0; GE Healthcare, Boston, MA, USA) and analyzed fully automatically by the GE Healthcare version of the MMA method using an incremental update factor of 1/32. Using a modified Bruce protocol in the treadmill exercise test, the heart rate of patients could be smoothly increased to 90 to 120 beats/min and maintained for at least 3 min. The treadmill was suspended if VT, syncope, ST-segment depression >0.2 mV, or weakness occurred in patients. The TWA values were calculated continuously during the entire exercise test from rest to recovery using all chest leads (V1–V6). The maximal TWA value at heart rate <120 beats/min among all pre-cordial leads of each subject was derived and designated the MaxValt. The MTWA value was independently analyzed by two blinded experts. Electrocardiographic characteristics All patients underwent ECG at rest (25 mm/s, 10 mm/mV) with standard lead positions. The ECGs were independently analyzed by two blinded experienced doctors. Differences in electrocardiographic interpretation were adjudicated by a third experienced doctor, and a final conclusion was made by consensus. QRS duration, epsilon wave, T-wave inversions in inferior or pre-cordial leads, and T-wave inversions beyond leads V1 to V3 were measured according to current practice. Electrophysiologic studies All the enrolled patients participated in the EPS. The endpoints of the EPS were the induction of sustained VT or ventricular fibrillation (VF) or the lack of induction of any tachyarrhythmias. Intravenous isoproterenol (1–4 μg/min) was administered if sustained VT was not induced by the baseline EPS. Programmed stimulation consisting of up to three extra stimuli was performed at the right ventricular apex and right outflow tract by delivering current at twice the diastolic threshold. Coupling intervals were progressively shortened until a response was no longer elicited or to a minimum of 200 ms. Follow-up All patients received follow-up with electrocardiography, 24-h ambulatory electrocardiography, and ICD interrogation during clinical visits, which were scheduled every 3 to 6 months. The positive endpoint was the first occurrence of any of the following events: SCD, documented sustained VT, VF, or the administration of appropriate ICD therapy. SCD was defined as death within 1 h of the onset of symptoms or during sleep without any other identified cause. Appropriate ICD therapies were shock or anti-tachycardia pacing delivered for persistent ventricular tachyarrhythmia. Statistical analyses Categorical variables are expressed as frequencies (percentages). Continuous variables are expressed as the mean ± standard deviation (SD). Variables with skewed distributions are expressed as medians (interquartile range [IQR], range). Clinical characteristics were compared with the t test or Pearson Chi-squared test as appropriate. Univariate and multivariate Cox regression analyses were performed to analyze associations between follow-up events and clinical variables. The results of the event-free analyses are presented with hazard ratios (HRs) and 95% confidence intervals (CIs). Event-free survival curves were plotted according to the Kaplan-Meier method with statistical significance examined by the log-rank test. All statistical analyses were performed with SPSS statistical software (version 20.0; IBM Co., NY, USA). A P value <0.05 was considered statistically significant.
The study protocol was approved by the institutional ethics committee of the First Affiliated Hospital of Nanjing Medical University. All participants provided written informed consent.
Between January 2007 and November 2008, all in-hospital patients with ARVC were consecutively recruited. The diagnosis of ARVC was established on the basis of the criteria set by the Task Force of the Working Group of Myocardial and Pericardial Disease of the European Society of Cardiology and the Scientific Council on Cardiomyopathies of the International Society and Federation of Cardiology. All patients underwent MTWA tests and EPSs. Patients with acute infectious disease, acute myocardial ischemia, or uncontrolled heart failure or who were unable to complete the treadmill exercise test were excluded.
The exercise test was performed using a treadmill ergometer after enrollment. Continuous electrocardiograms (ECGs) were digitally recorded at 500 Hz with a CardioSoft exercise ECG system (version 4.0; GE Healthcare, Boston, MA, USA) and analyzed fully automatically by the GE Healthcare version of the MMA method using an incremental update factor of 1/32. Using a modified Bruce protocol in the treadmill exercise test, the heart rate of patients could be smoothly increased to 90 to 120 beats/min and maintained for at least 3 min. The treadmill was suspended if VT, syncope, ST-segment depression >0.2 mV, or weakness occurred in patients. The TWA values were calculated continuously during the entire exercise test from rest to recovery using all chest leads (V1–V6). The maximal TWA value at heart rate <120 beats/min among all pre-cordial leads of each subject was derived and designated the MaxValt. The MTWA value was independently analyzed by two blinded experts.
All patients underwent ECG at rest (25 mm/s, 10 mm/mV) with standard lead positions. The ECGs were independently analyzed by two blinded experienced doctors. Differences in electrocardiographic interpretation were adjudicated by a third experienced doctor, and a final conclusion was made by consensus. QRS duration, epsilon wave, T-wave inversions in inferior or pre-cordial leads, and T-wave inversions beyond leads V1 to V3 were measured according to current practice.
All the enrolled patients participated in the EPS. The endpoints of the EPS were the induction of sustained VT or ventricular fibrillation (VF) or the lack of induction of any tachyarrhythmias. Intravenous isoproterenol (1–4 μg/min) was administered if sustained VT was not induced by the baseline EPS. Programmed stimulation consisting of up to three extra stimuli was performed at the right ventricular apex and right outflow tract by delivering current at twice the diastolic threshold. Coupling intervals were progressively shortened until a response was no longer elicited or to a minimum of 200 ms.
All patients received follow-up with electrocardiography, 24-h ambulatory electrocardiography, and ICD interrogation during clinical visits, which were scheduled every 3 to 6 months. The positive endpoint was the first occurrence of any of the following events: SCD, documented sustained VT, VF, or the administration of appropriate ICD therapy. SCD was defined as death within 1 h of the onset of symptoms or during sleep without any other identified cause. Appropriate ICD therapies were shock or anti-tachycardia pacing delivered for persistent ventricular tachyarrhythmia.
Categorical variables are expressed as frequencies (percentages). Continuous variables are expressed as the mean ± standard deviation (SD). Variables with skewed distributions are expressed as medians (interquartile range [IQR], range). Clinical characteristics were compared with the t test or Pearson Chi-squared test as appropriate. Univariate and multivariate Cox regression analyses were performed to analyze associations between follow-up events and clinical variables. The results of the event-free analyses are presented with hazard ratios (HRs) and 95% confidence intervals (CIs). Event-free survival curves were plotted according to the Kaplan-Meier method with statistical significance examined by the log-rank test. All statistical analyses were performed with SPSS statistical software (version 20.0; IBM Co., NY, USA). A P value <0.05 was considered statistically significant.
Patient characteristics The study population consisted of 35 patients with ARVC (mean age 38.6 ± 11.0 years; 28 males). The baseline characteristics are shown in Table . The echocardiography of all these patients showed the characteristic manifestations of RV abnormalities, such as wide RV outflow tract, local hypokinesia, local bulging, or local thinning [Table ]. All patients had LV ejection fractions (LVEFs) >50%, and their heart function was better than the New York Heart Association (NYHA) II class. Among the 35 patients with ARVC, 28 had experienced sustained VT, four had syncope, and three had implanted ICD devices. Analysis of MMA-TWA All enrolled patients with ARVC successfully underwent the MWTA test with treadmill exercise. The MaxValt was 17.0 (11.0–27.0) μV in 35 patients. The MaxValt values were 25.0 (12.5–28.0) μV in 15 patients with positive events and 17.0 (9.25–19) μV in 20 patients without positive events. Fifteen patients were using anti-arrhythmic drugs (AADs) when the MTWA test was performed, including seven patients using beta-blockers, five using sotalol, one using propafenone, one using a beta-blocker and propafenone, and one using a beta-blocker and amiodarone. The MaxValt of the 15 patients who used AADs was similar to that of the other 20 patients who did not use AADs (median: 16 vs . 17.5, P = 0.36). Electrophysiologic studies All 35 patients underwent EPSs. Among them, 22 patients had induced VT, with 13 exhibiting one type of induced monomorphic sustained VT, and nine exhibiting two or more types. Sixteen of 22 patients underwent ablation, with an acute success rate of 6/16. Follow-up During a median follow-up period of 99.9 ± 7.7 months, 15 patients had positive clinical events, including one SCD and 14 ventricular tachyarrhythmias. During the follow-up period, all 15 event-positive patients continued taking anti-arrhythmic drugs, two of them suffered syncope due to VT, and two ICD patients received shocks for fast VT/VF. Six patients underwent a second ablation for frequent sustained VT recurrence, and five more patients underwent implantation of ICDs during the follow-up period. Predictors of clinical events Compared with patients without positive events, patients with positive events had a higher proportion of male gender (15/15 vs . 13/20, P = 0.03) and a higher proportion of VT inducible by EPS (13/15 vs . 9/20, P = 0.02) [Table ]. Survival analysis with the univariate Cox regression model indicated that inducible VT (HR, 5.23; 95% CI, 1.17–23.24; P = 0.02) and MaxValt (HR, 1.06; 95% CI, 1.00–1.12; P = 0.02) were risk factors for positive events. The multivariate Cox regression model for survival showed that inducible VT (HR, 5.98; 95% CI, 1.33–26.8; P = 0.01) and MaxValt (HR, 1.06; 95% CI, 1.01–1.11; P = 0.01) independently predicted positive clinical events [Table ]. To evaluate the predictive power of MTWA and inducible VT for the occurrence of ventricular arrhythmias, the receiver operating characteristic (ROC) curve and the area under the curve (AUC) were used. When inducible VT was combined with MTWA, the AUC improved from 0.739 to 0.797 [Figure ]. The ROC curve showed that a MaxValt of 23.5 μV was the optimal cutoff value to identify positive events. Patients with MaxValt values >23.5 μV had worse event-free survival according to the Kaplan-Meier analysis (log-rank P = 0.015) [Figure ]. The Kaplan-Meier survival curve analysis of patients categorized by MaxValt and inducible VT showed that patients with both a MaxValt >23.5 μV and inducible VT had the worst prognosis ( P = 0.002) [Figure ]. The positive predictive value (PPV) was 85.71% (odds ratio [OR], 9; P = 0.027). In contrast, patients with a MaxValt ≤23.5 μV and non-inducible VT had the highest cumulative survival rate, with a negative predictive value (NPV) of 100% ( P = 0.0015).
The study population consisted of 35 patients with ARVC (mean age 38.6 ± 11.0 years; 28 males). The baseline characteristics are shown in Table . The echocardiography of all these patients showed the characteristic manifestations of RV abnormalities, such as wide RV outflow tract, local hypokinesia, local bulging, or local thinning [Table ]. All patients had LV ejection fractions (LVEFs) >50%, and their heart function was better than the New York Heart Association (NYHA) II class. Among the 35 patients with ARVC, 28 had experienced sustained VT, four had syncope, and three had implanted ICD devices.
All enrolled patients with ARVC successfully underwent the MWTA test with treadmill exercise. The MaxValt was 17.0 (11.0–27.0) μV in 35 patients. The MaxValt values were 25.0 (12.5–28.0) μV in 15 patients with positive events and 17.0 (9.25–19) μV in 20 patients without positive events. Fifteen patients were using anti-arrhythmic drugs (AADs) when the MTWA test was performed, including seven patients using beta-blockers, five using sotalol, one using propafenone, one using a beta-blocker and propafenone, and one using a beta-blocker and amiodarone. The MaxValt of the 15 patients who used AADs was similar to that of the other 20 patients who did not use AADs (median: 16 vs . 17.5, P = 0.36).
All 35 patients underwent EPSs. Among them, 22 patients had induced VT, with 13 exhibiting one type of induced monomorphic sustained VT, and nine exhibiting two or more types. Sixteen of 22 patients underwent ablation, with an acute success rate of 6/16.
During a median follow-up period of 99.9 ± 7.7 months, 15 patients had positive clinical events, including one SCD and 14 ventricular tachyarrhythmias. During the follow-up period, all 15 event-positive patients continued taking anti-arrhythmic drugs, two of them suffered syncope due to VT, and two ICD patients received shocks for fast VT/VF. Six patients underwent a second ablation for frequent sustained VT recurrence, and five more patients underwent implantation of ICDs during the follow-up period.
Compared with patients without positive events, patients with positive events had a higher proportion of male gender (15/15 vs . 13/20, P = 0.03) and a higher proportion of VT inducible by EPS (13/15 vs . 9/20, P = 0.02) [Table ]. Survival analysis with the univariate Cox regression model indicated that inducible VT (HR, 5.23; 95% CI, 1.17–23.24; P = 0.02) and MaxValt (HR, 1.06; 95% CI, 1.00–1.12; P = 0.02) were risk factors for positive events. The multivariate Cox regression model for survival showed that inducible VT (HR, 5.98; 95% CI, 1.33–26.8; P = 0.01) and MaxValt (HR, 1.06; 95% CI, 1.01–1.11; P = 0.01) independently predicted positive clinical events [Table ]. To evaluate the predictive power of MTWA and inducible VT for the occurrence of ventricular arrhythmias, the receiver operating characteristic (ROC) curve and the area under the curve (AUC) were used. When inducible VT was combined with MTWA, the AUC improved from 0.739 to 0.797 [Figure ]. The ROC curve showed that a MaxValt of 23.5 μV was the optimal cutoff value to identify positive events. Patients with MaxValt values >23.5 μV had worse event-free survival according to the Kaplan-Meier analysis (log-rank P = 0.015) [Figure ]. The Kaplan-Meier survival curve analysis of patients categorized by MaxValt and inducible VT showed that patients with both a MaxValt >23.5 μV and inducible VT had the worst prognosis ( P = 0.002) [Figure ]. The positive predictive value (PPV) was 85.71% (odds ratio [OR], 9; P = 0.027). In contrast, patients with a MaxValt ≤23.5 μV and non-inducible VT had the highest cumulative survival rate, with a negative predictive value (NPV) of 100% ( P = 0.0015).
The major findings of this study were as follows: (1) MTWA measured using MMA analysis could have prognostic value in patients with ARVC with preserved LV function during long-term follow-up. (2) A MaxValt >23.5 μV and inducible VT independently predicted life-threatening ventricular tachyarrhythmias, appropriate ICD therapy and SCD in this cohort. Exercise-induced MMA-TWA as a high-risk marker for cardiovascular events Exercise, such as on a treadmill or bicycle, is one of the most important methods to increase the heart rate to the critical level needed to detect TWA. The MTWA test using the spectral method and exercise can sometimes be affected by noise or artifacts. The MTWA test is improved when the MMA method is used. Moreover, exercise-induced MTWA by the MMA method has been shown to be a marker of increased risk for cardiovascular events in patients with or without heart diseases. In the Finnish Cardiovascular Study (FINCAVAS), the largest TWA study conducted to date, consecutive patients ( n = 3598, 2164 males) underwent a clinically indicated exercise test with a bicycle ergometer, and their MMA-TWA was analyzed from the pre-cordial leads. The results demonstrated that the maximal TWA monitored from pre-cordial lead V5 was the strongest predictor of cardiovascular mortality and SCD, and higher TWA values indicated greater cardiovascular mortality and SCD risk. Predictors of ventricular tachyarrhythmia and SCD in patients with ARVC ARVC is a genetically determined disease that often pre-disposes patients to life-threatening ventricular tachyarrhythmias, which may occur as the first manifestation of the disease in young people without previous symptoms. The main goal of ARVC therapy is the prevention of SCD. Previous studies showed that a number of clinical characteristics were associated with an increased risk of major arrhythmic events in patients with ARVC, such as male gender, proband status, cardiac arrest, syncope, LV dysfunction, inducible VT/VF, T-wave inversion in inferior or pre-cordial leads, T-wave inversions beyond right pre-cordial leads (V1–V3), and QRS fragmentation. However, these studies were based on smaller samples of patients who were followed for relatively short periods. In the present study, we explored the predictive value of the previous indices of history of syncope and cardiac arrest, T-wave inversion in inferior or pre-cordial leads, T-wave inversion beyond leads V1 to V3, QRS duration, epsilon waves, inducible VT, and the new index, MTWA. Using Cox regression analysis, only the MaxValt and inducible VT could independently predict major tachyarrhythmic events. Patients with both a MaxValt >23.5 μV and inducible VT had the worst prognosis. The PPV was 85.71%. However, patients with a MaxValt ≤23.5 μV and non-inducible VT had the highest cumulative survival rate, with an NPV of 100%. Therefore, the MTWA test using MMA analysis in this cohort was more powerful for predicting prognosis than the previous indices. Combined with other indices, such as heart rate turbulence or heart rate variability, the MMA-TWA test has been shown to be effective in identifying post-myocardial infarction patients at high risk of arrhythmic events. Furthermore, TWA testing could provide high NPVs for the primary endpoint of arrhythmia-free survival (appropriate ICD therapy, documented VT/VF, or all-cause mortality), as well as programmed ventricular stimulation. Our present study has provided comparable outcomes. Measurement and clinical implication of MMA-MTWA in patients with ARVC The spectral and MMA methods are the only two techniques approved by the United States Food and Drug Administration to measure microvolt levels of TWA for risk stratification for arrhythmic death. Tests with TWA levels ≥1.9 μV and a signal-to-noise ratio of K = 3 sustained for 2 min are classified as positive in the spectral protocol. However, a high incidence (approximately 20–40%) of “indeterminate” test results often occurs due to many factors, such as excessive ectopy, HR <105 beats/min, unsustained MTWA, or noise. As a promising time-domain analytical technique, MMA-TWA allows TWA analysis during routine exercise stress testing and 24-h ambulatory ECG monitoring without requiring special electrodes or a target heart rate. MMA-based TWA analyses on ambulatory ECG have been found to predict cardiovascular mortality and SCD in thousands of patients with coronary artery disease, myocardial infarction, heart failure, and ischemic and non-ischemic cardiomyopathy with varying extents of LV dysfunction. MMA-TWA has also been found to be elevated in ambulatory ECG recordings in other disease states associated with the risk of ventricular arrhythmias, including sleep apnea patients with congestive heart failure, epilepsy, chronic renal disease, hemodialysis, Brugada syndrome, and long QT syndrome. In the ambulatory ECG-based TWA test, TWA ≥47 μV in standard pre-cordial leads is considered abnormal, and TWA ≥60 μV is considered severely abnormal. These cutoff points are based on algorithms that used an update factor of 1/8. However, MMA-TWA characteristics and cutoff values based on the exercise stress test in patients with ARVC have not been previously reported. In our previous study, we found that TWA values in all pre-cordial leads and the MaxValt in patients with ARVC were much higher than those in healthy controls (MaxValt: 17.0 μV vs . 7 μV), using an update factor value of 1/32. In our present study, the cutoff value was proposed to be >23.5 μV, which was different from that suggested in the study by Chung et al . Chung et al used the protocol of ambulatory ECG-based TWA and an update factor value of 1/8. Their study showed that compared with patients without tachyarrhythmic events, those with tachyarrhythmic events had higher TWA within the modified V5 and V1 channels. These channels suggested a cutoff value of >66 μV, which is approximately two times our cutoff value. MTWA combined with EPS has a greater prognostic value in patients with ARVC MTWA has been reported to be a strong independent predictor of spontaneous ventricular arrhythmias, death, and programmed ventricular stimulation. MTWA was shown to be better than signal-averaged electrocardiography for the risk stratification of patients with regard to life-threatening arrhythmias. The Alternans Before Cardioverter Defibrillator study (566 patients, median follow-up 1.9 years) was the first clinical trial to use MTWA measured with the spectral method with a graded exercise protocol to guide prophylactic ICD insertion in patients with ischemic cardiomyopathy and low LVEF. The results demonstrated that TWA testing appeared to be comparable to EPSs for guiding ICD implantation and that the two methods may be complementary, especially with a high NPV to identify the subset of patients least likely to benefit from ICD insertion. In our present study, we also combined MMA-TWA and EPS to predict the prognosis in patients with AVRC with preserved LVEF. Our results showed that MaxValt and inducible VT independently predicted life-threatening ventricular tachyarrhythmias. Furthermore, the NPV was 100% when MMA-TWA was combined with EPS in a long-term follow-up study. Study limitations Despite a follow-up period as long as 8 years, our study was limited by a small sample size affiliated with a single center. ICD implantation was limited in our recruited patients. Furthermore, in this study, we used an update factor of 1/32 when analyzing TWA, which was less sensitive than the recommended update factor of 1/8. As a result, the TWA values obtained were substantially lower than expected, and the capacity to predict VT might also be reduced. In conclusion, MTWA assessment using MMA analysis complemented with an EPS could improve the ability to predict the prognosis in patients with ARVC with preserved LV function during long-term follow-up.
Exercise, such as on a treadmill or bicycle, is one of the most important methods to increase the heart rate to the critical level needed to detect TWA. The MTWA test using the spectral method and exercise can sometimes be affected by noise or artifacts. The MTWA test is improved when the MMA method is used. Moreover, exercise-induced MTWA by the MMA method has been shown to be a marker of increased risk for cardiovascular events in patients with or without heart diseases. In the Finnish Cardiovascular Study (FINCAVAS), the largest TWA study conducted to date, consecutive patients ( n = 3598, 2164 males) underwent a clinically indicated exercise test with a bicycle ergometer, and their MMA-TWA was analyzed from the pre-cordial leads. The results demonstrated that the maximal TWA monitored from pre-cordial lead V5 was the strongest predictor of cardiovascular mortality and SCD, and higher TWA values indicated greater cardiovascular mortality and SCD risk.
ARVC is a genetically determined disease that often pre-disposes patients to life-threatening ventricular tachyarrhythmias, which may occur as the first manifestation of the disease in young people without previous symptoms. The main goal of ARVC therapy is the prevention of SCD. Previous studies showed that a number of clinical characteristics were associated with an increased risk of major arrhythmic events in patients with ARVC, such as male gender, proband status, cardiac arrest, syncope, LV dysfunction, inducible VT/VF, T-wave inversion in inferior or pre-cordial leads, T-wave inversions beyond right pre-cordial leads (V1–V3), and QRS fragmentation. However, these studies were based on smaller samples of patients who were followed for relatively short periods. In the present study, we explored the predictive value of the previous indices of history of syncope and cardiac arrest, T-wave inversion in inferior or pre-cordial leads, T-wave inversion beyond leads V1 to V3, QRS duration, epsilon waves, inducible VT, and the new index, MTWA. Using Cox regression analysis, only the MaxValt and inducible VT could independently predict major tachyarrhythmic events. Patients with both a MaxValt >23.5 μV and inducible VT had the worst prognosis. The PPV was 85.71%. However, patients with a MaxValt ≤23.5 μV and non-inducible VT had the highest cumulative survival rate, with an NPV of 100%. Therefore, the MTWA test using MMA analysis in this cohort was more powerful for predicting prognosis than the previous indices. Combined with other indices, such as heart rate turbulence or heart rate variability, the MMA-TWA test has been shown to be effective in identifying post-myocardial infarction patients at high risk of arrhythmic events. Furthermore, TWA testing could provide high NPVs for the primary endpoint of arrhythmia-free survival (appropriate ICD therapy, documented VT/VF, or all-cause mortality), as well as programmed ventricular stimulation. Our present study has provided comparable outcomes.
The spectral and MMA methods are the only two techniques approved by the United States Food and Drug Administration to measure microvolt levels of TWA for risk stratification for arrhythmic death. Tests with TWA levels ≥1.9 μV and a signal-to-noise ratio of K = 3 sustained for 2 min are classified as positive in the spectral protocol. However, a high incidence (approximately 20–40%) of “indeterminate” test results often occurs due to many factors, such as excessive ectopy, HR <105 beats/min, unsustained MTWA, or noise. As a promising time-domain analytical technique, MMA-TWA allows TWA analysis during routine exercise stress testing and 24-h ambulatory ECG monitoring without requiring special electrodes or a target heart rate. MMA-based TWA analyses on ambulatory ECG have been found to predict cardiovascular mortality and SCD in thousands of patients with coronary artery disease, myocardial infarction, heart failure, and ischemic and non-ischemic cardiomyopathy with varying extents of LV dysfunction. MMA-TWA has also been found to be elevated in ambulatory ECG recordings in other disease states associated with the risk of ventricular arrhythmias, including sleep apnea patients with congestive heart failure, epilepsy, chronic renal disease, hemodialysis, Brugada syndrome, and long QT syndrome. In the ambulatory ECG-based TWA test, TWA ≥47 μV in standard pre-cordial leads is considered abnormal, and TWA ≥60 μV is considered severely abnormal. These cutoff points are based on algorithms that used an update factor of 1/8. However, MMA-TWA characteristics and cutoff values based on the exercise stress test in patients with ARVC have not been previously reported. In our previous study, we found that TWA values in all pre-cordial leads and the MaxValt in patients with ARVC were much higher than those in healthy controls (MaxValt: 17.0 μV vs . 7 μV), using an update factor value of 1/32. In our present study, the cutoff value was proposed to be >23.5 μV, which was different from that suggested in the study by Chung et al . Chung et al used the protocol of ambulatory ECG-based TWA and an update factor value of 1/8. Their study showed that compared with patients without tachyarrhythmic events, those with tachyarrhythmic events had higher TWA within the modified V5 and V1 channels. These channels suggested a cutoff value of >66 μV, which is approximately two times our cutoff value.
MTWA has been reported to be a strong independent predictor of spontaneous ventricular arrhythmias, death, and programmed ventricular stimulation. MTWA was shown to be better than signal-averaged electrocardiography for the risk stratification of patients with regard to life-threatening arrhythmias. The Alternans Before Cardioverter Defibrillator study (566 patients, median follow-up 1.9 years) was the first clinical trial to use MTWA measured with the spectral method with a graded exercise protocol to guide prophylactic ICD insertion in patients with ischemic cardiomyopathy and low LVEF. The results demonstrated that TWA testing appeared to be comparable to EPSs for guiding ICD implantation and that the two methods may be complementary, especially with a high NPV to identify the subset of patients least likely to benefit from ICD insertion. In our present study, we also combined MMA-TWA and EPS to predict the prognosis in patients with AVRC with preserved LVEF. Our results showed that MaxValt and inducible VT independently predicted life-threatening ventricular tachyarrhythmias. Furthermore, the NPV was 100% when MMA-TWA was combined with EPS in a long-term follow-up study.
Despite a follow-up period as long as 8 years, our study was limited by a small sample size affiliated with a single center. ICD implantation was limited in our recruited patients. Furthermore, in this study, we used an update factor of 1/32 when analyzing TWA, which was less sensitive than the recommended update factor of 1/8. As a result, the TWA values obtained were substantially lower than expected, and the capacity to predict VT might also be reduced. In conclusion, MTWA assessment using MMA analysis complemented with an EPS could improve the ability to predict the prognosis in patients with ARVC with preserved LV function during long-term follow-up.
The authors thank professor Zhao Yang of the Department of Biostatistics, School of Public Health, Nanjing Medical University, for his assistance in statistical analyses.
This work was supported by grants from the National Natural Science Foundation of China (No. 81470457) and the Frontier Technology of Jiangsu Provincial Science and Technology Department (No. BE2016764).
None.
|
2,4-Dimethoxy-6-Methylbenzene-1,3-diol, a Benzenoid From | 7b371cfe-6de4-4b13-9995-50c617fd4aaa | 8162112 | Histology[mh] | Psoriasis is one of the most common autoimmune skin disorders. Patients with psoriasis are characterized by red and thick plaques covered with silver multilayered scales. IL-23/helper T cell type 17 (Th17) axis is recognized to have a key role in psoriasis ( ). Typical histopathology of psoriatic lesions includes epidermal hyperplasia, elongated rete ridge, and immune cell infiltration. The estimated global prevalence of psoriasis is 2%−3% ( ). Approximately 80% of the patients with psoriasis are classified as mild-to-moderate ( ), for which topical drug treatment is feasible. However, the topical treatment is far from satisfactory due to the time-consuming therapeutic course, frustration with efficacy, and side effects ( ). There is an urgent need to develop new antipsoriatic agents with improved therapeutic efficiency and safety. The development of antipsoriatic candidates from natural resources can potentially achieve the purposes of superior therapeutic effectiveness and fewer adverse effects ( ). Nearly 39%−62% patient population in Asia and the Middle East use complementary and alternative medicine for treating psoriasis ( ) And nearly 47% of the patients in South Europe use plant extracts as a remedy against psoriasis ( ). Many natural compounds derived from mushrooms are known to exhibit anti-inflammatory and immunomodulatory activities ( ). The mushroom Antrodia cinnamomea is used as a medicinal herb because of its biological properties. A. cinnamomea is traditionally used to treat diarrhea, abdominal pain, hypertension, cancers, and itchy skin ( ). The extracts and bioactive compounds from A. cinnamomea are reported to show biological effects such as anti-inflammatory, antioxidant, antitumor, antihyperlipidemic, and hepatoprotective activities ( ). The ethanolic extract derived from A. cinnamomea inhibits Th17 cell infiltration in the dermis of the psoriasiform lesion and can thus be a therapeutic option for psoriasis ( ). Previous investigations ( , ) suggest the anti-inflammatory activity of some benzenoids isolated from A. cinnamomea on the activated T cells and macrophages. Likewise, we demonstrated in a previous study ( ) that the benzenoid 2,4-dimethoxy-6-methylbenzene-1,3-diol (DMD) from A. cinnamomea exerts anti-inflammatory activity in atopic dermatitis-like skin in mice. In that study, we mainly explored the therapeutic potential of DMD on psoriasis treatment based upon the cell-based and in vivo animal studies. Psoriasis is generally regarded as a T cell-mediated disease. Nevertheless, there is increasing evidence indicating that macrophages also play an essential role in psoriasis pathogenesis ( ). Macrophages differentiate from monocytes in the blood, enter the host tissue, and are influenced by the local environment. Macrophages are largely infiltrated in the dermal layer of psoriasis to release cytokines IL-23, IL-6, and TNF-α during the development of the lesion ( ). We aimed to explore an effective strategy to treat psoriasis by regulating macrophage activation and to elucidate the possible mechanisms of DMD-mediated inhibition of inflammation using the macrophages (the differentiated THP-1 cells) as the cell model. Imiquimod (IMQ) is a Toll-like receptor (TLR)7 ligand which acts as an immune stimulator for macrophages ( ). We used IMQ to activate macrophages and induce psoriasis-like plaque in mice for evaluating the anti-inflammatory effect of DMD in psoriasis treatment.
Reagents and Antibodies Menadione, tert-butylhydroquinone (TBHQ) and phorbol 12-myristate 13-acetate (PMA) were purchased from Sigma-Aldrich (St. Louis, MO, USA). IMQ cream (Aldara ® , 5%) was acquired from 3M Pharmaceuticals (Leicestershire, UK). Phospho (p)-JNK, p-ERK, p-p38, p-p65, JNK, ERK, p38, CCR7, Drp1 and GAPDH antibodies were purchased from Santa Cruz Biotechnology (Santa Cruz, CA, USA). F4/80 and Ly6G antibodies were purchased from Abcam (Cambridge, MA, USA). The anti-Ly6G antibody was purchased from eBiosciences (San Diego, CA, USA). GDAP1L1 and mtHSP70 antibodies were purchased from Invitrogen (Carlsbad, CA, USA). The antibody targeting phospho-Drp1-S616 was obtained from Biorbyt (St. Louis, MO, USA). DMD From A. cinnamomea DMD was obtained by partitioning and silica gel column chromatography. The detailed information of the extraction and isolation was described earlier ( ). The chemical structure of DMD is illustrated in . Cell Lines, Primary Cells, and Cell Culture Human monocytic leukemia THP-1 cell line was maintained in RPMI 1640 supplemented with 10% heat-inactivated FBS and 100 U/mL penicillin and streptomycin. Before the experiments, THP-1 cells were differentiated into macrophages by treating with phorbol 12-myristate 13-acetate (100 ng/mL) for 36 h, followed by overnight incubation in a fresh medium. Bone marrow was collected from the femur and tibia of BALB/c mice to generate bone marrow-derived macrophages (BMDMs) following an earlier published protocol ( ). For testing the effect of DMD on activated THP-1 and BMDMs, the cells were pretreated with DMD for 1 h and then incubated with IMQ (10 μg/mL) for 24 h. HaCaT cells, the immortalized cell line of human keratinocytes, were cultured in DMEMs supplemented with 10% FBS and 100 U/mL penicillin-streptomycin at 37°C. Human primary neutrophils were obtained from healthy, 20-30 years old volunteers using a protocol approved by the Institutional Review Board at Chang Gung Memorial Hospital (201701925B0). All volunteers provided written informed consent for participation. The neutrophils were purified by sedimentation prior to centrifugation and erythrocyte lysis following the protocol in a previous report ( ). Cytotoxicity Assay The cytotoxicity of DMD was studied by 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) analysis. The macrophages were cultured in DMEM at a density of 2 x 10 6 cells/well and incubated at 37°C for 24 h. Then DMD (1−40 μg/mL) was added into the cell suspension and incubated for 24 h. The cells in a blank medium were used as the control. MTT (0.5 mg/mL) present in culture medium (200 μL) was incorporated in the cell suspension, which was further incubated at 37°C for 4 h. The THP-1 viability was detected by a spectrophotometer at 570 nm. The trypan blue assay was also used to evaluate the cell viability. After treating DMD for 24 h, the cells were removed and stained with 0.4% trypan blue. The number of unstained viable cells was counted in a hemocytometer under light microscope (Leica DMi8). Total RNA Extraction and Real-Time Quantitative Polymerase Chain Reaction (RT-qPCR) Total cellular RNA was extracted using the Direct-zol kit with RNase-free DNase 1 digestion to remove genomic DNA contamination according to the manufacturer’s instructions. Reverse transcription to cDNA was performed by iScript cDNA Synthesis kit. RT-qPCR was carried out by a CFX Connect RT PCR Detection System using iQ SYBR Green Supermix. The level of GAPDH was used to normalize the mRNA level. The primer sequences used for amplification from mouse and human species are listed in and , respectively. Enzyme-Linked Immunosorbent Assay (ELISA) The level of cytokines and chemokines in the supernatant of the cell medium was quantified using ELISA kits (BioLegend) according to the manufacturer’s instructions. The absorbance was measured at 450 nm on a microplate spectrophotometer. The concentration of cytokines and chemokines was estimated based on the corresponding standard curves. Immunoblotting The level of mitogen-activated protein kinases (MAPKs), NF-κB, ganglioside-induced differentiation-associated protein 1 like 1 (GDAP1L1), dynamin-related protein 1 (Drp1), and Drp1 S616 were estimated by western blotting. The cells were collected and added to the lysis buffer. The nuclear pellets were obtained after centrifugation at 400 x g and 4°C for 5 min. After probe sonication, the protein fraction was obtained by centrifugation at 8,000 x g and 4°C for 10 min. For quantification, protein assay dye was mixed with the protein fraction, and separated by 10% acrylamide SDS-PAGE, and transferred to a polyvinylidene difluoride membrane. The membrane was incubated with the primary antibody (1:1000 dilution) at 4°C overnight. Subsequently, the membrane was washed using tris-buffered saline and incubated with horseradish peroxidase-conjugated secondary antibody (1:5000 dilution) for 1 h. The immunoreactive bands were detected by Western Lightning Plus-ECL. Anti-GAPDH or anti-actin antibody was used as the loading control. Wound Healing Assay Fresh neutrophils (4 x 10 5 cells/well) were seeded in a six-well plate and cultured in the complete DMEM medium. The cells were scraped using a 200 μL pipette tip to achieve a noncellular region, and the culture medium was then replaced by the conditioned medium of THP-1 cells after IMQ stimulation with or without DMD intervention (10 μg/mL). After 4 h, the migration number of the neutrophils was measured by using ImageJ software. Chemotaxis Assay The neutrophil migration initiated by the conditioned medium of THP-1 cells was evaluated by the Boyden chamber migration analysis. Briefly, the isolated human neutrophils were added to DMEM supplemented with 0.25% BSA. The neutrophils (4 x 10 5 cells/well) were added to the upper well of the Boyden chamber. The conditioned medium harvested from macrophages was added into the lower well. The plate was stored at 37°C for 4 h before placing it on ice, and 100 μL of 0.5 M EDTA was incorporated into the well at 4°C for 10 min. The well insert was removed and the cell suspension was collected. Then neutrophil count was estimated using a Moxi Z Mini-Automated Cell Counter kit. Isolation of Cytosolic and Mitochondrial Fractions GDAP1L1 and Drp1 levels in cytosol and mitochondria of macrophages were determined by western blotting. The separation of mitochondria from cytoplasm was performed as described previously ( ). In brief, the macrophages were collected, then the Mitochondria Isolation kit was used to separate the cellular components according to the manufacturer’s instructions. Cytoplasmic and mitochondrial proteins were quantified by western blotting using the same method described in the section of immunoblotting. The p38 and mitochondrial HSP70 (mtHSP70) were employed as the loading control for cytosolic and mitochondrial fractions, respectively. Immunofluorescence Staining To appraise the expression of GDAP1L1 and Drp1 S616 in THP-1 cells, immunofluorescence staining was carried out. The cells were cultured on coverslips and then labeled with MitoTracker Red for staining mitochondria. The macrophages were fixed with 4% paraformaldehyde for 15 min at room temperature. The fixed macrophages were permeabilized with Triton X-100 for 5 min and then blocked with 1% BSA for 30 min. For immunofluorescence staining, primary antibodies against GDAP1L1 or Drp1 S616 were used followed by the incubation of secondary antibody conjugated with Alexa Fluor 488 or 594. The cells were monitored under a confocal microscope (Zeiss LSM780). Quantification and measurement of mitochondrial length was performed with MetaMorph (Molecular Devices) by an investigator blinded to experimental groups. Mitochondrial length was classified into three categories; <0.5 μm, 0.5−2 μm, and >2 μm. The resulting data were visualized using GraphPad Prism software. Immunoprecipitation Total mitochondrial fraction extract (100 μg) was incubated with GDAP1L1 antibody overnight at 4°C. Protein A beads were added to the mixtures and incubated for 1 h at 4°C. After five washings with lysis buffer, the immunoprecipitants were subjected to SDS/PAGE and analyzed by immunoblotting as indicated antibodies. Animals Female BALB/c mice (8-weeks old) were purchased from the National Laboratory Animal Center (Taipei, Taiwan). All animal experiments were approved by the Institutional Animal Care and Use Committee of Chang Gung University and complied with Directive 86/109/EEC from the European Commission (CGU108-101). IMQ-Induced Psoriasiform Skin in Mice A psoriasis-like lesion was evoked on the back of the BALB/c mouse following a protocol from van der Fits et al., with modifications ( ). The mouse received a daily topical dose of 62.5 mg 5% IMQ cream (Aldara, 3M) on the shaved area of the dorsal skin for five consecutive days. Before 30 min of IMQ cream treatment, 100 μL DMD (1 mg/mL) in PEG400/PBS (3:7) was topically administered on the back. The vehicle was dried during this 30-min period. The skin surface appearance was visualized using a portable digital magnifier (Mini Scope-V, M&T Optics). The subjective evaluation of the severity was estimated by the cumulative score (scaling plus erythema) with a scale from 0 to 8 based on the Psoriasis Area and Severity Index (PASI). Transepidermal water loss (TEWL) was detected by a Tewameter TM300 (Courage and Khazaka). The animals were sacrificed on day 6 for further histological examination and detection of cytokine/chemokine expression. The skin was extracted following a previously described method ( ). The skin extract was used to determine proinflammatory mediators by RT-qPCR. The protocol for RT-qPCR was the same as that mentioned in the section for the in vitro THP-1 study. Histology The skin sample was fixed in 10% formaldehyde and embedded in paraffin and then stained with hematoxylin and eosin (H&E). The unstained slice of skin sample was prepared for immunohistochemistry (IHC). After dewaxing and rehydration, the skin section was subjected to heat-induced epitope retrieval using the Bond Epitope Retrieval Solution 2 according to the manufacturer’s instructions, followed by the blocking with diluted normal serum. The section was incubated with rabbit polyclonal anti-mouse Ki67, Ly6G, F4/80, GDAP1L1, or Drp1 S616 antibody (1:100 dilution) for 1 h at room temperature, washed with 0.5% Tween 20 in saline, and subsequently incubated with biotinylated donkey anti-rabbit IgG at ambient temperature for 20 min. The photomicrographs were observed under an optical microscope (Leica DMi8). Quantification of the IHC-stained sections was performed by AlphaView software. Each section was examined independently by two investigators in a blinded manner. Numbers of positive cells were evaluated by counting the numbers of cells (original manifestation x200) for three sections of three mice per group. Statistical Analysis The statistical differences in the data of different treatment groups were measured using the one-way analysis of variance followed by Tukey’s multiple comparison test. The data distribution was checked by Kolmogorov-Smirnov test. The levels of probability including 0.05, 0.01, and 0.001 were considered statistically significant.
Menadione, tert-butylhydroquinone (TBHQ) and phorbol 12-myristate 13-acetate (PMA) were purchased from Sigma-Aldrich (St. Louis, MO, USA). IMQ cream (Aldara ® , 5%) was acquired from 3M Pharmaceuticals (Leicestershire, UK). Phospho (p)-JNK, p-ERK, p-p38, p-p65, JNK, ERK, p38, CCR7, Drp1 and GAPDH antibodies were purchased from Santa Cruz Biotechnology (Santa Cruz, CA, USA). F4/80 and Ly6G antibodies were purchased from Abcam (Cambridge, MA, USA). The anti-Ly6G antibody was purchased from eBiosciences (San Diego, CA, USA). GDAP1L1 and mtHSP70 antibodies were purchased from Invitrogen (Carlsbad, CA, USA). The antibody targeting phospho-Drp1-S616 was obtained from Biorbyt (St. Louis, MO, USA).
A. cinnamomea DMD was obtained by partitioning and silica gel column chromatography. The detailed information of the extraction and isolation was described earlier ( ). The chemical structure of DMD is illustrated in .
Human monocytic leukemia THP-1 cell line was maintained in RPMI 1640 supplemented with 10% heat-inactivated FBS and 100 U/mL penicillin and streptomycin. Before the experiments, THP-1 cells were differentiated into macrophages by treating with phorbol 12-myristate 13-acetate (100 ng/mL) for 36 h, followed by overnight incubation in a fresh medium. Bone marrow was collected from the femur and tibia of BALB/c mice to generate bone marrow-derived macrophages (BMDMs) following an earlier published protocol ( ). For testing the effect of DMD on activated THP-1 and BMDMs, the cells were pretreated with DMD for 1 h and then incubated with IMQ (10 μg/mL) for 24 h. HaCaT cells, the immortalized cell line of human keratinocytes, were cultured in DMEMs supplemented with 10% FBS and 100 U/mL penicillin-streptomycin at 37°C. Human primary neutrophils were obtained from healthy, 20-30 years old volunteers using a protocol approved by the Institutional Review Board at Chang Gung Memorial Hospital (201701925B0). All volunteers provided written informed consent for participation. The neutrophils were purified by sedimentation prior to centrifugation and erythrocyte lysis following the protocol in a previous report ( ).
The cytotoxicity of DMD was studied by 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) analysis. The macrophages were cultured in DMEM at a density of 2 x 10 6 cells/well and incubated at 37°C for 24 h. Then DMD (1−40 μg/mL) was added into the cell suspension and incubated for 24 h. The cells in a blank medium were used as the control. MTT (0.5 mg/mL) present in culture medium (200 μL) was incorporated in the cell suspension, which was further incubated at 37°C for 4 h. The THP-1 viability was detected by a spectrophotometer at 570 nm. The trypan blue assay was also used to evaluate the cell viability. After treating DMD for 24 h, the cells were removed and stained with 0.4% trypan blue. The number of unstained viable cells was counted in a hemocytometer under light microscope (Leica DMi8).
Total cellular RNA was extracted using the Direct-zol kit with RNase-free DNase 1 digestion to remove genomic DNA contamination according to the manufacturer’s instructions. Reverse transcription to cDNA was performed by iScript cDNA Synthesis kit. RT-qPCR was carried out by a CFX Connect RT PCR Detection System using iQ SYBR Green Supermix. The level of GAPDH was used to normalize the mRNA level. The primer sequences used for amplification from mouse and human species are listed in and , respectively.
The level of cytokines and chemokines in the supernatant of the cell medium was quantified using ELISA kits (BioLegend) according to the manufacturer’s instructions. The absorbance was measured at 450 nm on a microplate spectrophotometer. The concentration of cytokines and chemokines was estimated based on the corresponding standard curves.
The level of mitogen-activated protein kinases (MAPKs), NF-κB, ganglioside-induced differentiation-associated protein 1 like 1 (GDAP1L1), dynamin-related protein 1 (Drp1), and Drp1 S616 were estimated by western blotting. The cells were collected and added to the lysis buffer. The nuclear pellets were obtained after centrifugation at 400 x g and 4°C for 5 min. After probe sonication, the protein fraction was obtained by centrifugation at 8,000 x g and 4°C for 10 min. For quantification, protein assay dye was mixed with the protein fraction, and separated by 10% acrylamide SDS-PAGE, and transferred to a polyvinylidene difluoride membrane. The membrane was incubated with the primary antibody (1:1000 dilution) at 4°C overnight. Subsequently, the membrane was washed using tris-buffered saline and incubated with horseradish peroxidase-conjugated secondary antibody (1:5000 dilution) for 1 h. The immunoreactive bands were detected by Western Lightning Plus-ECL. Anti-GAPDH or anti-actin antibody was used as the loading control.
Fresh neutrophils (4 x 10 5 cells/well) were seeded in a six-well plate and cultured in the complete DMEM medium. The cells were scraped using a 200 μL pipette tip to achieve a noncellular region, and the culture medium was then replaced by the conditioned medium of THP-1 cells after IMQ stimulation with or without DMD intervention (10 μg/mL). After 4 h, the migration number of the neutrophils was measured by using ImageJ software.
The neutrophil migration initiated by the conditioned medium of THP-1 cells was evaluated by the Boyden chamber migration analysis. Briefly, the isolated human neutrophils were added to DMEM supplemented with 0.25% BSA. The neutrophils (4 x 10 5 cells/well) were added to the upper well of the Boyden chamber. The conditioned medium harvested from macrophages was added into the lower well. The plate was stored at 37°C for 4 h before placing it on ice, and 100 μL of 0.5 M EDTA was incorporated into the well at 4°C for 10 min. The well insert was removed and the cell suspension was collected. Then neutrophil count was estimated using a Moxi Z Mini-Automated Cell Counter kit.
GDAP1L1 and Drp1 levels in cytosol and mitochondria of macrophages were determined by western blotting. The separation of mitochondria from cytoplasm was performed as described previously ( ). In brief, the macrophages were collected, then the Mitochondria Isolation kit was used to separate the cellular components according to the manufacturer’s instructions. Cytoplasmic and mitochondrial proteins were quantified by western blotting using the same method described in the section of immunoblotting. The p38 and mitochondrial HSP70 (mtHSP70) were employed as the loading control for cytosolic and mitochondrial fractions, respectively.
To appraise the expression of GDAP1L1 and Drp1 S616 in THP-1 cells, immunofluorescence staining was carried out. The cells were cultured on coverslips and then labeled with MitoTracker Red for staining mitochondria. The macrophages were fixed with 4% paraformaldehyde for 15 min at room temperature. The fixed macrophages were permeabilized with Triton X-100 for 5 min and then blocked with 1% BSA for 30 min. For immunofluorescence staining, primary antibodies against GDAP1L1 or Drp1 S616 were used followed by the incubation of secondary antibody conjugated with Alexa Fluor 488 or 594. The cells were monitored under a confocal microscope (Zeiss LSM780). Quantification and measurement of mitochondrial length was performed with MetaMorph (Molecular Devices) by an investigator blinded to experimental groups. Mitochondrial length was classified into three categories; <0.5 μm, 0.5−2 μm, and >2 μm. The resulting data were visualized using GraphPad Prism software.
Total mitochondrial fraction extract (100 μg) was incubated with GDAP1L1 antibody overnight at 4°C. Protein A beads were added to the mixtures and incubated for 1 h at 4°C. After five washings with lysis buffer, the immunoprecipitants were subjected to SDS/PAGE and analyzed by immunoblotting as indicated antibodies.
Female BALB/c mice (8-weeks old) were purchased from the National Laboratory Animal Center (Taipei, Taiwan). All animal experiments were approved by the Institutional Animal Care and Use Committee of Chang Gung University and complied with Directive 86/109/EEC from the European Commission (CGU108-101).
A psoriasis-like lesion was evoked on the back of the BALB/c mouse following a protocol from van der Fits et al., with modifications ( ). The mouse received a daily topical dose of 62.5 mg 5% IMQ cream (Aldara, 3M) on the shaved area of the dorsal skin for five consecutive days. Before 30 min of IMQ cream treatment, 100 μL DMD (1 mg/mL) in PEG400/PBS (3:7) was topically administered on the back. The vehicle was dried during this 30-min period. The skin surface appearance was visualized using a portable digital magnifier (Mini Scope-V, M&T Optics). The subjective evaluation of the severity was estimated by the cumulative score (scaling plus erythema) with a scale from 0 to 8 based on the Psoriasis Area and Severity Index (PASI). Transepidermal water loss (TEWL) was detected by a Tewameter TM300 (Courage and Khazaka). The animals were sacrificed on day 6 for further histological examination and detection of cytokine/chemokine expression. The skin was extracted following a previously described method ( ). The skin extract was used to determine proinflammatory mediators by RT-qPCR. The protocol for RT-qPCR was the same as that mentioned in the section for the in vitro THP-1 study.
The skin sample was fixed in 10% formaldehyde and embedded in paraffin and then stained with hematoxylin and eosin (H&E). The unstained slice of skin sample was prepared for immunohistochemistry (IHC). After dewaxing and rehydration, the skin section was subjected to heat-induced epitope retrieval using the Bond Epitope Retrieval Solution 2 according to the manufacturer’s instructions, followed by the blocking with diluted normal serum. The section was incubated with rabbit polyclonal anti-mouse Ki67, Ly6G, F4/80, GDAP1L1, or Drp1 S616 antibody (1:100 dilution) for 1 h at room temperature, washed with 0.5% Tween 20 in saline, and subsequently incubated with biotinylated donkey anti-rabbit IgG at ambient temperature for 20 min. The photomicrographs were observed under an optical microscope (Leica DMi8). Quantification of the IHC-stained sections was performed by AlphaView software. Each section was examined independently by two investigators in a blinded manner. Numbers of positive cells were evaluated by counting the numbers of cells (original manifestation x200) for three sections of three mice per group.
The statistical differences in the data of different treatment groups were measured using the one-way analysis of variance followed by Tukey’s multiple comparison test. The data distribution was checked by Kolmogorov-Smirnov test. The levels of probability including 0.05, 0.01, and 0.001 were considered statistically significant.
DMD Inhibits Cytokine/Chemokine Expression in Activated Macrophages IMQ was used as the activator to evoke the macrophage stimulation in this study. We tested the effect of IMQ on THP-1 viability to determine its cytotoxicity to the macrophages. Although slight cytotoxicity determined by MTT assay was found after IMQ treatment, the viability could still surpass 80% for the activator concentrations between 1 and 20 μg/mL (the left panel of ). In addition to the evaluation of cell viability, MTT assay is a method to assess cell metabolic activity. The MTT assay is dependent on the mitochondrial respiration. The slight reduction of MTT by IMQ may suggest the involvement of IMQ in the metabolic activity of THP-1 cells. No cell viability reduction by IMQ was detected using trypan blue analysis (the right panel of ), demonstrating a minimal cytotoxicity of IMQ. Then, an IMQ dose of 10 μg/mL was used in the subsequent experiments. Up to a concentration of 20 μg/mL, DMD had no effect on the viability for both MTT and trypan blue analyses ( ), while following the treatment with 40 μg/mL DMD the viability percentage was 94% in the MTT assay. The non-cytotoxic dose of 10 μg/mL DMD was selected to investigate the anti-inflammatory effect on macrophages. To further elucidate the role of DMD on the cytokine expression, RT-qPCR analysis was carried out for the five genes of IL-23, IL-6, IL-24, TNF, and IL-8. IMQ was able to increase the expression of these inflammatory factors with a statistical significance ( ). The mRNA detection verified a decrease in cytokine level after DMD versus IMQ stimulation. At 10 μg/mL, DMD totally inhibited IL-23, IL-6, IL-24, and TNF to the baseline control. The expression of the cytokines in macrophages was further confirmed by ELISA. The protein and mRNA expression showed a consistent trend ( ). BMDMs, the primary macrophages derived from the mouse, were used as another macrophage model for treating the effect of DMD on inflammation attenuation. Similar to THP-1 cells, IMQ induced some cytotoxicity against BMDMs ( ). DMD produced a limited cytotoxicity on BMDMs. The viability of BMDMs could be maintained to 75% after DMD treatment at the highest concentration (40 μg/mL). In vitro , BMDMs also expressed IL-23 and IL-6 at high levels after IMQ stimulation ( ), and this upregulation could be reduced in the DMD-treated groups. To understand the capacity of chemokines, we analyzed mRNA expression of nine chemokines in IMQ-activated macrophages and observed a high level of CCR7, CCL1, CCL2, CCL4L2, CCL20, CCL22, CXCL3, CXCL6, and CXCR5 in stimulated THP-1 ( ), and this could be abolished by DMD treatment. This result suggests the involvement of DMD in the regulation of macrophage chemotaxis. We also observed upregulated GDAP1L1 mRNA by IMQ and a 70-fold decrease to that of the baseline after DMD treatment ( ). GDAP1L1 is a paralog of GDAP1 and predominantly governs mitochondrial dynamics. DMD Inhibits Cytokine/Chemokine Expression by Blocking the MAPK/NF-κB Signaling To elucidate the signaling pathways through which DMD attenuated the inflammation, we checked the possible role of NF-κB. Western blot analysis displayed that IMQ upregulated the expression of p-p65 that was a subunit of p-NF-κB ( ), while a significantly opposite effect was observed after DMD treatment in macrophages. DMD inhibited the phosphorylation of p65 by 2-fold as measured by densitometry ( ). To examine the possible role of MAPKs in the inflammation suppression by DMD, we studied the IMQ-induced phosphorylation of JNK, ERK, and p38 in macrophages. The maximum expression of the phosphorylated MAPKs was detected at 2 h after IMQ stimulation ( and ), while DMD could reverse the phosphorylation, clearly indicating the anti-inflammatory activity of DMD through the downregulation of MAPKs and NF-κB. GDAP1L1 is an important marker that showed a significant decrease after DMD treatment of activated THP-1 macrophages. Immunoblotting revealed that GDAP1L1 protein level increased after IMQ and arrested by DMD treatments in a dose-dependent manner ( and ). The Conditioned Medium of DMD-Treated Macrophages Prevents Neutrophil Migration and Keratinocyte Proliferation In psoriasis, macrophage chemotaxis may mediate the interplay between the macrophages and the other cells. To gain an in-depth understanding of the chemotaxis, we validated the effect of IMQ-activated macrophage conditioned medium on neutrophil invasion and keratinocyte proliferation. The wound-healing assay showed that DMD-treated macrophage medium suppressed wound closure in neutrophils ( ). Moreover, the DMD-treated group had lowered the average number of migrating neutrophils permeating the transwell membrane ( ). The conditioned medium from IMQ-stimulated macrophages promoted keratinocyte proliferation by about 2-fold as measured by both MTT and trypan blue assays ( ). This effect was inhibited by DMD at 10 μg/mL. The keratinocyte proliferation in psoriasis is in close association with the signal transducer and activator of transcription 3 (STAT3) pathway. The IMQ-treated conditioned medium enhanced STAT3 activation in keratinocytes, which was observed in the form of increased p-STAT3 ( and ), which was considerably reduced upon the intervention of DMD. Our data demonstrated that DMD was effective in suppressing neutrophil and keratinocyte activation via the inhibition of macrophage chemotaxis. GDAP1L1/Drp1 Axis Participates in DMD-Induced Inhibition on Macrophage Activation According to the cell-based study, we found a possible role of GDAP1L1 in the inhibition of activated macrophage by DMD. GDAP1L1 is a mitochondrial fission factor. Mitochondrial fission and fusion are required for maintaining mitochondrial functions related to biogenesis, apoptosis, neurodegeneration, and inflammation. Following IMQ stimulation, the GDAP1L1 level was significantly increased at the indicated times as verified by densitometry ( and ). DMD treatment at 10 μg/mL blocked IMQ-stimulated GDAP1L1 expression. The activity of GDAP1L1 to induce mitochondrial fission depends upon another fission factor Drp1. The fission needs the phosphorylated Drp1 at the S616 site and translocation of Drp1 from the cytoplasm to mitochondria. We found an increase in IMQ-mediated Drp1 phosphorylation in THP-1 ( ), and this elevation could be reduced by DMD. GDAP1L1 is a cytosolic protein that is also delivered to the mitochondria in response to external stimuli. We found the association between GDAP1L1 and mitochondria after IMQ treatment ( ), which increased following the increase in treatment duration. To appraise whether DMD affected GDAP1L1 and Drp1 location, cytosol and mitochondria were isolated from IMQ-treated macrophages and more GDAP1L1 was expressed in the mitochondria after IMQ stimulation ( ). We found a minimal expression of the cytosolic specific marker p38 in the mitochondrial fraction. Menadione is a quinone-related derivative used as a positive control in oxidative stress-induction studies. Mitochondrial translocation of GDAP1L1 was found after menadione treatment. DMD not only reduced GDAP1L1 production but also decreased IMQ- and menadione-induced GDAP1L1 translocation. Menadione is conjugated to form menadione-S-glutathione by glutathione S-transferase to induce stress. Glutathione conjugation detoxifies tertiary butylhydroquinone (TBHQ). TBHQ could reduce the GDAP1L1 transfer to mitochondria sensitized by menadione. IMQ and menadione also strongly increased the expression of Drp1 in mitochondria without a significant change of total Drp1. DMD treatment abolished the elevated GDAP1L1 in mitochondria. However, this phenomenon was not detected in the case of menadione-treated cells intervened by TBHQ. We then examined the phosphorylated Drp1 at S616 in mitochondria. The level of mitochondrial Drp1 and Drp1 S616 increased in THP-1 stimulated with IMQ and menadione ( ) and this increase in the active form of Drp1 in mitochondria was blocked by DMD, as was confirmed by immunofluorescence analysis (the left panel of ). Colocalization of Drp1 S616 with mitochondria was observed using double immunofluorescence staining. The estimation of mitochondrial length showed a decreased average length after IMQ intervention (the upper right panel of ), suggesting a fission of mitochondria after an inflammation stimulation. DMD treatment could reverse this reduction to the control baseline level. We also found an increased proportion of mitochondria with short length (<0.5 μm) by IMQ stimulation (the lower right panel of ). Again, this increase could be reversed to baseline control by DMD. Based on the above-mentioned results, we speculated that Drp1 might complex with GDAP1L1. We employed immunoprecipitation to check if GDAP1L1 and Drp1 were associated to interact with each other. The GDAP1L1 antibody could successfully pull down both GDAP1L1 and Drp1 ( ). GDAP1L1 indeed coprecipitated with Drp1 when transiently expressed in THP-1 cells. Both Drp1 S616 and GDAP1L1 expression was upregulated in mitochondria after IMQ stimulation, as shown by the immunofluorescence image ( ). An appreciable overlap was observed between both fission factors and mitochondria. GDAP1L1 possibly colocalized with phosphorylated Drp1 in mitochondria. We verified that the DMD-induced inflammation suppression was mediated by regulating GDAP1L1/Drp1 translocation to mitochondria. DMD Alleviates Psoriasiform Lesion in IMQ-Induced Mouse Model IMQ was topically applied on mouse skin to generate psoriasis-like plaque for evaluating the anti-inflammatory activity of DMD. The pure compound from A. cinnamomea was topically administered on the skin of the back treated with IMQ cream for five consecutive days ( ). Compared to the healthy skin, scaling, erythema, and thickening in the IMQ-treated skin were observed in the macroscopic and microscopic visualization of the skin surface ( ). These symptoms suggest a typical psoriasis feature. DMD-treated groups displayed the clearance of these signs compared to the only IMQ-treated group. The cumulative score estimated by scaling and redness was significantly reduced by DMD as compared to the only IMQ treatment group ( ). TEWL, as an indicator of cutaneous barrier property, exhibited a 4-fold increase after IMQ stimulation ( ). No improvement of skin barrier dysfunction was detected after DMD intervention. The epidermal thickness increased by about 3-fold by IMQ activation as compared to normal skin ( ). This histological sign was relieved by DMD. The epidermal thickness of IMQ-treated skin was reduced from 77 to 50 μm after DMD treatment. The Munro’s microabscess in IMQ-stimulated lesion was reduced by 6.8-fold after treatment of DMD ( ). In vivo efficacy of DMD on psoriasiform lesions was qualitatively monitored using histology. Representative H&E staining from healthy skin revealed normal morphology with no damage to the epidermis and dermis ( ). H&E-stained IMQ-treated skin showed typical histological hallmarks of psoriasis, including acanthosis, hyperkeratosis, elongated rete ridge, and immune cell infiltration. IHC staining for Ki67, a proliferation biomarker revealed a remarkable increase of Ki67 in the basal layer after IMQ treatment (red arrows in ). The higher power magnification of the histology is revealed in the upper left corner of each image. DMD effectively suppressed epidermal hyperproliferation in mice. The neutrophil infiltration in the stratum corneum and dermis could be visualized by Ly6G staining (red arrows in ). The IHC showed clouds of Ly6G expression in the dermis after the IMQ challenge. In addition, DMD inhibited infiltrating neutrophils to a certain level. To detect macrophages, the skin was immunostained with F4/80 and its expression was observed in epidermal and dermal layers in IMQ-treated mouse (red arrows in ), suggesting that macrophages were recruited to the lesional plaque. DMD could mitigate the macrophage infiltration. Strong immunoreactivity for GDAP1L1 and Drp1 S616 was observed in the psoriasiform lesion (red arrows in ). Like the macrophage distribution, GDAP1L1 mainly expressed throughout the viable epidermis and dermis. Expression of both fission factors decreased after and DMD-treatment of the skin compared to IMQ stimulation alone. All quantification of the antibody-positive cell count was estimated by AlphaView software and depicted in the right panel of . A statistically significant reduction of antibody-positive cell count was observed after topical administration of DMD on IMQ-treated mouse skin. The mRNA level of the inflammation-related cytokines and chemokines including IL-23, IL-6, IL-17A, IL-24, TNF, and CXCL2 in mouse skin was determined through RT-qPCR and a dramatic increase in proinflammatory mediator expression was observed after the IMQ challenge ( ). The cytokines/chemokines in psoriasiform skin could be restrained to baseline control by DMD. These data clearly demonstrate the ability of DMD to reduce inflammation caused by IMQ. F4/80 mRNA level was significantly increased in IMQ-activated mouse skin and decreased after DMD application ( ). Collectively, the in vivo data prove the anti-inflammatory potential of DMD on the psoriasis-like lesions, especially on macrophages and IL-23/Th17 signaling.
IMQ was used as the activator to evoke the macrophage stimulation in this study. We tested the effect of IMQ on THP-1 viability to determine its cytotoxicity to the macrophages. Although slight cytotoxicity determined by MTT assay was found after IMQ treatment, the viability could still surpass 80% for the activator concentrations between 1 and 20 μg/mL (the left panel of ). In addition to the evaluation of cell viability, MTT assay is a method to assess cell metabolic activity. The MTT assay is dependent on the mitochondrial respiration. The slight reduction of MTT by IMQ may suggest the involvement of IMQ in the metabolic activity of THP-1 cells. No cell viability reduction by IMQ was detected using trypan blue analysis (the right panel of ), demonstrating a minimal cytotoxicity of IMQ. Then, an IMQ dose of 10 μg/mL was used in the subsequent experiments. Up to a concentration of 20 μg/mL, DMD had no effect on the viability for both MTT and trypan blue analyses ( ), while following the treatment with 40 μg/mL DMD the viability percentage was 94% in the MTT assay. The non-cytotoxic dose of 10 μg/mL DMD was selected to investigate the anti-inflammatory effect on macrophages. To further elucidate the role of DMD on the cytokine expression, RT-qPCR analysis was carried out for the five genes of IL-23, IL-6, IL-24, TNF, and IL-8. IMQ was able to increase the expression of these inflammatory factors with a statistical significance ( ). The mRNA detection verified a decrease in cytokine level after DMD versus IMQ stimulation. At 10 μg/mL, DMD totally inhibited IL-23, IL-6, IL-24, and TNF to the baseline control. The expression of the cytokines in macrophages was further confirmed by ELISA. The protein and mRNA expression showed a consistent trend ( ). BMDMs, the primary macrophages derived from the mouse, were used as another macrophage model for treating the effect of DMD on inflammation attenuation. Similar to THP-1 cells, IMQ induced some cytotoxicity against BMDMs ( ). DMD produced a limited cytotoxicity on BMDMs. The viability of BMDMs could be maintained to 75% after DMD treatment at the highest concentration (40 μg/mL). In vitro , BMDMs also expressed IL-23 and IL-6 at high levels after IMQ stimulation ( ), and this upregulation could be reduced in the DMD-treated groups. To understand the capacity of chemokines, we analyzed mRNA expression of nine chemokines in IMQ-activated macrophages and observed a high level of CCR7, CCL1, CCL2, CCL4L2, CCL20, CCL22, CXCL3, CXCL6, and CXCR5 in stimulated THP-1 ( ), and this could be abolished by DMD treatment. This result suggests the involvement of DMD in the regulation of macrophage chemotaxis. We also observed upregulated GDAP1L1 mRNA by IMQ and a 70-fold decrease to that of the baseline after DMD treatment ( ). GDAP1L1 is a paralog of GDAP1 and predominantly governs mitochondrial dynamics.
To elucidate the signaling pathways through which DMD attenuated the inflammation, we checked the possible role of NF-κB. Western blot analysis displayed that IMQ upregulated the expression of p-p65 that was a subunit of p-NF-κB ( ), while a significantly opposite effect was observed after DMD treatment in macrophages. DMD inhibited the phosphorylation of p65 by 2-fold as measured by densitometry ( ). To examine the possible role of MAPKs in the inflammation suppression by DMD, we studied the IMQ-induced phosphorylation of JNK, ERK, and p38 in macrophages. The maximum expression of the phosphorylated MAPKs was detected at 2 h after IMQ stimulation ( and ), while DMD could reverse the phosphorylation, clearly indicating the anti-inflammatory activity of DMD through the downregulation of MAPKs and NF-κB. GDAP1L1 is an important marker that showed a significant decrease after DMD treatment of activated THP-1 macrophages. Immunoblotting revealed that GDAP1L1 protein level increased after IMQ and arrested by DMD treatments in a dose-dependent manner ( and ).
In psoriasis, macrophage chemotaxis may mediate the interplay between the macrophages and the other cells. To gain an in-depth understanding of the chemotaxis, we validated the effect of IMQ-activated macrophage conditioned medium on neutrophil invasion and keratinocyte proliferation. The wound-healing assay showed that DMD-treated macrophage medium suppressed wound closure in neutrophils ( ). Moreover, the DMD-treated group had lowered the average number of migrating neutrophils permeating the transwell membrane ( ). The conditioned medium from IMQ-stimulated macrophages promoted keratinocyte proliferation by about 2-fold as measured by both MTT and trypan blue assays ( ). This effect was inhibited by DMD at 10 μg/mL. The keratinocyte proliferation in psoriasis is in close association with the signal transducer and activator of transcription 3 (STAT3) pathway. The IMQ-treated conditioned medium enhanced STAT3 activation in keratinocytes, which was observed in the form of increased p-STAT3 ( and ), which was considerably reduced upon the intervention of DMD. Our data demonstrated that DMD was effective in suppressing neutrophil and keratinocyte activation via the inhibition of macrophage chemotaxis.
According to the cell-based study, we found a possible role of GDAP1L1 in the inhibition of activated macrophage by DMD. GDAP1L1 is a mitochondrial fission factor. Mitochondrial fission and fusion are required for maintaining mitochondrial functions related to biogenesis, apoptosis, neurodegeneration, and inflammation. Following IMQ stimulation, the GDAP1L1 level was significantly increased at the indicated times as verified by densitometry ( and ). DMD treatment at 10 μg/mL blocked IMQ-stimulated GDAP1L1 expression. The activity of GDAP1L1 to induce mitochondrial fission depends upon another fission factor Drp1. The fission needs the phosphorylated Drp1 at the S616 site and translocation of Drp1 from the cytoplasm to mitochondria. We found an increase in IMQ-mediated Drp1 phosphorylation in THP-1 ( ), and this elevation could be reduced by DMD. GDAP1L1 is a cytosolic protein that is also delivered to the mitochondria in response to external stimuli. We found the association between GDAP1L1 and mitochondria after IMQ treatment ( ), which increased following the increase in treatment duration. To appraise whether DMD affected GDAP1L1 and Drp1 location, cytosol and mitochondria were isolated from IMQ-treated macrophages and more GDAP1L1 was expressed in the mitochondria after IMQ stimulation ( ). We found a minimal expression of the cytosolic specific marker p38 in the mitochondrial fraction. Menadione is a quinone-related derivative used as a positive control in oxidative stress-induction studies. Mitochondrial translocation of GDAP1L1 was found after menadione treatment. DMD not only reduced GDAP1L1 production but also decreased IMQ- and menadione-induced GDAP1L1 translocation. Menadione is conjugated to form menadione-S-glutathione by glutathione S-transferase to induce stress. Glutathione conjugation detoxifies tertiary butylhydroquinone (TBHQ). TBHQ could reduce the GDAP1L1 transfer to mitochondria sensitized by menadione. IMQ and menadione also strongly increased the expression of Drp1 in mitochondria without a significant change of total Drp1. DMD treatment abolished the elevated GDAP1L1 in mitochondria. However, this phenomenon was not detected in the case of menadione-treated cells intervened by TBHQ. We then examined the phosphorylated Drp1 at S616 in mitochondria. The level of mitochondrial Drp1 and Drp1 S616 increased in THP-1 stimulated with IMQ and menadione ( ) and this increase in the active form of Drp1 in mitochondria was blocked by DMD, as was confirmed by immunofluorescence analysis (the left panel of ). Colocalization of Drp1 S616 with mitochondria was observed using double immunofluorescence staining. The estimation of mitochondrial length showed a decreased average length after IMQ intervention (the upper right panel of ), suggesting a fission of mitochondria after an inflammation stimulation. DMD treatment could reverse this reduction to the control baseline level. We also found an increased proportion of mitochondria with short length (<0.5 μm) by IMQ stimulation (the lower right panel of ). Again, this increase could be reversed to baseline control by DMD. Based on the above-mentioned results, we speculated that Drp1 might complex with GDAP1L1. We employed immunoprecipitation to check if GDAP1L1 and Drp1 were associated to interact with each other. The GDAP1L1 antibody could successfully pull down both GDAP1L1 and Drp1 ( ). GDAP1L1 indeed coprecipitated with Drp1 when transiently expressed in THP-1 cells. Both Drp1 S616 and GDAP1L1 expression was upregulated in mitochondria after IMQ stimulation, as shown by the immunofluorescence image ( ). An appreciable overlap was observed between both fission factors and mitochondria. GDAP1L1 possibly colocalized with phosphorylated Drp1 in mitochondria. We verified that the DMD-induced inflammation suppression was mediated by regulating GDAP1L1/Drp1 translocation to mitochondria.
IMQ was topically applied on mouse skin to generate psoriasis-like plaque for evaluating the anti-inflammatory activity of DMD. The pure compound from A. cinnamomea was topically administered on the skin of the back treated with IMQ cream for five consecutive days ( ). Compared to the healthy skin, scaling, erythema, and thickening in the IMQ-treated skin were observed in the macroscopic and microscopic visualization of the skin surface ( ). These symptoms suggest a typical psoriasis feature. DMD-treated groups displayed the clearance of these signs compared to the only IMQ-treated group. The cumulative score estimated by scaling and redness was significantly reduced by DMD as compared to the only IMQ treatment group ( ). TEWL, as an indicator of cutaneous barrier property, exhibited a 4-fold increase after IMQ stimulation ( ). No improvement of skin barrier dysfunction was detected after DMD intervention. The epidermal thickness increased by about 3-fold by IMQ activation as compared to normal skin ( ). This histological sign was relieved by DMD. The epidermal thickness of IMQ-treated skin was reduced from 77 to 50 μm after DMD treatment. The Munro’s microabscess in IMQ-stimulated lesion was reduced by 6.8-fold after treatment of DMD ( ). In vivo efficacy of DMD on psoriasiform lesions was qualitatively monitored using histology. Representative H&E staining from healthy skin revealed normal morphology with no damage to the epidermis and dermis ( ). H&E-stained IMQ-treated skin showed typical histological hallmarks of psoriasis, including acanthosis, hyperkeratosis, elongated rete ridge, and immune cell infiltration. IHC staining for Ki67, a proliferation biomarker revealed a remarkable increase of Ki67 in the basal layer after IMQ treatment (red arrows in ). The higher power magnification of the histology is revealed in the upper left corner of each image. DMD effectively suppressed epidermal hyperproliferation in mice. The neutrophil infiltration in the stratum corneum and dermis could be visualized by Ly6G staining (red arrows in ). The IHC showed clouds of Ly6G expression in the dermis after the IMQ challenge. In addition, DMD inhibited infiltrating neutrophils to a certain level. To detect macrophages, the skin was immunostained with F4/80 and its expression was observed in epidermal and dermal layers in IMQ-treated mouse (red arrows in ), suggesting that macrophages were recruited to the lesional plaque. DMD could mitigate the macrophage infiltration. Strong immunoreactivity for GDAP1L1 and Drp1 S616 was observed in the psoriasiform lesion (red arrows in ). Like the macrophage distribution, GDAP1L1 mainly expressed throughout the viable epidermis and dermis. Expression of both fission factors decreased after and DMD-treatment of the skin compared to IMQ stimulation alone. All quantification of the antibody-positive cell count was estimated by AlphaView software and depicted in the right panel of . A statistically significant reduction of antibody-positive cell count was observed after topical administration of DMD on IMQ-treated mouse skin. The mRNA level of the inflammation-related cytokines and chemokines including IL-23, IL-6, IL-17A, IL-24, TNF, and CXCL2 in mouse skin was determined through RT-qPCR and a dramatic increase in proinflammatory mediator expression was observed after the IMQ challenge ( ). The cytokines/chemokines in psoriasiform skin could be restrained to baseline control by DMD. These data clearly demonstrate the ability of DMD to reduce inflammation caused by IMQ. F4/80 mRNA level was significantly increased in IMQ-activated mouse skin and decreased after DMD application ( ). Collectively, the in vivo data prove the anti-inflammatory potential of DMD on the psoriasis-like lesions, especially on macrophages and IL-23/Th17 signaling.
Macrophages are thought to have a vital role in the induction of inflammation and autoimmune skin diseases such as psoriasis ( , ). Therefore, we employed macrophages as the main cell model for evaluating the effect of DMD on the attenuation of psoriasis. The cutaneous macrophages in psoriatic plaque can release cytokines IL-6, TNF, and IL-1β for developing inflammation ( ). Cytokines influence cell proliferation in psoriasis ( ). Multiple cytokine-signaling including IL-23, IL-6, IL-24, and TNF is known to be important in psoriatic pathogenesis ( ). The TLR stimulated by IMQ can initiate macrophage activation to express proinflammatory cytokines ( ). Among these cytokines, IL-23 plays a preliminary role in macrophages to initiate psoriasis development ( ). IL-23 is important for inducing IL-6-dependent epidermal hyperplasia and cytokine production in psoriasis. DMD could restrain the upregulation of IL-23 in the activated macrophages in the early stage, followed by the inhibition of IL-6 expression. In the psoriatic skin, one of the major sources of TNF is macrophage ( , ). Upregulation of TNF was also largely inhibited by DMD. Psoriasis development can be triggered by TNF receptor 1-dependent upregulation of IL-24 ( ). IL-24 is a member of the IL-20 family induced by IL-6, TNF, and IL-1β in psoriasis. Macrophages and T lymphocytes are the primary cells expressing IL-24 ( ). Here, we confirmed the repression of IL-24 by DMD. The cytokines including IL-24 and TNF-α mediate the crosstalk between immune cells and keratinocytes. The release of these cytokines stimulates keratinocyte proliferation and amplifies the inflammation in the psoriatic lesions ( ). Macrophages accumulate in the psoriatic lesions and release cytokines to prompt the excessive proliferation of keratinocytes ( ). We found that the conditioned medium of DMD-treated THP-1 could successfully inhibit keratinocyte proliferation and p-STAT3 expression. Hyperproliferation of keratinocytes in psoriasis correlates with the STAT3 pathway ( ). STAT3 activation in keratinocytes is regarded as the main effector producing cytokines, which in turn leads to the positive feedback loop of psoriasis. DMD not only suppressed macrophage activation but also restrained the following interplay between macrophages and keratinocytes to restrict inflammation. Another key factor generating the crosstalk between immune cells and keratinocytes is chemotaxis. Immune cell recruitment in psoriatic lesions depends on the chemotactic proteins, including chemokines and their receptors ( ). The function of chemokines and cytokines shows some overlaps, but chemokines are primarily recognized for their capacity to modulate cell migration. Among the chemokines, CCL20 is a major contributor to immune cell infiltration in psoriasis ( ). CCL20 expression is largely increased by TNF-α, IL-24, and IFN-γ ( , ). CCL20 attracts dendritic cells and Th17 cells to sustain the inflammatory response through a feedback loop ( ). DMD was found to significantly suppress CCL20, resulting in the blockage of the vicious inflammatory cycle. Besides CCL20, DMD was capable of inhibiting a series of chemokines and the receptors including CC, and CXC chemokines. CC chemokines majorly recruit T cells and monocytes, whereas CXC chemokines predominantly recruit neutrophils in psoriasis ( ). For instance, enhanced CCL2 release by macrophages induces monocyte migration to the psoriasiform lesions in mice ( ). CCL4L2 and CCL22 are pivotal chemokines in patients with psoriasis to stimulate the chemotactic infiltration of dendritic cells and macrophages ( , ). A principal histopathological feature of psoriasis is the inflammatory infiltration of neutrophils in the stratum corneum (Munro’s microabscess) and dermis. Chemokine production by the stimulated macrophages acts as a chemoattractant for neutrophil accumulation ( ). Our in vitro wound healing and transwell assays demonstrated the blockage of neutrophil migration after the suppression of chemokine subset in macrophages treated by DMD. The binding of chemokines and some stimulators to the corresponding receptors can activate the downstream signaling pathways including those of MAPKs and NF-κB. IMQ is reported to bind with TLR, resulting in MAPKs phosphorylation and NF-κB to generate cytokines and chemokines ( ). Overexpression of p-JNK, p-ERK, and p-p38 was inhibited by DMD, suggesting the possible role of this compound in MAPK regulation. Among MAPKs, ERK regulates IL-24 expression in psoriatic lesions ( ). Our result verified IL-24 downregulation by DMD through the ERK-dependent signaling. MAPKs regulate the downstream signaling of the transcription factor NF-κB in macrophages to produce cytokines and chemokines for psoriasis development ( ). We speculated that both MAPKs and NF-κB pathways participate in DMD-induced cytokine/chemokine arrest. The RNA sequencing study showed a potential role of GDAP1L1 in the anti-inflammatory activity of DMD against macrophages. GDAP1L1 is a mitochondrial fission factor governing the mitochondrial dynamics. Under oxidative stress, GDAP1L1 translocates to the mitochondria. Mitochondrial fission and fusion need to be controlled to keep the balance required for the persistence of mitochondrial morphology. Mitochondrial dynamics mediate energy output, quality control, and the generation of reactive oxygen species (ROS) ( ). The defect in mitochondrial membrane dynamics predominantly affects neuron functions to cause neurodegenerative diseases ( ). Mitochondrial fission raised as a result of a high level of stress may be one of the main causes of inflammation and immune dysregulation ( ). Due to its central role in mitochondrial fission, Drp1 is a prime target for regulatory pathways. Drp1 mainly exists in the cytosol but partially moves to the outer membrane of mitochondria for mediating the process of fission ( ). Drp1-mediated fragmentation leads to ROS generation, which is implicated in the pathogenesis of Alzheimer’s and Parkinson’s diseases ( ). Translocation of Drp1 to mitochondria depends on the phosphorylation at S616. Phosphorylation regulates the cycling of Drp1 between the cytoplasm and mitochondrial membrane to prompt fission ( ). In this study, IMQ treatment-induced GDAP1L1 expression and translocation, which is involved in the proinflammatory factor expression. GDAP1L1 could induce inflammation in the presence of Drp1 and inhibition of GDAP1L1 and Drp1 S616 translocation to mitochondria mediated by DMD prevented macrophage activation. Some fission factors are associated with apoptotic induction, whereas GDAP1L1 elicits mitochondrial fragmentation without inducing apoptosis ( ). Thus, we ruled out the apoptosis pathway mediating the effect of DMD on macrophage activation. Mitochondrial fission may be fundamental for dominating the production of proinflammatory mediators. A previous study ( ) demonstrated that downregulation of Drp1 attenuates proinflammatory factors via the reduced MAPK and NF-κB signaling in microglial cells. ERK can trigger mitochondrial fission through phosphorylation of Drp1 at S616 and its translocation to mitochondria ( ). Our experimental data inferred that DMD arrested MAPK signaling, blocking the translocation of Drp1 S616 and the subsequent fission. The mechanisms by which GDAP1L1/Drp1 axis regulates macrophage activation are unclear. There is little evidence to link GDAP1L1/Drp1 with inflammation. Zhang et al. ( ) reported that Drp1 gene expression increases in atopic dermatitis, and other autoimmune skin diseases. However, Therianou et al. ( ) demonstrated a contrary result in lesional psoriatic skin. Further study is needed to yield detailed insight into the molecular pathways that mediate GDAP1L1/Drp1 translocation in psoriasis. The topical treatment of IMQ cream on murine skin creates inflammation resembling the symptoms of human psoriasis ( ). IMQ caused phenotypic changes in psoriasis, including epidermal hyperplasia, inflammatory cell infiltration, and IL-23/Th17 axis activation. Relieving these pathological events is critical for psoriasis management. DMD effectively ameliorated IMQ-triggered inflammation and hyperproliferation. The epidermal thickness of IMQ-treated mouse skin could be reduced by 47% after DMD application. Based on the same protocol of psoriasis-like lesion induction by IMQ, our previous data ( ) demonstrated that betamethasone as a positive control could decrease epidermal thickness by 43%. This indicates a comparable therapeutic efficacy between DMD and the drug used in clinics. IHC and RT-qPCR results manifested a large accumulation of macrophages and neutrophils in viable skin of the psoriasiform lesion. This observation confirmed the macrophage recruitment in the epidermis and dermis of IMQ-treated mouse skin as was proved in the previous study ( ). Our experimental data convincingly demonstrated that a decrease in macrophage recruitment by topical DMD contributed to the resolution of psoriasiform inflammation. We found that DMD application inhibited IMQ-induced overexpression of GDAP1L1 and Drp1 in mouse skin. IL-23 and IL-17 are highly involved in the IMQ-induced animal model of psoriasis ( ). In psoriatic lesions, IL-23 is greatly expressed in dendritic cells and macrophages ( ). IL-17A is important in relieving psoriasis by upregulating cytokines IL-6, IL-1β, and TNF-α ( ). Overexpressed IL-17A in psoriatic skin leads to the proliferation and abnormal differentiation of keratinocytes ( ). Macrophages express low levels of IL-17A ( ). Psoriasis is a complex disease with a dynamic interaction between immune cells and keratinocytes, as well as their cytokines and chemokines ( ). Activation of a confederacy of cell types in psoriasis includes T cells, mast cells, dendritic cells, neutrophils, macrophages, and keratinocytes. As psoriasis develops, T cells and neutrophils are reported to release IL-17A ( , ). Our data manifested downregulation of IL-17A in IMQ-treated skin by DMD, which might directly or indirectly act on different immune cells with or without macrophage intervention. IL-17A further recruits dendritic cells and Th17 cells to the psoriatic lesion. The production of IL-1β and TNF-α from macrophages can be stimulated by IL-17A ( ), which also acts on keratinocytes to increase their proliferation and chemokine expression. Blocking IL-17A by DMD led to the inhibition of epidermal thickening as detected in our study. IL-24 is another cytokine largely produced by macrophages to direct keratinocyte proliferation ( ). In addition to macrophages, IL-24 is produced by T cells, mast cells, and keratinocytes in psoriatic plaque ( ). The inhibition of IL-24 by topically applied DMD could be beneficial in diminishing keratinocyte proliferation according to IHC. Neutralization of cytokines by DMD not only decreased the number of macrophages in viable skin, but also the number of neutrophils in the stratum corneum and dermis. Inhibition of neutrophil recruitment impeded the aberrant interplay between neutrophils and keratinocytes, thus blocking the keratinocyte hyperproliferation induced by neutrophil cytokines/chemokines. The significant reduction of GDAP1L1 and Drp1 S616 in IMQ-stimulated skin by DMD corroborated with the in vitro result of translocation inhibition. The findings in the mechanistic and animal studies suggest the multiple antipsoriatic mechanisms of DMD. This compound suppressed IMQ-induced expression of proinflammatory cytokines/chemokines by obstructing of MAPK and NF-κB phosphorylation in macrophages. We also verified that DMD downregulated proinflammatory factors via the inhibition of GDAP1L1 and Drp1 translocation and Drp1 phosphorylation. This inhibition might be mediated by decreasing the overexpression of p-NF-κB. Further neutrophil recruitment and keratinocyte hyperproliferation could be prevented by inhibiting macrophage activation after DMD management. The possible mechanisms of action of DMD are depicted in . The safety of topically applied DMD was examined earlier and shows a negligible irritation in healthy mouse skin ( ). A satisfactory therapeutic efficacy and safety could be achieved for topical DMD application.
The present work investigated the anti-inflammatory potential of a pure compound (DMD) derived from A. cinnamomea for mitigating psoriasiform plaque. The in vitro assay on THP-1 cells presented DMD-mediated inhibition of overexpressed cytokine/chemokine. The conditioned medium of the activated macrophages treated with DMD could suppress neutrophil migration and keratinocyte proliferation. The in vivo animal study showed relief in psoriasiform symptoms after DMD treatment. In addition, cytokine/chemokine upregulation and macrophage recruitment in the psoriasis-like lesion were alleviated in vivo . DMD downregulated proinflammatory mediators through the MAPKs and NF-κB pathways. The finding in this study also revealed that the GDAP1L1/Drp1 signaling axis plays a critical role in macrophages as an activator of IMQ-induced proinflammatory factors. This axis could be a target of anti-inflammatory DMD to reduce the lesions. This natural compound, DMD, sheds new light on the therapeutic approach against psoriasis.
The datasets of transcriptomic analysis for this study can be found on the website Gene Expression Omnibus ( https://www.ncbi.nlm.nih.gov/geo/ ). The accession number is GSE171667.
The studies involving human participants were reviewed and approved by Institutional Review Board at Chang Gung Memorial Hospital. The patients/participants provided their written informed consent to participate in this study.
S-YC initiated the study and drafted the manuscript. C-YC involved in the design of all experiments. S-YC and S-CY carried out the experiments. S-YC and AA analyzed data and wrote the manuscript. C-HL and J-YF supervised the entire project. J-YF reviewed critically and approved the final manuscript. All authors contributed to the article and approved the submitted version.
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
|
Industry Payments Received by Residents During Training | 9e5617b8-f337-402b-a27f-2241ad8bbaf2 | 10580108 | Gynaecology[mh] | Professional identity formation is a crucial component of residency training. The National Academy of Medicine (NAM, formerly the Institute of Medicine) recognized that medical residents’ ethical formation is susceptible to the influence of their mentors, especially in cases of potential conflicts of interest. Although they acknowledge the importance of industry-practitioner relationships that usher in valuable scientific advancements, the NAM expressed concern that “financial ties to industry may unduly influence professional judgments involving the primary interests and goals of medicine.” The NAM’s concerns are more than conjecture. The pharmaceutical industry spent nearly US $30 billion on marketing in 2016, two-thirds of which was direct to health care practitioners. Industry payments to physicians are correlated with higher tendencies for opiate prescribing , and writing for name-brand drugs over less-expensive generics. , , Industry interactions with trainees are associated with a lower inclination of residents to rely on evidence-based prescribing choices. As far as we know, no peer-reviewed publication examining industry payments to physicians has failed to find an association between payments and professional decisions working to the benefactor’s advantage. In fact, DeJong and colleagues found payments, such as meals worth as little as $20, are correlated with prescription behaviors favoring the entities who make the payments. During medical training, much of the hidden curriculum that residents learn comes by intuiting what is right from wrong by observing others in the learning environment. , The tendencies of mentors to accept industry payments is correlated with the propensity of fellows in internal medicine (IM) subspecialty training to follow suit. The presence of device and drug industry influence in medical education is neither a secret, nor a recent revelation. , Survey research has addressed the frequency of contact between industry and learners and resulting attitudes about interactions. , But we know of no study that has used the federal Open Payments Program (OPP) data set in an attempt to examine the factors associated with residents accepting industry payments during training. We examined the patterns of industry payments accepted by residents using the OPP database, which is housed by the Centers for Medicare and Medicaid Services. Federal law requires industry representatives to document their payments to health care professionals. We aim to identify whether this data set can shed light on industry-learner relationships and reveal factors in the learning environment that may be associated with residents’ receipt habits. Specifically, we hypothesized that payment receipt would vary according to specialty, the type of the sponsoring institution, and program director (PD) payment receipt behavior. The OPP database, established under the Patient Protection and Affordable Care Act of 2013, is intended to provide a comprehensive record of payments made by pharmaceutical and medical device companies to health care practitioners. The OPP database includes information on the recipient’s name and profession (ie, physicians, chiropractors, dentists, and so forth), organizational affiliation, the entity making the payment, the nature and timing of the payment, and its market value. Payments are categorized in the OPP database as general, research, and ownership. Research payments refer to consideration for facilitating a clinical study. Ownership payments include dividends paid for ownership in a business. This study focuses on general payments made to residents and PDs. General payments include any transfers such as gifts, honoraria, consulting fees, meals, travel, entertainment, or fee waivers. Although OPP requires reporting of payments to most health professionals, including fellows and independently practicing physicians, , residents are an exception. Although the law does not mandate reporting of payments to residents, it does not prohibit it. During our study period, July 1, 2020, until June 30, 2021, 124 715 residents (excluding fellows) were training in Accreditation Council for Graduate Medical Education (ACGME)–accredited programs. Twelve percent (14 817) of them appear in the OPP database as having received a combined $6.38 million in industry payments. Since regulations do not require reporting of payments to residents, these figures are likely an underestimation. Despite these limitations, our hypothesis is that by using the OPP database, we can identify patterns of payments to residents that align with their specialty, gender, PD behavior, and sponsoring institutions. Participants This study had a cross-sectional, retrospective design. The acceptance of general payments to residents was analyzed and compared by the type of the sponsoring institution, specialty, resident’s gender, and PD’s payment acceptance. The institutional review board at the American Institute for Research exempted this study from review. Because the OPP data are publicly available, participant consent is not required. This study followed Strengthening the Reporting of Observational Studies in Epidemiology ( STROBE ) reporting guidelines for cross-sectional studies. We focused on 65 992 residents in the 6 large specialties who were training in an ACGME-accredited program during the academic year 2020 to 2021. These residents accounted for 52.9% of all residents in training and represented the 3 largest primary-care specialties: IM, obstetrics and gynecology (OBGYN), and family medicine (FM); and the 3 largest surgical specialties: general surgery (GS), orthopedic surgery (OS), and urology. To link residents with their OPP records, we used the national provider identification (NPI) number. Seven residents switched programs during the study period, and we aligned them with the programs they belonged to for the longest term. PDs were identified through ACGME records, which included their NPIs and their starting or ending dates in that role. We found 168 programs that had more than 1 individual serving as PD during our time frame. We attributed payments to the program that the PD was affiliated with at the time of receipt. For example, a PD could have been employed with 1 program between July and December and received payments then. The same individual could have moved to another organization in the subsequent January through June and accepted no payments at the new program. By using the date of payment, we could attach the association of the PD’s acceptance to the program in which the PD was employed in the July to December time frame. When any program switched PDs, the sum of industry payments to the program was treated as the number of receptions for the PD of the individual program, regardless of who the individual person was. We used data from the ACGME to locate information about the residents’ training programs, including the specialty and the type of sponsoring institution (for-profit, nonprofit, local, state, or federal government agencies). We categorized residents’ gender as men, women, and other, which encompasses not reported, nonbinary, or another gender identity. Resident, PD, and sponsoring institution characteristics (such as ownership) were provided by the ACGME. Data about receipt of industry payment came from the OPP. Resident gender was provided by the Association of American Medical Colleges. PDs and residents are matched to the OPP data by use of NPI number. The outcome variable is the receipt of general payments during the 2020 to 2021 academic year. We did not include research and ownership payments. Residents who appeared in the general payments database were coded as 1, and those who did not were coded as 0. We kept a continuous count of the number of payments received by residents and PDs and the dollar value of each transfer. Statistical Analysis Modified Poisson regression was used to examine each variable’s association with the residents’ acceptance of a payment according to a generalized estimating equation (GEE) model. , Residents were nested within each program, and programs were embedded within each cross-classified cluster by specialties and sponsoring institutions. This data clustering was accounted for in the GEE model, using an exchangeable covariance matrix. The matrix assumes that associations of outcomes are equal in size among peer residents within clusters but are specified as independent between clusters. In the regression model, we measure the increased risk of residents in 1 group to receive a payment over those in a reference group. We chose as a reference the group that received the fewest industry payments. For specialty, that was IM. For gender, women received the least. For sponsoring institution type, it was federally owned. Results are reported as relative risk and 95% CIs for ownership type of sponsoring institution, medical specialty, resident gender, and PD receipt. Two-tailed P values were considered statistically significant at P < .05. All statistical analyses were performed using SAS Enterprise Guide, version 7.15 (SAS Institute). This study had a cross-sectional, retrospective design. The acceptance of general payments to residents was analyzed and compared by the type of the sponsoring institution, specialty, resident’s gender, and PD’s payment acceptance. The institutional review board at the American Institute for Research exempted this study from review. Because the OPP data are publicly available, participant consent is not required. This study followed Strengthening the Reporting of Observational Studies in Epidemiology ( STROBE ) reporting guidelines for cross-sectional studies. We focused on 65 992 residents in the 6 large specialties who were training in an ACGME-accredited program during the academic year 2020 to 2021. These residents accounted for 52.9% of all residents in training and represented the 3 largest primary-care specialties: IM, obstetrics and gynecology (OBGYN), and family medicine (FM); and the 3 largest surgical specialties: general surgery (GS), orthopedic surgery (OS), and urology. To link residents with their OPP records, we used the national provider identification (NPI) number. Seven residents switched programs during the study period, and we aligned them with the programs they belonged to for the longest term. PDs were identified through ACGME records, which included their NPIs and their starting or ending dates in that role. We found 168 programs that had more than 1 individual serving as PD during our time frame. We attributed payments to the program that the PD was affiliated with at the time of receipt. For example, a PD could have been employed with 1 program between July and December and received payments then. The same individual could have moved to another organization in the subsequent January through June and accepted no payments at the new program. By using the date of payment, we could attach the association of the PD’s acceptance to the program in which the PD was employed in the July to December time frame. When any program switched PDs, the sum of industry payments to the program was treated as the number of receptions for the PD of the individual program, regardless of who the individual person was. We used data from the ACGME to locate information about the residents’ training programs, including the specialty and the type of sponsoring institution (for-profit, nonprofit, local, state, or federal government agencies). We categorized residents’ gender as men, women, and other, which encompasses not reported, nonbinary, or another gender identity. Resident, PD, and sponsoring institution characteristics (such as ownership) were provided by the ACGME. Data about receipt of industry payment came from the OPP. Resident gender was provided by the Association of American Medical Colleges. PDs and residents are matched to the OPP data by use of NPI number. The outcome variable is the receipt of general payments during the 2020 to 2021 academic year. We did not include research and ownership payments. Residents who appeared in the general payments database were coded as 1, and those who did not were coded as 0. We kept a continuous count of the number of payments received by residents and PDs and the dollar value of each transfer. Modified Poisson regression was used to examine each variable’s association with the residents’ acceptance of a payment according to a generalized estimating equation (GEE) model. , Residents were nested within each program, and programs were embedded within each cross-classified cluster by specialties and sponsoring institutions. This data clustering was accounted for in the GEE model, using an exchangeable covariance matrix. The matrix assumes that associations of outcomes are equal in size among peer residents within clusters but are specified as independent between clusters. In the regression model, we measure the increased risk of residents in 1 group to receive a payment over those in a reference group. We chose as a reference the group that received the fewest industry payments. For specialty, that was IM. For gender, women received the least. For sponsoring institution type, it was federally owned. Results are reported as relative risk and 95% CIs for ownership type of sponsoring institution, medical specialty, resident gender, and PD receipt. Two-tailed P values were considered statistically significant at P < .05. All statistical analyses were performed using SAS Enterprise Guide, version 7.15 (SAS Institute). There were 65 992 residents training in accredited programs in the 6 specialties included in this study during the 2020 to 2021 academic year. Of them, 30 749 identified as women (46.6%), 34 258 identified as men (51.9%), and the balance, 985, another gender identity or declined to answer (1.5%). IM had the largest number of residents (30 286). The next largest specialty was FM, followed by GS, OBGYN, and then OS. FM had the largest number of programs with 678. Most programs were situated in private, nonprofit sponsoring institutions (1464 of 2182 programs [67.1%]); followed by state systems; then private, for-profit organizations; the federal government; and local governments. summarizes the characteristics of residents, specialties, and sponsoring institutions. Overall, 14 817 residents were reported to have received nearly $6.4 million in a single academic year, and 8750 residents among the 6 specialties in the analysis received $3.9 million. Payment Amounts A total of 26 611 payments were made to residents in 2020 to 2021. Of the 65 992 residents, 616 residents were excluded because their NPIs were missing or duplicate. The 7 residents who changed programs were analyzed using data for the program they were enrolled in for the greatest period of time, resulting in 26 526 separate payments used for analysis. describes payments received by residents and PDs. Of the 65 376 residents training during the study period, 8750 (13.4%) reportedly accepted at least 1 industry payment. Of the 2466 individuals who served as PD, 976 (39.6%) had accepted industry payments worth a total of $5.2 million. OS residents and PDs received the most total payments at $2.2 million and $4 million, respectively. OBGYN residents and PDs received the lowest total payments at $68 110 and $185 817, respectively. The median (IQR) value of all payments received by residents was $64.40 ($21.04-$193.68). The median (IQR) value of any particular payment was $31.27 ($17.47-$89.42) with a maximum value of $50 000 and a minimum of $0.79. Consistent with previous findings of early career graduates in clinical practice, OS had the highest proportion (39.0%) who accepted payments, and IM had the smallest (9.0%). Of those residents in the OPP database, the median (IQR) total payment to the recipients was highest for OS ($526.32 [$131.20-$1400.75]) and lowest for FM ($32.50 [$16.78-$80.58]). The median (IQR) value of payments to PDs was $119.55 ($35.76-$614.21). The median (IQR) value of any particular payment was $17.68 ($13.55-$33.68) with a maximum value of $418 127.10 and a minimum of $0.17. OS had the highest proportion (153 of 213 participants [71.8%]) who accepted industry payments, and FM had the smallest (219 of 775 participants [28.3%]). Of those PDs in the OPP database, the median (IQR) total payment to the recipients was highest for OS ($980.48 [$151.62-$4060.27]) and lowest for FM ($62.08 [$19.85-$240.85]). For programs whose directors accepted payments, the maximum was $1.9 million (OS), and the minimum was $3.82 (urology). Relative Risk of Accepting Payments Sponsoring institution control was significantly associated with the probability of a resident to appear in the OPP data set. reports the regression analysis indicating that that the relative risk for trainees with for–profit sponsoring institutions was 3.50 times higher than those in federal sponsoring institutions to accept industry payments (95% CI, 2.32-5.28; P < .001). Residents affiliated with nonprofit sponsoring institutions were 2.00 times more likely than those in federal sponsoring institutions to have accepted payments (95% CI, 1.36-2.93; P < .001). The difference between the probability of residents training in federal sponsoring institutions and those in programs operated by state or local government was not statistically significant. Specialty was significantly associated with the risk of residents’ acceptance of industry payments. The relative risk of OS residents accepting industry payments was 3.21 times higher than IM residents (95% CI, 2.73-3.77; P < .001). Urology residents were 2.95 times more likely than IM residents to accept industry payments (95% CI, 2.44-3.56; P < .001). OBGYN residents were 1.30 times more likely (95% CI, 1.05-1.62). GS residents were 1.21 times more likely to accept industry payments (95% CI, 1.00-1.45). FM was 1.15 times more likely to accept industry payments (95% CI, 0.97-1.36). The difference in the risk of accepting a payment between FM and IM residents was not statistically significant. We found that the probability of men to accept payments was 9% higher than women, whereas those who did not identify a gender, whose data were missing, or who were nonbinary (others) were 15% lower than women. The number of the PD’s receipt of payments had a small but significant association for residents in that training program to accept payments. The number of payments PDs accepted slightly elevated the risk of residents to accept a payment by 1.01 (95% CI, 1.01-1.01). Types of Payment The OPP records broad categories for the nature of these payments (see eTable 1 in for a count of the types of payments received by residents in the 6 specialties in our study and eTable 2 in for all specialties). eTable 3 in defines the OPP categories. Among those categories, refreshments were the largest type of payment received. Nine in 10 payments (23 864 payments [89.96%]) were for food and drink. This was followed by education (2050 payments [7.7%]) and travel and lodging (528 payments [2.0%]). Group meals accounted for 15 822 payments (59.6%), with a total dollar value of $579 259.41. We inferred group meals by identifying records for food and beverages made on the same dates for residents in the same program with identical dollar values. reports that several residents received general payments as high as $50 000 in a single year. The largest individual payments were in the form of grants. Two urology residents each received $50 000 grants, while 1 IM resident received a grant of $46 875. One GS resident received a $25 000 grant. Each of 5 OS residents accepted $20 000 grants. One FM resident received a $10 000 grant. One OBGYN resident accepted a $5200 consulting fee. A total of 26 611 payments were made to residents in 2020 to 2021. Of the 65 992 residents, 616 residents were excluded because their NPIs were missing or duplicate. The 7 residents who changed programs were analyzed using data for the program they were enrolled in for the greatest period of time, resulting in 26 526 separate payments used for analysis. describes payments received by residents and PDs. Of the 65 376 residents training during the study period, 8750 (13.4%) reportedly accepted at least 1 industry payment. Of the 2466 individuals who served as PD, 976 (39.6%) had accepted industry payments worth a total of $5.2 million. OS residents and PDs received the most total payments at $2.2 million and $4 million, respectively. OBGYN residents and PDs received the lowest total payments at $68 110 and $185 817, respectively. The median (IQR) value of all payments received by residents was $64.40 ($21.04-$193.68). The median (IQR) value of any particular payment was $31.27 ($17.47-$89.42) with a maximum value of $50 000 and a minimum of $0.79. Consistent with previous findings of early career graduates in clinical practice, OS had the highest proportion (39.0%) who accepted payments, and IM had the smallest (9.0%). Of those residents in the OPP database, the median (IQR) total payment to the recipients was highest for OS ($526.32 [$131.20-$1400.75]) and lowest for FM ($32.50 [$16.78-$80.58]). The median (IQR) value of payments to PDs was $119.55 ($35.76-$614.21). The median (IQR) value of any particular payment was $17.68 ($13.55-$33.68) with a maximum value of $418 127.10 and a minimum of $0.17. OS had the highest proportion (153 of 213 participants [71.8%]) who accepted industry payments, and FM had the smallest (219 of 775 participants [28.3%]). Of those PDs in the OPP database, the median (IQR) total payment to the recipients was highest for OS ($980.48 [$151.62-$4060.27]) and lowest for FM ($62.08 [$19.85-$240.85]). For programs whose directors accepted payments, the maximum was $1.9 million (OS), and the minimum was $3.82 (urology). Sponsoring institution control was significantly associated with the probability of a resident to appear in the OPP data set. reports the regression analysis indicating that that the relative risk for trainees with for–profit sponsoring institutions was 3.50 times higher than those in federal sponsoring institutions to accept industry payments (95% CI, 2.32-5.28; P < .001). Residents affiliated with nonprofit sponsoring institutions were 2.00 times more likely than those in federal sponsoring institutions to have accepted payments (95% CI, 1.36-2.93; P < .001). The difference between the probability of residents training in federal sponsoring institutions and those in programs operated by state or local government was not statistically significant. Specialty was significantly associated with the risk of residents’ acceptance of industry payments. The relative risk of OS residents accepting industry payments was 3.21 times higher than IM residents (95% CI, 2.73-3.77; P < .001). Urology residents were 2.95 times more likely than IM residents to accept industry payments (95% CI, 2.44-3.56; P < .001). OBGYN residents were 1.30 times more likely (95% CI, 1.05-1.62). GS residents were 1.21 times more likely to accept industry payments (95% CI, 1.00-1.45). FM was 1.15 times more likely to accept industry payments (95% CI, 0.97-1.36). The difference in the risk of accepting a payment between FM and IM residents was not statistically significant. We found that the probability of men to accept payments was 9% higher than women, whereas those who did not identify a gender, whose data were missing, or who were nonbinary (others) were 15% lower than women. The number of the PD’s receipt of payments had a small but significant association for residents in that training program to accept payments. The number of payments PDs accepted slightly elevated the risk of residents to accept a payment by 1.01 (95% CI, 1.01-1.01). The OPP records broad categories for the nature of these payments (see eTable 1 in for a count of the types of payments received by residents in the 6 specialties in our study and eTable 2 in for all specialties). eTable 3 in defines the OPP categories. Among those categories, refreshments were the largest type of payment received. Nine in 10 payments (23 864 payments [89.96%]) were for food and drink. This was followed by education (2050 payments [7.7%]) and travel and lodging (528 payments [2.0%]). Group meals accounted for 15 822 payments (59.6%), with a total dollar value of $579 259.41. We inferred group meals by identifying records for food and beverages made on the same dates for residents in the same program with identical dollar values. reports that several residents received general payments as high as $50 000 in a single year. The largest individual payments were in the form of grants. Two urology residents each received $50 000 grants, while 1 IM resident received a grant of $46 875. One GS resident received a $25 000 grant. Each of 5 OS residents accepted $20 000 grants. One FM resident received a $10 000 grant. One OBGYN resident accepted a $5200 consulting fee. About 12% of all residents were reported to have received nearly $6.4 million in a single academic year, despite the fact that such reporting is optional. Although there were instances of residents accepting as much as $50 000 in industry payments in a single year, the vast majority of payments were for refreshments, and many of those were in group settings. Bearing in mind that the typical year-1 resident is carrying $200 000 or more in student debt on a salary of about $60 000, one may find it difficult to criticize a resident for participating in a sponsored lunch. It is not a state of impecunity alone that may attract the resident to a free lunch. Our data suggest that their colleagues and faculty are attending the same events, thus creating a social environment that condones acceptance of industry payments. The influence of such payments has been associated with prescribing behaviors that benefit the giving pharmaceutical firm. , , , , , These findings highlight the importance of revising the federal exception for trainees in OPP. In 2013, the Centers for Medicare and Medicaid Services concluded that extending the OPP regulations to residents would lead to uneven reporting because too few had an NPI and that state licensure for residents was variable. Today, the ACGME has NPIs for 99.1% of active residents; thus, the rationale for excluding them no longer holds true. Including residents would provide a more comprehensive understanding of the extent of industry payments in the health care sector, their potential influence on patient care, and the patterns of formation of industry-physician financial relationships during medical education. Given the long-lasting effects of residency training, specialty societies, medical educators, and sponsoring institutions may elect to reevaluate their policies on conflicts of interest. Our study did not explore the effects of industry payments on residents’ decision-making. However, they suggest that there is variation in reporting by specialty, sponsoring institution control, and PD behavior. The consequences of these factors of the training environment support the concerns raised by the NAM regarding the potential outcomes of industry payments on the ethical formation of physicians during their training. Although the effect size was small in this study, the association between PDs and resident behavior is present and warrants further investigation. It is possible that the reporting exception for residents may obscure the true effect of the PDs. We found specialty was associated with payment patterns, with the association most pronounced in OS. This is consistent with other cross-specialty studies of the OPP data set, which find proceduralists, particularly orthopedic surgeons, have a higher tendency to receive payments. Residents in programs affiliated with privately controlled sponsoring institutions (for-profit and nonprofit) had a higher likelihood of accepting industry payments than those in government-controlled institutions. Although residents in programs affiliated with state and local health systems seem to have appeared in the OPP more than those in federal sponsoring institutions, the difference is not statistically significant. It remains to be determined why this is. State-level gifting policies vary widely, but states with stringent drug detailing regulations see slower uptake of new, expensive drugs. , Limitations This study had limitations. The federal government does not require reporting of industry payments to residents; thus, we suspect there is coverage error in the OPP. This may result in understating the extent to which transfers are made to residents in certain specialties, programs, or sponsoring institutions. With more complete coverage it is possible that there would be a stronger relative risk associated with gender, as our findings were somewhat lower than those found in studies of industry payments to independently practicing physicians. , The minimal relative risk of resident receipt given PD behavior might also come more in line with findings in studies of fellowship programs had the OPP systematically collected resident payments. Our findings, though, concur with existing literature that suggests procedural specialists, orthopedics in particular, receive more industry payments than primary-care practitioners. Our study spanned only 1 academic year, and resident interactions with industry representatives may have been diminished by the COVID-19 outbreak as industry representatives’ in-person access to medical facilities was curtailed. This study had limitations. The federal government does not require reporting of industry payments to residents; thus, we suspect there is coverage error in the OPP. This may result in understating the extent to which transfers are made to residents in certain specialties, programs, or sponsoring institutions. With more complete coverage it is possible that there would be a stronger relative risk associated with gender, as our findings were somewhat lower than those found in studies of industry payments to independently practicing physicians. , The minimal relative risk of resident receipt given PD behavior might also come more in line with findings in studies of fellowship programs had the OPP systematically collected resident payments. Our findings, though, concur with existing literature that suggests procedural specialists, orthopedics in particular, receive more industry payments than primary-care practitioners. Our study spanned only 1 academic year, and resident interactions with industry representatives may have been diminished by the COVID-19 outbreak as industry representatives’ in-person access to medical facilities was curtailed. In this cross-sectional study of industry payments to residents, approximately 12% of residents were reported to have accepted industry payments in an academic year. That proportion could be an understatement because the available reporting is strictly voluntary. Specialty, PD behavior, and sponsoring institution control were significantly associated with residents accepting industry payments. Sponsored group meals appear to account for a large proportion of payments, suggesting that programs condone these trainee-industry interactions. Specialty societies, medical educators, and sponsoring institutions should reexamine how they address potential conflicts of interest in training programs. |
Recent Advances of Microfluidic Platforms for Controlled Drug Delivery in Nanomedicine | 00253829-a3f5-4255-9bb5-752839de1c4f | 8439440 | Pharmacology[mh] | Nanomedicine is a branch of medicine that aims to use nanotechnology—that is, the manipulation and manufacture of materials and devices with a diameter of 1 to 100 nanometers—to prevent disease and to image, diagnose, monitor, treat, repair, and regenerate biological systems. Medicine nanotechnology, termed nanomedicine, have provided essential new strategies for the evolution of pharmaceuticals and biotech products. Following ground-breaking discoveries in the use of nanoscale materials, nanomedicine has begun to play an important position in pharmaceutical studies and product development and has resulted in significant commercialization efforts. Nanoparticles (NPs) are the main fringe of the growing area of nanotechnology and are expected to conquer continual demanding situations of inefficient drug delivery strategies. In fact, nano-particulate drug delivery was discovered to effectively affect nanomedicine because of their small size, tunable chemical surface properties, high volume-to-surface ratio, and, basically, the potential to load active pharmaceutical ingredients and imaging agents. Furthermore, nanomedicine drug delivery mediums have been shown to improve the useful and favorable result or effect and reduce side-effects associated with drugs that have already received market approval, to enable new therapeutics, and to inspire further investigation of active biological products undesirable pharmaceutical properties that were previously thought to be incapable of development. Microfluidic technologies employ nano and microscale fabrication techniques to develop highly controllable and reproducible fluidic microenvironments. , Lead compounds with controlled physicochemical properties can be produced using microfluidics, characterized in a high throughput fashion, and evaluated in an in vitro biomimetic fashion for human organ organ-on-a-chip. , The microfluidic generation has emerged as an effective device for the fabrication of microparticles with controlled morphologies and preferred properties because of its capacity to exactly control the emulsification procedure and generate monodispersed compound droplets in the microchannels. The ability to generate double emulsions containing one, two, three, or more quantities of droplets with extreme accuracy demonstrates the level of control offered by microfluidics. To exactly control the release of payloads, it is far important to put together polymeric debris with regarded sizes and size distributions, because the particle size strongly influences the payload release rate. By tailoring their inner structures, the loading of therapeutics into polymeric debris and the release of payloads also can be tuned. Co-delivery of multiple drugs can be attained through adjusting the size and number of internal compartments. , Another method to control the release of payloads is with the aid of using the usage of stimulus-responsive substances to synthesize polymeric debris. After publicity of the environmental triggers, which include pH, temperature or ionic strength, the debris goes through a physicochemical alternative after which it releases the payloads. , Microfluidics offers advantages in terms of small size distribution with less polydispersibility index, higher encapsulation and loading efficiencies, better batch-to-batch uniformity, and easy scale-up possibilities. Interestingly, the preparation of microfluidic chips is simple and easy, leading to the economical production of nano-carriers. Various microfluidic chips have been fabricated to synthesize organic, inorganic, polymeric, lipid-based vesicular, and hybrid nano-carriers. Taking it all together, the microfluidic technology provides a potential platform for the quick synthesis of different novel drug delivery systems. The production methods of polymer microparticles have become increasingly important for applications such as controlled drug delivery, medical diagnostic tests, obtaining super-hydrophobic surfaces, optimum design of toughened polymeric composites and food technology. Polymeric microparticles are made using a variety of processes, including suspension or emulsion polymerization, solvent evaporation, spray drying, spraying a polymer solution through a small hole, and the Shirasu Porous Glass (SPG) membrane emulsification technique. Traditional production processes, on the other hand, have a number of disadvantages, including the fact that they take time, cause particle coalescence, and result in non-homogeneous particle sizes and shape non-uniformity. The electrospray approach can be used to circumvent these constraints. Furthermore, the electrospray process has numerous advantages over previous methods, including minimal residue, the use of fewer solvents, cheap cost, and the use of high molecular weight polymers. The micro devices are comprised of two flow focusing junctions that work together to create double emulsions in a two-stage process. The aqueous phase is symmetrically pinched off at the first junction at low flow rates, forming monodisperse aqueous monomer plugs. The oil phase encapsulates liquids 1 and 2 at the second junction, generating double droplets of aqueous and monomer phases. The compound droplets then reach a third junction, where the channel cross-section is enlarged, causing them to take on spherical shapes. In the large section, mass conservation forces the droplets to slow down significantly, reducing the distances between successive droplets and thus reducing the distances between consecutive droplets, thus reducing the distances between consecutive droplets. Drug delivery technology advancements can improve pharmacological factors, including efficacy and bioavailability, resulting in the discovery and development of more effective drugs for better patient outcomes and quality of life. Fabrication quality controls, product batch-to-batch fluctuation, and the inability to obtain physiologically relevant test results in traditional in vitro prescreening platforms are all obstacles to nanoparticle drug delivery. Microfluidics has evolved from micro to nanoliter fluid handling to include a multidisciplinary approach that may be used for a wide range of applications. , In comparison to traditional methods, microfluidics provides a mechanism to build highly controllable, reproducible, and scalable fabrication methods for nanoparticle synthesis. When compared to traditional in vitro culture methods, organ-on-chip microfluidic technology provides highly relevant organ-specific testing platforms capable of biologically relevant experimental time scales while employing a fraction of the sample and media volumes , The application of innovative microfluidic techniques to nanoparticle development processes may be able to address key challenges in nanoparticle drug carrier clinical translation.
The aim of controlled drug delivery is to deliver the drug to the exact anatomical location at the preferred rate and time to enhance the efficacy, pharmacokinetics, and bioavailability of the drug at the same time as preserving the minimum side effects. Due to a few specific characteristics (specific or even more than one dosing, perfect and target-specific release, sustainable and controlled delivery, and mild facet effects) of the current microfluidic techniques, unparalleled possibilities exist to manipulate drug delivery. Controlled drug delivery systems help to improve the ease of administration of effective drugs and are pharmacokinetically superior for drugs formulated in this dosage form, including biological products and synthetic drugs. They enhance the amount released and that reaches the systemic circulation from the therapeutic drugs with the aid of increasing the uptake, preventing presystemic metabolism, retaining drugs at the safe therapeutic level, and lowering the side effects by targeting drugs to particular cells or tissues. , Drug delivery may be controlled to release the active drug in the favored quantity that might allow individuals suffering from certain diseases to obtain reproducible, in urgent need, and tunable dosing at the desired time. Such a device permits correct regulation of dosage for favored effects, minimizes the associated side effects, minimizes the related side effects, and averts repeated drug administration or implantation of devices, ultimately increasing patient compliance. Such drug delivery strategies are elaborated in . Microfluidic technologies provide low-cost, simple-to-use platforms for fluid flow control. Emulsions generated in microfluidic systems have been employed in bioanalysis, organic synthesis, fluidic optics, and controlled drug delivery. , T-junctions and flow-focusing nozzles are two types of devices used to generate emulsions in microfluidic platforms. , Both approaches allow for the creation of monodisperse particles and provide versatility in the size of the emulsions produced. Monodisperse polymer particles, both spherical and non-spherical, are widely made by utilizing flow-focusing (FF) machines. It has been established that FF devices may be used to make photocurable polymeric particles, , ion-crosslinkable thermosensitive gels, , polymer-encapsulated cells, , and other particles. , The review by Brzeziński et al presented the utility of microfluidics technique for the preparation of polylactide (PLA)-based particles towards developing novel drug delivery systems. Brzeziński et al developed a novel microfluidic approach for the preparation of (co) polymeric and hybrid nanoparticles (NPs) composed of (co) polymers/tannic acid (TA) in the microfluidic flow-focusing glass-capillary device. However, there has been little systematic attempt to evaluate the controlled drug release kinetics from particles generated by microfluidic devices in most previous research.
An extensive range of fabrication strategies has been explored and advanced for generating microfluidic elements and systems. Cast molding and associated soft lithography techniques, patterning of spin-on polymers, embossing and combined methods are popular. Microfluidic Devices Microfluidic devices for particle production are mainly of two types, namely microchannels and microcapillaries. Microchannel-based devices are commonly fabricated by processes such as micromilling, micromachining, lithography, and mold replication. In such devices, the interfacial vicinity minimization brings about spontaneous droplet formation, and therefore, while preserving the oil-phase flow rate inside an optimum range, the droplet dimension is best depending on the microchannel geometry. Manufacturing of microchannel-primarily based totally overall devices is costly and takes a long time, but it allows for the fabrication of microsystems with particle size as small as a few tens of microns. Furthermore, in such devices, the microchannels may be flawlessly aligned, and in addition to uniform flow and highly liquefying of specific droplets or splitting droplets to uniform size, may serve to equip them, and the systems may be extended to produce a large number of products. Besides, capillary-primarily based structures are frequently made from low-cost components available in the market and have the capability to be microchannels in particle manufacturing. Importantly, the ones systems can be fabricated in a great deal much less time and can be operated in aggressive process conditions. In microchannel-based complete devices, the dispersed phase is in proximity to the tool’s wall before being emulsified by the continuous phase, which may result in a phase inversion. While the affinity of the dispersed phase for matter is greater than that of the continuous phase, the dispersed phase preferentially wets the partitions of the tool. This turns the choice of the substance for the production of the device has superior importance to all others. Yet, phase inversion can be avoided via deciding on the proper device for aqueous or organic droplets. Preferentially, the polymers used may be changed on the vicinity wherein droplets of the dispersed phase are arise (develop); however, this interprets into a further step withinside the micro-fabrication process. By contrast, capillary-primarily-based totally devices are tremendous for such terms. Here, the dispersed levels may be brought to the centerline of the non-stop segment flow, stopping the droplets from assembling the device’s partitions. Furthermore, capillary-primarily based totally devices do away with clogging troubles not unusual place in microchannel-primarily based totally devices and permit the manufacturing of oil-in-water (O/W) or water-in-oil (W/O) emulsions with a unit microsystem. These developments have allowed microfluidic devices to be fabricated using a wide range of materials and geometries, enabling new and advantageous physical behaviors and qualities in microfluidic devices. For laminate microfluidic devices, each layer is cut individually. The cutting method has a significant impact on the dimensions and the device functionality. For prototyping and laboratory settings, cutting is usually done with a knife plotter (ie, xurography) or laser cutter because of the speed and simplicity that each tool offers. A knife plotter works by precisely cutting a material with a blade to create the geometry, while a laser cutter uses a focused beam (traditionally CO2 lasers are used). , Under these conditions, the reduction of the diameter of droplets could be obtained by increasing flow rate, density ratio, and viscosity. Concerning the shape of devices used Bottom-up technologies that rely on emulsions or self-assembly cannot always provide fine, pre-designed control over particle geometry (shape, aspect ratio) and composition. Bicudo and co-workers investigated the manufacture of HANPs crosslinked with adipic dihydrazide (ADH) and chloride carbodiimide utilizing a microchannel flow-focusing system (EDCl). The work is focused on analyzing the process parameters of the unique method, which is a continuous nanoprecipitation at the water–organic solvent interface. The effect of the type of organic solvent employed, non-solvent flow rate, and Hyaluronic Acid (HA) content on the characteristics of (hyaluronic acid nanoparticles) HANPs was investigated by the authors. The findings revealed that through water diffusion and nanoprecipitation rates, the affinity between water and organic solvent regulated the mean diameter of the nanoparticles (NPs). When the non-solvent exhibited an intermediate affinity for water, the polydispersity was narrowed. Furthermore, because the process was convection regulated, lower HA concentrations and higher isopropyl alcohol flow rates resulted in smaller particles. Regardless of the organic solvent, flow rate, or HA concentration, somewhat stable NPs were formed. The procedure was found to be straightforward, repeatable, and quick. The authors found that this process was promising for manufacturing oil-free HANPs, which are important for medical, pharmacological, and aesthetic applications. , The benefits and drawbacks of these methods are mentioned in . Synthesis of Microfluidic Nanocarriers Beyond that, there has been great attraction of pharmaceutical formulators to synthetic nanocarriers and natural nanocarriers and colloidal systems have been of little interest. Significant attention has recently been paid to the production of organic nano-carriers, particularly in pharmaceuticals, as pharmaceutical scientists have begun to recognize the important properties conferred on nano-carriers by the microfluidic methods. , In general, nano-carriers are obtained through the dispersion of preformed polymers or through the monomer reactions to form polymers. These nanocarriers may be advanced via unique strategies and are categorized into classes relying on processes involved. Emulsification of materials is required in the primary group, whereas it is not always required in the other groups. As a result, it provides a clean and simple method of synthesis. When those strategies are utilized in traditional device, lack of control over uniform blending, formation, and improved effects on formulation ingredients and, as a consequence, very few products have an excessive particle size distribution. On the other hand, microfluidics controlling structures are able to supply control over the above-mentioned elements as a consequence of giving them uniform size particles. Lipid polymer hybrid nanoparticles (LPHNPs) have been merged as capacity nanocarriers. LPHNPs were created with the help of the microfluidic co-flow nanoprecipitation method. The internal fluid changed into organized with the aid of using dissolving poly (lactic-co-glycolic acid) (2 mg/mL) into acetonitrile as a natural phase. The outer fluid contained lecithin and Distearoyl-sn-glycero-3-phosphoethanolamine-polyethylene glycol 2000 in a two-to-three mass ratio dissolved in 4% ethanol and answered in water. , These defined compatibilities, higher internalization, protection profile and a relative better anticancer interest within breast cancers with controlled drug release behavior. , , Li et al conducted research on open-channel, water-in-oil emulsification in a paper-based microfluidic drug delivery system. Providing novel open-channel microfluidics that is primarily based on surfaces with designed wettability and has the ability to control the floating of all excessive and occasional surface tension of liquids. The open channel devices have been capable of constraining numerous smaller surface tension oils at excessive, but sufficient flow rates to enable water-in-oil microfluidic emulsification in an open channel tool. It ought to extrude the dimensions of the emulsified droplets formed inside the open channel tool with the aid of adjusting the speed of each of the dispersed aqueous and natural continuous phases. Finally, a fabricated tool has been changed to be used efficaciously to synthesize surprisingly monodisperse hydrogel microparticles that might contain a drug molecule. Additional investigation of the drug delivery properties of the fabricated products had promising results with an open-channel microfluidic devices having the potential to achieve a high level of fluid manipulation while being manufactured quickly and cheaply. Another study conducted by Raphael and Martin utilized dynamic microfluidic method; dopamine was used as a model drug to quantify the electrochemical flow on paper-based devices. They combined an electrochemical method with a microfluidic device for the time-resolved detection of dopamine from neuron-like PC12 cells cultured on filter paper. After an investigation of cellular adherence to the exterior of the paper with fluorescence microscopy; dopamine drug delivery after stimulation with acetylcholine without/with drugs has been investigated As a result, the data collected with the device agreed with single-cell statistics, demonstrating the technique’s validity for higher-throughput quantification of chemical targets on tissues- or organs-on-a-chip. Generally, microfluidic devices maintain numerous qualities in the pharmaceutical sciences, which consist of suitable dosage, ideal drug delivery, site-targeted delivery, extended and controlled release, less repetitive dosing, and minimum side effects. These benefits are important qualities of drug delivery systems. Microfluidic techniques have been regularly employed in the production of polymers as carriers of many active moieties, direct drug delivery systems, high-throughput screening, and additives and excellent carriers of drugs. Cheaper and effortlessly produced paper-based materials are good substrates that truly mitigate several challenges associated with transportation, filtration, and storage, concentrators, valving, and multiplexing. Hereafter, paper-based microfluidic generation in controlled drug delivery programs might carry thrilling possibilities to amplify the frame of understanding within the subject and support the improvement of the scientific translation of drug delivery systems. Huang et al prepared a stimuli-responsive controlled Vinblastine (VBL) drug release device from magnetically sensitive chitosan capsules. A magnetically responsive controllable drug delivery device has been designed by way of a means of embedding superparamagnetic iron oxide (SPIO) nanoparticles (NPs) in a chitosan matrix and an exterior magnet. In addition, droplet microfluidics, that is a completely unique technique for producing polymer spheres, has grown to be used for the manufacturing of monodispersed chitosan microparticles. The prepared VBL and SPIO NPs-loaded chitosan microparticles were characterized and showed individual and distinctive controlled-release patterns. Thus, the release rate, time, and dose of VBL release have become controlled through an exterior magnet. The outcomes presume that the usage of a magnetically responsive controlled drug delivery device brings a precious possibility for VBL drug release, wherein the delivery device is an energetic contributor, instead of a passive vehicle, within the optimization of most cancer treatments. This method actively focused magnetic drug delivery device to bring many benefits over traditional drug delivery structures through enhancing the precision and time of release, smooth operation, and better compliance for pharmaceutical applications. Because of their distinct physicochemical behavior and synergetic impact in preventing and inhibiting the progression of colorectal cancer, Liu et al chose atorvastatin and celecoxib as version dosage forms for their microfluidic collection of monodisperse multistage pH-responsive polymer/porous silicon composites for exactly controlled multi-drug delivery. To be efficient in prevention and inhibition of the acceleration of colon and rectal cancer, the fabricated microcomposite incorporating hypromellose acetate succinate, which is not soluble in acidic conditions, however, is incredibly soluble in basic (alkaline) pH environments. The benefit of the larger pore volume of porous silicon (PSi) is atorvastatin was initially loaded into the PSi matrix before being encapsulated into pH-responsive polymer microparticles containing celecoxib through microfluidics to acquire multi-drug loaded polymer/PSi microcomposites. The manufactured microcomposites confirmed monodisperse size distribution, multistage pH-response, particular ratiometric controllable loading extent closer to the concurrently loaded drug molecules, and tailor-made drug release kinetics of the loaded materials. This appealing microcomposite platform protects the payloads from being delivered at low pH values and improves the drug delivery at better pH values, which may be similarly used in preventing and treating colon and rectum cancer. Overall, the pH-responsive polymer/PSi-primarily based completely microcomposites may be used as a common platform for the combined drug delivery system of various drug molecules. Monodisperse biodegradable polymer microparticles were prepared by Xu et al using a microfluidic flow-focusing tool for controlled drug delivery and have described the fabrication of monodisperse, drug-loaded microparticles from biodegradable polymers, the use of microfluidic flow-focusing devices and the drug-delivery properties of these particles. Particles are engineered in a variety of sizes ranging from 10mm to 50mm. These particles are nearly monodispersed with a polydispersity index of 3.9%. Bupivacaine (amphiphilic) is included in the biodegradable matrix of the debris to characterize the formulation as a model drug. The kinetic evaluation suggests that the release of the drug from those monodisperse microparticles is slower than that from traditional strategies of identical average size, but reveals a larger particle size distribution and, more importantly, a substantial decrease in the primary burst than that discovered with traditional particles. , The distinction in the preliminary kinetics of drug release is attributed to the uniform distribution of the drug within the particles generated with the assistance of microfluidic strategies. These outcomes showed the application of microfluidic flow-focusing on the technology of homogenous systems of particles for drug delivery. Meng and co-workers prepared thermal-sensitive liposomal controlled release by using a disposable microfluidic tool. The release of an encapsulated drug from a nano-carrier consisting of a liposome must increase local drug delivery while reducing the toxicity consequences of a temperature increase. High-intensity focused ultrasound (HIFU), particularly micro-HIFU (MHIFU) microfluidic devices, allows us to imitate the bulky HIFU transmission tool with the help of lower power consumption and control the release of low temperature-sensitive liposomes (LTSL) investigated. , Furthermore, at the transition to a local temperature of between 41°C and 43°C, the structure changes from a gel to a liquid crystalline phase, releasing the encapsulated drugs with an external source of hyperthermia, such as microwaves, and an infrared radiation laser with the structure of a lipid membrane of low-temperature sensitive liposomes (LTSL). The microfluidic era may also provide a promising approach to studying the complex dynamics of ultrasound and organisms at the ultramicroscopic level. A principal task within the improvement of polymeric nanoparticles for numerous programs is the specific engineering of preferred physicochemical properties in a reproducible manner. Shamsi and co-workers developed microfluidic self-assembly of polymeric nanoparticles with tunable compactness for controlled drug delivery. , It is proven that the compactness of nanoparticles primarily based on self-assembled hydrophobically modified chitosan (HMC) biopolymers may be dictated by tunable fast blending through hydrodynamic flow focusing in microfluidic channels. It has been demonstrated through various flow rates, as well as the hydrophobicity of the chitosan chains, that the self-assembly properties of the chains can be controlled by optimizing the dimensions and compactness of the species, as well as a greater limited particle size distribution of the nanoparticles. , The particle size of the formulation components increased with increasing blending time, while chitosan produced smaller and more compact nanoparticles with a much smaller variety of aggregated chains and a higher degree of hydrophobicity. The investigation revealed that, to the greatest extent possible, despite the lack of affinity for the aqueous medium and at blending times longer than the time of aggregation, nanoparticles with nearly equal forms of hydrophobic adhesion were formed. Furthermore, exploring the effectiveness of microfluidics directed to organizing HMCs and encapsulating paclitaxels, a common anticancer drug, has discovered remarkably higher encapsulation efficiency overall performance in comparison to the conventional bulk method. The in-vitro release of the paclitaxel from the synthetic nanoparticles was evaluated to analyze the effect of the compactness of the formulation components on the drug release properties. The anticipated values of the diffusion coefficient of paclitaxel as high as 50% at drug release implied sustainability in controlling drug release regarding the compactness of the nanoparticles, and an outstanding outcome as compared to the traditional bulk blending method. , These outcomes suggest the excessive capability of the microfluidic method for specific bottom-up controlling of physicochemical properties of polymeric nanoparticles for numerous programs, which include controlled drug delivery. , , Electrokinetic microfluidic devices for rapid, low electricity drug release in self-sustaining microsystems evolved by Chung et al as a low electricity and strong electroactive microwell-primarily based totally implantable drug delivery system, meant to be used with self-sustaining microsystems, are presented. The tool has a top silicon-primary-based shape that describes the drug storage location and an electrically functionalized PDMS (polydimethylsiloxane) as a polymer. The drug release mechanism evolved right here exploits localized electrokinetic consequences of controlled drug release time and rate of chemical compounds saved in an unbiased, proper storage area. , It is proven how this could lessen the dosage time from hours to seconds over preceding diffusion, primarily based on the usage of as low as 20 mJ of strength in step with the dose. It turned into determined that the release technique can be completed in much less than 2 min or the usage of as low as 20 mJ of energy, each of which in comparison favorably to the state of the artwork microsystems. For the version of the electrokinetic delivery concerned with the inside of the release technique, detailed three-dimensional numerical simulations were used. The simulated model showed that much of the contents are released early from this technique. It further offers a physical point of view of the delivery process. ,
Microfluidic devices for particle production are mainly of two types, namely microchannels and microcapillaries. Microchannel-based devices are commonly fabricated by processes such as micromilling, micromachining, lithography, and mold replication. In such devices, the interfacial vicinity minimization brings about spontaneous droplet formation, and therefore, while preserving the oil-phase flow rate inside an optimum range, the droplet dimension is best depending on the microchannel geometry. Manufacturing of microchannel-primarily based totally overall devices is costly and takes a long time, but it allows for the fabrication of microsystems with particle size as small as a few tens of microns. Furthermore, in such devices, the microchannels may be flawlessly aligned, and in addition to uniform flow and highly liquefying of specific droplets or splitting droplets to uniform size, may serve to equip them, and the systems may be extended to produce a large number of products. Besides, capillary-primarily based structures are frequently made from low-cost components available in the market and have the capability to be microchannels in particle manufacturing. Importantly, the ones systems can be fabricated in a great deal much less time and can be operated in aggressive process conditions. In microchannel-based complete devices, the dispersed phase is in proximity to the tool’s wall before being emulsified by the continuous phase, which may result in a phase inversion. While the affinity of the dispersed phase for matter is greater than that of the continuous phase, the dispersed phase preferentially wets the partitions of the tool. This turns the choice of the substance for the production of the device has superior importance to all others. Yet, phase inversion can be avoided via deciding on the proper device for aqueous or organic droplets. Preferentially, the polymers used may be changed on the vicinity wherein droplets of the dispersed phase are arise (develop); however, this interprets into a further step withinside the micro-fabrication process. By contrast, capillary-primarily-based totally devices are tremendous for such terms. Here, the dispersed levels may be brought to the centerline of the non-stop segment flow, stopping the droplets from assembling the device’s partitions. Furthermore, capillary-primarily based totally devices do away with clogging troubles not unusual place in microchannel-primarily based totally devices and permit the manufacturing of oil-in-water (O/W) or water-in-oil (W/O) emulsions with a unit microsystem. These developments have allowed microfluidic devices to be fabricated using a wide range of materials and geometries, enabling new and advantageous physical behaviors and qualities in microfluidic devices. For laminate microfluidic devices, each layer is cut individually. The cutting method has a significant impact on the dimensions and the device functionality. For prototyping and laboratory settings, cutting is usually done with a knife plotter (ie, xurography) or laser cutter because of the speed and simplicity that each tool offers. A knife plotter works by precisely cutting a material with a blade to create the geometry, while a laser cutter uses a focused beam (traditionally CO2 lasers are used). , Under these conditions, the reduction of the diameter of droplets could be obtained by increasing flow rate, density ratio, and viscosity. Concerning the shape of devices used Bottom-up technologies that rely on emulsions or self-assembly cannot always provide fine, pre-designed control over particle geometry (shape, aspect ratio) and composition. Bicudo and co-workers investigated the manufacture of HANPs crosslinked with adipic dihydrazide (ADH) and chloride carbodiimide utilizing a microchannel flow-focusing system (EDCl). The work is focused on analyzing the process parameters of the unique method, which is a continuous nanoprecipitation at the water–organic solvent interface. The effect of the type of organic solvent employed, non-solvent flow rate, and Hyaluronic Acid (HA) content on the characteristics of (hyaluronic acid nanoparticles) HANPs was investigated by the authors. The findings revealed that through water diffusion and nanoprecipitation rates, the affinity between water and organic solvent regulated the mean diameter of the nanoparticles (NPs). When the non-solvent exhibited an intermediate affinity for water, the polydispersity was narrowed. Furthermore, because the process was convection regulated, lower HA concentrations and higher isopropyl alcohol flow rates resulted in smaller particles. Regardless of the organic solvent, flow rate, or HA concentration, somewhat stable NPs were formed. The procedure was found to be straightforward, repeatable, and quick. The authors found that this process was promising for manufacturing oil-free HANPs, which are important for medical, pharmacological, and aesthetic applications. , The benefits and drawbacks of these methods are mentioned in .
Beyond that, there has been great attraction of pharmaceutical formulators to synthetic nanocarriers and natural nanocarriers and colloidal systems have been of little interest. Significant attention has recently been paid to the production of organic nano-carriers, particularly in pharmaceuticals, as pharmaceutical scientists have begun to recognize the important properties conferred on nano-carriers by the microfluidic methods. , In general, nano-carriers are obtained through the dispersion of preformed polymers or through the monomer reactions to form polymers. These nanocarriers may be advanced via unique strategies and are categorized into classes relying on processes involved. Emulsification of materials is required in the primary group, whereas it is not always required in the other groups. As a result, it provides a clean and simple method of synthesis. When those strategies are utilized in traditional device, lack of control over uniform blending, formation, and improved effects on formulation ingredients and, as a consequence, very few products have an excessive particle size distribution. On the other hand, microfluidics controlling structures are able to supply control over the above-mentioned elements as a consequence of giving them uniform size particles. Lipid polymer hybrid nanoparticles (LPHNPs) have been merged as capacity nanocarriers. LPHNPs were created with the help of the microfluidic co-flow nanoprecipitation method. The internal fluid changed into organized with the aid of using dissolving poly (lactic-co-glycolic acid) (2 mg/mL) into acetonitrile as a natural phase. The outer fluid contained lecithin and Distearoyl-sn-glycero-3-phosphoethanolamine-polyethylene glycol 2000 in a two-to-three mass ratio dissolved in 4% ethanol and answered in water. , These defined compatibilities, higher internalization, protection profile and a relative better anticancer interest within breast cancers with controlled drug release behavior. , , Li et al conducted research on open-channel, water-in-oil emulsification in a paper-based microfluidic drug delivery system. Providing novel open-channel microfluidics that is primarily based on surfaces with designed wettability and has the ability to control the floating of all excessive and occasional surface tension of liquids. The open channel devices have been capable of constraining numerous smaller surface tension oils at excessive, but sufficient flow rates to enable water-in-oil microfluidic emulsification in an open channel tool. It ought to extrude the dimensions of the emulsified droplets formed inside the open channel tool with the aid of adjusting the speed of each of the dispersed aqueous and natural continuous phases. Finally, a fabricated tool has been changed to be used efficaciously to synthesize surprisingly monodisperse hydrogel microparticles that might contain a drug molecule. Additional investigation of the drug delivery properties of the fabricated products had promising results with an open-channel microfluidic devices having the potential to achieve a high level of fluid manipulation while being manufactured quickly and cheaply. Another study conducted by Raphael and Martin utilized dynamic microfluidic method; dopamine was used as a model drug to quantify the electrochemical flow on paper-based devices. They combined an electrochemical method with a microfluidic device for the time-resolved detection of dopamine from neuron-like PC12 cells cultured on filter paper. After an investigation of cellular adherence to the exterior of the paper with fluorescence microscopy; dopamine drug delivery after stimulation with acetylcholine without/with drugs has been investigated As a result, the data collected with the device agreed with single-cell statistics, demonstrating the technique’s validity for higher-throughput quantification of chemical targets on tissues- or organs-on-a-chip. Generally, microfluidic devices maintain numerous qualities in the pharmaceutical sciences, which consist of suitable dosage, ideal drug delivery, site-targeted delivery, extended and controlled release, less repetitive dosing, and minimum side effects. These benefits are important qualities of drug delivery systems. Microfluidic techniques have been regularly employed in the production of polymers as carriers of many active moieties, direct drug delivery systems, high-throughput screening, and additives and excellent carriers of drugs. Cheaper and effortlessly produced paper-based materials are good substrates that truly mitigate several challenges associated with transportation, filtration, and storage, concentrators, valving, and multiplexing. Hereafter, paper-based microfluidic generation in controlled drug delivery programs might carry thrilling possibilities to amplify the frame of understanding within the subject and support the improvement of the scientific translation of drug delivery systems. Huang et al prepared a stimuli-responsive controlled Vinblastine (VBL) drug release device from magnetically sensitive chitosan capsules. A magnetically responsive controllable drug delivery device has been designed by way of a means of embedding superparamagnetic iron oxide (SPIO) nanoparticles (NPs) in a chitosan matrix and an exterior magnet. In addition, droplet microfluidics, that is a completely unique technique for producing polymer spheres, has grown to be used for the manufacturing of monodispersed chitosan microparticles. The prepared VBL and SPIO NPs-loaded chitosan microparticles were characterized and showed individual and distinctive controlled-release patterns. Thus, the release rate, time, and dose of VBL release have become controlled through an exterior magnet. The outcomes presume that the usage of a magnetically responsive controlled drug delivery device brings a precious possibility for VBL drug release, wherein the delivery device is an energetic contributor, instead of a passive vehicle, within the optimization of most cancer treatments. This method actively focused magnetic drug delivery device to bring many benefits over traditional drug delivery structures through enhancing the precision and time of release, smooth operation, and better compliance for pharmaceutical applications. Because of their distinct physicochemical behavior and synergetic impact in preventing and inhibiting the progression of colorectal cancer, Liu et al chose atorvastatin and celecoxib as version dosage forms for their microfluidic collection of monodisperse multistage pH-responsive polymer/porous silicon composites for exactly controlled multi-drug delivery. To be efficient in prevention and inhibition of the acceleration of colon and rectal cancer, the fabricated microcomposite incorporating hypromellose acetate succinate, which is not soluble in acidic conditions, however, is incredibly soluble in basic (alkaline) pH environments. The benefit of the larger pore volume of porous silicon (PSi) is atorvastatin was initially loaded into the PSi matrix before being encapsulated into pH-responsive polymer microparticles containing celecoxib through microfluidics to acquire multi-drug loaded polymer/PSi microcomposites. The manufactured microcomposites confirmed monodisperse size distribution, multistage pH-response, particular ratiometric controllable loading extent closer to the concurrently loaded drug molecules, and tailor-made drug release kinetics of the loaded materials. This appealing microcomposite platform protects the payloads from being delivered at low pH values and improves the drug delivery at better pH values, which may be similarly used in preventing and treating colon and rectum cancer. Overall, the pH-responsive polymer/PSi-primarily based completely microcomposites may be used as a common platform for the combined drug delivery system of various drug molecules. Monodisperse biodegradable polymer microparticles were prepared by Xu et al using a microfluidic flow-focusing tool for controlled drug delivery and have described the fabrication of monodisperse, drug-loaded microparticles from biodegradable polymers, the use of microfluidic flow-focusing devices and the drug-delivery properties of these particles. Particles are engineered in a variety of sizes ranging from 10mm to 50mm. These particles are nearly monodispersed with a polydispersity index of 3.9%. Bupivacaine (amphiphilic) is included in the biodegradable matrix of the debris to characterize the formulation as a model drug. The kinetic evaluation suggests that the release of the drug from those monodisperse microparticles is slower than that from traditional strategies of identical average size, but reveals a larger particle size distribution and, more importantly, a substantial decrease in the primary burst than that discovered with traditional particles. , The distinction in the preliminary kinetics of drug release is attributed to the uniform distribution of the drug within the particles generated with the assistance of microfluidic strategies. These outcomes showed the application of microfluidic flow-focusing on the technology of homogenous systems of particles for drug delivery. Meng and co-workers prepared thermal-sensitive liposomal controlled release by using a disposable microfluidic tool. The release of an encapsulated drug from a nano-carrier consisting of a liposome must increase local drug delivery while reducing the toxicity consequences of a temperature increase. High-intensity focused ultrasound (HIFU), particularly micro-HIFU (MHIFU) microfluidic devices, allows us to imitate the bulky HIFU transmission tool with the help of lower power consumption and control the release of low temperature-sensitive liposomes (LTSL) investigated. , Furthermore, at the transition to a local temperature of between 41°C and 43°C, the structure changes from a gel to a liquid crystalline phase, releasing the encapsulated drugs with an external source of hyperthermia, such as microwaves, and an infrared radiation laser with the structure of a lipid membrane of low-temperature sensitive liposomes (LTSL). The microfluidic era may also provide a promising approach to studying the complex dynamics of ultrasound and organisms at the ultramicroscopic level. A principal task within the improvement of polymeric nanoparticles for numerous programs is the specific engineering of preferred physicochemical properties in a reproducible manner. Shamsi and co-workers developed microfluidic self-assembly of polymeric nanoparticles with tunable compactness for controlled drug delivery. , It is proven that the compactness of nanoparticles primarily based on self-assembled hydrophobically modified chitosan (HMC) biopolymers may be dictated by tunable fast blending through hydrodynamic flow focusing in microfluidic channels. It has been demonstrated through various flow rates, as well as the hydrophobicity of the chitosan chains, that the self-assembly properties of the chains can be controlled by optimizing the dimensions and compactness of the species, as well as a greater limited particle size distribution of the nanoparticles. , The particle size of the formulation components increased with increasing blending time, while chitosan produced smaller and more compact nanoparticles with a much smaller variety of aggregated chains and a higher degree of hydrophobicity. The investigation revealed that, to the greatest extent possible, despite the lack of affinity for the aqueous medium and at blending times longer than the time of aggregation, nanoparticles with nearly equal forms of hydrophobic adhesion were formed. Furthermore, exploring the effectiveness of microfluidics directed to organizing HMCs and encapsulating paclitaxels, a common anticancer drug, has discovered remarkably higher encapsulation efficiency overall performance in comparison to the conventional bulk method. The in-vitro release of the paclitaxel from the synthetic nanoparticles was evaluated to analyze the effect of the compactness of the formulation components on the drug release properties. The anticipated values of the diffusion coefficient of paclitaxel as high as 50% at drug release implied sustainability in controlling drug release regarding the compactness of the nanoparticles, and an outstanding outcome as compared to the traditional bulk blending method. , These outcomes suggest the excessive capability of the microfluidic method for specific bottom-up controlling of physicochemical properties of polymeric nanoparticles for numerous programs, which include controlled drug delivery. , , Electrokinetic microfluidic devices for rapid, low electricity drug release in self-sustaining microsystems evolved by Chung et al as a low electricity and strong electroactive microwell-primarily based totally implantable drug delivery system, meant to be used with self-sustaining microsystems, are presented. The tool has a top silicon-primary-based shape that describes the drug storage location and an electrically functionalized PDMS (polydimethylsiloxane) as a polymer. The drug release mechanism evolved right here exploits localized electrokinetic consequences of controlled drug release time and rate of chemical compounds saved in an unbiased, proper storage area. , It is proven how this could lessen the dosage time from hours to seconds over preceding diffusion, primarily based on the usage of as low as 20 mJ of strength in step with the dose. It turned into determined that the release technique can be completed in much less than 2 min or the usage of as low as 20 mJ of energy, each of which in comparison favorably to the state of the artwork microsystems. For the version of the electrokinetic delivery concerned with the inside of the release technique, detailed three-dimensional numerical simulations were used. The simulated model showed that much of the contents are released early from this technique. It further offers a physical point of view of the delivery process. ,
Pattern production is required for many organic and chemical assays, hence mixing is an important stage in the production of microfluidic platforms. Diffusion-based amalgam techniques fall short of meeting the current demand for quick and uniform blending. Scholars have studied advances in crucial blending enhancement tactics, including blending with power sources, as well as difficult channel geometry. Real-time tracking and the capacity to mix with diverse continuous-flow properties are also advantages of continuous-flow microfluidic separation. An appropriate outside pressure for the group components can be chosen based on the particular signature of the group components, and an appropriate outside pressure may be selected for the separation process. Cutting-area advances in continuous-stream microfluidic separation strategies, which include magneto-fluidics, inertial microfluidics, acoustic-fluidics, dielectrophoretic, and optofluidics, have been developed. Emerging programs of mixed continuous-flow separation and combining technology for extra superior microfluidic platforms, including diagnostic and therapeutic micro bioreactors, lab-on-a-chip, and microfluidic chromatography for protein purification, have been investigated. Droplet-primarily based microfluidic strategies (DMF), including electrowetting-on dielectric (EWOD), dielectrophoresis, and magnetic strategies have been explained. Programs for more advanced combinatorial DMF devices have also been introduced. In addition, manipulation strategies for liquid marble as a microbioreactor have been demonstrated. Recent advances in microfluidics suggest that extra complicated microfluidic structures, specifically for blending programs, may be fabricated with three-D printing. The design freedom provided by three-D printing will enable novel designs of nanomedicine formulations and preparations that were previously not possible with planar micromachining strategies such as soft lithography with poly-di-methyl-siloxane. Microfluidic cell way of life may be taken into consideration because of the next-technology method for biomedical and pharmaceutical programs. Liquid marble became as a promising digital microfluidics strategy. Continuous-flow microfluidics will remain used for programs that require excessive throughput. However, the trouble of cumbersome outside liquid transport and the requirement of optical microscopy for characterization makes continuous-flow microfluidics much less appropriate for programs with constrained pattern sizes. Digital microfluidics with droplets and liquid marbles is the answer to the problems of cumbersome outside structures, in addition to the rather big pattern volume. As the latest work is best at the proof-of-idea of liquid-marble-primarily based on totally virtual microfluidics, computerized structures for developing liquid marble, and the controlled manipulation of liquid marble, including coalescence and splitting, are areas of interest for bringing this platform toward realistic use. In summary, microfluidic technology allows for extremely precise liquid administration. It can be connected to an actuator system for on-demand or continuous drug release. Microfluidics has revolutionized the manufacture of drug carriers and the development of direct drug administration chips in general. Producing drug carriers that can generate a repeatable release profile as well as the controlled release of many compounds with varied release profiles needs the use of microfluidic technology.
|
Retrospective Analysis of the Thai Percutaneous Coronary Intervention Registry: Impacts of Center Volume and Operator Experience on Outcomes | c801c4c2-fa31-4fae-8761-fa42aba07c7a | 11788963 | Surgical Procedures, Operative[mh] | Introduction Outcomes of percutaneous coronary intervention (PCI) procedures can vary due to multiple factors, including patient clinical condition, complexity of coronary lesions, expertise of operators, and quality of the PCI center. The significance of operator skill and of a center's PCI volume becomes more important when managing complex anatomical conditions (e.g., left main disease or chronic total occlusion), critical clinical scenarios (like primary PCI), or procedural complications . Previous studies have suggested that there is a direct relationship between favorable PCI outcomes and high‐volume operators and centers, often referred to as the “volume‐outcome” relationship . In, the American College of Cardiology Foundation (ACCF), American Heart Association (AHA), Society for Cardiovascular Angiography and Interventions (SCAI) and the European Society of Cardiology (ESC)/European Association for Cardio‐Thoracic Surgery (EACTS) have all endorsed recommendations for minimum institutional and operator annual volumes to maintain satisfactory outcomes, particularly in patients with ST elevation myocardial infarction (STEMI). These volume‐outcome guidelines have been globally adopted to maximize quality and safety of the PCI procedure. While past studies indicated that there are positive correlations between both operator experience and PCI center volume, and PCI outcome, more recent studies find this association to be weaker than in those earlier studies . Such a trend could be attributed to the increased standardization of PCI techniques, advancements in stent technology enhancing deliverability and performance and/or the development of specialized equipment enabling easier completion of complex PCI procedures. Thus, the procedure may have become less reliant on individual operator skills and more influenced by technical advancements. Notably, a study from the US Veterans Affairs’ healthcare system finds no association between operator nor facility volumes and 30‐day mortality . Similarly, prior publications involving Asian populations presented inconclusive findings regarding an association between operator volume and in‐hospital outcome following PCI . The direct mechanisms by which operator experience and hospital volume may impact outcomes remains an area of research and debate within the interventional community, both in Asia and the West. Additionally, short‐term PCI outcomes, including in‐hospital mortality, might not reflect an operator's skill level, particularly when assessed in isolation . Thus, there is a need to identify or refine variables which better predict patient outcome. Extending the period of outcome observation may better reflect an operator's skill and the influence of a PCI center's volume. Before 2010, qualified interventional cardiologists in Thailand typically gained their skills through overseas training or self‐directed learning, as formal training programs were not established until later. Since then, numerous cardiac catheterization laboratories have been operating across the country over the past decade. The government's policy also enhances accessibility to PCI treatment by leveraging various health insurance schemes. These efforts aimed to enhance patient accessibility to PCI treatment. To further improve PCI outcomes and patient safety, ensuring operator proficiency and PCI center volume remain important. Therefore, this study utilized data from the Thai PCI Registry to assess the relationships of PCI center volume and operator experience to PCI outcomes, specifically PCI failure, procedural complications, and both PCI‐related in‐hospital and 1‐year all‐cause mortalities.
Methods This retrospective cohort study utilized a nationwide, prospective multicenter PCI Registry initiated in 2018 by the Cardiac Intervention Association of Thailand. The study protocol was detailed in a previous publication . Briefly, all Thai PCI facilities were invited to participate in the Registry and 39 out of 72 facilities voluntarily joined in this registry initiative. These included university, government, and private PCI centers which were situated across all five regions of the country. Patients included in this study were aged 18 years or older and underwent PCI during the periods May 1, 2018–April 2, 2019, and June 21–August 1, 2019. Patients involved in other clinical trials, unable to complete follow‐up, or who declined participation were deemed ineligible. The study received approval from the Central Research Ethics Committee of Thailand (COA‐CREC # 006/2018) and the Ethics Committee of the Faculty of Medicine, Ramathibodi Hospital, Mahidol University (COA‐MURA 2024/129). Written informed consent was given by each patient (or his/her legal representative) before the PCI intervention. Note that patients who underwent procedures and subsequently died were maintained in this analysis. 2.1 Data Collection All electronic databases were stored at a central data management unit (DMU), in the Department of Clinical Epidemiology and Biostatistics, Faculty of Medicine, Ramathibodi Hospital. Data were initially recorded on case record forms (CRFs), and then entered into the electronic databases by trained staff from the catheterization labs. The authors audited data from all study sites to validate accuracy and completeness, randomly selecting 10% of the total number of PCI patients from each site. Medical records and hard copies of CRFs were meticulously reviewed during these audits. Follow‐up audits were conducted in cases where the central DMU had questions regarding the correctness of the data. Following demographic, clinical, and angiographic characteristics, along with procedural data were extracted from the DMU. The patient data analyzed encompassed various factors: age, gender, health insurance (universal coverage, government service/state enterprise, social security service, uninsured or self‐pay), body‐mass index (BMI), presence of cardiovascular risk factors (diabetes mellitus [DM], hypertension, dyslipidemia, smoking, and chronic kidney disease [CKD, defined as eGFR < 60 ml/min/1.73m 2 ]), and history of relevant underlying cardiovascular diseases (cerebrovascular disease [CVD], peripheral arterial disease [PAD], myocardial infarction [MI], previous PCI/coronary artery bypass graft [CABG]). The clinical and angiographic data compiled included: clinical presentation (ST‐elevation myocardial infarction [STEMI], non‐ST‐elevation myocardial infarction [NSTEMI]/unstable angina [UA] and stable CAD), left ventricular ejection fraction (LVEF), presence of cardiogenic shock before PCI, prior thrombolytic treatment, type of PCI (urgent/emergent, primary PCI, rescue PCI), number of diseased vessels [single‐vessel disease (SVD), double vessel disease (DVD), triple vessel disease (TVD), left main (LM)], SYNTAX score (< 22, 23–32, > 33%), lesion complexity types (A, B1, B2, C), number of lesions treated, and stent deployed, mechanical support devices (intra‐aortic balloon pump [IABP] insertion, extracorporeal membrane oxygenation [ECMO], ventricular assist device [VAD]), radial access, stent type (drug eluting stent [DES], bare metal stent, bioabsorbable stent), mode of lesion severity assessment (intravascular ultrasound study [IVUS], optical coherence tomography [OCT] or fractional flow reserve [FFR] wire), and use (or not) of plaque modification devices (rotational atherectomy, cutting/scoring balloon or laser atherectomy), glycoprotein IIb/IIIa inhibitor, and post‐PCI medication treatments such as aspirin, statin, clopidogrel, prasugrel, and ticagrelor. 2.2 Definitions of PCI Center and Operator Volume The determinations of PCI center and operator volumes (number of PCIs performed annually) were extracted from the Thai PCI Registry databases. The duration of operator experience was calculated by subtracting the year of current practice from the year of completion of interventional cardiology training. We established the stratification for PCI center and operator volumes in accordance with the ACCF/AHA/SCAI 2011 (and 2013 Update) Clinical Competence Statement on Coronary Artery Interventional Procedures . For PCI center volume, groupings of annual PCI volumes were defined as follows: low (< 200), intermediate (200–499), and high (≥ 500). Operator experience was evaluated based on years of practice (low [< 5] and high [≥ 5]) and number of PCI cases performed per year (low [< 75] and high [≥ 75]). 2.3 PCI Outcomes of Interest The PCI outcomes of primary interest were PCI‐related in‐hospital mortality and 1‐year all‐cause mortality. Additionally, instances of PCI failure (defined as the inability to cross lesions with a wire, or to advance a balloon or stent, resulting in > 50% residual stenosis) and procedural complications were carefully documented. Procedural complications included nonfatal MI, nonfatal stroke, cardiogenic shock, heart failure, new requirements for dialysis, major bleeding requiring blood transfusion, endotracheal intubation, cardioversion/defibrillation, and in‐hospital CABG. Over the 1‐year follow‐up period, instances of nonfatal MI, nonfatal stroke, and unplanned repeat PCI were also recorded. In cases where a patient experienced multiple cardiovascular events, only the first event was considered. The definition of MI was an increase in cardiac troponin (cTn) coupled with one of the following: (1) evidence of prolonged ischemia characterized by chest pain lasting over 20 min; (2) ischemic ST‐segment changes or the emergence of new pathological Q waves; (3) angiographic evidence of coronary occlusion or no‐reflow/slow flow; (4) imaging demonstrating new loss of viable myocardium or new regional wall motion abnormality. Stroke was defined as a history of ischemic or hemorrhagic stroke, or a transient ischemic attack, validated through CT or MRI scans. 2.4 Statistical Analysis Data are presented as mean and standard deviation (SD), or median and interquartile range (IQR) where appropriate for continuous data; frequency and percentage for categorical data. Characteristics of patients, PCI centers, and operators were then compared between groups using t‐test or the Mann‐Whitney U test where appropriate for continuous data; using Chi‐square test for categorical data. Multivariate logistic regression was applied to assess associations between PCI center and operator factors, and PCI outcomes after adjustment for co‐variables. All analysis were performed using STATA version 17.0. P‐values of less than 0.05 were considered statistically significant.
Data Collection All electronic databases were stored at a central data management unit (DMU), in the Department of Clinical Epidemiology and Biostatistics, Faculty of Medicine, Ramathibodi Hospital. Data were initially recorded on case record forms (CRFs), and then entered into the electronic databases by trained staff from the catheterization labs. The authors audited data from all study sites to validate accuracy and completeness, randomly selecting 10% of the total number of PCI patients from each site. Medical records and hard copies of CRFs were meticulously reviewed during these audits. Follow‐up audits were conducted in cases where the central DMU had questions regarding the correctness of the data. Following demographic, clinical, and angiographic characteristics, along with procedural data were extracted from the DMU. The patient data analyzed encompassed various factors: age, gender, health insurance (universal coverage, government service/state enterprise, social security service, uninsured or self‐pay), body‐mass index (BMI), presence of cardiovascular risk factors (diabetes mellitus [DM], hypertension, dyslipidemia, smoking, and chronic kidney disease [CKD, defined as eGFR < 60 ml/min/1.73m 2 ]), and history of relevant underlying cardiovascular diseases (cerebrovascular disease [CVD], peripheral arterial disease [PAD], myocardial infarction [MI], previous PCI/coronary artery bypass graft [CABG]). The clinical and angiographic data compiled included: clinical presentation (ST‐elevation myocardial infarction [STEMI], non‐ST‐elevation myocardial infarction [NSTEMI]/unstable angina [UA] and stable CAD), left ventricular ejection fraction (LVEF), presence of cardiogenic shock before PCI, prior thrombolytic treatment, type of PCI (urgent/emergent, primary PCI, rescue PCI), number of diseased vessels [single‐vessel disease (SVD), double vessel disease (DVD), triple vessel disease (TVD), left main (LM)], SYNTAX score (< 22, 23–32, > 33%), lesion complexity types (A, B1, B2, C), number of lesions treated, and stent deployed, mechanical support devices (intra‐aortic balloon pump [IABP] insertion, extracorporeal membrane oxygenation [ECMO], ventricular assist device [VAD]), radial access, stent type (drug eluting stent [DES], bare metal stent, bioabsorbable stent), mode of lesion severity assessment (intravascular ultrasound study [IVUS], optical coherence tomography [OCT] or fractional flow reserve [FFR] wire), and use (or not) of plaque modification devices (rotational atherectomy, cutting/scoring balloon or laser atherectomy), glycoprotein IIb/IIIa inhibitor, and post‐PCI medication treatments such as aspirin, statin, clopidogrel, prasugrel, and ticagrelor.
Definitions of PCI Center and Operator Volume The determinations of PCI center and operator volumes (number of PCIs performed annually) were extracted from the Thai PCI Registry databases. The duration of operator experience was calculated by subtracting the year of current practice from the year of completion of interventional cardiology training. We established the stratification for PCI center and operator volumes in accordance with the ACCF/AHA/SCAI 2011 (and 2013 Update) Clinical Competence Statement on Coronary Artery Interventional Procedures . For PCI center volume, groupings of annual PCI volumes were defined as follows: low (< 200), intermediate (200–499), and high (≥ 500). Operator experience was evaluated based on years of practice (low [< 5] and high [≥ 5]) and number of PCI cases performed per year (low [< 75] and high [≥ 75]).
PCI Outcomes of Interest The PCI outcomes of primary interest were PCI‐related in‐hospital mortality and 1‐year all‐cause mortality. Additionally, instances of PCI failure (defined as the inability to cross lesions with a wire, or to advance a balloon or stent, resulting in > 50% residual stenosis) and procedural complications were carefully documented. Procedural complications included nonfatal MI, nonfatal stroke, cardiogenic shock, heart failure, new requirements for dialysis, major bleeding requiring blood transfusion, endotracheal intubation, cardioversion/defibrillation, and in‐hospital CABG. Over the 1‐year follow‐up period, instances of nonfatal MI, nonfatal stroke, and unplanned repeat PCI were also recorded. In cases where a patient experienced multiple cardiovascular events, only the first event was considered. The definition of MI was an increase in cardiac troponin (cTn) coupled with one of the following: (1) evidence of prolonged ischemia characterized by chest pain lasting over 20 min; (2) ischemic ST‐segment changes or the emergence of new pathological Q waves; (3) angiographic evidence of coronary occlusion or no‐reflow/slow flow; (4) imaging demonstrating new loss of viable myocardium or new regional wall motion abnormality. Stroke was defined as a history of ischemic or hemorrhagic stroke, or a transient ischemic attack, validated through CT or MRI scans.
Statistical Analysis Data are presented as mean and standard deviation (SD), or median and interquartile range (IQR) where appropriate for continuous data; frequency and percentage for categorical data. Characteristics of patients, PCI centers, and operators were then compared between groups using t‐test or the Mann‐Whitney U test where appropriate for continuous data; using Chi‐square test for categorical data. Multivariate logistic regression was applied to assess associations between PCI center and operator factors, and PCI outcomes after adjustment for co‐variables. All analysis were performed using STATA version 17.0. P‐values of less than 0.05 were considered statistically significant.
Results A total of 19,701 patients who underwent PCI were included in the analysis; 1‐year follow‐up data were available for 17,432. Their average age was 64.1 ± 11.7 years; 69.1% were male and 60% classified as overweight (Table ). Atherosclerotic risk factors were common, with 67.4% having hypertension, 65.3% dyslipidemia, 44.2% DM, 28.9% CKD without dialysis, and 3.5% CKD requiring dialysis; 55.4% were current or ex‐smokers. Notably, 23.3% had a previous history of MI, and 20.7% had undergone a prior PCI. The majority (63.6%) of patients were covered by universal health coverage, followed by those covered under government service/state enterprise (26.6%) or social security service (6.7%). Regarding clinical presentation, about 58.1% of patients showed acute coronary syndrome, comprising either STEMI or NSTEMI/UA, while the remainder exhibited stable CAD. Nearly half of the patients had triple‐vessel or left main disease, and 22% were found to have impaired left ventricular systolic function (LVEF < 40%), with 8.7% presenting with cardiogenic shock. Of the 39 participating hospitals, the distribution of PCI volumes was as follows: 17 hospitals (43.6%) were categorized as high‐volume PCI centers, nine hospitals (23.1%) as intermediate‐volume centers, and 13 hospitals (33.3%) as low‐volume centers. There were 135 operators involved; of them, 115 (85.2%) had over 5 years of interventional practice. When considering the number of PCIs done annually, 79 operators (58.5%) performed a high number (> 75) of cases. 3.1 Comparison of Baseline Characteristics Among Groups The majority of patients underwent PCI procedures at high‐volume PCI centers (77.2%). Likewise, most patients received their PCI from experienced operators with over 5 years of interventional practice (80.3%) and a high number of PCIs annually (93.6%). A comparative analysis of baseline demographic, clinical, and angiographic characteristics, and procedural data stratified by both PCI center and operator volume, is presented in Table . In summary, no specific patterns were observed among groups in regard to age, gender and comorbidities. However, distinct angiographic and procedural features were found in each group. For instance, patients with high‐risk profiles and referred cases (e.g., STEMI, cardiogenic shock, rescue PCI, emergency/urgent PCI, post‐thrombolytic therapy, high SYNTAX score, and complex lesions [i.e., type B2 and C, including chronic total occlusion (CTO)]) tended to be treated at high‐volume centers by operators with more years of experience and a higher annual PCI volume. Conversely, primary PCI and emergent/urgent PCI procedures tended to be conducted at low‐volume PCI centers, by interventionists with fewer years of practice and lower annual PCI volumes. Regarding procedural features, approaches seemed inconsistent, potentially based on operator discretion. However, radial access was commonly used across all groups of operators and PCI centers, accounting for 30%–40% of cases. Lesion severity assessment primarily utilized IVUS (14.4%); the use of plaque modification devices (5.6%) was common in high‐volume PCI centers and among operators with greater years of practice. Drug‐eluting stents were used in nearly 90% of procedures, while the remaining 10% deployed either bare‐metal or bioabsorbable stents. Most patients were administered aspirin and statins (99%, and 93.0%, respectively) after PCI. Interestingly, clopidogrel was more frequently prescribed in high‐volume PCI centers (93.8%), whereas ticagrelor and prasugrel were commonly prescribed in low‐volume centers and by operators conducting fewer annual PCIs. Glycoprotein IIb/IIIa inhibitor was more frequently prescribed in high‐volume PCI centers (6.6%) followed by intermediate and low‐volume centers (4.4% and 2.9%, respectively). 3.2 In‐Hospital and 1 Year PCI Outcomes Overall, the percentage of PCI failures was 4.9%; with 5.1% experiencing procedural complications and 1.0% requiring blood transfusion due to bleeding. The in‐hospital all‐cause mortality rate was 2.7%. Despite patients with higher‐risk profiles being treated at high‐volume PCI centers and by experienced operators, there were no significant differences in PCI failure and in‐hospital mortality based on center volumes (Table ). However, high‐volume PCI centers had a higher rate of procedural complications (4.7%) compared to intermediate‐ (3.9%) and low‐volume centers (2.5%) ( p < 0.001). Similarly, high‐volume PCI centers experienced more major bleeds (1.1% compared to 0.4% and 0.7%, respectively; p < 0.001), especially among operators who conducted a higher number of PCIs annually. Additionally, there was a trend toward increased procedural complications among operators with more years of practice (4.6% vs. 3.9%, p = 0.055). At the 1‐year follow‐up, all‐cause mortality was increased (from 2.7% to 11.8%). Again, there were no significant differences in all‐cause of mortality among the groups based on PCI center volume nor operator experience. Patients treated at low‐volume PCI centers were more likely to develop nonfatal MI (11.8%) and experience unplanned repeat revascularization (2.5%) compared to those treated at intermediate and high‐volume centers (4.8%, 7.6%, p < 0.001, and 1.1%, 1.9%, respectively; p = 0.002). Interestingly, patients undergoing PCI by operators with more years of practice showed a higher incidence of nonfatal MI (7.5 vs. 6.5%, p = 0.036) and nonfatal stroke (1.2 vs. 0.8%, p = 0.027) compared to those with fewer years. Unplanned repeat revascularizations were also more common among operators conducting fewer PCIs annually (2.8 vs. 1.7%, p = 0.009). 3.3 Impact of PCI Center Volume and Operator Experience on PCI Outcomes Univariate analyses were conducted for all variables associated with procedural failure, procedural complications, PCI‐related in‐hospital mortality, and 1‐year all‐cause mortality (Table ). No significant associations were found between PCI center volume or operator experience and PCI failure or 1‐year all‐cause mortality. However, high‐ and intermediate‐ volume PCI centers were associated with increased procedural complications. But notably, intermediate PCI volume centers were associated with decreased in‐hospital mortality. After adjusting for confounding factors, no significant association was found between PCI center volume and any aspect of PCI outcome. Similarly, regardless of the number of years of practice and PCIs performed annually, operator experience was not significantly related to procedural complications nor 1‐year all‐cause mortality. However, operators performing a higher number of PCIs annually did tend to have fewer PCI failures (odds ratio [95% CIs] of 0.77 (0.57, 1.02), p = 0.066) (Table ). And those operators with more years of practice had lower PCI‐related in‐hospital mortality rates (odds ratio [95% CIs] of 0.76 (0.58, 1.00), p = 0.045) (Table ).
Comparison of Baseline Characteristics Among Groups The majority of patients underwent PCI procedures at high‐volume PCI centers (77.2%). Likewise, most patients received their PCI from experienced operators with over 5 years of interventional practice (80.3%) and a high number of PCIs annually (93.6%). A comparative analysis of baseline demographic, clinical, and angiographic characteristics, and procedural data stratified by both PCI center and operator volume, is presented in Table . In summary, no specific patterns were observed among groups in regard to age, gender and comorbidities. However, distinct angiographic and procedural features were found in each group. For instance, patients with high‐risk profiles and referred cases (e.g., STEMI, cardiogenic shock, rescue PCI, emergency/urgent PCI, post‐thrombolytic therapy, high SYNTAX score, and complex lesions [i.e., type B2 and C, including chronic total occlusion (CTO)]) tended to be treated at high‐volume centers by operators with more years of experience and a higher annual PCI volume. Conversely, primary PCI and emergent/urgent PCI procedures tended to be conducted at low‐volume PCI centers, by interventionists with fewer years of practice and lower annual PCI volumes. Regarding procedural features, approaches seemed inconsistent, potentially based on operator discretion. However, radial access was commonly used across all groups of operators and PCI centers, accounting for 30%–40% of cases. Lesion severity assessment primarily utilized IVUS (14.4%); the use of plaque modification devices (5.6%) was common in high‐volume PCI centers and among operators with greater years of practice. Drug‐eluting stents were used in nearly 90% of procedures, while the remaining 10% deployed either bare‐metal or bioabsorbable stents. Most patients were administered aspirin and statins (99%, and 93.0%, respectively) after PCI. Interestingly, clopidogrel was more frequently prescribed in high‐volume PCI centers (93.8%), whereas ticagrelor and prasugrel were commonly prescribed in low‐volume centers and by operators conducting fewer annual PCIs. Glycoprotein IIb/IIIa inhibitor was more frequently prescribed in high‐volume PCI centers (6.6%) followed by intermediate and low‐volume centers (4.4% and 2.9%, respectively).
In‐Hospital and 1 Year PCI Outcomes Overall, the percentage of PCI failures was 4.9%; with 5.1% experiencing procedural complications and 1.0% requiring blood transfusion due to bleeding. The in‐hospital all‐cause mortality rate was 2.7%. Despite patients with higher‐risk profiles being treated at high‐volume PCI centers and by experienced operators, there were no significant differences in PCI failure and in‐hospital mortality based on center volumes (Table ). However, high‐volume PCI centers had a higher rate of procedural complications (4.7%) compared to intermediate‐ (3.9%) and low‐volume centers (2.5%) ( p < 0.001). Similarly, high‐volume PCI centers experienced more major bleeds (1.1% compared to 0.4% and 0.7%, respectively; p < 0.001), especially among operators who conducted a higher number of PCIs annually. Additionally, there was a trend toward increased procedural complications among operators with more years of practice (4.6% vs. 3.9%, p = 0.055). At the 1‐year follow‐up, all‐cause mortality was increased (from 2.7% to 11.8%). Again, there were no significant differences in all‐cause of mortality among the groups based on PCI center volume nor operator experience. Patients treated at low‐volume PCI centers were more likely to develop nonfatal MI (11.8%) and experience unplanned repeat revascularization (2.5%) compared to those treated at intermediate and high‐volume centers (4.8%, 7.6%, p < 0.001, and 1.1%, 1.9%, respectively; p = 0.002). Interestingly, patients undergoing PCI by operators with more years of practice showed a higher incidence of nonfatal MI (7.5 vs. 6.5%, p = 0.036) and nonfatal stroke (1.2 vs. 0.8%, p = 0.027) compared to those with fewer years. Unplanned repeat revascularizations were also more common among operators conducting fewer PCIs annually (2.8 vs. 1.7%, p = 0.009).
Impact of PCI Center Volume and Operator Experience on PCI Outcomes Univariate analyses were conducted for all variables associated with procedural failure, procedural complications, PCI‐related in‐hospital mortality, and 1‐year all‐cause mortality (Table ). No significant associations were found between PCI center volume or operator experience and PCI failure or 1‐year all‐cause mortality. However, high‐ and intermediate‐ volume PCI centers were associated with increased procedural complications. But notably, intermediate PCI volume centers were associated with decreased in‐hospital mortality. After adjusting for confounding factors, no significant association was found between PCI center volume and any aspect of PCI outcome. Similarly, regardless of the number of years of practice and PCIs performed annually, operator experience was not significantly related to procedural complications nor 1‐year all‐cause mortality. However, operators performing a higher number of PCIs annually did tend to have fewer PCI failures (odds ratio [95% CIs] of 0.77 (0.57, 1.02), p = 0.066) (Table ). And those operators with more years of practice had lower PCI‐related in‐hospital mortality rates (odds ratio [95% CIs] of 0.76 (0.58, 1.00), p = 0.045) (Table ).
Discussion This study investigated the correlations among PCI center volumes, operator experience, and short‐ and long‐term clinical outcomes by utilizing a nationwide PCI registry. Our findings revealed no significant associations between PCI center volume and PCI outcome such as PCI failure, procedural complications, PCI‐related in‐hospital mortality, and 1‐year all‐cause mortality. However, the study highlighted the significance of operator experience as a pivotal factor. We observed that more years of practice was associated with lower in‐hospital mortality. Additionally, operators performing a higher number of PCIs annually tended to have lower PCI failure rates. Despite minor absolute differences in risk among operators, which might partly be explained by unmeasured variations in case complexity, an inverse relationship persisted even in risk‐adjusted analyses. Several other countries have established registries to enable the analysis of clinical outcomes, which have subsequently guided their local recommendations and contributed to improvements in their national programs . Recent analyses have shown that operators with greater experience tend to have fewer complications, improved procedural success rates, and reduced mortality . Additionally, experienced operators often exhibit better proficiency in handling complex lesions (e.g., left main or CTO) resulting in better outcomes for challenging cases . PCI volume has frequently served as a surrogate for measuring quality due to its ease of measurement, with previous studies indicating correlations with outcomes at both operator and center levels . Analyzing volume outcomes also offers a means of benchmarking performance at both the center and operator levels. Nevertheless, it is crucial to acknowledge that discrepancies may occur in center‐ or operator‐volume outcomes at regional levels. Disparities in PCI volumes across different geographical regions might explain variation in the volume‐outcome relationships in existing literature , especially since what defines a high‐volume center in one country may not represent the same volume status in another country. Many PCI operators in the United States have been performing fewer PCI procedures annually than originally recommended, which led to revised ACCF/AHA/SCAI guidelines in 2013, replacing the 2011 recommendations , reducing the threshold for PCI center volume from > 400/year to > 200/year, and the annual number of PCIs per operator from > 75/year to > 50/year. In contrast, the ESC/EACTS guidelines in 2018 for myocardial revascularization state that a minimum of 75 PCIs per year is necessary to maintain proficiency. Meanwhile, the number of PCI procedures done annually in Thailand rises steadily. To uniformly classify PCI centers, we have adjusted the criteria for PCI center volume, setting it higher (> 500 PCIs/year) for high‐volume centers while maintaining < 200 PCIs/year for low‐volume PCI centers. Furthermore, regarding operator case volume, we have retained the criterion of > 75 PCIs/year for high volume instead of > 50/year as outlined in the 2013 guidelines. This adjustment aligns with a previous report that showed an association between center volumes of < 400 PCIs/year and operator volumes of < 75 PCIs/year to be consistently linked to higher rates of inpatient mortality and adverse events . In our study, skilled operators performing more than 75 PCIs/year tended to have lower PCI failure rates. However, there were no differences in procedural complications, PCI‐related in‐hospital mortality, nor 1‐year all‐cause mortality. In addition to considering operator case volume, our analysis addresses a gap in current data by examining the interplay between volume‐outcome relationships and the lifetime experience of operators. The existing evidence lacks insight into how an operator's lifetime experience impacts PCI outcomes . Therefore, we included years of experience as an interventionist as one of the factors to assess PCI outcomes. It seems logical to assume that lifetime learning may play an important role in reducing adverse events and complications, potentially beyond the experience gained through procedural volume. Our findings indicated a direct correlation between operators with more years of experience and reduced PCI‐related in‐hospital mortality. However, these experienced operators, more than those with less experience, may have encountered more challenging cases involving complex lesions or patients with unstable hemodynamics which necessitated hemodynamic support devices or plaque modification devices combined with lesion severity assessment . Consequently, patients treated by these experienced operators may have included more with procedural complications and major bleeding, and potentially more negative long‐term outcomes such as nonfatal MI and stroke. Nevertheless, after adjusting for all confounding risk factors, the PCI‐related in‐hospital mortality of patients treated by these experienced operators remained less, as compared to those treated by other operators. In terms of PCI center volume, evidence supports a relationship between volume and PCI outcome . Hospitals handling higher volumes of PCI cases generally demonstrate lower mortality rates, fewer complications, and improved procedural success. Furthermore, high‐volume PCI centers often possess experienced operators, specialized resources, experienced teams, established protocols, and streamlined processes. Conversely, low‐volume PCI centers typically had less experienced operators, often younger interventionists who had just finished their training programs. However, in our study, we observed no significant impact of either the main effect of PCI center volume or the interaction between PCI volume and operator experience on any aspect of PCI outcomes after adjusting for confounding factors. Several explanations may account for positive PCI outcomes in our low‐volume PCI centers. First, case selection played a pivotal role. Young operators often initiated their practice with simpler cases and referred more complex cases to high‐volume PCI centers with more experienced operators. Second, the presence of on‐site proctorship was noteworthy. Recently graduated interventionists frequently maintained relationships with senior mentors and invited them to assist with challenging cases, providing invaluable guidance. Our results support a recent article highlighting the value of operator experience and the added safety benefits for less experienced operators due to considerable support from more senior operators . Third, an efficient training program contributes significantly. The integration of advanced stent technology, specialized equipment, and increased utilization of radial access can streamline procedures , even within low‐volume centers. Fourth, swift access to treatment is facilitated by a rapid referral system and universal health insurance coverage, which is particularly important for initiating treatment of prevalent STEMI cases in low‐volume centers. Despite the fact that a higher annual hospital volume of primary PCI (> 36/year) is usually associated with lower mortality compared to those with lower volumes , our findings revealed no differences of the in‐hospital and 1‐year mortality rates based on either PCI volume or operator experience. Lastly, the frequent use of potent P2Y12 inhibitors , more common in low‐volume PCI centers, is associated with the younger interventionists who conduct fewer PCIs. The better outcome after primary PCIs is likely due to a combination of these explanations. Our study identified room for improvement for operators in low‐volume PCI centers. Over the 1‐year follow‐up period, the incidence of recurrent nonfatal MI and unplanned repeat PCIs (excluding staged PCI) was higher in low‐volume centers compared to intermediate‐ and high‐volume centers. Patients in these centers may have experienced in‐stent restenosis or stent thrombosis, necessitating repeat revascularization. Unfortunately, due to the need for further adjudication, comprehensive information regarding subsequent events could not be definitely established. Furthermore, not all centers had onsite cardiac surgery. Although there are data suggesting this may not influence outcome , Thailand's national policy encourages all hospitals to have onsite surgery, which can handle the more complex cases. Moreover, the primary mechanical support device used is the IABP, while less than 0.1% utilized ECMO or VAD; the Impella® heart pump was not used, as it was not available in the country. Such devices should be accessible to provide backup support when performing PCIs on high‐risk patients and those who develop unstable hemodynamics. 4.1 Study Limitations While studies utilizing PCI registries can provide valuable insights into the impact of operator experience and hospital volume on PCI outcome, there are some limitations that need to be considered when interpreting the results. First, the study focused on a Thai population and hospital system; generalizability of findings may be limited to where health and hospital system contexts are similar to those in Thailand. Second, not all PCI facilities are required to participate, and among those that do, not all consecutive procedures were registered. This lack of comprehensive registration could introduce selection bias. Third, the study's criteria regarding PCI center volume and the number of PCIs/operator limit comparability with other studies. Additionally, being a retrospective cohort study, there was some incomplete data and the analysis could not consider variables such as socioeconomic status, psychosocial factors, and concurrent treatments which may have confounded PCI outcomes. Fourth, the Registry primarily gathered quantitative procedural and outcome data. It tended to lack qualitative insights into procedural intricacies, operator proficiency, hospital resources, and team dynamics that may have impacted outcomes. Lastly, 1.2% of subjects did not complete the 12‐month assessment despite the team's effort and this incomplete follow‐up, though a small percentage, may have introduced bias into the findings.
Study Limitations While studies utilizing PCI registries can provide valuable insights into the impact of operator experience and hospital volume on PCI outcome, there are some limitations that need to be considered when interpreting the results. First, the study focused on a Thai population and hospital system; generalizability of findings may be limited to where health and hospital system contexts are similar to those in Thailand. Second, not all PCI facilities are required to participate, and among those that do, not all consecutive procedures were registered. This lack of comprehensive registration could introduce selection bias. Third, the study's criteria regarding PCI center volume and the number of PCIs/operator limit comparability with other studies. Additionally, being a retrospective cohort study, there was some incomplete data and the analysis could not consider variables such as socioeconomic status, psychosocial factors, and concurrent treatments which may have confounded PCI outcomes. Fourth, the Registry primarily gathered quantitative procedural and outcome data. It tended to lack qualitative insights into procedural intricacies, operator proficiency, hospital resources, and team dynamics that may have impacted outcomes. Lastly, 1.2% of subjects did not complete the 12‐month assessment despite the team's effort and this incomplete follow‐up, though a small percentage, may have introduced bias into the findings.
Conclusions In this national cohort of patients who underwent PCIs, center volumes did not significantly influence PCI outcomes after risk adjustment. However, it was evident that operator experience played a pivotal role. Operators with more years of practice were associated with decreased in‐hospital mortality rates, while those performing a greater number of PCIs annually tended to have fewer PCI failures. These findings offer valuable insights that can highlight areas for improvement and aid in the refinement of strategies within the Thai national PCI system, at both the center and operator level.
The authors declare no conflicts of interest.
Supporting information.
|
99b11f75-55e3-476f-ab79-32e49c0d7365 | 9564454 | Anatomy[mh] | INTRODUCTION Programed cell death ligand 1 (PD‐L1) is an immune regulatory molecule which can act as a coregulatory signal through binding to the inhibitory programed cell death 1 (PD‐1) receptor. Binding leads to inhibition of cytokine production and cytolytic activity of PD‐1‐expressing tumor infiltrating CD4 + and CD8 + T cells. During pregnancy, this mechanism has a crucial role in suppressing immune responses and allows development of the partly allogenic fetus and placenta. , Previous studies showed that PD‐L1 is strongly expressed by trophoblast cells in different types of gestational trophoblastic disease (GTD), , , evading immune responses, and thereby allowing proliferation of these (pre‐)malignant cells. GTD comprises a group of pregnancy‐related disorders, originating from placental tissue. It potentially progresses into gestational trophoblastic neoplasia (GTN), a malignant form of GTD. Although GTN is highly curable with current chemotherapies, the mortality rate is still 5% due to chemotherapy resistance, necessitating novel treatment approaches. Blockade of the PD‐1/PD‐L1 pathway with immune checkpoint inhibitors (ICI) has emerged as a novel therapy for various types of cancer. , In cancer patients with, for example, head and neck malignancies or melanoma, durable responses after treatment with PD‐1/PD‐L1 ICIs of up to 2 years in 20%–40% of patients have been reported. , In patients with unresectable chemotherapy‐resistant GTN, pembrolizumab, an antibody against PD‐1, can induce complete responses as described in several case reports. , , , , Selection of patients eligible for treatment with anti PD‐L1 ICIs, however, is challenging. Currently, there are multiple methods to assess expression of PD‐L1 on tumor cells using different antibodies, platforms, scoring systems and cutoff values, often linked to a specific ICI. , , In most studies, patients eligible for anti‐PD‐L1 treatment are selected based on the immunohistochemical (IHC) expression of PD‐L1 on tumor cells, inflammatory cells, or both. However, substantial response rates in patients with tumors lacking PD‐L1, and minimal response rates in patients with tumors highly expressing PD‐L1 have been reported. , Studies analyzing PD‐L1 expression in patients with GTD showed considerable variation in reported PD‐L1 expression patterns. , This variation may be explained not only by tumor heterogeneity but also by the use of different commercially available PD‐L1 antibodies for IHC analysis. , , , Therefore, it is currently insufficiently clear which PD‐L1 antibody is most suitable for detection of PD‐L1‐expressing trophoblast cells, to assess whether PD‐L1 would be a good marker for response to ICI in GTD patients. Therefore, the aim of this study was to identify the antibody most specific for detection of PD‐L1‐expressing cells, by evaluating the most frequently used commercially available PD‐L1 antibodies for IHC.
MATERIAL AND METHODS 2.1 Patient material Formalin‐fixed paraffin embedded (FFPE) samples of four patients with a complete hydatidiform mole (CHM) which progressed to post‐molar GTN after suction curettage, and four patients with a choriocarcinoma were selected from the Radboudumc pathology archives. Selection was based on the availability of FFPE specimens of the first uterine suction curettage at the Radboudumc, and concerned trophoblast tissue removed from patients before start of chemotherapy. Post‐molar GTN was defined according to the FIGO 2000 guideline (ie mola hydatidosa with serum hCG plateauing for three consecutive weeks or rising over a period of two consecutive weeks). 2.2 Flow cytometry Chinese hamster ovarian (CHO) cells (85 050 302, Sigma Aldrich) were transfected using Lipofectamine 3000 (L3000‐015; Invitrogen) with the cDNA constructs cloned into the mammalian expression vector pcDNA3.1+/C‐(K)‐DYK encoding for PD‐L1 and PD‐L2 (OHu22144D and OHu04434D, respectively, both from Genscript) as described before. CHO cells were used since these cells have been the most commonly used mammalian host for (large)‐scale commercial production of therapeutic proteins in the past decades. Flow cytometry was performed on (transfected) CHO cells with conjugated antibodies suitable for flow cytometric assays: anti–PD‐L1‐BV421 (clone MIH1, 563 738, BD Biosciences) and anti‐PD‐L2‐PE (clone MIH18, 558 066, BD Biosciences), with the FACS Verse (BD Biosciences). Flow cytometry data were analyzed using FlowJo software (v10; Tree Star). Transfection efficiency was ~80% for both PD‐L1 and PD‐L2 (Figure ). The remainder of cells were embedded in paraffin with AgarCyto cell block preparation and used for staining. 2.3 Immunohistochemistry Sections of 4‐μm thickness were cut and mounted on glass slides (900 226, VWR). Seven commercially available PD‐L1 antibodies were validated on wildtype Chinese ovarian hamster cells (CHO), and CHO cells transfected with PD‐L1 and PD‐L2. Tonsil tissue was used as a positive control, since PD‐L1 and PD‐L2 are expressed in lymphoid tissue. Primary antibodies included; anti‐PD‐L1 (clone E1L3N, 13 684, Cell Signaling), PD‐L1 Antibody Panel (clones 73–10, CAL10, SP142, 28–8, ab239749, Abcam), anti‐PD‐L1 (clone 22C3, M365329, DAKO) and anti‐PD‐L1 (clone SP263, Roche). IHC with anti‐PD‐L1 clone SP263 was performed on a VENTANA BenchMark ULTRA automated slide stainer as described by Hurkmans et al. Other IHC protocols were carried out manually with antibody dilutions and epitope retrievals according to the manufacturers’ instructions (Appendix ). The IHC‐stained CHO cell and tonsil slides were assessed individually by two pathologists specialized in gynecological oncology. Positivity of the PD‐L1 protein was defined by membranous staining. The percentage of positive cells was determined. Additionally, cytoplasmatic staining and staining intensity were documented. We scored staining intensity as weak (+/−), mild , strong or very strong . Scoring results of the slides were revealed after both pathologists scored all slides individually. Antibodies considered specific for PD‐L1 protein staining based on the results of the tonsil and transfected CHO samples, were used to stain four FFPE tumor tissue samples of patients with a CHM, and four FFPEs of patients with a choriocarcinoma. Samples were again analyzed by the two pathologists, and scoring results were revealed after they both scored all slides individually. To evaluate the agreement between the two pathologists in PD‐L1‐positive scored cells of the CHM and choriocarcinoma samples, we determined the intraclass correlation coefficient (ICC) for all tested antibodies individually. 2.4 Statistical analyses To calculate ICC, the two‐way random‐effects model with consistency was used. The ICCs were determined with both average and single measures. Average measures are used when measures of ≥2 raters are averaged to derive the result of a test. Single measures are used when a single rater performs the test, and is likely to be used in daily clinical practice. An ICC value of <0.50 is considered poor, between 0.50 and 0.75 moderate, between 0.75 and 0.90 good, and >0.90 red excellent. Analyses were performed with IBM SPSS statistics, version 25. 2.5 Ethics statement This study was approved by the local ethical committee of the Radboud University Medical Center (reference number 2018–4132) on June 7, 2018.
Patient material Formalin‐fixed paraffin embedded (FFPE) samples of four patients with a complete hydatidiform mole (CHM) which progressed to post‐molar GTN after suction curettage, and four patients with a choriocarcinoma were selected from the Radboudumc pathology archives. Selection was based on the availability of FFPE specimens of the first uterine suction curettage at the Radboudumc, and concerned trophoblast tissue removed from patients before start of chemotherapy. Post‐molar GTN was defined according to the FIGO 2000 guideline (ie mola hydatidosa with serum hCG plateauing for three consecutive weeks or rising over a period of two consecutive weeks).
Flow cytometry Chinese hamster ovarian (CHO) cells (85 050 302, Sigma Aldrich) were transfected using Lipofectamine 3000 (L3000‐015; Invitrogen) with the cDNA constructs cloned into the mammalian expression vector pcDNA3.1+/C‐(K)‐DYK encoding for PD‐L1 and PD‐L2 (OHu22144D and OHu04434D, respectively, both from Genscript) as described before. CHO cells were used since these cells have been the most commonly used mammalian host for (large)‐scale commercial production of therapeutic proteins in the past decades. Flow cytometry was performed on (transfected) CHO cells with conjugated antibodies suitable for flow cytometric assays: anti–PD‐L1‐BV421 (clone MIH1, 563 738, BD Biosciences) and anti‐PD‐L2‐PE (clone MIH18, 558 066, BD Biosciences), with the FACS Verse (BD Biosciences). Flow cytometry data were analyzed using FlowJo software (v10; Tree Star). Transfection efficiency was ~80% for both PD‐L1 and PD‐L2 (Figure ). The remainder of cells were embedded in paraffin with AgarCyto cell block preparation and used for staining.
Immunohistochemistry Sections of 4‐μm thickness were cut and mounted on glass slides (900 226, VWR). Seven commercially available PD‐L1 antibodies were validated on wildtype Chinese ovarian hamster cells (CHO), and CHO cells transfected with PD‐L1 and PD‐L2. Tonsil tissue was used as a positive control, since PD‐L1 and PD‐L2 are expressed in lymphoid tissue. Primary antibodies included; anti‐PD‐L1 (clone E1L3N, 13 684, Cell Signaling), PD‐L1 Antibody Panel (clones 73–10, CAL10, SP142, 28–8, ab239749, Abcam), anti‐PD‐L1 (clone 22C3, M365329, DAKO) and anti‐PD‐L1 (clone SP263, Roche). IHC with anti‐PD‐L1 clone SP263 was performed on a VENTANA BenchMark ULTRA automated slide stainer as described by Hurkmans et al. Other IHC protocols were carried out manually with antibody dilutions and epitope retrievals according to the manufacturers’ instructions (Appendix ). The IHC‐stained CHO cell and tonsil slides were assessed individually by two pathologists specialized in gynecological oncology. Positivity of the PD‐L1 protein was defined by membranous staining. The percentage of positive cells was determined. Additionally, cytoplasmatic staining and staining intensity were documented. We scored staining intensity as weak (+/−), mild , strong or very strong . Scoring results of the slides were revealed after both pathologists scored all slides individually. Antibodies considered specific for PD‐L1 protein staining based on the results of the tonsil and transfected CHO samples, were used to stain four FFPE tumor tissue samples of patients with a CHM, and four FFPEs of patients with a choriocarcinoma. Samples were again analyzed by the two pathologists, and scoring results were revealed after they both scored all slides individually. To evaluate the agreement between the two pathologists in PD‐L1‐positive scored cells of the CHM and choriocarcinoma samples, we determined the intraclass correlation coefficient (ICC) for all tested antibodies individually.
Statistical analyses To calculate ICC, the two‐way random‐effects model with consistency was used. The ICCs were determined with both average and single measures. Average measures are used when measures of ≥2 raters are averaged to derive the result of a test. Single measures are used when a single rater performs the test, and is likely to be used in daily clinical practice. An ICC value of <0.50 is considered poor, between 0.50 and 0.75 moderate, between 0.75 and 0.90 good, and >0.90 red excellent. Analyses were performed with IBM SPSS statistics, version 25.
Ethics statement This study was approved by the local ethical committee of the Radboud University Medical Center (reference number 2018–4132) on June 7, 2018.
RESULTS 3.1 IHC on CHO and tonsil tissue All antibodies showed positive membranous staining of the PD‐L1‐transfected CHO cells varying from 1% to 90%. The highest percentages of cells that scored positive for PD‐L1 in the PD‐L1‐transfected CHO cells were observed for the EL13N antibody (70%–90%) and 22C3 antibody (60%–70%),whereas the 73–10, CAL10, SP142, 28–8 and SP263 antibodies scored low percentages (Figure ; Table ). None of the PD‐L1 antibodies showed positive membranous staining on the CHO cells transfected with PD‐L2, or the wildtype CHO cells; however, the E1L3N antibody showed slight cytoplasmatic staining in the CHO PD‐L2 cells and, according to one pathologist, also in the wildtype CHO cells. The E1L3N, 73–10, 22C3, SP142 and SP263 antibodies showed strong membranous staining of the tonsil tissue (Figure ; Table ). The CAL10 and 28–8 antibodies showed weak to mild membranous staining of the tonsil tissue (Figure ; Table ). 3.2 PD‐L1 staining of complete hydatidiform moles and choriocarcinomas We observed qualitative differences in PD‐L1 staining patterns between the antibodies in the same CHM samples (Figure ; Table ). Expression of PD‐L1 was predominantly seen on the membrane of syncytiotrophoblast cells. Staining intensity was strongest with the SP263 antibody; however, besides membranous staining, strong cytoplasmatic staining was observed for all CHM samples with this antibody (Figure ). The 73–10, 22C3 and SP142 antibodies showed mild staining intensity, and staining intensity was weakest with the CAL10 and 28–8 antibodies (Figure ). Comparable to the results in CHM tissue, expression of PD‐L1 on choriocarcinoma tissue was predominantly seen on the membrane of syncytiotrophoblast cells. Percentages of tumor cells positive for PD‐L1 appeared to be higher in the choriocarcinoma samples than the CHM samples (Table ). Staining intensity was comparable to the results in the CHM samples: intensity was strongest with the SP263 antibody, mild with the 73–10, 22C3 and SP142 antibodies, and weakest with the CAL10 and 28–8 antibodies (Figure ). Again, cytoplasmatic staining was observed for choriocarcinoma samples stained with the SP263 antibody. 3.3 Pathologists’ agreement The ICCs determined with average measures were excellent for the 73–10, 22C3, 28–8 and SP263 antibodies, and good for the CAL10 and SP142 antibodies (Table ). ICCs determined with single measures were considered excellent for the 22C3 and SP263 antibodies, good for the 73–10 and 28–8 antibodies, and moderate for the CAL10 and SP142 antibodies (Table ).
IHC on CHO and tonsil tissue All antibodies showed positive membranous staining of the PD‐L1‐transfected CHO cells varying from 1% to 90%. The highest percentages of cells that scored positive for PD‐L1 in the PD‐L1‐transfected CHO cells were observed for the EL13N antibody (70%–90%) and 22C3 antibody (60%–70%),whereas the 73–10, CAL10, SP142, 28–8 and SP263 antibodies scored low percentages (Figure ; Table ). None of the PD‐L1 antibodies showed positive membranous staining on the CHO cells transfected with PD‐L2, or the wildtype CHO cells; however, the E1L3N antibody showed slight cytoplasmatic staining in the CHO PD‐L2 cells and, according to one pathologist, also in the wildtype CHO cells. The E1L3N, 73–10, 22C3, SP142 and SP263 antibodies showed strong membranous staining of the tonsil tissue (Figure ; Table ). The CAL10 and 28–8 antibodies showed weak to mild membranous staining of the tonsil tissue (Figure ; Table ).
PD‐L1 staining of complete hydatidiform moles and choriocarcinomas We observed qualitative differences in PD‐L1 staining patterns between the antibodies in the same CHM samples (Figure ; Table ). Expression of PD‐L1 was predominantly seen on the membrane of syncytiotrophoblast cells. Staining intensity was strongest with the SP263 antibody; however, besides membranous staining, strong cytoplasmatic staining was observed for all CHM samples with this antibody (Figure ). The 73–10, 22C3 and SP142 antibodies showed mild staining intensity, and staining intensity was weakest with the CAL10 and 28–8 antibodies (Figure ). Comparable to the results in CHM tissue, expression of PD‐L1 on choriocarcinoma tissue was predominantly seen on the membrane of syncytiotrophoblast cells. Percentages of tumor cells positive for PD‐L1 appeared to be higher in the choriocarcinoma samples than the CHM samples (Table ). Staining intensity was comparable to the results in the CHM samples: intensity was strongest with the SP263 antibody, mild with the 73–10, 22C3 and SP142 antibodies, and weakest with the CAL10 and 28–8 antibodies (Figure ). Again, cytoplasmatic staining was observed for choriocarcinoma samples stained with the SP263 antibody.
Pathologists’ agreement The ICCs determined with average measures were excellent for the 73–10, 22C3, 28–8 and SP263 antibodies, and good for the CAL10 and SP142 antibodies (Table ). ICCs determined with single measures were considered excellent for the 22C3 and SP263 antibodies, good for the 73–10 and 28–8 antibodies, and moderate for the CAL10 and SP142 antibodies (Table ).
DISCUSSION We observed substantial heterogeneity in the percentage of cells stained positive for PD‐L1 within the same tonsil, CHO, CHM and choriocarcinoma samples between the different PD‐L1 antibodies. Based on our results, the 22C3 antibody is the most suitable for adequate detection of PD‐L1‐positive cells in tissues of CHM, and in choriocarcinoma patients. With the use of the 22C3 antibody, we observed a high to very high membranous staining intensity positive for PD‐L1 in 60%–70% within the, for 80% PD‐L1‐transfected CHO cells. No positive staining on the CHO cells transfected with PD‐L2 or wildtype CHO cells was observed. The latter two observations underscore the specificity of the antibody. Additionally, with the 22C3 antibody, ICCs were excellent for the CHO, CHM and choriocarcinoma samples. Based on the results with the transfected CHO cells, we excluded the E1L3N antibody, since unexpected positive cytoplasmatic staining was observed in the CHO cells transfected with PD‐L2. Although according to the manufacturer cells with pure cytoplasmic immunoreaction (without membranous staining) should be ignored, we cannot fully exclude the possibility that E1L3N recognizes either a PD‐L1 variant, or another structurally related protein like for instance PD‐L2, and may therefore incorrectly indicate the presence of PD‐L1‐positive cells. Concerning this cytoplasmatic staining with the E1L3N antibody of the CHO cells that were transfected with PD‐L2, the pathologists scored 30%–40% of CHO cells that were transfected with PD‐L2 as falsely positive for PD‐L1, underlining the risk of overestimation and incorrect observation of PD‐L1‐positive cells. Low percentages of PD‐L1‐positive cells with PD‐L1‐transfected CHO cells were observed upon staining with the antibodies 73–10, CAL10, SP142 and SP263. This suggests that these antibodies have a lower affinity for the PD‐L1 protein than do 22C3 and 28–8 antibodies. Many ongoing studies in patients with GTD, analyzing PD‐L1 expression by trophoblast cells in order to predict clinical outcome or to select patients for treatment with PD‐L1 ICIs, are based on percentage of PD‐L1 expression in tumor samples of these patients, stained with, among others, the E1L3N, 28–8, SP263 and 22C3 PD‐L1 antibodies. , , , , , , Since different antibodies are used to stain the same type of tumor samples, we compared the percentage of PD‐L1‐positive cells within the same CHM and choriocarcinoma samples using the 73–10, 22C3, CAL10, SP142, 28–8 and SP263 antibodies, to assess whether comparable results are obtained. As expected, PD‐L1 expression was different in the CHM and choriocarcinoma samples. Fewer PD‐L1‐positive cells were present in CHM samples compared with the choriocarcinoma samples. Strikingly, we also observed a substantial variety in detection of PD‐L1‐expressing cells within the same CHM and choriocarcinoma samples when we compared staining results with the different antibodies. These results suggest that the reported varieties in staining patterns and trophoblast subtypes expressing PD‐L1 can be explained by the use of different PD‐L1 antibodies. , , As treatment eligibility of patients is frequently based on the percentage of PD‐L1 expression within tumor samples, , , , , and since different PD‐L1 antibodies showed various levels of PD‐L1‐expressing cells within the same samples, patient eligibility for treatment with anti‐PD‐L1 ICIs is actually determined by the choice of PD‐L1 antibody, rather than by the actual expression percentages of PD‐L1. This is the first study to compare seven frequently used commercially available antibodies in order to select the antibody most suitable for detection of PD‐L1‐expressing cells in CHO cells and tissue of CHM and choriocarcinoma patients. , , , , To assess the specificity of the PD‐L1 antibodies, we needed a “model” whose PD‐L1 and PD‐L2 expression percentages were known beforehand, in order afterwards to assess the percentage of PD‐L1‐expressing cells the antibody stained positive for expression of PD‐L1. By transfecting CHO cells with PD‐L1 and PD‐L2, we were able to compare the IHC results with the flow cytometry results, here used as the gold standard. This gold standard is needed, since choriocarcinoma cell lines already express PD‐L1 and PD‐L2, and we cannot visualize the difference between cells transfected with PD‐L1 and PD‐L2, and the non‐transfected (ie negative control) cells. Additionally, since the true percentage of PD‐L1‐expressing cells within the trophoblast tissues is unknown, it is not feasible to determine antibody specificity based on staining results in trophoblast samples. Selection of the 22C3 antibody was primarily based on the staining results of the CHO samples, and was underlined by the staining results of the CHM and choriocarcinoma samples. Although we used a low number of CHM and choriocarcinoma cases, we observed substantial variety in staining patterns using different PD‐L1 antibodies within the same CHM and choriocarcinoma samples, emphasizing the need for uniformity in usage of a single PD‐L1 antibody to detect PD‐L1‐expressing cells in patients with GTN. One could argue that the variety in staining patterns may partially be explained by PD‐L1 heterogeneity within the tissues. This was also observed in other studies in which PD‐L1 expression was determined on different tumor or trophoblast cells. , , , We anticipated this possible inter‐sample heterogeneity using subsequent samples of the same tissue‐blocks for staining with the different antibodies. A possible limitation of our study is that except for SP263, samples were stained manually and therefore were more prone to staining inaccuracies. However, manufacturer protocols were strictly followed, and all samples were stained simultaneously. Secondly, in clinical settings, antibody staining protocols are frequently optimized to enhance staining intensity for easier detection of positively stained cells. For example, the CAL10 antibody hardly stained the tonsil or the PD‐L1‐transfected CHO cells and may have improved optimization. A limitation of “in‐house” optimization of immunohistochemical staining methods is that these procedures should be described thoroughly to facilitate comparison between published studies, and to facilitate reproducibility in future studies. Therefore, we adhered to the manufacturer’s recommended protocols to determine the most optimal and reliable PD‐L1 antibody, resulting in 22C3 being our recommended choice for future studies.
CONCLUSION Using different antibodies to detect PD‐L1‐expressing cells in CHM and choriocarcinoma samples leads to substantial variation in observed tumor cells expressing PD‐L1. Based on our results, the 22C3 antibody seems most suitable for detecting PD‐L1‐positive cells in CHO cells transfected with PD‐L1, and should therefore be used in future studies analyzing expression of PD‐L1 in samples of patients with GTD. Clinical correlation with treatment response should be the next step.
All authors have been involved in the study design. YH collected the data. YH, MS, JB, MG, NO, JdeV and FS interpreted the data. YH wrote the manuscript and all authors were equally involved in reviewing and editing the manuscript.
None.
Appendix S1 Click here for additional data file.
|
|
Etude rétrospective sur 70 cas d´hystérectomie d´hémostase dans le département de gynécologie obstétrique de l´Hôpital de Ben Arous, Tunisie | 59c195f1-328c-441c-94df-0cb33e043d46 | 9482212 | Gynaecology[mh] | L´hémorragie du postpartum immédiat (HPPI) constitue l´une des complications les plus redoutées en obstétrique. Elle est définie par un saignement provenant du tractus génital, dépassant les 500 ml et survenant dans les 24 heures qui suivent l´accouchement . Elle constitue la principale cause de mortalité maternelle, malgré les progrès en matière de prise en charge médicale, obstétricale et en radiologie interventionnelle . L´hystérectomie d´hémostase constitue le traitement radical, elle a l´avantage d´offrir le maximum de sécurité mais au prix d´une stérilité définitive surtout pour les femmes jeunes désireuses d´autres grossesses. À ce jour, les nombreux travaux comparatifs publiés dans la littérature entre les différentes techniques chirurgicales ne permettent pas d´affirmer une supériorité statistiquement significative de l´hystérectomie totale vis-à-vis de l´hystérectomie subtotale. Au cours de ce travail, nous avons précisé les facteurs de risque, les indications, les complications, le pronostic maternel après une hystérectomie d´hémostase et déterminé les facteurs influençant le choix du type d´hystérectomie. Conception de l´étude: nous avons mené une étude rétrospective, monocentrique, descriptive et analytique s´étendant de janvier 2003 à décembre 2019, dans le service de gynécologie obstétrique de l´hôpital régional de Ben Arous. Cadre de l´étude : il s´agit d´une maternité de niveau IIB avec une moyenne de 4500 accouchements par an. Population de l´étude: il s´agissait d´une étude portant sur les patientes ayant eu une hystérectomie d´hémostase entre 2003 et 2019. Critères d´inclusion: nous avons inclus toutes les patientes ayant accouché par voie basse ou par césarienne au-delà de 28 semaines d´aménorrhée, ayant présenté une hémorragie grave du postpartum dans les 24 heures suivant l´accouchement et ayant nécessité une hystérectomie d´hémostase d´emblée ou après échec du traitement chirurgical conservateur. Critères de non-inclusion: les patientes non incluses étaient les femmes qui ont accouché par voie basse ou césarienne avant 28 semaines d´aménorrhée, celles qui ont présenté une hémorragie du postpartum n´ayant pas nécessité une hystérectomie et celles qui avaient eu une hystérectomie à la suite d´une hémorragie dans le postabortum ou une hémorragie d´origine autre que la sphère génitale. Critères d´exclusion: nous avons exclu tous les dossiers inexploitables avec des données manquantes. Critères de jugements: le critère de jugement principal était la relation entre l´indication de l´hystérectomie d´hémostase et le choix de la technique opératoire. Les critères de jugements secondaires associaient l´âge, l´utérus cicatriciel, le mode d´accouchement, les pertes sanguines, les besoins transfusionnels, le temps opératoire et les complications per et post opératoires. Variables étudiées: âge, gestité, parité, antécédents médicaux et chirurgicaux, bilan prénatal, circonstances de découverte, état hémodynamique, globe, abondance du saignement, diurèse, taux d´hémoglobine, taux d´hématocrite, taux de plaquettes, taux de prothrombine, temps de céphaline activé, fibrinémie, utilisation des utérotoniques, utilisation de produits sanguins, examen sous valves, révision utérine, traitement conservateur réalisé en première intention, en deuxième intention et en troisième intention, indication de l´hystérectomie d´hémostase, durée de l´intervention, pertes sanguines, complications opératoires, rétablissement du transit, durée de l´hospitalisation, transfert en réanimation, complications post opératoires et la mortalité maternelle. Analyse et traitement des données: les variables qualitatives ont été décrites en fonction de leur répartition en pourcentage avec les intervalles de confiance 95% (IC 95%). On a utilisé le test de chi deux de Pearson. Pour les variables quantitatives, l´analyse s´est faite par présentation de moyennes et écart-types quand la distribution est normale. Nous avons utilisé le test t de Student, Une analyse multivariée en régression logistique méthode descendante (Wald) a été réalisée en introduisant tous les facteurs dont les p sont < 0,05. Chaque facteur identifié a été alors présenté avec son odds-ratio ajusté (ORa) et son IC 95% Dans tous les tests, le seuil de p a été fixé à 5% (p < 0,05) et l´analyse multivariée a été réalisée à l´aide du logiciel SPSS 23.0 Fr. Considérations éthiques: compte tenu de son caractère rétrospectif, le consentement n´était pas exigé. Afin de garantir la confidentialité des informations personnelles des patients, les données ont été recueillies sur des fiches d´enquête anonyme. Le protocole d´étude a été validé par le comité d´éthique de l´hôpital. Données épidémiologiques: durant la période d´étude, 51896 accouchements ont eu lieu. Parmi ces accouchements, nous avons recensé 70 cas d´hystérectomie d´hémostase ce qui fait un taux de 1,3 ‰, soit une hystérectomie d´hémostase pour 769 accouchements. L´âge moyen était de 34,5 ans (± 5,1) . Soixante-dix pour cent des patientes avaient un bas niveau socio-économique. Déroulement de la grossesse: dans la majorité des cas (96%), la grossesse était mono-fœtale. Sept patientes (10%) avaient des ATCD d´hémorragie du postpartum. Dix patientes (14%) seulement avaient une grossesse bien suivie selon le programme national de périnatalité, soit 5 ou plus de consultations prénatales. Treize patientes (19%) avaient une anémie au bilan prénatal. Une hypertension artérielle gravidique a été observée chez neuf patientes (13%) et onze patientes (16%) avaient une grossesse compliquée de diabète gestationnel, un retentissement fœtal à type d´hydramnios et macrosomie a été objectivé chez six patientes (9%). Déroulement de l´accouchement: vingt-quatre patientes (34%) avaient un travail spontané. Un déclenchement du travail a été noté chez 11 patientes (16%). Parmi ces patientes, huit (12%) ont reçu un quart de comprimé de Misoprostol et trois (4%) ont eu une sonde extra-amniotique. Un travail rapide a été noté dans 57% des cas. La majorité des patientes ont accouché par césarienne (67%). Vingt-trois patientes (33%) ont accouché par voie basse. Données cliniques: les facteurs de risque de survenue d´HGPP étaient surtout la multiparité dans 73% des cas, l´accouchement par césarienne (67%) et l´utérus cicatriciel (60%). Le diagnostic de l´HPPI est clinique. Le maitre symptôme de l´HGPP dans notre population était le saignement observé chez 84% des patientes. Une rupture utérine a été découverte à la révision utérine dans 16% des cas. Prise en charge médico-obstétricale: chez les patientes qui ont accouché par voie basse, l´examen sous valves et la révision utérine étaient systématiques. Chez 55 patientes (79%) nous avons eu recours à une transfusion par des culots globulaires (CGR). Le nombre de CGR médian était de 4[0-8]. Cinquante-quatre patientes (78%) ont été transfusées par du plasma frais congelé (PFC) en association avec les CGR. Le nombre de PFC médian était de 6[0-24]. Cinquante et une patientes (73%) ont eu du fibrinogène. Par manque de moyen, aucune patiente n´a reçu le facteur VII activé recombinant. L´ocytocine était administrée chez 39% des parturientes, le sulprostone (Nalador) était nécessaire chez 18 patientes (26%) toujours après échec de l´ocytocine et deux patientes (3%) ont reçu 800 milligrammes de Misoprostol par voie rectale. Prise en charge chirurgicale: concernant le traitement chirurgical, 41 patientes (59%) ont eu une hystérectomie d´hémostase après échec du traitement conservateur et 29 patientes (41%) ont nécessité une hystérectomie d´hémostase d´emblée. Le traitement conservateur de première intention était dans la majorité des cas une ligature des artères utérines (70%). En deuxième intention, on a eu recours à une deuxième technique conservatrice chez 27 patientes (66%) soit, une triple ligature de Tsirulnikov chez 70% des cas (19/27) et une ligature bilatérale des artères hypogastriques (LBAH) chez 30% des patientes. Une hystérectomie d´hémostase a été réalisée dans un deuxième temps chez 14 patientes (34%). Indications de l´hystérectomie d´hémostase: les indications d´hystérectomie d´hémostase étaient multiples et dominées par les anomalies de l´implantation placentaire en particulier le placenta accreta chez 27 patientes (39%) et l´inertie utérine chez 24 patientes (34%). Les autres indications d´hystérectomie d´hémostase étaient: la rupture utérine (16%), le placenta prævia (1%), l´hématome rétro-placentaire associé à une coagulopathie (6%) et les déchirures cervicales complexes (3%). Une hystérectomie d´hémostase a été faite suite à l´impossibilité de suturer les berges d´un fibrome prævia avec hémorragie de grande abondance chez une patiente. La durée moyenne de l´intervention chirurgicale était de 180,86 minutes [± 8,34]. Le taux de morbidité peropératoire était de 34%. Les complications les plus fréquentes étaient, par ordre de fréquence, l´état de choc hémorragique (17%), la CIVD (6%) et les lésions vésicales (6%). La Delta hémoglobine moyenne était de 2,54 g/dl. Onze patientes (16%) ont été transférées en réanimation. Les complications postopératoires étaient observées chez 11 patientes soit 16% des cas. Ces complications étaient dominées par l´insuffisance rénale aigue et l´état de choc hémorragique. Aucun accident transfusionnel n´a été noté. La durée médiane de l´hospitalisation était de sept jours . Nous avons déploré six cas de décès maternel (8%). Trois décès étaient survenus en peropératoire et trois décès étaient recensées en postopératoire . Une hystérectomie subtotale (HST) était réalisée chez 55 patientes (79%). Le reste ont eu une hystérectomie totale (HT). Il n´y avait pas de différence statistiquement significative dans l´incidence des complications entre les groupes HST et HT. La comparaison des données des patientes par l´analyse multivariée par régression logistique a permis de retenir le placenta accreta et la durée d´intervention comme facteurs influençant le choix du type de l´hystérectomie. Ainsi le placenta accreta était associé de manière significative au groupe HT (ORa:6,93, IC 95%: 1,07-44,80, p=0,042) et la durée moyenne de l´opération était significativement plus courte pour le groupe HST que pour le groupe HT (ORa: 1,023, IC 95%: 1,009-1,03, p=0,01) . Dans notre travail, 70 patientes ont été incluses. L´analyse a montré que le placenta accreta était l´indication la plus fréquente de l´HH. Quarante et une patientes ont eu une HH après échec du traitement conservateur et 29 patientes ont nécessité une HH d´emblée. Une HST était réalisée chez 55 patientes. Les facteurs influençant le choix du type de l´hystérectomie d´hémostase étaient le placenta accréta et la durée d´intervention. La fréquence de l´HH reste encore élevée dans le monde malgré les progrès réalisés dans la prise en charge de l´HGPP, avec une prédominance dans les pays du sud . Durant la période de notre étude, nous avons recensé un taux de 0,13%. Cette fréquence est inférieure à celle rapportée dans les pays sous-développés mais encore élevée par rapport aux pays développés . L´étude italienne de Barillari et al . a décrit un âge moyen de 34,5 ans . Les résultats de notre étude étaient concordants à ceux avancés par la littérature. Le niveau socio-économique prend toute sa valeur dans les pays en voie de développement où un faible niveau socio-économique est associé à une précarité du système de santé, à une carence nutritionnelle et à l´anémie influençant négativement l´incidence des HGPP . Selon les données épidémiologiques rapportées par Ahmadi et al . , 75% des parturientes avaient un niveau socio-économique précaire. Dans notre série, 70% des parturientes étaient issues d´un milieu socio-économique défavorable. Zelop et al . ont démontré que l´incidence de l´HH augmente de façon significative avec la parité. De même Zelop et al . ont montré que la multiparité est un facteur de risque d´HH . En effet la multiparité est un facteur de fragilisation de l´utérus favorisant la rupture utérine, l´inertie utérine et l´inversion utérine ainsi que l´hémorragie du postpartum . Ces résultats sont similaires à ceux retrouvés dans notre travail, où 63% des patientes étaient des multipares. Par ailleurs, Combs et al . ont constaté que la primiparité est un facteur de risque de prééclampsie, de travail anormal, de lésion de la filière génitale pouvant ainsi être responsable aussi de l´hémorragie du postpartum. Quatre pour cent de nos patientes étaient primipares, alors que dans la littérature ce taux est disparate . Selon les données de la littérature, l´utérus cicatriciel est un facteur de risque de l´HH . En outre, l´utérus cicatriciel augmente le risque d´anomalies d´adhésion et d´insertion placentaire majorant ainsi le risque de survenue d´une hémorragie de la délivrance et d´HH. En effet, l´incidence du placenta accréta semble être augmentée et cette augmentation est proportionnelle au nombre de césariennes antérieures chez les patientes présentant un placenta prævia. L´étude de Clark et al . a décrit que le risque de placenta accréta était de 24% en cas d´un utérus unicicatriciel, et atteint 67% en cas d´utérus quadricicatriciel . Notre étude est venue également réconforter ces résultats, l´antécédent d´utérus cicatriciel était retrouvé chez 60% des patientes. L´existence d´antécédents d´HGPP est un des facteurs de risque les plus associés, mais reste peu prévalent. Ford et al . ont rapporté que le risque d´HGPP est multiplié par trois en cas d´antécédent d´HPP . Les études récentes conduisent à penser que les consultations prénatales de suivi des grossesses permettent de mettre en place des interventions d´efficacité prouvée pour la mère et le nouveau-né. Ces consultations ont un rôle fondamental dans la réduction de la morbidité et la mortalité maternelle et l´absence d´un suivi prénatal est un facteur identifié d´hystérectomie d´hémostase . D´autant plus que dans notre étude seulement dix patientes (14%) avaient une grossesse bien suivie selon le programme national de périnatalité. Nwobodo et al . ont rapporté que la survenue des événements indésirables comme l´HH était statistiquement liée au suivi prénatal, notant une incidence de 1,82% chez les patientes n´ayant pas eu de suivi prénatal comparée à 0,07% chez les patientes suivies . Dans l´étude de Nyflot et al ., l´anémie au cours de la grossesse était considérée comme facteur de risque d´HH. Ceci peut être expliqué par la baisse du seuil de tolérance à des pertes sanguines même moyennes en cas d´anémie . Plusieurs pathologies gravidiques telles que la pré-éclampsie, la mort fœtale in utero (MFIU), la stéatose hépatique aigue gravidique (SHAG) et l´hématome rétro-placentaire (HRP) constituent des situations favorisantes d´un trouble acquis de l´hémostase. Combs et Reyal considéraient que la toxémie gravidique est l´un des principaux facteurs de risque d´HH avec des OR respectifs de 5,02 et 6,92 . Des études expérimentales ont prouvé qu´une exposition prolongée à l´ocytocine pendant le travail pouvait entrainer la désensibilisation et la saturation de ses récepteurs. Ceci entraine une interruption de la cascade de signalisation intracellulaire limitant la contraction médiée par l´ocytocine . Dans notre série, une direction de travail a été faite chez 21% des patientes. La transfusion fait l´objet de recommandations générales. Différentes études concluent que la transfusion de produits sanguin labiles (PSL) est un élément primordial dans la prise en charge des HPPI . Les indications d´HH sont multiples. La littérature rapporte que les anomalies d´insertion placentaire est actuellement le premier motif de réalisation d´une HH . L´étude récente de Flood et al . a étudié le changement des indications de l´HH en montrant une augmentation significative des anomalies d´adhésion placentaire de 5,4% à 46,5% (p <0,001) et une réduction significative du taux de l´atonie utérine de 40,5% à 9,3% (p <0,001) . L´incidence du placenta accréta a été multipliée par dix dans les cinquante dernières années et serait de 1/2500 accouchements . Cette élévation est expliquée par l´augmentation constante du taux de césarienne dans le monde et elle est proportionnelle au nombre de césariennes antérieures chez les patientes ayant un placenta prævia . En raison du fort taux d´échec des mesures conservatrices, la césarienne-hystérectomie en un temps est le traitement de référence du placenta accréta selon le Collège Américain de Gynécologie Obstétrique . Le diagnostic anténatal du placenta accréta est capital. Il doit être réalisé chez toute femme présentant un ou plusieurs facteurs de risque. Il se base essentiellement sur l´échographie couplée au doppler et l´imagerie par résonnance magnétique en cas de placenta postérieur . En effet, Warshak et al . ont montré que les patientes dont le placenta accréta était diagnostiqué en anténatal et chez qui une césarienne hystérectomie sans tentative de délivrance artificielle était alors programmée à 34-35 SA, recevaient moins de culots globulaires et avaient tendance à moins saigner en comparaison avec les patientes dont le placenta accréta n´avait pas été diagnostiqué . En termes de fréquence, la deuxième indication d´hystérectomie d´hémostase dans notre étude était l´atonie utérine avec 34%. L´étude de Habek et al . a montré que le taux d´HH due à une inertie utérine était de 25% . La diminution de l´incidence de l´inerte utérine rapportée par plusieurs auteurs est expliquée par l´efficacité du traitement médical (les utérotoniques et les prostaglandines) et le développement des techniques chirurgicales conservatrices . Dans notre série, le taux de rupture utérine était de 16%. Ce taux rejoint celui de Rahman et al . qui était de 18% . Flood et al . ont décrit une diminution significative au cours de cette décennie de cette indication de 40% à 9% . Cette baisse significative est probablement le résultat de l´évolution de la pratique obstétricale, la diminution de la parité des patientes et l´utilisation plus judicieuse de l´ocytocine. Dans notre travail, une HST était réalisée chez 79% des patientes. Smith et al . n´ont pas trouvé une différence significative concernant le temps opératoire, les indications et les complications per et postopératoires entre les deux groupes HST et HT . Zhang et al . ont trouvé que la durée moyenne de l´opération était significativement plus courte pour le groupe HST par rapport au groupe HT et que les besoins transfusionnels en CGR étaient plus importants dans le groupe HT . Cependant, Flood et al . ont montré que l´HT était significativement plus efficace en matière de contrôle de saignement d´origine cervicale . Les résultats contradictoires de la littérature peuvent être expliqués par la randomisation extrêmement difficile à envisager dans un contexte d´urgence vitale, mais aussi la stratégie globale de prise en charge et en particulier les mesures médicales, obstétricales et radiologiques variant en fonction des équipes, du plateau technique et des moyens humains. Les patientes et les indications sont également très différentes, notamment en ce qui concerne la gravité de l´hémorragie au moment où est décidée l´intervention chirurgicale. Les études ont de ce fait des références de niveau de preuve faible et les stratégies proposées reposent le plus souvent sur un consensus professionnel obtenu au sein d´un groupe de travail ou d´une équipe . Il semble en revanche plus intéressant de considérer que c´est l´indication même de l´hystérectomie qui guide l´opérateur en fonction de son appréciation clinique et de son expérience. L´indication d´une hystérectomie totale est indispensable en cas d´hémorragie d´origine segmentaire inférieure ou cervicale qui se voit surtout dans le placenta prævia accréta et les déchirures cervicales complexes. Dans les autres cas, et notamment en cas d´atonie utérine, l´hystérectomie subtotale apparaît comme une alternative acceptable . En peropératoire, les plaies des organes avoisinants sont des accidents rencontrés au cours de l´hystérectomie d´hémostase. Selon Harris, les complications les plus fréquentes sont les plaies vésicales suivies des plaies intestinales . Dans la littérature, des lésions vésicales et urétérales en peropératoire ont été retrouvées dans 4 à 16% . Les lésions vésicales étaient plus élevées de manière significative chez les femmes présentant un placenta accréta que chez les femmes présentant une atonie utérine . Pradhan et al . ont montré que les complications peropératoires étaient dominées par la CIVD dans 37,7% des cas suivies de l´état de choc hémorragique dans 26,2% des cas . Dans notre travail, 6% des patientes ont présenté une insuffisance rénale aiguë nécessitant des séances d´hémodialyse. Cette insuffisance rénale est liée à une hypo perfusion des reins faisant suite à la spoliation sanguine. Ces résultats sont similaires à ceux trouvés dans la série de Ducarme et al ou le taux d´insuffisance rénale était de 6.2% . Dans les pays développés, le taux de la mortalité après HH est faible . Alors que dans les pays en voie de développement, ce taux est encore élevé . Dans notre étude, nous avons enregistré six décès maternels soit un taux de 8%. Les causes de décès maternel les plus fréquentes, dans notre série, étaient la CIVD et l´état de choc hémorragique. Ces résultats sont en concordance avec les constatations décrites par Zeteroglu et al . et Chawla et al . . Notre étude présente les points forts suivants: la population étudiée dans notre travail est considérée comme un échantillon assez représentatif diminuant le biais de sélection avec un recul suffisant pour évaluer les résultats à long terme. Néanmoins, elle présente certaines limites qui sont surtout d´ordre méthodologique liées notamment au caractère rétrospectif et monocentrique de l´étude. Le taux d´HH dans notre série reste élevé par rapport aux pays développés. Les facteurs de risque d´hystérectomie d´hémostase les plus fréquents étaient la multiparité, l´accouchement par césarienne, et l´utérus cicatriciel. Le placenta accréta représente la principale indication d´hystérectomie dans notre étude. Bien que le temps opératoire fût plus long, l´hystérectomie totale n´était pas associée à un risque accru de complications par rapport à l´hystérectomie subtotale. Seulement 10% de nos patientes avaient un bon suivi de leurs grossesses. Ainsi, nous insistons sur l´intérêt du suivi prénatal. Etat des connaissances sur le sujet L´hystérectomie est indispensable précisément dans certains cas d´hémorragie sévère du post partum, malgré les progrès en matière de prise en charge médicale, obstétricale et en radiologie interventionnelle de l´hémorragie grave du postpartum; Les données publiées dans la littérature entre l´hystérectomie totale et subtotale ne permettent pas d´affirmer une supériorité statistiquement significative d´une technique par rapport à l´autre. Contribution de notre étude à la connaissance La fréquence de l´hystérectomie d´hémostase est encore élevée par rapport aux pays développés; Le placenta accréta représente la principale indication d´hystérectomie; Bien que le temps opératoire soit plus long, l´hystérectomie totale n´est pas associée à un risque accru de complications par rapport à l´hystérectomie subtotale. L´hystérectomie est indispensable précisément dans certains cas d´hémorragie sévère du post partum, malgré les progrès en matière de prise en charge médicale, obstétricale et en radiologie interventionnelle de l´hémorragie grave du postpartum; Les données publiées dans la littérature entre l´hystérectomie totale et subtotale ne permettent pas d´affirmer une supériorité statistiquement significative d´une technique par rapport à l´autre. La fréquence de l´hystérectomie d´hémostase est encore élevée par rapport aux pays développés; Le placenta accréta représente la principale indication d´hystérectomie; Bien que le temps opératoire soit plus long, l´hystérectomie totale n´est pas associée à un risque accru de complications par rapport à l´hystérectomie subtotale. |
Building a Boot Camp: Pediatric Residency Preparatory Course Design Workshop and Tool Kit | 0d2c0ff5-9f0c-47d2-91d5-755637b1c819 | 7010200 | Pediatrics[mh] | By the end of this activity, learners will be able to: 1. Discuss the literature pertaining to residency preparatory courses or boot camps. 2. Perform a boot camp needs assessment related to available resources, identified gaps, and institutional requirements. 3. Design boot camp course schedules aligned with individualized needs assessments. 4. Develop module ideas to use in boot camp courses that address core entrustable professional activities. 5. Identify barriers and strategies to successfully implement boot camp courses at home institutions.
Arising from an identified gap between the starting expectations for interns and the variability of fourth-year medical school curricula, boot camps (also referred to as residency preparatory courses) have gained interest in medical education as a way to further prepare students for the transition from medical school to internship. – A meta-analysis on the effectiveness of boot camps for transitions into residency concluded that completion of such a course was associated with improved clinical skills, knowledge acquisition, and perceived confidence. Designing a boot camp course is complex, as there is wide variability in content, duration, and timing of courses, as well as institution-specific requirements. Faculty are frequently asked to develop courses without any guidance, and clinical faculty who are developing the courses often have limited formal training in medical education. The American College of Surgeons, the Association of Program Directors in Surgery, and the Association for Surgical Education have jointly developed a modular resident preparatory curriculum recommended for students prior to the beginning of surgical internship. Otherwise, there are few published discipline-specific boot camp content recommendations or faculty development resources. , Although there is a recent publication on pediatric boot camp program development and evaluation, based on a conceptual framework, there is no published literature for specifically designing pediatric boot camp curricula. Numerous published modular sessions could easily be incorporated into boot camps, – and there are multiple publications describing institutional experiences with boot camp implementation. , This workshop introduced faculty to the pediatric boot camp course design tool kit, which serves as a framework for faculty tasked with boot camp course development. It was designed to guide participants through the initial stages of Kern's six steps of curriculum development and to assist faculty in designing a course that addresses entrustable professional activities (EPAs) for entering residency through development of new sessions or adaptation of published modules. Participants were encouraged to prioritize course content based on the duration of boot camp, physical resources, and local expertise to design a course meeting individual institutional needs. The materials provided in the appendices include the slide presentation used in the workshop, both blank and completed sample worksheets for each section of the tool kit, and a facilitator guide. There is also a review of the existing boot camp literature for additional preparation for facilitators. The target audience of the workshop and associated tool kit is faculty tasked with designing and implementing boot camp courses that are competency driven and provide maximal preparation for students entering a pediatric or pediatric-involved residency.
Curricular Context We developed this workshop and tool kit to guide physicians and educators involved in undergraduate medical education through the steps of designing a pediatric boot camp for fourth-year medical students. We introduced the tool kit as part of a 2-hour workshop during the Council on Medical Student Education in Pediatrics (COMSEP) Annual Meeting in April 2018. Participants primarily included pediatricians involved in undergraduate medical education with an interest in boot camps. We used Kern's six steps of curriculum development as a theoretical model for content development. These steps include the following: 1. Problem identification and general needs assessment. 2. Targeted needs assessment. 3. Goals and objectives. 4. Educational strategies. 5. Implementation. 6. Evaluation and feedback. Of note, program evaluation was not specifically addressed within the workshop or tool kit because individual institutions often have local practices for gathering course feedback. Prerequisite knowledge about objective writing, basic curricular design, the core EPAs, and pediatric milestones was beneficial but not required. This tool kit could be tailored for use by faculty in other specialties as well. Implementation The workshop included a mixture of short didactics, guided reflective exercises, and both small- and large-group discussions. Four facilitators with experience in boot camp development created the tool kit and facilitated the workshop. We organized tables and chairs in the conference room to allow for optimal small-group discussion. Materials included a computer and projector, a PowerPoint presentation , pens, and the tool kit materials . We used the PowerPoint presentation throughout the workshop to emphasize teaching points, transition between exercises, and guide activities. Timing and sequence are detailed in . We also created a facilitator guide , which reviews the tool kit materials provided in the appendices and includes recommendations and practical instructions for planning the workshop, facilitating each of the workshop sections using these materials, adapting the workshop for participants either planning de novo boot camps or revising an existing course, and conducting the follow-up survey. Initially, we reviewed existing background literature pertaining to the current state of the fourth year of medical school, – with particular emphasis on boot camps, both general and pediatric specific. We then guided participants in performing a needs assessment or adapting one they may have previously performed. Next, they created sample schedules tailored to the identified needs of their home institution. Participants then designed a sample module targeting specific EPAs. Finally, we discussed barriers to boot camp implementation and possible solutions. The workshop was intended to provide participants with knowledge and resources to develop pediatric boot camp–style courses de novo or improve on existing courses. Introduction, audience identification, and review of objectives Initially, each of the facilitators gave a brief introduction, sharing personal experience with boot camp courses and curriculum development. We then asked participants to share their own experiences with the development and implementation of boot camps. This allowed us to gauge our audience's level of background knowledge and experience, as well as leverage participants’ experiences for future discussion. Finally, one facilitator reviewed the objectives for the workshop. Review of existing background literature We provided background on the variability of the fourth year of medical school; perceived gaps in interns’ knowledge, skills, and attitudes; and the use of boot camps for graduating medical students across different specialties, with emphasis on pediatric-specific courses. This allowed participants to understand the context of boot camps in medical education and the variability of the courses across institutions. A summary and references for this review are provided in . Needs assessment The goal of this exercise was to guide the participants in planning a boot camp–specific needs assessment, taking into account institutional, trainee, and residency program director perspectives. First, we instructed participants to reflect on any boot camp needs assessment conducted previously, including the specific methods used and stakeholders involved. Using the institutional needs assessment worksheet , participants then created a plan for performing an assessment or improving on an existing assessment at their own institution, including strategies for soliciting input from stakeholders. Participants identified some of the basic considerations for their courses, including the number of instructional days, students needing to enroll, and faculty required. They also listed basic resources available to them, including physical space, simulation centers, standardized patients, among others. Finally, participants completed a matrix relating to core EPAs currently addressed in the curriculum to identify potential gaps that could be addressed in a boot camp. Following brief individual reflection, participants discussed ideas in small groups of three to six people. Facilitators visited tables throughout the exercise to provide additional direction as needed. Schedule development After a brief introduction, we asked participants to individually brainstorm general session topics identified during the needs assessment portion. Using the recommended content list and session prioritization worksheet , they prioritized sessions based on core EPA topics and identified educational gaps, as well as the resources and time allotment available, at their institutions. The recommended content areas represented a consensus list from our own experience considering the knowledge, skill, and behavioral gaps identified in the literature review included in the background . Since creation of the workshop, this content area has been reviewed, and a curriculum has been vetted through COMSEP that is now published on the COMSEP website. After completing their individual prioritization lists, participants formed new groups of three to six members based on boot camp course length (3 days or less, 5 days, 10 days, more than 10 days). Each group created a draft schedule using the schedule worksheet with sample schedules from the facilitators’ institutions as examples . In a large-group format, facilitators led a focused debriefing session about the experience of developing a course schedule. Module development We began this activity with a brief discussion about how to identify instructional methods to address learning objectives within practical constraints such as time and physical resources. We then assigned each small group up to three EPAs. Using the module design worksheet and planning resources , participants considered which educational method could best be applied to their assigned EPAs. Each group outlined a module to address one of its assigned EPAs by creating learning objectives, identifying the number of learners that could be accommodated per session, describing the faculty/resource needs and time required, and identifying an instructional method. Additional resources used for this exercise were a list of relevant resources available in MedEdPORTAL . Each small group then presented its module to the large group and received feedback from participants and facilitators. Barriers During a large-group discussion, we asked participants to identify barriers that they had encountered or anticipated encountering while designing and implementing boot camps. Common themes surfaced, including limitations with physical and monetary resources, coordinator or administrative support, faculty volunteers, institutional support, student participation, and objective assessment of students. We facilitated a brief discussion about potential solutions. Wrap-up We closed the workshop with a summary of the key points and encouraged participants to seek out existing resources and to collaborate across institutions. Assessment Immediately following the workshop, we emailed an anonymous electronic survey via SurveyMonkey (San Mateo, California) to all participants asking about their personal and institutional experience with boot camp courses, as well as their self-assessed confidence in the components covered in the workshop, including the following: 1. Discussing the literature pertaining to boot camp courses; 2. Performing a needs assessment related to available resources, identified gaps, and institutional requirements; 3. Designing a boot camp schedule aligned with individualized needs assessments; 4. Developing modules to use in a boot camp that address core EPAs; and 5. Identifying barriers and strategies to successfully implement a boot camp at their home institutions. The survey collected retrospective pre-/postworkshop self-assessed confidence responses using visual analog scales measuring from 0 to 100 for each of the five components. The survey was open for 1 month following the workshop, and we sent one reminder email 3 weeks after the initial request. We compared the mean level of self-assessed confidence pre- and postworkshop for the five components of course planning and design addressed in the boot camp course design tool kit. Participants' change in confidence level was then compared using a paired two-tailed Student t test. Four months following the workshop, we emailed a second anonymous electronic survey to all participants asking about the impact of the workshop on their institution's boot camp course .
We developed this workshop and tool kit to guide physicians and educators involved in undergraduate medical education through the steps of designing a pediatric boot camp for fourth-year medical students. We introduced the tool kit as part of a 2-hour workshop during the Council on Medical Student Education in Pediatrics (COMSEP) Annual Meeting in April 2018. Participants primarily included pediatricians involved in undergraduate medical education with an interest in boot camps. We used Kern's six steps of curriculum development as a theoretical model for content development. These steps include the following: 1. Problem identification and general needs assessment. 2. Targeted needs assessment. 3. Goals and objectives. 4. Educational strategies. 5. Implementation. 6. Evaluation and feedback. Of note, program evaluation was not specifically addressed within the workshop or tool kit because individual institutions often have local practices for gathering course feedback. Prerequisite knowledge about objective writing, basic curricular design, the core EPAs, and pediatric milestones was beneficial but not required. This tool kit could be tailored for use by faculty in other specialties as well.
The workshop included a mixture of short didactics, guided reflective exercises, and both small- and large-group discussions. Four facilitators with experience in boot camp development created the tool kit and facilitated the workshop. We organized tables and chairs in the conference room to allow for optimal small-group discussion. Materials included a computer and projector, a PowerPoint presentation , pens, and the tool kit materials . We used the PowerPoint presentation throughout the workshop to emphasize teaching points, transition between exercises, and guide activities. Timing and sequence are detailed in . We also created a facilitator guide , which reviews the tool kit materials provided in the appendices and includes recommendations and practical instructions for planning the workshop, facilitating each of the workshop sections using these materials, adapting the workshop for participants either planning de novo boot camps or revising an existing course, and conducting the follow-up survey. Initially, we reviewed existing background literature pertaining to the current state of the fourth year of medical school, – with particular emphasis on boot camps, both general and pediatric specific. We then guided participants in performing a needs assessment or adapting one they may have previously performed. Next, they created sample schedules tailored to the identified needs of their home institution. Participants then designed a sample module targeting specific EPAs. Finally, we discussed barriers to boot camp implementation and possible solutions. The workshop was intended to provide participants with knowledge and resources to develop pediatric boot camp–style courses de novo or improve on existing courses. Introduction, audience identification, and review of objectives Initially, each of the facilitators gave a brief introduction, sharing personal experience with boot camp courses and curriculum development. We then asked participants to share their own experiences with the development and implementation of boot camps. This allowed us to gauge our audience's level of background knowledge and experience, as well as leverage participants’ experiences for future discussion. Finally, one facilitator reviewed the objectives for the workshop. Review of existing background literature We provided background on the variability of the fourth year of medical school; perceived gaps in interns’ knowledge, skills, and attitudes; and the use of boot camps for graduating medical students across different specialties, with emphasis on pediatric-specific courses. This allowed participants to understand the context of boot camps in medical education and the variability of the courses across institutions. A summary and references for this review are provided in . Needs assessment The goal of this exercise was to guide the participants in planning a boot camp–specific needs assessment, taking into account institutional, trainee, and residency program director perspectives. First, we instructed participants to reflect on any boot camp needs assessment conducted previously, including the specific methods used and stakeholders involved. Using the institutional needs assessment worksheet , participants then created a plan for performing an assessment or improving on an existing assessment at their own institution, including strategies for soliciting input from stakeholders. Participants identified some of the basic considerations for their courses, including the number of instructional days, students needing to enroll, and faculty required. They also listed basic resources available to them, including physical space, simulation centers, standardized patients, among others. Finally, participants completed a matrix relating to core EPAs currently addressed in the curriculum to identify potential gaps that could be addressed in a boot camp. Following brief individual reflection, participants discussed ideas in small groups of three to six people. Facilitators visited tables throughout the exercise to provide additional direction as needed. Schedule development After a brief introduction, we asked participants to individually brainstorm general session topics identified during the needs assessment portion. Using the recommended content list and session prioritization worksheet , they prioritized sessions based on core EPA topics and identified educational gaps, as well as the resources and time allotment available, at their institutions. The recommended content areas represented a consensus list from our own experience considering the knowledge, skill, and behavioral gaps identified in the literature review included in the background . Since creation of the workshop, this content area has been reviewed, and a curriculum has been vetted through COMSEP that is now published on the COMSEP website. After completing their individual prioritization lists, participants formed new groups of three to six members based on boot camp course length (3 days or less, 5 days, 10 days, more than 10 days). Each group created a draft schedule using the schedule worksheet with sample schedules from the facilitators’ institutions as examples . In a large-group format, facilitators led a focused debriefing session about the experience of developing a course schedule. Module development We began this activity with a brief discussion about how to identify instructional methods to address learning objectives within practical constraints such as time and physical resources. We then assigned each small group up to three EPAs. Using the module design worksheet and planning resources , participants considered which educational method could best be applied to their assigned EPAs. Each group outlined a module to address one of its assigned EPAs by creating learning objectives, identifying the number of learners that could be accommodated per session, describing the faculty/resource needs and time required, and identifying an instructional method. Additional resources used for this exercise were a list of relevant resources available in MedEdPORTAL . Each small group then presented its module to the large group and received feedback from participants and facilitators. Barriers During a large-group discussion, we asked participants to identify barriers that they had encountered or anticipated encountering while designing and implementing boot camps. Common themes surfaced, including limitations with physical and monetary resources, coordinator or administrative support, faculty volunteers, institutional support, student participation, and objective assessment of students. We facilitated a brief discussion about potential solutions. Wrap-up We closed the workshop with a summary of the key points and encouraged participants to seek out existing resources and to collaborate across institutions.
Initially, each of the facilitators gave a brief introduction, sharing personal experience with boot camp courses and curriculum development. We then asked participants to share their own experiences with the development and implementation of boot camps. This allowed us to gauge our audience's level of background knowledge and experience, as well as leverage participants’ experiences for future discussion. Finally, one facilitator reviewed the objectives for the workshop.
We provided background on the variability of the fourth year of medical school; perceived gaps in interns’ knowledge, skills, and attitudes; and the use of boot camps for graduating medical students across different specialties, with emphasis on pediatric-specific courses. This allowed participants to understand the context of boot camps in medical education and the variability of the courses across institutions. A summary and references for this review are provided in .
The goal of this exercise was to guide the participants in planning a boot camp–specific needs assessment, taking into account institutional, trainee, and residency program director perspectives. First, we instructed participants to reflect on any boot camp needs assessment conducted previously, including the specific methods used and stakeholders involved. Using the institutional needs assessment worksheet , participants then created a plan for performing an assessment or improving on an existing assessment at their own institution, including strategies for soliciting input from stakeholders. Participants identified some of the basic considerations for their courses, including the number of instructional days, students needing to enroll, and faculty required. They also listed basic resources available to them, including physical space, simulation centers, standardized patients, among others. Finally, participants completed a matrix relating to core EPAs currently addressed in the curriculum to identify potential gaps that could be addressed in a boot camp. Following brief individual reflection, participants discussed ideas in small groups of three to six people. Facilitators visited tables throughout the exercise to provide additional direction as needed.
After a brief introduction, we asked participants to individually brainstorm general session topics identified during the needs assessment portion. Using the recommended content list and session prioritization worksheet , they prioritized sessions based on core EPA topics and identified educational gaps, as well as the resources and time allotment available, at their institutions. The recommended content areas represented a consensus list from our own experience considering the knowledge, skill, and behavioral gaps identified in the literature review included in the background . Since creation of the workshop, this content area has been reviewed, and a curriculum has been vetted through COMSEP that is now published on the COMSEP website. After completing their individual prioritization lists, participants formed new groups of three to six members based on boot camp course length (3 days or less, 5 days, 10 days, more than 10 days). Each group created a draft schedule using the schedule worksheet with sample schedules from the facilitators’ institutions as examples . In a large-group format, facilitators led a focused debriefing session about the experience of developing a course schedule.
We began this activity with a brief discussion about how to identify instructional methods to address learning objectives within practical constraints such as time and physical resources. We then assigned each small group up to three EPAs. Using the module design worksheet and planning resources , participants considered which educational method could best be applied to their assigned EPAs. Each group outlined a module to address one of its assigned EPAs by creating learning objectives, identifying the number of learners that could be accommodated per session, describing the faculty/resource needs and time required, and identifying an instructional method. Additional resources used for this exercise were a list of relevant resources available in MedEdPORTAL . Each small group then presented its module to the large group and received feedback from participants and facilitators.
During a large-group discussion, we asked participants to identify barriers that they had encountered or anticipated encountering while designing and implementing boot camps. Common themes surfaced, including limitations with physical and monetary resources, coordinator or administrative support, faculty volunteers, institutional support, student participation, and objective assessment of students. We facilitated a brief discussion about potential solutions.
We closed the workshop with a summary of the key points and encouraged participants to seek out existing resources and to collaborate across institutions.
Immediately following the workshop, we emailed an anonymous electronic survey via SurveyMonkey (San Mateo, California) to all participants asking about their personal and institutional experience with boot camp courses, as well as their self-assessed confidence in the components covered in the workshop, including the following: 1. Discussing the literature pertaining to boot camp courses; 2. Performing a needs assessment related to available resources, identified gaps, and institutional requirements; 3. Designing a boot camp schedule aligned with individualized needs assessments; 4. Developing modules to use in a boot camp that address core EPAs; and 5. Identifying barriers and strategies to successfully implement a boot camp at their home institutions. The survey collected retrospective pre-/postworkshop self-assessed confidence responses using visual analog scales measuring from 0 to 100 for each of the five components. The survey was open for 1 month following the workshop, and we sent one reminder email 3 weeks after the initial request. We compared the mean level of self-assessed confidence pre- and postworkshop for the five components of course planning and design addressed in the boot camp course design tool kit. Participants' change in confidence level was then compared using a paired two-tailed Student t test. Four months following the workshop, we emailed a second anonymous electronic survey to all participants asking about the impact of the workshop on their institution's boot camp course .
During the COMSEP 2018 Annual Meeting, 27 of the conference attendees self-selected for participation in the workshop. There were response rates of 41% (11 out of 27) to the initial postworkshop electronic survey and 27% to the follow-up survey 4 months later. presents demographics of course participants responding to the initial survey. There was a statistically significant increase in self-assessed confidence for all five components of course planning and design addressed in the workshop . The lowest preconference confidence levels were in the areas of discussing the literature around boot camps and developing modules to address EPAs. These categories were also associated with the greatest average increase in confidence following participation in the workshop. Following is a representative sample of qualitative responses from the initial postworkshop survey: • Prompt: Name one thing you plan to implement from this workshop in the next year. ○ “Needs assessment” (three responses). ○ “Even if institutionalized curriculum does not get up and running, I plan on trying to offer an experience for interested 4th years that may be a 1–2 week experience depending on my availability.” ○ “More on interdisciplinary communication and ‘giving bad news.’” ○ “Choosing more engaging activities for certain existing modules.” ○ “Investigate online modules such as choosing wisely to use.” ○ “A boot camp!” • Prompt: Is there any other feedback you would like to offer the presenters? ○ “Great generalized discussion, appreciate sharing resources!” ○ “It was great to hear how others approach these. All are different and walked away with some possible additions to our curriculum.” ○ “Wonderful job. The literature and schedules shared were so very valuable. Sharing the barriers and some novel solutions as well as innovative teaching methods was greatly appreciated.” ○ “Well done. The resources you provided are exceptional.” On the follow-up electronic survey 4 months after the workshop, all respondents who did not have a boot camp before participating ( n = 3) reported that completing the workshop assisted them in planning or implementing a new boot camp. All respondents ( n = 7) ranked the materials and content from the workshop as being moderately helpful or very helpful as a self-contained resource for faculty to use independently when planning a boot camp based on a 5-point Likert scale (1 = not helpful at all , 2 = slightly helpful , 3 = moderately helpful , 4 = very helpful , 5 = extremely helpful ), with a weighted average of the responses of 3.57 and a range of 3 to 4. Responses from participants without a preexisting boot camp ( n = 3) included the following: • Prompt: Which of the workshop exercises and materials were helpful in planning or implementing a new boot camp? ○ Boot camp background literature review ( n = 1). ○ Example schedules from facilitators’ institutions ( n = 2). ○ Sample schedule worksheet exercise ( n = 1). ○ Example MedEdPORTAL boot camp module list ( n = 2). ○ Module design worksheet exercise based on selected EPAs ( n = 1). ○ Implementation strategies and barriers discussion ( n = 2). • Prompt: Please specify how your participation in the workshop helped you develop a boot camp. ○ “Bootcamp was already being planned by the medical school, but I was able to bring some useful ideas to the planning group due to workshop participation.” ○ “Utilizing ideas from workshop discussions in creation of bootcamp at my institution.” Responses from participants with a preexisting boot camp ( n = 4) included the following: • Prompt: Which of the workshop exercises and materials were helpful in making changes to your boot camp? ○ Needs assessment worksheet exercise ( n = 1). ○ Module design worksheet exercise ( n = 1). • Prompt: What specific changes did you make to your existing boot camp as a result of your participation in the workshop? ○ “ For the upcoming year will incorporate more on advocacy, poverty and wellness. ”
Preresidency boot camps for graduating medical students are rapidly growing in popularity and, in some cases, are required courses. This resource aims to provide guidance to faculty so that they can develop the courses expeditiously, paying attention to the particular needs and requirements of their institutions. Workshop participants used the pediatric boot camp course design tool kit to develop new courses and refine ongoing curricula at their home institutions, guided by the initial stages of Kern's six steps for curriculum development. By working through a needs assessment, developing targeted goals and objectives, designing modules, and considering barriers to implementation, participants successfully used the tool kit to begin the design and improvement of pediatric boot camps. Participants reported a statistically significant increase in confidence for all five components of the workshop following participation, and several reported using components of the tool kit to continue designing and improving boot camps following the workshop. All survey respondents reported that the tool kit would be at least moderately helpful as a self-contained resource for faculty to use in designing boot camp courses. Our findings are important because there are a very few published resources to help faculty who are designing boot camp courses. Faculty could use the materials to facilitate a similar workshop, thereby assisting others in designing pediatric boot camp courses. Alternatively, considering that the pediatric boot camp course design tool kit is a self-contained resource, faculty could use it independently to design a single course. Last, although this workshop and the associated tool kit were designed with a focus on pediatric boot camps, the steps of curriculum development are the same for any discipline, and with minor adaptations, the materials could be used to design either generalized boot camps for an entire senior medical school class or specialty-specific boot camps for students pursuing a nonpediatric residency. The pediatric boot camp course design workshop and tool kit add to the growing literature surrounding development of preresidency boot camps. The field of surgery has a robust national curriculum being pilot tested across dozens of institutions ; however, other disciplines have not developed supporting materials as quickly, despite similar institutional requirements to provide training in this area. This resource is an effort to continue to build and develop coordinated resources to maximize intern preparedness. For faculty with existing boot camps, this workshop and tool kit provide structured materials that can guide users to reassess and revise their curricula. This is especially important following initial boot camp implementation or significant institutional curricular changes that could affect stakeholder needs and content goals. Included in the survey results presented above, workshop attendees with existing boot camps planned adjustments to their courses, including incorporation of new content and revision of module design to make activities more engaging. Given the rapidly evolving literature surrounding the transition from medical school to residency, revisiting an existing curriculum through the lens of this tool kit may help users continually improve on the relevancy and efficacy of their courses. There are some limitations of this workshop and tool kit. First, considering that the workshop was presented nationally at a single venue, the number of participants was relatively small. Although we plan on further dissemination through future workshops and faculty development sessions, we felt that it was important to publicize the work so that faculty would be able to access it. Also, the response rate for the surveys was relatively low, likely reflecting survey fatigue. In the future, we plan to provide paper surveys at the end of the workshop (not sent over email following the workshop) and also hold focused interviews about the tool kit to discern the most useful aspects. In addition, we learned several specific lessons related to the tool kit that may be helpful for people using it. During the needs assessment, we found that peer sharing helped participants identify potential needs and topics they had not considered previously. For faculty using this tool kit outside of the workshop format, it may be useful to enlist others within the institution to provide this form of peer sharing. When creating sample schedules, participants ranked most choices in the highest, need-this category, making it difficult for them to prioritize content, especially in the shorter boot camps. An additional step of dividing content into general and pediatric specific could help participants with shorter boot camps to prioritize pediatric content. Our workshop did not specifically address whether boot camps should be an elective or required course, although the topic was discussed during the barriers and solutions segment of the presentation. This status varies within our own institutions and remains an area for additional exploration in the literature. Finally, the workshop could be led by a single facilitator; however, having multiple facilitators allowed for more interaction and guidance during the small-group exercises. By implementing this workshop and disseminating the tool kit for widespread use, it is our hope that educators will be able to efficiently develop pediatric boot camps that address the specific requirements and needs of individual institutions while simultaneously uniformly preparing students for the crucial transition to intern physician. It is also our hope that through continued collaboration, educators will develop and share resources related to residency preparedness so as to build robust learning resources for use by all. Future work includes collaborating with both COMSEP and the Association of Pediatric Program Directors to establish standard resources for programs to use in pediatric residency preparatory courses.
A. Boot Camp Workshop Presentation.pptx B. Review of Existing Boot Camp Literature.docx C. Institutional Needs Assessment Worksheet.docx D. Recommended Content List and Session Prioritization Worksheet.docx E. Schedule Worksheet and Sample Schedules.docx F. Module Design Worksheet and Planning Resources.docx G. Selected MedEdPORTAL Boot Camp Resources.docx H. Workshop Feedback Surveys.docx I. Facilitator Guide.docx All appendices are peer reviewed as integral parts of the Original Publication.
|
Proteomic Insights into the Effects of Jianweixiaoshi Tablets on Functional Dyspepsia with Spleen Deficiency in Rats | d79848ee-5379-4b61-a50f-f0cf4257e9c9 | 11568854 | Biochemistry[mh] | Functional dyspepsia (FD) is a common disorder of gut-brain interaction that originates in the gastroduodenal region, without a structural abnormality explaining the problems. The prevalence of FD is about 16% worldwide. According to the Rome IV consensus, FD can be categorized into postprandial distress syndrome with postprandial fullness or early satiation, and epigastric pain syndrome with epigastric pain and/or burning. Currently, the pathogenesis of FD is complex and includes factors such as abnormal gastric emptying, increased visceral sensitivity, mild duodenal inflammation, and disturbances in the gut-brain axis. Unfortunately, the symptomatic treatment of FD has limited success in fully improving patients’ symptoms and quality of life, accompanied by adverse reactions. Traditional Chinese medicine (TCM), known for its wide range of components and targets, high safety, and fewer toxic side effects, has shown promising results in treating FD. In TCM, FD is defined by the TCM terms “epigastric pain” and “distension and fullness” based on clinical symptoms and signs. Based on the principle of syndrome differentiation in treatment, the Chinese consensus on the diagnosis and treatment of FD considers spleen deficiency as the primary syndrome type of FD. A clinical survey involving 565 patients with FD revealed that 63.6% of the patients were diagnosed with syndromes related to spleen deficiency. Spleen deficiency is primarily characterized by weakened digestive function, abnormal secretion of gastrointestinal (GI) hormones, and decreased immune function, which overlap with the symptoms of FD. Clarifying the symptoms of FD in Chinese medicine facilitates the selection of appropriate Chinese medicines. Jianweixiaoshi tablets (JWXS), a traditional Chinese medicine used to treat dyspepsia caused by spleen and stomach weakness, has been used in the clinical treatment of FD in China. It is a TCM modified from the classical prescription “Spleen-invigorating pill”, which was recorded in the ancient Chinese medicine book “Zheng Zhi Zhun Sheng” by Wang Kentang of the Ming Dynasty in China. JWXS contains five botanical drugs, including Taizishen (Radix Pseudostellariae), Shanyao (Dioscorea opposita Thunb.), Chenpi (Citri Reticulatae Pericarpium), Shanzha (Crataegus pinnatifida Bunge), and fired Maiya (Hordei Fructus Germinatus). Pharmacological research has demonstrated that polysaccharides extracted from the raw materials for JWXS can promote gastric emptying function in mice. Additionally, previous reports have indicated that Chenpi, Shanzha, and fired Maiya extracts may have beneficial effects on the clinical symptoms of FD. Extracts from Taizishen and Shanyao can regulate immunity and alleviate the syndromes of spleen deficiency. , Although JWXS is extensively utilized for FD treatment in China, there is insufficient information on the key targets and underlying mechanisms through which JWXS relieves FD. Herbal medicines consist of a variety of components and usually have multiple pharmacological effects, exerting synergistic therapeutic effects. Proteomics plays a key role in investigating disease biomarkers and drug targets and has been extensively utilized in TCM studies to uncover the pharmacological mechanisms of traditional medicines. , In this study, we established a rat model of functional dyspepsia with spleen deficiency (SD-FD) by using intragastric iodoacetamide, combined with the modified multiple-platform method. To evaluate the pharmacological effects of JWXS in the treatment of SD-FD, this study examined the effects of JWXS on GI motility, hormones, and immune function in rats with SD-FD. Quantitative proteomic analysis was employed to examine alterations in the protein profiles of gastric and duodenal tissues in both SD-FD rats and those treated with JWXS. These findings provide valuable insights for the clinical application of JWXS in managing FD.
Drugs and Reagents JWXS consists of five different herbs ( ). These ingredients are prepared by Jiangzhong Pharmaceutical Co., Ltd., China. UPLC-MS/MS was used to identify the major compounds of the JWXS extract (batch number: 22030005). Domperidone (batch number: KDJ3YSP) and iodoacetamide were obtained from Xi’an Janssen Pharmaceutical Ltd. (Xi’an, China) and Sigma-Aldrich, respectively (St. Louis, Missouri, United States). Concentrations of the gastrin ELISA kit (CSB-E12743r), motilin ELISA kit (CSB-E08208r), vasoactive intestinal polypeptide (VIP) ELISA kit (CSB-E08355r), ghrelin ELISA kit (CSB-E09816r), and somatostatin (SST) ELISA kit (CSB-E08204r) were purchased from Cusabio Biotech Co. Ltd. (Wuhan, China). The cholecystokinin-octapeptide (CCK-8) ELISA kit (CEB044Ra) was acquired from USCN Life Science Inc (Wuhan, China). Examination of JWXS Extract by UPLC-MS/MS About 2.0 g of the JWXS extract was accurately weighed and extracted with 25 mL of 80% v/v methanol for 30 minutes. The 80% methanol extracts were centrifuged at 13000 rpm for 15 minutes at 4°C and the supernatants were filtered through a 0.22 μm filter membrane. The filtrate was collected and stored at 4°C before UPLC-MS/MS analysis. Sample separation and analysis were performed on a UHPLC-Q Exactive system (Thermo Scientific, Bremen, Germany). The samples were separated at 40°C using a Waters UPLC BEH C18 column (100 × 2.1 mm, 1.7 µm). The mobile phase consisted of 0.1% formic acid in water (A) and acetonitrile (B). The flow rate was set to 0.4 mL/min, and the sample injection volume was set to 5 µL. The gradient program was as follows: 0–11 min, 85–25% A; 11–12 min, 25–2% A; 12–14 min, 2–2% A; 14–14.1 min, 2–8% A; 14.1–16 min, 85–85% A. Mass spectrometry analysis was conducted using a Q-Exactive mass spectrometer (Thermo Scientific, Bremen, Germany) coupled with Xcalibur software (Thermo Scientific). The samples underwent ionization through electrospray ionization, and the mass spectral signals were collected in positive and negative ion scanning modes, respectively. The data were collected in the m/z range of 70–1050. The mass spectrometric detection mode was Full MS/dd-MS 2 with a resolution of 70,000 for Full MS and 17,500 for dd-MS 2 . The optimized source parameters in positive (negative) mode included a capillary temperature of 320°C, auxiliary gas temperature at 400°C, sheath gas flow at 40 arb, auxiliary gas flow at 10 arb, and a spray voltage of 3.5 kV (positive) or −2.8 kV (negative) with collision energies of 20, 40, and 60 eV. Animals 7-day-old male Sprague-Dawley rats were purchased from SPF (Beijing) Biotechnology Co., Ltd. [SCXK (Jing) 2019–0010]. Rats were housed in a climate-controlled environment with an ambient temperature of 20–25°C and a relative humidity of 40–70%, following a 12-hour light and dark cycle. All rats had ad libitum access to food and water throughout the experiment. All animal care and experimental procedures were conducted in strict accordance with the National Institutes of Health guidelines for the care and use of laboratory animals. This study was approved by the Laboratory Animal Ethics Committee of Jiangzhong Pharmaceutical Co., Ltd. (Nanchang, China). SD-FD Model Establishment and Treatment After acclimatization feeding, the rats were randomly assigned to either the healthy control group (Control) or the model group (Model). SD-FD model rats were established by gastric perfusion with an iodoacetamide solution combined with the modified multiple platform method (MMPM). In brief, 10-day-old model rats were orally administered a mixture of 0.1% iodoacetamide and 2% sucrose at a dosage of 0.2 mL/day for 7 consecutive days. At week 3, the SD-FD rats were randomly divided into the following groups: Model group (distilled water, 10 mL/kg), low-dose JWXS group (JWXS-L, 0.28 g/kg/d, clinical equivalent dose), high-dose JWXS group (JWXS-H, 1.4 g/kg/d) and positive control group (Domperidone, 3 mg/kg/d, clinical equivalent dose). Domperidone, as a dopamine receptor antagonist, is a widely utilized treatment for FD that enhances GI motility. Research has substantiated its effectiveness, and it was utilized as a positive control drug. At week 6, these groups were placed in a water-filled platform box for 14 consecutive days (16 hours per day) to induce spleen deficiency. Concomitantly, the rats received distilled water, low-dose JWXS, high-dose JWXS, or Domperidone by oral gavage for 14 days. The schematic diagram of the experimental design is shown in . At the end of the experiment, the rats were fasted for 18 h with free access to water. The following day, rats were orally administered 2 mL of 10% hydroxymethyl cellulose solution containing 0.04% phenol red, one hour after the last administration. Then the gastric emptying rate and intestinal propulsion rate were determined (refer to the “Gastric emptying and intestine propulsion test”). After 20 minutes, the rats were euthanized, and blood samples were collected. Blood samples were allowed to clot at room temperature for 2 h and then centrifuged at 4000 rpm for 10 minutes at 4°C. The serum was collected and stored at −80°C. Gastric Emptying and Intestine Propulsion Test The rate of gastric emptying and intestinal propulsion was assessed using a phenol red meal, as described in previous studies. After overnight fasting, the rats were given 2 mL of 10% hydroxymethyl cellulose containing 0.04% phenol red by gavage 1 h after the last administration. Twenty minutes later, the rats were euthanized, and their stomachs and small intestines were picked out. After making an incision along the greater curvature of the stomach and thoroughly collecting the gastric contents with 30 mL of 0.5 mol/L NaOH solution, the mixture was centrifuged at 3000 rpm for 10 minutes. The absorbance of the supernatant was measured at 560 nm using a UV-Vis spectrophotometer. The gastric emptying rate was calculated using the following formula (1). (1) [12pt]{minimal}
[substack]{amsmath}
[mathscr]{eucal}
{linotext }{}
$${}\ {}\ {}( % ) = 100 % - {{{}560\ {}\ {}\ {}\ {}} {{}560\ {}\ {}\ {}\ {}}} 100\% $$ To determine the rate of intestinal propulsion, we measured the distance traveled by phenol red through the small intestine and the total length of the small intestine. The intestinal propulsion rate was calculated using the following formula (2). (2) [12pt]{minimal}
[substack]{amsmath}
[mathscr]{eucal}
{linotext }{}
$${{ Intestinal\ propulsion\ rate }}({{ \% }} ){{ = }}{{{{ distance\ traveled\ by\ phenol\ red}}} {{{ length\ of\ the\ small\ intestine}}}}{{ 100\% }}$$ Immune Organ Index Collecting and weighing the immune organs (thymus and spleen) in rats, excess liquid was removed using filter paper to minimize calculation errors. The immune organ index was calculated by the following formula (3). (3) [12pt]{minimal}
[substack]{amsmath}
[mathscr]{eucal}
{linotext }{}
$${{ organ\ index }}({{ \% }} ){{ = }}{{{{ organ\ mass}}} {{{ body\ weight}}}}{{ 100\% }}$$ Hematological Analyses Whole blood was collected from anesthetized animals and temporarily stored in EDTA-coated tubes. Lymphocytes (Lym), white blood cells (WBC), and neutrophils (Neu) were evaluated using an automatic blood cell analyzer (BC-5000VET, Mindray, Shenzhen, China). Biochemical Analyses Serum lactate dehydrogenase (LDH), L-lactic acid (LA), and urea nitrogen (BUN) were measured using the kit from Nanjing Jiancheng Bioengineering Institute (Cat No. A020-2-2, A019-2-1, and C013-2-1, respectively), following the manufacturer’s instructions. Gastrointestinal Hormones Measurement Serum levels of gastrin, motilin, CCK-8, ghrelin, VIP, and SST were detected using ELISA kits according to the manufacturer’s instructions. Briefly, each sample (100 μL) was mixed sequentially with Biotin-antibody, HRP-avidin, TMB Substrate, and stop solution, following the manufacturer’s instructions. The mixture was then incubated at the designated temperature for the specified duration. Finally, the absorbance of the sample was measured at 450 nm, and the concentration of the tested GI hormones was determined using the standard curve. Quantitative Proteomics Sample Preparation Data-independent acquisition-based (DIA-based) quantitative proteomics methods were used to analyze the protein changes in the gastric and duodenal tissues of SD-FD rats that were treated with JWXS. Proteomic analysis was conducted at Beijing Qinglian Baio Biotechnology Co. (Beijing, China), following the manufacturer’s protocols. Briefly, proteins were extracted from gastric and duodenal tissues, and the concentration of the extracted proteins was determined using the Bradford method. 100 μg of protein was prepared and adjusted to a final concentration of 5 mm DTT (M109-5G, Amresco, USA). The mixture was then incubated at 37°C for 1 h and subsequently returned to room temperature. Next, iodoacetamide IAM (M216-30G, Amresco, USA) was added to achieve a final concentration of 10 mm at room temperature for 45 min. The samples were then diluted 4 times with 25 mm ammonium bicarbonate (A6141-500G, Sigma-Aldrich, USA), and trypsin was added at a ratio of 50:1 of protein to trypsin (V5280/100ug, Promega, USA). The mixture was incubated overnight at 37°C. Formic acid (T79708, Sigma-Aldrich, USA) was added to terminate the reaction and adjusted to pH < 3. Finally, the sample was desalted using a C18 column. LC-MS/MS Analysis High-performance liquid chromatography (HPLC) with a RIGOL L-3000 system (Beijing Puyuan Precision Electric Technology, China) was employed for gradient elution. The eluted peptides were subjected to mass spectrometry analysis using a Q Exactive HF-X mass spectrometer (Thermo Fisher Scientific, USA). Mobile phase A consisted of 100% mass spectrometry water with 0.1% formic acid, while mobile phase B comprised 80% acetonitrile (34851 MSDS, J.T. Baker, USA) with 0.1% formic acid. The separation flow rate was set at 0.7 mL/min, and the elution procedure consisted of the following steps: 0–5 min, 5–8% B; 5–40 min, 8–18% B; 40–62 min, 18–32% B; 62–64 min, 32–95% B; 64–68 min, 95% B. For the mass spectrometry analysis, a Nanospray Flex (NSI) ion source was utilized with an ion spray voltage of 2.0 kV, and the temperature of the ion transfer tube was set to 320 °C. The full scanning range was set as m/z 350–1500 with a primary mass spectrometry resolution of 120,000 (200 m/z). The AGC was set to 300%, and the maximum injection time was set to 50 ms. The secondary mass spectrometry resolution was set at 30,000 (200 m/z), with an AGC of 100%, and the maximum injection time was set to 54 ms. The Spectronaut software was used to search the Rattus norvegicus database. Proteomics Data Analysis Difference analysis using the t -test method was conducted to analyze proteomic data. Differential proteins were selected based on the criteria of p -value < 0.05 and Fold change > 1.2. Gene ontology (GO) analysis of the non-redundant protein database was conducted using InterProScan-5. We analyzed the protein families and pathways by using the Kyoto Encyclopedia of Genes and Genomes (KEGG) database. Data and Statistical Analysis Data were presented as mean ± standard error of the mean (SEM). Statistical significance among multiple groups was analyzed using one-way ANOVA followed by Dunnett’s multiple comparisons test, performed with GraphPad Prism 8 (GraphPad Software, San Diego, CA, USA). An unpaired two-tailed Student’s t -test was employed to determine statistical significance between the two groups. A significance level of p < 0.05 was considered statistically significant for all analyses.
JWXS consists of five different herbs ( ). These ingredients are prepared by Jiangzhong Pharmaceutical Co., Ltd., China. UPLC-MS/MS was used to identify the major compounds of the JWXS extract (batch number: 22030005). Domperidone (batch number: KDJ3YSP) and iodoacetamide were obtained from Xi’an Janssen Pharmaceutical Ltd. (Xi’an, China) and Sigma-Aldrich, respectively (St. Louis, Missouri, United States). Concentrations of the gastrin ELISA kit (CSB-E12743r), motilin ELISA kit (CSB-E08208r), vasoactive intestinal polypeptide (VIP) ELISA kit (CSB-E08355r), ghrelin ELISA kit (CSB-E09816r), and somatostatin (SST) ELISA kit (CSB-E08204r) were purchased from Cusabio Biotech Co. Ltd. (Wuhan, China). The cholecystokinin-octapeptide (CCK-8) ELISA kit (CEB044Ra) was acquired from USCN Life Science Inc (Wuhan, China).
About 2.0 g of the JWXS extract was accurately weighed and extracted with 25 mL of 80% v/v methanol for 30 minutes. The 80% methanol extracts were centrifuged at 13000 rpm for 15 minutes at 4°C and the supernatants were filtered through a 0.22 μm filter membrane. The filtrate was collected and stored at 4°C before UPLC-MS/MS analysis. Sample separation and analysis were performed on a UHPLC-Q Exactive system (Thermo Scientific, Bremen, Germany). The samples were separated at 40°C using a Waters UPLC BEH C18 column (100 × 2.1 mm, 1.7 µm). The mobile phase consisted of 0.1% formic acid in water (A) and acetonitrile (B). The flow rate was set to 0.4 mL/min, and the sample injection volume was set to 5 µL. The gradient program was as follows: 0–11 min, 85–25% A; 11–12 min, 25–2% A; 12–14 min, 2–2% A; 14–14.1 min, 2–8% A; 14.1–16 min, 85–85% A. Mass spectrometry analysis was conducted using a Q-Exactive mass spectrometer (Thermo Scientific, Bremen, Germany) coupled with Xcalibur software (Thermo Scientific). The samples underwent ionization through electrospray ionization, and the mass spectral signals were collected in positive and negative ion scanning modes, respectively. The data were collected in the m/z range of 70–1050. The mass spectrometric detection mode was Full MS/dd-MS 2 with a resolution of 70,000 for Full MS and 17,500 for dd-MS 2 . The optimized source parameters in positive (negative) mode included a capillary temperature of 320°C, auxiliary gas temperature at 400°C, sheath gas flow at 40 arb, auxiliary gas flow at 10 arb, and a spray voltage of 3.5 kV (positive) or −2.8 kV (negative) with collision energies of 20, 40, and 60 eV.
7-day-old male Sprague-Dawley rats were purchased from SPF (Beijing) Biotechnology Co., Ltd. [SCXK (Jing) 2019–0010]. Rats were housed in a climate-controlled environment with an ambient temperature of 20–25°C and a relative humidity of 40–70%, following a 12-hour light and dark cycle. All rats had ad libitum access to food and water throughout the experiment. All animal care and experimental procedures were conducted in strict accordance with the National Institutes of Health guidelines for the care and use of laboratory animals. This study was approved by the Laboratory Animal Ethics Committee of Jiangzhong Pharmaceutical Co., Ltd. (Nanchang, China).
After acclimatization feeding, the rats were randomly assigned to either the healthy control group (Control) or the model group (Model). SD-FD model rats were established by gastric perfusion with an iodoacetamide solution combined with the modified multiple platform method (MMPM). In brief, 10-day-old model rats were orally administered a mixture of 0.1% iodoacetamide and 2% sucrose at a dosage of 0.2 mL/day for 7 consecutive days. At week 3, the SD-FD rats were randomly divided into the following groups: Model group (distilled water, 10 mL/kg), low-dose JWXS group (JWXS-L, 0.28 g/kg/d, clinical equivalent dose), high-dose JWXS group (JWXS-H, 1.4 g/kg/d) and positive control group (Domperidone, 3 mg/kg/d, clinical equivalent dose). Domperidone, as a dopamine receptor antagonist, is a widely utilized treatment for FD that enhances GI motility. Research has substantiated its effectiveness, and it was utilized as a positive control drug. At week 6, these groups were placed in a water-filled platform box for 14 consecutive days (16 hours per day) to induce spleen deficiency. Concomitantly, the rats received distilled water, low-dose JWXS, high-dose JWXS, or Domperidone by oral gavage for 14 days. The schematic diagram of the experimental design is shown in . At the end of the experiment, the rats were fasted for 18 h with free access to water. The following day, rats were orally administered 2 mL of 10% hydroxymethyl cellulose solution containing 0.04% phenol red, one hour after the last administration. Then the gastric emptying rate and intestinal propulsion rate were determined (refer to the “Gastric emptying and intestine propulsion test”). After 20 minutes, the rats were euthanized, and blood samples were collected. Blood samples were allowed to clot at room temperature for 2 h and then centrifuged at 4000 rpm for 10 minutes at 4°C. The serum was collected and stored at −80°C.
The rate of gastric emptying and intestinal propulsion was assessed using a phenol red meal, as described in previous studies. After overnight fasting, the rats were given 2 mL of 10% hydroxymethyl cellulose containing 0.04% phenol red by gavage 1 h after the last administration. Twenty minutes later, the rats were euthanized, and their stomachs and small intestines were picked out. After making an incision along the greater curvature of the stomach and thoroughly collecting the gastric contents with 30 mL of 0.5 mol/L NaOH solution, the mixture was centrifuged at 3000 rpm for 10 minutes. The absorbance of the supernatant was measured at 560 nm using a UV-Vis spectrophotometer. The gastric emptying rate was calculated using the following formula (1). (1) [12pt]{minimal}
[substack]{amsmath}
[mathscr]{eucal}
{linotext }{}
$${}\ {}\ {}( % ) = 100 % - {{{}560\ {}\ {}\ {}\ {}} {{}560\ {}\ {}\ {}\ {}}} 100\% $$ To determine the rate of intestinal propulsion, we measured the distance traveled by phenol red through the small intestine and the total length of the small intestine. The intestinal propulsion rate was calculated using the following formula (2). (2) [12pt]{minimal}
[substack]{amsmath}
[mathscr]{eucal}
{linotext }{}
$${{ Intestinal\ propulsion\ rate }}({{ \% }} ){{ = }}{{{{ distance\ traveled\ by\ phenol\ red}}} {{{ length\ of\ the\ small\ intestine}}}}{{ 100\% }}$$
Collecting and weighing the immune organs (thymus and spleen) in rats, excess liquid was removed using filter paper to minimize calculation errors. The immune organ index was calculated by the following formula (3). (3) [12pt]{minimal}
[substack]{amsmath}
[mathscr]{eucal}
{linotext }{}
$${{ organ\ index }}({{ \% }} ){{ = }}{{{{ organ\ mass}}} {{{ body\ weight}}}}{{ 100\% }}$$
Whole blood was collected from anesthetized animals and temporarily stored in EDTA-coated tubes. Lymphocytes (Lym), white blood cells (WBC), and neutrophils (Neu) were evaluated using an automatic blood cell analyzer (BC-5000VET, Mindray, Shenzhen, China).
Serum lactate dehydrogenase (LDH), L-lactic acid (LA), and urea nitrogen (BUN) were measured using the kit from Nanjing Jiancheng Bioengineering Institute (Cat No. A020-2-2, A019-2-1, and C013-2-1, respectively), following the manufacturer’s instructions.
Serum levels of gastrin, motilin, CCK-8, ghrelin, VIP, and SST were detected using ELISA kits according to the manufacturer’s instructions. Briefly, each sample (100 μL) was mixed sequentially with Biotin-antibody, HRP-avidin, TMB Substrate, and stop solution, following the manufacturer’s instructions. The mixture was then incubated at the designated temperature for the specified duration. Finally, the absorbance of the sample was measured at 450 nm, and the concentration of the tested GI hormones was determined using the standard curve.
Sample Preparation Data-independent acquisition-based (DIA-based) quantitative proteomics methods were used to analyze the protein changes in the gastric and duodenal tissues of SD-FD rats that were treated with JWXS. Proteomic analysis was conducted at Beijing Qinglian Baio Biotechnology Co. (Beijing, China), following the manufacturer’s protocols. Briefly, proteins were extracted from gastric and duodenal tissues, and the concentration of the extracted proteins was determined using the Bradford method. 100 μg of protein was prepared and adjusted to a final concentration of 5 mm DTT (M109-5G, Amresco, USA). The mixture was then incubated at 37°C for 1 h and subsequently returned to room temperature. Next, iodoacetamide IAM (M216-30G, Amresco, USA) was added to achieve a final concentration of 10 mm at room temperature for 45 min. The samples were then diluted 4 times with 25 mm ammonium bicarbonate (A6141-500G, Sigma-Aldrich, USA), and trypsin was added at a ratio of 50:1 of protein to trypsin (V5280/100ug, Promega, USA). The mixture was incubated overnight at 37°C. Formic acid (T79708, Sigma-Aldrich, USA) was added to terminate the reaction and adjusted to pH < 3. Finally, the sample was desalted using a C18 column. LC-MS/MS Analysis High-performance liquid chromatography (HPLC) with a RIGOL L-3000 system (Beijing Puyuan Precision Electric Technology, China) was employed for gradient elution. The eluted peptides were subjected to mass spectrometry analysis using a Q Exactive HF-X mass spectrometer (Thermo Fisher Scientific, USA). Mobile phase A consisted of 100% mass spectrometry water with 0.1% formic acid, while mobile phase B comprised 80% acetonitrile (34851 MSDS, J.T. Baker, USA) with 0.1% formic acid. The separation flow rate was set at 0.7 mL/min, and the elution procedure consisted of the following steps: 0–5 min, 5–8% B; 5–40 min, 8–18% B; 40–62 min, 18–32% B; 62–64 min, 32–95% B; 64–68 min, 95% B. For the mass spectrometry analysis, a Nanospray Flex (NSI) ion source was utilized with an ion spray voltage of 2.0 kV, and the temperature of the ion transfer tube was set to 320 °C. The full scanning range was set as m/z 350–1500 with a primary mass spectrometry resolution of 120,000 (200 m/z). The AGC was set to 300%, and the maximum injection time was set to 50 ms. The secondary mass spectrometry resolution was set at 30,000 (200 m/z), with an AGC of 100%, and the maximum injection time was set to 54 ms. The Spectronaut software was used to search the Rattus norvegicus database. Proteomics Data Analysis Difference analysis using the t -test method was conducted to analyze proteomic data. Differential proteins were selected based on the criteria of p -value < 0.05 and Fold change > 1.2. Gene ontology (GO) analysis of the non-redundant protein database was conducted using InterProScan-5. We analyzed the protein families and pathways by using the Kyoto Encyclopedia of Genes and Genomes (KEGG) database.
Data-independent acquisition-based (DIA-based) quantitative proteomics methods were used to analyze the protein changes in the gastric and duodenal tissues of SD-FD rats that were treated with JWXS. Proteomic analysis was conducted at Beijing Qinglian Baio Biotechnology Co. (Beijing, China), following the manufacturer’s protocols. Briefly, proteins were extracted from gastric and duodenal tissues, and the concentration of the extracted proteins was determined using the Bradford method. 100 μg of protein was prepared and adjusted to a final concentration of 5 mm DTT (M109-5G, Amresco, USA). The mixture was then incubated at 37°C for 1 h and subsequently returned to room temperature. Next, iodoacetamide IAM (M216-30G, Amresco, USA) was added to achieve a final concentration of 10 mm at room temperature for 45 min. The samples were then diluted 4 times with 25 mm ammonium bicarbonate (A6141-500G, Sigma-Aldrich, USA), and trypsin was added at a ratio of 50:1 of protein to trypsin (V5280/100ug, Promega, USA). The mixture was incubated overnight at 37°C. Formic acid (T79708, Sigma-Aldrich, USA) was added to terminate the reaction and adjusted to pH < 3. Finally, the sample was desalted using a C18 column.
High-performance liquid chromatography (HPLC) with a RIGOL L-3000 system (Beijing Puyuan Precision Electric Technology, China) was employed for gradient elution. The eluted peptides were subjected to mass spectrometry analysis using a Q Exactive HF-X mass spectrometer (Thermo Fisher Scientific, USA). Mobile phase A consisted of 100% mass spectrometry water with 0.1% formic acid, while mobile phase B comprised 80% acetonitrile (34851 MSDS, J.T. Baker, USA) with 0.1% formic acid. The separation flow rate was set at 0.7 mL/min, and the elution procedure consisted of the following steps: 0–5 min, 5–8% B; 5–40 min, 8–18% B; 40–62 min, 18–32% B; 62–64 min, 32–95% B; 64–68 min, 95% B. For the mass spectrometry analysis, a Nanospray Flex (NSI) ion source was utilized with an ion spray voltage of 2.0 kV, and the temperature of the ion transfer tube was set to 320 °C. The full scanning range was set as m/z 350–1500 with a primary mass spectrometry resolution of 120,000 (200 m/z). The AGC was set to 300%, and the maximum injection time was set to 50 ms. The secondary mass spectrometry resolution was set at 30,000 (200 m/z), with an AGC of 100%, and the maximum injection time was set to 54 ms. The Spectronaut software was used to search the Rattus norvegicus database.
Difference analysis using the t -test method was conducted to analyze proteomic data. Differential proteins were selected based on the criteria of p -value < 0.05 and Fold change > 1.2. Gene ontology (GO) analysis of the non-redundant protein database was conducted using InterProScan-5. We analyzed the protein families and pathways by using the Kyoto Encyclopedia of Genes and Genomes (KEGG) database.
Data were presented as mean ± standard error of the mean (SEM). Statistical significance among multiple groups was analyzed using one-way ANOVA followed by Dunnett’s multiple comparisons test, performed with GraphPad Prism 8 (GraphPad Software, San Diego, CA, USA). An unpaired two-tailed Student’s t -test was employed to determine statistical significance between the two groups. A significance level of p < 0.05 was considered statistically significant for all analyses.
Main Components of JWXS The total ion chromatogram of JWXS was obtained by UPLC-MS/MS analysis, and a total of 115 main compounds were identified ( and ), including 50 flavonoids, 18 terpenoids, and some other compounds. JWXS Alleviates the Symptoms of the Rats with SD-FD The SD-FD model was established by gastric infusion of iodoacetamide combined with the modified multi-platform method (MMPM) to evaluate the intervention effect of JWXS in the SD-FD model. At the end of the experiment, the model rats showed significant weight loss and decreased food consumption ( p < 0.05 or p < 0.001, and ), as well as symptoms such as lackluster hair and fatigue, which were consistent with the spleen deficiency syndrome. Furthermore, the body weight of rats in the JWXS treatment group increased compared to the model group; however, no statistically significant difference was observed ( p > 0.05). Food consumption significantly increased in all treatment groups compared to the model group ( p < 0.05). The gastric emptying rate and intestinal propulsive rate of the model group were significantly decreased compared to the control group ( p < 0.05, or p < 0.001, and ). After JWXS and Domperidone treatment, the gastric emptying rate and intestinal propulsion rate of the treatment group significantly increased ( p < 0.05, or p < 0.01). Concomitantly, the spleen and thymus indices in the model group of rats decreased, suggesting a potential impairment in immune function for the SD-FD rats ( and ). In contrast, the spleen and thymus indices of rats in the high and low-dose groups showed significant improvement compared to the model group ( p < 0.05, or p < 0.01). Effects of JWXS on Fatigue-Related Biochemical Indexes and Inflammatory Cell Numbers in SD-FD Rats The characteristics of the spleen deficiency model include low appetite, fatigue, and reduced immune function. Fatigue in modern medicine encompasses damage to the central nervous system and decreased function of the motor system. To assess the effects of JWXS on fatigue and immune function in FD-SD rats, we measured serum levels of LDH, LA, and BUN, along with blood inflammatory cell count. As shown in , serum levels of LDH, LA, and BUN were significantly increased ( p < 0.05, p < 0.01, or p < 0.001) in the model group compared to the normal group. Additionally, blood WBC count and %Lym were significantly decreased ( p < 0.05 or p < 0.001), while %Neu was significantly increased ( p < 0.001). In comparison to the model group, LDH, BUN, and %Neu were significantly decreased in the JWXS-L and JWXS-H groups ( p < 0.05, p < 0.01, or p < 0.001), whereas %Lym was extremely significantly increased ( p < 0.05). Furthermore, LA was significantly decreased in the JWXS-H treatment group ( p < 0.05), while WBC count was significantly increased ( p < 0.01). JWXS Improves SD-FD via the Regulation of Gastrointestinal Hormones The balance of hormones related to GI regulation plays an important role in regulating GI motility. Our study quantified the levels of six hormones associated with GI regulation: gastrin, motilin, cholecystokinin-octapeptide (CCK8), vasoactive intestinal peptide (VIP), somatostatin (SST), and ghrelin. As shown in , serum levels of gastrin, motilin, and ghrelin were significantly increased in all treatment groups ( p < 0.05, p < 0.01, or p < 0.001), while CCK8 and SST levels were significantly decreased ( p < 0.05 or p < 0.01). JWXS-H and Domperidone significantly decreased VIP levels ( p < 0.05). Proteomics Analysis of Differentially Expressed Proteins in Gastric Tissues The studies mentioned above demonstrated that JWXS-L significantly enhanced gastric motility, and the effect was similar to positive drug. Consequently, in the subsequent experiments, JWXS was orally administered at a dose of 0.45 g/kg (JWXS-L). To further elucidate the mechanisms behind the therapeutic effect of JWXS on SD-FD, we conducted DIA-based quantitative proteomic analysis on stomach tissues from three groups of rats: Control, Model, and JWXS (JWXS-L). A total of 28,647 peptides and 5400 proteins were detected using DIA-based quantitative proteomics. Volcano plots were used to visualize the distribution of differentially expressed proteins (DEPs) by applying filtering criteria of fold change (>1.2 for up-regulated or <0.83 for down-regulated) and p -value < 0.05. Between the Model and Control groups, 728 DEPs were identified, comprising 516 upregulated proteins and 212 downregulated proteins ( and ). Furthermore, a total of 333 DEPs (120 upregulated and 213 downregulated) were identified in the JWXS group compared to the Model group. The Venn plot displayed an overlap of 126 DEPs between the Model vs Control and JWXS vs Model groups ( ). Cluster analysis showed significant differences between the model and control groups ( ). Additionally, the protein expression was significantly changed after JWXS intervention in SD-FD rats. Bioinformatics Analysis of DEPs in Gastric Tissues To enhance comprehension of the biological properties of the differentially expressed proteins, we conducted a bioinformatics analysis ( ). The results of the GO enrichment analysis of differential proteins are presented in the form of a bar chart, which displays the top 10 terms for cellular component (CC), biological process (BP), and molecular function (MF). In biological processes, JWXS could improve energy metabolism by regulating the tricarboxylic acid cycle, succinyl-CoA metabolic process, respiratory electron transport chain, and the regulation of cytochrome-c oxidase activity. The results of the KEGG enrichment analysis of differential proteins are also displayed as a bar chart, which shows the top 20 pathways ( ). The JWXS-regulated differential proteins are primarily associated with the Citrate cycle (TCA cycle), Glutathione metabolism, the calcium signaling pathway, and others. In the calcium signaling pathway, the sequencing results revealed that the main differential proteins were SERCA2, PLC-γ2, Vdac1, Vdac2, and Vdac3. Among these proteins, the abundance of PLC-γ2 significantly increased after JWXS treatment ( , p < 0.05), while the abundance of SERCA2, Vdac1, Vdac2, and Vdac3 markedly decreased ( p < 0.05). Proteomics Analysis of Differentially Expressed Proteins in Duodenal Tissues The duodenum is a key region in the pathophysiology of FD. Disruption of the duodenal barrier has been demonstrated in FD patients, and the underlying damage-associated molecular pattern can enter the compromised intestinal barrier and trigger host innate immunity. To elucidate the therapeutic mechanism of JWXS on the duodenum of SD-FD rats, we conducted a DIA-based quantitative proteomic analysis on the duodenal tissues of three groups of rats. A total of 29682 peptides and 4421 proteins were detected using DIA-based quantitative proteomics. Volcano plots were used to visualize the distribution of differentially expressed proteins (DEPs) by applying filtering criteria of fold change (>1.5 for up-regulated or <0.67 for down-regulated) and p -value < 0.05 ( ). There was a total of 276 differentially expressed proteins (DEPs) identified between the Model and Control groups, of which 108 proteins were upregulated and 51 proteins were downregulated ( ). Meanwhile, a total of 732 DEPs (376 up-regulated and 277 down-regulated) were identified in the JWXS group compared to the Model group. The Venn plot showed that there were 64 overlapping DEPs found between the Model vs Control and the JWXS vs Model groups ( ). Cluster analysis showed significant differences between the model and control groups ( ). Additionally, the protein expression was significantly changed after JWXS intervention in SD-FD rats. Bioinformatics Analysis of DEPs in Duodenal Tissues To better classify the functional properties of differential proteins, the DEPs were subjected to GO analysis ( ). In biological processes, JWXS could respond to immune activation via antigen processing and presentation of exogenous peptide antigen via MHC class II, negative regulation of hydrogen peroxide-induced cell death, positive regulation of T cell-mediated cytotoxicity, protection from natural killer cell-mediated cytotoxicity. Furthermore, we performed the KEGG pathway analysis of differentially expressed proteins ( ). The results indicated that the KEGG pathways enriched in the JWXS group mainly included antigen processing and presentation, staphylococcus aureus infection, oxidative phosphorylation, intestinal immune network for IgA production, among others. In the antigen processing and presentation and intestinal immune network for IgA production, the sequencing results revealed that the main differential proteins were PIgR, RT1-A1, RT1-CE1, MHC II, Tap1, and Tap2. Among these proteins, the abundance of PIgR significantly increased after JWXS treatment ( , p < 0.05), while the abundance of RT1-A1, RT1-CE1, MHC II, Tap1, and Tap2 markedly decreased ( p < 0.05).
The total ion chromatogram of JWXS was obtained by UPLC-MS/MS analysis, and a total of 115 main compounds were identified ( and ), including 50 flavonoids, 18 terpenoids, and some other compounds.
The SD-FD model was established by gastric infusion of iodoacetamide combined with the modified multi-platform method (MMPM) to evaluate the intervention effect of JWXS in the SD-FD model. At the end of the experiment, the model rats showed significant weight loss and decreased food consumption ( p < 0.05 or p < 0.001, and ), as well as symptoms such as lackluster hair and fatigue, which were consistent with the spleen deficiency syndrome. Furthermore, the body weight of rats in the JWXS treatment group increased compared to the model group; however, no statistically significant difference was observed ( p > 0.05). Food consumption significantly increased in all treatment groups compared to the model group ( p < 0.05). The gastric emptying rate and intestinal propulsive rate of the model group were significantly decreased compared to the control group ( p < 0.05, or p < 0.001, and ). After JWXS and Domperidone treatment, the gastric emptying rate and intestinal propulsion rate of the treatment group significantly increased ( p < 0.05, or p < 0.01). Concomitantly, the spleen and thymus indices in the model group of rats decreased, suggesting a potential impairment in immune function for the SD-FD rats ( and ). In contrast, the spleen and thymus indices of rats in the high and low-dose groups showed significant improvement compared to the model group ( p < 0.05, or p < 0.01).
The characteristics of the spleen deficiency model include low appetite, fatigue, and reduced immune function. Fatigue in modern medicine encompasses damage to the central nervous system and decreased function of the motor system. To assess the effects of JWXS on fatigue and immune function in FD-SD rats, we measured serum levels of LDH, LA, and BUN, along with blood inflammatory cell count. As shown in , serum levels of LDH, LA, and BUN were significantly increased ( p < 0.05, p < 0.01, or p < 0.001) in the model group compared to the normal group. Additionally, blood WBC count and %Lym were significantly decreased ( p < 0.05 or p < 0.001), while %Neu was significantly increased ( p < 0.001). In comparison to the model group, LDH, BUN, and %Neu were significantly decreased in the JWXS-L and JWXS-H groups ( p < 0.05, p < 0.01, or p < 0.001), whereas %Lym was extremely significantly increased ( p < 0.05). Furthermore, LA was significantly decreased in the JWXS-H treatment group ( p < 0.05), while WBC count was significantly increased ( p < 0.01).
The balance of hormones related to GI regulation plays an important role in regulating GI motility. Our study quantified the levels of six hormones associated with GI regulation: gastrin, motilin, cholecystokinin-octapeptide (CCK8), vasoactive intestinal peptide (VIP), somatostatin (SST), and ghrelin. As shown in , serum levels of gastrin, motilin, and ghrelin were significantly increased in all treatment groups ( p < 0.05, p < 0.01, or p < 0.001), while CCK8 and SST levels were significantly decreased ( p < 0.05 or p < 0.01). JWXS-H and Domperidone significantly decreased VIP levels ( p < 0.05).
The studies mentioned above demonstrated that JWXS-L significantly enhanced gastric motility, and the effect was similar to positive drug. Consequently, in the subsequent experiments, JWXS was orally administered at a dose of 0.45 g/kg (JWXS-L). To further elucidate the mechanisms behind the therapeutic effect of JWXS on SD-FD, we conducted DIA-based quantitative proteomic analysis on stomach tissues from three groups of rats: Control, Model, and JWXS (JWXS-L). A total of 28,647 peptides and 5400 proteins were detected using DIA-based quantitative proteomics. Volcano plots were used to visualize the distribution of differentially expressed proteins (DEPs) by applying filtering criteria of fold change (>1.2 for up-regulated or <0.83 for down-regulated) and p -value < 0.05. Between the Model and Control groups, 728 DEPs were identified, comprising 516 upregulated proteins and 212 downregulated proteins ( and ). Furthermore, a total of 333 DEPs (120 upregulated and 213 downregulated) were identified in the JWXS group compared to the Model group. The Venn plot displayed an overlap of 126 DEPs between the Model vs Control and JWXS vs Model groups ( ). Cluster analysis showed significant differences between the model and control groups ( ). Additionally, the protein expression was significantly changed after JWXS intervention in SD-FD rats.
To enhance comprehension of the biological properties of the differentially expressed proteins, we conducted a bioinformatics analysis ( ). The results of the GO enrichment analysis of differential proteins are presented in the form of a bar chart, which displays the top 10 terms for cellular component (CC), biological process (BP), and molecular function (MF). In biological processes, JWXS could improve energy metabolism by regulating the tricarboxylic acid cycle, succinyl-CoA metabolic process, respiratory electron transport chain, and the regulation of cytochrome-c oxidase activity. The results of the KEGG enrichment analysis of differential proteins are also displayed as a bar chart, which shows the top 20 pathways ( ). The JWXS-regulated differential proteins are primarily associated with the Citrate cycle (TCA cycle), Glutathione metabolism, the calcium signaling pathway, and others. In the calcium signaling pathway, the sequencing results revealed that the main differential proteins were SERCA2, PLC-γ2, Vdac1, Vdac2, and Vdac3. Among these proteins, the abundance of PLC-γ2 significantly increased after JWXS treatment ( , p < 0.05), while the abundance of SERCA2, Vdac1, Vdac2, and Vdac3 markedly decreased ( p < 0.05).
The duodenum is a key region in the pathophysiology of FD. Disruption of the duodenal barrier has been demonstrated in FD patients, and the underlying damage-associated molecular pattern can enter the compromised intestinal barrier and trigger host innate immunity. To elucidate the therapeutic mechanism of JWXS on the duodenum of SD-FD rats, we conducted a DIA-based quantitative proteomic analysis on the duodenal tissues of three groups of rats. A total of 29682 peptides and 4421 proteins were detected using DIA-based quantitative proteomics. Volcano plots were used to visualize the distribution of differentially expressed proteins (DEPs) by applying filtering criteria of fold change (>1.5 for up-regulated or <0.67 for down-regulated) and p -value < 0.05 ( ). There was a total of 276 differentially expressed proteins (DEPs) identified between the Model and Control groups, of which 108 proteins were upregulated and 51 proteins were downregulated ( ). Meanwhile, a total of 732 DEPs (376 up-regulated and 277 down-regulated) were identified in the JWXS group compared to the Model group. The Venn plot showed that there were 64 overlapping DEPs found between the Model vs Control and the JWXS vs Model groups ( ). Cluster analysis showed significant differences between the model and control groups ( ). Additionally, the protein expression was significantly changed after JWXS intervention in SD-FD rats.
To better classify the functional properties of differential proteins, the DEPs were subjected to GO analysis ( ). In biological processes, JWXS could respond to immune activation via antigen processing and presentation of exogenous peptide antigen via MHC class II, negative regulation of hydrogen peroxide-induced cell death, positive regulation of T cell-mediated cytotoxicity, protection from natural killer cell-mediated cytotoxicity. Furthermore, we performed the KEGG pathway analysis of differentially expressed proteins ( ). The results indicated that the KEGG pathways enriched in the JWXS group mainly included antigen processing and presentation, staphylococcus aureus infection, oxidative phosphorylation, intestinal immune network for IgA production, among others. In the antigen processing and presentation and intestinal immune network for IgA production, the sequencing results revealed that the main differential proteins were PIgR, RT1-A1, RT1-CE1, MHC II, Tap1, and Tap2. Among these proteins, the abundance of PIgR significantly increased after JWXS treatment ( , p < 0.05), while the abundance of RT1-A1, RT1-CE1, MHC II, Tap1, and Tap2 markedly decreased ( p < 0.05).
Gastric motor dysfunction is considered the primary mechanism that causes symptoms of FD. Previous studies have shown that administering iodoacetamide to neonatal rats for gastric stimulation may result in gastric hypersensitivity and gastric motor dysfunction in adult rats. In TCM, the term “spleen” encompasses not only the anatomical spleen but also functions related to digestion and absorption, hematopoiesis, muscle growth, energy metabolism, and the regulation of brain-gut peptides and neurotransmitters. Spleen deficiency is a common clinical syndrome that involves the deterioration of digestion, absorption, lassitude, energy conversion, and immune system function. Under the guidance of the “fatigue hurts the spleen” theory in TCM, MMPM may induce spleen deficiency by causing fatigue. In our study, we observed weight loss, fatigue, inactivity, and dry hair in model rats, which are similar to the symptoms of spleen deficiency in TCM. The lower gastric emptying rate and intestinal propulsive rate of model rats, compared to healthy rats, suggest the successful construction of the FD model. In addition, SD-FD rats also had lower spleen and thymus weights than healthy rats, decreased WBC count and %Lym in blood, and increased %Neu, suggesting immune dysfunction. The administration of JWXS enhanced the aforementioned symptoms in SD-FD rats, implying an improvement in both GI motility as well as immune function. Gastrointestinal (GI) absorption and movement are regulated by GI hormones secreted by endocrine cells located within the intestines of the GI tract. Numerous studies have examined the impact of FD on GI hormones, highlighting their significance in this context. Motilin and gastrin promote smooth muscle contraction, increase GI motility, stimulate gastric acid secretion and pepsin secretion, and regulate food intake. , Meanwhile, ghrelin increases appetite, aids in gastric emptying, and protects GI mucosa. VIP, SST, and CCK serve as inhibitory neurotransmitters in both the central and enteric nervous systems, effectively reducing GI motility, gastric emptying, and gastric acid secretion. In comparison to healthy rats, SD-FD rats exhibited an increase in serum levels of SST, VIP, and CCK, and a decrease in the expression of motilin, gastrin, and ghrelin. After the administration of JWXS treatment, all of the aforementioned phenomena were almost completely restored to their normal levels. We hypothesize that JWXS enhances digestion by regulating the homeostasis of GI hormones. Proteomics offers a comprehensive understanding of protein composition, which facilitates the identification of signaling pathways for environmental stimuli and molecular mechanisms of drug action. The potential mechanism of JWXS in the treatment of SD-FD was further investigated using DIA-based proteomic analysis of gastric tissues. Interestingly, our proteomic analysis identified significant alterations in proteins related to the calcium signaling pathway in gastric tissues. These findings align with previous animal studies that have demonstrated the crucial role of intracellular Ca 2+ concentration ([Ca2 + ] i ) in regulating GI smooth muscle contraction. The process of elevated intracellular Ca 2+ concentration involves Ca 2+ entering from the extracellular into the intracellular and the release of Ca 2+ from intracellular stores. An increase in [Ca 2+ ] i initiates smooth muscle contraction, while a decrease in [Ca 2+ ] i leads to smooth muscle relaxation and delayed gastric emptying. When comparing the JWXS group with the Model group, we observed a decrease in the abundance of sarcoplasmic/endoplasmic reticulum calcium ATPase 2 (SERCA2) and an increase in the abundance of phospholipase C-γ (PLC-γ) in the JWXS group. SERCA2 is primarily located in the endoplasmic reticulum (ER) and is responsible for transporting Ca 2+ from the cytoplasm into the ER against a concentration gradient, requiring significant ATP consumption. On the other hand, the activation of PLC-γ leads to the hydrolysis of phosphatidylinositol 4.5-bisphosphate in the cell membrane, generating inositol trisphosphate (IP3) and diacylglycerol as secondary messengers. The released IP3 then binds to receptors on the ER membrane, triggering the release of intracellular stored Ca 2+ . These findings suggest that JWXS intervention in SD-FD rats may improve gastric emptying disorders by reducing SERCA2 expression, inhibiting Ca 2+ storage, and increasing PLC-γ expression thereby enhancing IP3-mediated Ca 2+ release and contributing further to the regulation of intracellular [Ca 2+ ] i . However, the mechanism by which JWXS improves gastric emptying delay by regulating calcium signaling pathways needs to be further verified. An increasing number of studies indicate that duodenal immune activation is crucially involved in the pathophysiology of FD. To examine the regulatory impact of JWXS on intestinal immunity, the current study utilized a proteomic approach to investigate the underlying molecular mechanism of JWXS in SD-FD rats. After analyzing the proteomic profile of the duodenum, we observed that the administration of JWXS has an impact on proteins involved in the antigen processing and presentation pathway, as well as the Intestinal immune network for IgA production pathway. Duodenal mucosal integrity impairment and low-grade inflammation are associated with systemic immune activation, which can ultimately lead to dyspeptic symptoms. Previous clinical studies have also indicated that inflammation of the duodenal mucosa and the presence of increased T-cells in the small intestine indicates intestinal inflammation, which is associated with delayed gastric emptying and the severity of dyspeptic symptoms. , In the present study, JWXS may play a therapeutic role by modulating the expression of two proteins, polymeric immunoglobulin receptor (pIgR) and major histocompatibility complex (MHC) class II antigen. PIgR is an immune factor produced on the surface of mucous membranes (such as intestinal tissues), which can combine with immunoglobulin IgA and transport it to the mucosal surface, leading to the formation of secretory IgA (SIgA). SIgA plays a crucial role in maintaining the immune function of the intestinal mucosa. One of its main functions is to prevent the adhesion of pathogens, thereby inhibiting their colonization and spread on the surface of the intestinal mucosa, and limiting infection. In the present study, the expression of pIgR was reduced in SD-FD rats, while treatment with JWXS restored pIgR expression to normal levels. MHC II is a significant group of immune system molecules primarily found on the surface of antigen-presenting cells. The elevated expression of MHC II in the intestine has been firmly linked to intestinal inflammation. The study results indicate that the expression of MHC II in the model rats was increased, and the administration of JWXS was able to reduce MHC II expression. However, there are some limitations to this study. For example, whether impaired intestinal immunity affects intestinal barrier function requires further study.
Our study suggests that JWXS may be a promising drug for the treatment of FD-SD, as it appears capable of improving gastric motility disorders and regulating immune function. We hypothesize that its mechanism of action involves the regulation of gastric smooth muscle contraction in FD-SD through calcium signaling pathways and the modulation of duodenal immune function. However, the precise mechanism by which JWXS exerts its effects remains unclear, necessitating further validation studies to explore specific targets and pathways.
|
Postgraduate medical education in obstetrics and gynaecology: Where are we now and what do we need for the future? A study on postgraduate training in obstetrics and gynaecology in Germany, Austria and Switzerland | daa785c1-9091-4ed5-8965-29ff413ca4bf | 9585411 | Gynaecology[mh] | High quality training is essential to become a specialist in obstetrics and gynaecology (OBGYN) and to ensure optimal patient care in the future. Every country in Europe defined specific goals and qualifications of training for graduate and postgraduate medical education (PGME) . Being individually defined and implemented in the countries, the curricula considerably differ in most parts of Europe, e.g. in the catalogue of requirements. Differences in training are often caused by differing medical infrastructure of most countries, diverging clinical responsibilities of subspecialties and the lack of protected timeslots for practical and theoretical training. For example, breast surgery which is mandatory content of training curricula in some countries, is not even being trained in others. In 2009, Rodriguez et al. found major differences in definition of training content and outcome among European trainees in OBGYN of every level of experience . The resulting need for harmonization of European training outcomes is reflected by the establishment of the pan-European curriculum for training in OBGYN by the European Board and College of Obstetrics and Gynaecology (EBCOG). The neighbouring German-speaking countries Germany, Austria and Switzerland vary substantially regarding their training curricula; however, they are highly comparable regarding medical infrastructure and health care systems. Therefore, these countries are ideal models to study the effects of different training curricula on educational outcomes and satisfaction of trainees. In this study, we aim to provide a representative picture of the current situation of training in OBGYN in Germany, Austria and Switzerland. Furthermore, we intend to determine transferable advantages of the different systems. We considered a survey to be the most appropriate method to get a comprehensive overview of the general situation and satisfaction of trainees, since it allows a large group of people to participate, even in case of limited time and financial resources and even if they are geographically diversified. Participants were enrolled anonymously. We performed the survey through a digital questionnaire with a total of 30 questions (see attachment 1 ). The questionnaire was designed by consensus of experts, all board members of a cooperation, including representatives of trainee networks of the German (DGGG), Swiss (SGGG) and Austrian (OEGGG) society for obstetrics and gynaecology. Advice was given by eight gynaecologists (FMW, KW, GB, MF, PF, MF, BK, MW) from Germany, Austria and Switzerland, who are partially involved in the development of national OBGYN curricula. The experts chose to focus on certain topics (such as simulation programs, training regulations or training in subspecialties) in order to provide a representative picture of the current situation of trainees but prevent a question overload of the survey. The questionnaire was set-up in German as online questionnaire allowing wide accessibility. The questions were presented sequentially. A new question was shown only if the preceding was answered. Thus, for completion of the survey the respondent had to answer all 30 questions. The survey was controlled and circulated by the Swiss Federal Institute of Technology in Zurich (ETH), Department for Health Sciences and Technology, Consumer Behavior and was advertised via communication channels of the societies. Data collection was performed from August to September 2020. Multiple participations by one person were excluded by an anonymous IP address check. The evaluation of data was carried out with SPSS 26.0 (IBM Corp.). Characteristics of participants A total of 422 trainees took part in the survey. Of these 209 (49.5%) were currently trained in Germany, 116 (27.5%) in Switzerland and 97 (23%) in Austria. 88.9% were female and the median age was 32 years. Around three-fourths of the study participants were in the third year of training or more. 42% worked at a large hospital and training facility with at least 500 beds. 77% of participants worked full-time. Part-time work and work sharing Although the largest number of participants worked full time, the majority (70%) rated part-time workloads between 80 and 95 % as the most attractive employment obligation. However, this did not seem to reflect a general desire to work less, as part-time work with less than 40% pensum was classified as “not attractive or not at all attractive” by 82% and 40-55% pensum were perceived as considerably less attractive by 45% of participants. Overall, part-time working models seemed to be widespread already, as 94% of the participants confirmed that their hospital offered some form of reduced pensum (see table 1 ). Workload Notably, we found distinct country-specific differences in the performance of medical procedures by non-medical healthcare professionals. Certain invasive interventions no longer need to be carried out by doctors only (see figure 1 ). For example, 98% of the trainees in Switzerland stated that they “never or rarely” insert an intravenous line, whereas 85% of German trainees “always to often” performed this intervention. Generally, the trainees described to spend the majority of their daily working hours on documentation. 27% perceived this as “not or not at all efficient” and only a quarter of the trainees indicated to have an assistant available (e.g. ward assistant), supporting them in documentation and organization tasks. Training of sub-specialties in OBGYN Regarding the training of specific sub-areas in OBGYN, 76% of participants rated obstetrics as “well to very well” represented in all participating countries (see figure 2 ). Major differences between the countries were observed in gynaecological oncology and senology as well as in prenatal care. Interestingly, only 5% of participants stated Endocrinology being “well or very well” covered within their curricula. Paediatric Gynaecology, as well as Sexual Health and Reproductive Medicine were seen as underrepresented sub-areas in OBGYN training by a vast majority of participants (see attachment 2 ). Assessment and implementation of current training regulations Interestingly, only 22% of participants felt “well to very well” prepared for their work as a specialist in the hospital setting and only 11% felt “well to very well” prepared for working in private practice. The majority of trainees assessed themselves only “moderately prepared” for the further work in a hospital (66%) or private practice (57%). In addition, this was associated with the fact that only 47% of trainees stated that they regularly fulfilled the required obligatory numbers of self-performed interventions (see figure 3 ). 53% of participants stated that they were actually faced with notable to serious difficulties to fulfil the required obligatory numbers of self-performed interventions being documented. As a result, two-thirds of the participants agreed that an electronically kept logbook is useful when documenting the obligatory interventions. However, the survey showed substantial, country-specific differences in the actual implementation of an electronically kept logbook. Whereas 86% of the participants from Switzerland answered that their logbook is kept electronically, only 5% from Germany and 1% from Austria currently kept their logbook electronically. Furthermore, an annual evaluation interview being documented in written form was offered to only 48% of participants. 54% of the trainees stated that they had a supervisor giving advice on questions with medical content or about career planning. Confidence during intervention To determine if the lack of self-performed interventions and the corresponding documentation have an impact on the feeling of security among the trainees, participants were asked how confident they feel in standard situations and interventions in OBGYN. Thereby, the intrinsic feeling of safety when they perform standard surgery is an important parameter for determining the quality of training. Interestingly, around two-thirds of trainees felt “confident to very confident” during standard interventions like curettage, Caesarean section and hysteroscopy. Among other interventions like simple laparoscopy and vacuum extraction as well as the management of emergencies in obstetrics like postpartum bleeding, or shoulder dystocia more than half of participants felt only "moderately" confident. When it comes to rare situations like breech birth or forceps delivery, most of the participants felt “not” or “not at all” confident (see attachment 3 ). Considering the years of specialty training, the feeling of security among frequently performed interventions such as the curettage, hysteroscopy and Caesarean section increased over time. However, divided into groups with and without a simulation training in obstetrics (44% with simulation training) or gynaecology (20%) offered in their hospital there was a noticeable increase of safety among the trainees who could use simulation training for their further education. Particularly, in these interventions (simple laparoscopy, management of postpartum bleeding or shoulder dystocia and vacuum extraction) that can arise in a hospital at any time, the trainees with simulation training feel up to 12% more confident than those participants without (see figure 4 ; box with broken line: noticeable differences between “with” and “without” simulation training). A total of 422 trainees took part in the survey. Of these 209 (49.5%) were currently trained in Germany, 116 (27.5%) in Switzerland and 97 (23%) in Austria. 88.9% were female and the median age was 32 years. Around three-fourths of the study participants were in the third year of training or more. 42% worked at a large hospital and training facility with at least 500 beds. 77% of participants worked full-time. Part-time work and work sharing Although the largest number of participants worked full time, the majority (70%) rated part-time workloads between 80 and 95 % as the most attractive employment obligation. However, this did not seem to reflect a general desire to work less, as part-time work with less than 40% pensum was classified as “not attractive or not at all attractive” by 82% and 40-55% pensum were perceived as considerably less attractive by 45% of participants. Overall, part-time working models seemed to be widespread already, as 94% of the participants confirmed that their hospital offered some form of reduced pensum (see table 1 ). Workload Notably, we found distinct country-specific differences in the performance of medical procedures by non-medical healthcare professionals. Certain invasive interventions no longer need to be carried out by doctors only (see figure 1 ). For example, 98% of the trainees in Switzerland stated that they “never or rarely” insert an intravenous line, whereas 85% of German trainees “always to often” performed this intervention. Generally, the trainees described to spend the majority of their daily working hours on documentation. 27% perceived this as “not or not at all efficient” and only a quarter of the trainees indicated to have an assistant available (e.g. ward assistant), supporting them in documentation and organization tasks. Training of sub-specialties in OBGYN Regarding the training of specific sub-areas in OBGYN, 76% of participants rated obstetrics as “well to very well” represented in all participating countries (see figure 2 ). Major differences between the countries were observed in gynaecological oncology and senology as well as in prenatal care. Interestingly, only 5% of participants stated Endocrinology being “well or very well” covered within their curricula. Paediatric Gynaecology, as well as Sexual Health and Reproductive Medicine were seen as underrepresented sub-areas in OBGYN training by a vast majority of participants (see attachment 2 ). Assessment and implementation of current training regulations Interestingly, only 22% of participants felt “well to very well” prepared for their work as a specialist in the hospital setting and only 11% felt “well to very well” prepared for working in private practice. The majority of trainees assessed themselves only “moderately prepared” for the further work in a hospital (66%) or private practice (57%). In addition, this was associated with the fact that only 47% of trainees stated that they regularly fulfilled the required obligatory numbers of self-performed interventions (see figure 3 ). 53% of participants stated that they were actually faced with notable to serious difficulties to fulfil the required obligatory numbers of self-performed interventions being documented. As a result, two-thirds of the participants agreed that an electronically kept logbook is useful when documenting the obligatory interventions. However, the survey showed substantial, country-specific differences in the actual implementation of an electronically kept logbook. Whereas 86% of the participants from Switzerland answered that their logbook is kept electronically, only 5% from Germany and 1% from Austria currently kept their logbook electronically. Furthermore, an annual evaluation interview being documented in written form was offered to only 48% of participants. 54% of the trainees stated that they had a supervisor giving advice on questions with medical content or about career planning. Confidence during intervention To determine if the lack of self-performed interventions and the corresponding documentation have an impact on the feeling of security among the trainees, participants were asked how confident they feel in standard situations and interventions in OBGYN. Thereby, the intrinsic feeling of safety when they perform standard surgery is an important parameter for determining the quality of training. Interestingly, around two-thirds of trainees felt “confident to very confident” during standard interventions like curettage, Caesarean section and hysteroscopy. Among other interventions like simple laparoscopy and vacuum extraction as well as the management of emergencies in obstetrics like postpartum bleeding, or shoulder dystocia more than half of participants felt only "moderately" confident. When it comes to rare situations like breech birth or forceps delivery, most of the participants felt “not” or “not at all” confident (see attachment 3 ). Considering the years of specialty training, the feeling of security among frequently performed interventions such as the curettage, hysteroscopy and Caesarean section increased over time. However, divided into groups with and without a simulation training in obstetrics (44% with simulation training) or gynaecology (20%) offered in their hospital there was a noticeable increase of safety among the trainees who could use simulation training for their further education. Particularly, in these interventions (simple laparoscopy, management of postpartum bleeding or shoulder dystocia and vacuum extraction) that can arise in a hospital at any time, the trainees with simulation training feel up to 12% more confident than those participants without (see figure 4 ; box with broken line: noticeable differences between “with” and “without” simulation training). Although the largest number of participants worked full time, the majority (70%) rated part-time workloads between 80 and 95 % as the most attractive employment obligation. However, this did not seem to reflect a general desire to work less, as part-time work with less than 40% pensum was classified as “not attractive or not at all attractive” by 82% and 40-55% pensum were perceived as considerably less attractive by 45% of participants. Overall, part-time working models seemed to be widespread already, as 94% of the participants confirmed that their hospital offered some form of reduced pensum (see table 1 ). Notably, we found distinct country-specific differences in the performance of medical procedures by non-medical healthcare professionals. Certain invasive interventions no longer need to be carried out by doctors only (see figure 1 ). For example, 98% of the trainees in Switzerland stated that they “never or rarely” insert an intravenous line, whereas 85% of German trainees “always to often” performed this intervention. Generally, the trainees described to spend the majority of their daily working hours on documentation. 27% perceived this as “not or not at all efficient” and only a quarter of the trainees indicated to have an assistant available (e.g. ward assistant), supporting them in documentation and organization tasks. Regarding the training of specific sub-areas in OBGYN, 76% of participants rated obstetrics as “well to very well” represented in all participating countries (see figure 2 ). Major differences between the countries were observed in gynaecological oncology and senology as well as in prenatal care. Interestingly, only 5% of participants stated Endocrinology being “well or very well” covered within their curricula. Paediatric Gynaecology, as well as Sexual Health and Reproductive Medicine were seen as underrepresented sub-areas in OBGYN training by a vast majority of participants (see attachment 2 ). Interestingly, only 22% of participants felt “well to very well” prepared for their work as a specialist in the hospital setting and only 11% felt “well to very well” prepared for working in private practice. The majority of trainees assessed themselves only “moderately prepared” for the further work in a hospital (66%) or private practice (57%). In addition, this was associated with the fact that only 47% of trainees stated that they regularly fulfilled the required obligatory numbers of self-performed interventions (see figure 3 ). 53% of participants stated that they were actually faced with notable to serious difficulties to fulfil the required obligatory numbers of self-performed interventions being documented. As a result, two-thirds of the participants agreed that an electronically kept logbook is useful when documenting the obligatory interventions. However, the survey showed substantial, country-specific differences in the actual implementation of an electronically kept logbook. Whereas 86% of the participants from Switzerland answered that their logbook is kept electronically, only 5% from Germany and 1% from Austria currently kept their logbook electronically. Furthermore, an annual evaluation interview being documented in written form was offered to only 48% of participants. 54% of the trainees stated that they had a supervisor giving advice on questions with medical content or about career planning. To determine if the lack of self-performed interventions and the corresponding documentation have an impact on the feeling of security among the trainees, participants were asked how confident they feel in standard situations and interventions in OBGYN. Thereby, the intrinsic feeling of safety when they perform standard surgery is an important parameter for determining the quality of training. Interestingly, around two-thirds of trainees felt “confident to very confident” during standard interventions like curettage, Caesarean section and hysteroscopy. Among other interventions like simple laparoscopy and vacuum extraction as well as the management of emergencies in obstetrics like postpartum bleeding, or shoulder dystocia more than half of participants felt only "moderately" confident. When it comes to rare situations like breech birth or forceps delivery, most of the participants felt “not” or “not at all” confident (see attachment 3 ). Considering the years of specialty training, the feeling of security among frequently performed interventions such as the curettage, hysteroscopy and Caesarean section increased over time. However, divided into groups with and without a simulation training in obstetrics (44% with simulation training) or gynaecology (20%) offered in their hospital there was a noticeable increase of safety among the trainees who could use simulation training for their further education. Particularly, in these interventions (simple laparoscopy, management of postpartum bleeding or shoulder dystocia and vacuum extraction) that can arise in a hospital at any time, the trainees with simulation training feel up to 12% more confident than those participants without (see figure 4 ; box with broken line: noticeable differences between “with” and “without” simulation training). This study aimed to identify the current situation of OBGYN training as well as transferable advantages of the different training systems in Germany, Austria and Switzerland. Anonymous surveys on “customer satisfaction” with their training and work situation, but also the assessment of the heads of the facilities are centrally recorded in some countries. Since 1996, the Swiss Institute for Medical Training (SIWF) has been carried out an annual survey among Swiss trainees which has served as a model for the current international survey. For decades, the results of the SIWF survey have served as feedback in order to recognize and promote successful concepts or to promptly uncover weak issues. Annually, the results of the survey are published online, and thus, offer young doctors an assessment basis for choosing an attractive employment and training position. At the same time, the data provide an annual benchmarking of the institutional training quality . Cross-border cooperation offers a great opportunity to learn and benefit from other training systems. PGME is teamwork that requires shared commitment to innovation, shared responsibility, supportive frameworks, and a teaching culture . Besides a few regional and interdisciplinary evaluation projects , , to our best knowledge this survey is the first cross-border project with special focus on OBGYN as specific subject area. This data serves as a valuable basis for further research and development in the field of supranational PGME in OBGYN. To overcome country-specific differences in training there has been great effort to harmonize training standards in OBGYN. The tendency of pan-European harmonization is the result of the increasing mobility of medical specialists and patients and the need for quality assurance of training throughout Europe , , . However, within the European Union, all countries have mutually recognized training qualifications for graduate and postgraduate medical education. This mutual recognition mostly is not content-related but based on minimum requirements, including training sites (recognized teaching hospitals) and duration of training , , . A push in the direction of a common, harmonized, European curriculum for advanced training in OBGYN is the Project for Achieving Consensus in Training (PACT) of the European Board and College of Obstetrics and Gynaecology . The curriculum defines content and competencies during a three-year basic training course (so-called “core”), which is the same for every gynaecologist. This is followed by a two-year advanced training phase with elective modules that can be chosen depending on the desired profile (so-called “electives”). Therefore, the EBCOG PACT offers sufficient opportunities to overcome country-specific burdens of PGME. Within our study group, Austria has already implemented the EBCOG PACT structure in OBGYN PGME. An important step towards improvement and maintenance of high quality PGME is to distribute the limited time resources as best as possible and to restructure non-medical tasks or to evaluate bureaucratic processes . According to a recent study by Trezzini et al., a trainee spends 167 minutes per day on documenting patient records. This corresponds to 27% of their working time. Instead of reduction, the medical bureaucracy has substantially increased in recent years . Also, within the present survey, the participants complained that bureaucracy takes up a large part of their everyday working time. Medical documentation and organization of standard procedures are seen to be inefficient and do drastically reduce the satisfaction of trainees . If additional tasks as venipuncture or insertion of an intravenous line are performed by trainees, this has major impact on patient care and quality of PGME. Simple tools such as digitized Dictaphones with voice recognition, but also major structural changes such as clinical nurses or physician assistants assuming tasks of clinical routine could considerably relieve the workload and increase time dedicated to PGME. In light of increasing workload and bureaucracy, it is not surprising that required obligatory numbers of interventions can barely be fulfilled during the standard length of training. However, instead of adjusting numbers or structure, the results suggest that missing interventions were subsequently attested and documented. Are the required numbers of different interventions too high for the existing number of cases? Can certain interventions and diagnostic measures only be carried out in specialized centres? Is the current routine clinical workload and bureaucratic effort incomparable to former generations of trainees? Comparing the logbooks of the three countries regarding the required number of interventions and diagnostic measures, serious differences are detectable. Whereas in Austria and Switzerland a total of 85 and 80 obstetric interventions performed by the trainee are required, respectively, German trainees only need 25 Caesarean sections and “contribution” in further obstetric interventions. 50 colposcopies are required in Germany and Switzerland, however, only 20 in Austria. A total of 275 gynaecological surgeries are required in Austria, 255 in Switzerland, and only 200 in Germany. We have to strike out new paths in order to ensure that not numbers, but practical verifications attest the level of training in OBGYN. So-called “Entrustable Professional Activities” (EPAs) can support the relationship between trainer and trainee being part of competence-based medical education. An EPA is a detailed description of a medical activity, e.g. a Caesarean section, which combines the knowledge, skills and attitudes required for this procedure. In countries such as the Netherlands and Canada, EPAs have already found their way into continuing medical education in various specialties ( Einleitung, . They support the change in PGME, away from an “on-off knowledge-based” examination at the end of training to a modern, practice-adapted and competence-oriented training concept. Trainees receive timely feedback on their activities and annual goals can be defined and evaluated. EPAs thus also form a valuable basis for annual evaluation meetings. An electronically kept logbook is also indispensable for recording such advanced training competencies and target-oriented evaluation discussions. In addition to written documentation, the digital form also enables a timely evaluation of the level of training and should be an essential part of a modern training program. Simulation training covers a wide range of training opportunities from high-tech team simulation training and skill drills to low fidelity training units. Each of these methods has its justification, as they train completely different abilities. While the team simulation training, which is increasingly established in obstetrics, is primarily about consolidation and training of treatment coordination, low fidelity models help to understand concepts based on simple technical repetitions. Although, simulation units are often accompanied by high costs, the present survey illustrates the positive impact of this additional training. In our survey, trainees who confirmed participation in any type of simulation training felt more confident especially in situations and interventions that are part of the basic training of OBGYN such as simple laparoscopy or postpartum bleeding. Simulation training offers a great benefit for modern PGME by increasing the efficiency in gaining experience and thus, improving the patients' safety. Like the SIWF survey, this study can not cover all aspects of the current training situation in the three countries. The study group chose to focus on certain topics that shape the daily worklife of trainees. The selection of topics and data was done to our best knowledge but is a limitation to the study. Certainly, further studies that cover more aspects of the basic training in OBGYN are needed to create a broader picture of the current situation of training in OBGYN in Germany, Austria and Switzerland. The current postgraduate training for OBGYN is already at a very high level in Germany, Austria and Switzerland. The aim is to jointly further develop this advanced training to be future-oriented. With the help of this survey, current weak points can be identified. Projects and ideas such as EBCOG PACT, EPAs, the reduction of bureaucracy through digitization and deepening skills through simulation training make a valuable contribution to compensate for these deficits and to adapt to future requirements. In this way, it is possible to secure the high level of European postgraduate training in OBGYN for future generations. We thank Larissa Luchsinger and Jeanine Ammann, research assistant at ETH Zurich, for the helpful evaluations, discussions and additions to this survey. We thank everyone who participated in the study. The study was funded with support of the DGGG, OEGGG and SGGG, however, the societies had no further involvement except for the financial support. The authors declare that they have no competing interests. Supplementary material – original ouestionnaire Sub-specialties represented in PGME in OBGYN in Germany, Austria and Switzerland Intrinsic feeling of safety among trainees during standard situations and interventions |
The gender-affirming model of care is incompatible with competent, ethical medical practice | 52fdfea8-44b1-4f0a-b7cb-d41bbdfc3967 | 11103900 | Psychiatry[mh] | The nature of gender diversity is unclear because the terms and concepts used to understand it continually change. , Drescher summarised the history, starting with the mid-20 th century diagnosis transsexualism, a form of sexual deviance associated with homosexuality, defined as living as a member of the opposite to one’s biological sex. The subsequent variety of presentations indicated that sexuality and preferred gender were substantially independent, leading to the replacement of transsexualism by the diagnosis gender identity disorder in the late 1980s. The gender diverse community welcomed the separation of gender identity from sexual deviance, alongside the reconceptualisation of homosexuality itself as a healthy form of human behaviour. However, many interpreted the introduction of gender identity disorders into the DSM-III and ICD-10 as a pathologization of their sense of self. This triggered the depathologization movement which continues to apply pressure to the American Psychiatric Association and World Health Organization to remove all gender diversity diagnoses from the DSM and ICD. The movement was instrumental in changing gender identity disorder to gender dysphoria in DSM-5, and categorising gender incongruence as a form of sexual health condition rather than a mental disorder in ICD-11. , The influence of activists is concerning given that both categories are based on clinician consensus rather than empirical evidence, due to the small number of patients involved.
The driving principle of the GAMOC is that health care professionals cannot assess but must affirm patient-reported gender identity. , The emergence of non-binary and fluid genders means there are no boundaries to self-reported gender identity, which may include a gender consistent with one of the two biological sexes; a combination of features consistent with both sexes; the absence of features of gender; an identity as a voluntarily/involuntarily castrated eunuch; or arbitrary and rapidly changing variations. , , , The principle of unquestioning affirmation of gender identity critically relies upon the assumption that pathology plays no part in the development of gender diversity. If it is admitted there are some pathological causes of gender diversity, then it becomes necessary to assess the health or illness of all presentations. Despite the existential reliance of the GAMOC upon this assumption, it has never been tested, or even questioned, by GAMOC advocates. The World Professional Association for Transgender Health (WPATH) endorses the leading international standards of care for treatment of gender diverse patients. They assert that ‘[g]ender diversity is a natural variation in people and is not inherently pathological’ (pS34). However, no evidence is presented and the supporting reference leads back through the previous version of the Guidelines to a statement by the WPATH Board of Directors. , Not only do the guidelines rely on a circular reference to an evidence-free assertion of this core assumption of their model, GAMOC advocates reject the possibility of testing the model using randomised control trials as unethical.
From a psychiatric perspective, the proposition that psychopathology plays no role in gender diversity is absurd. The most detailed personal description of the experiences of psychosis is that of Daniel Paul Schreber, a German judge who minutely described his belief that God had turned him into a woman and was sending ribbons from the sun through his body to impregnate him and repopulate the earth. It is difficult to imagine a more pathological aetiology for gender diversity, yet the GAMOC provides no framework for assessing such a patient, and does not view Schreber’s case as an absolute contraindication to social, medical, or surgical transition. , , While GAMOC advocates have argued transition is safe in patients with psychosis because it is easy to differentiate psychotic from non-psychotic aetiologies of gender diversity, they have provided no guidance on how to do so, and no empirical evidence that it is safe to try. To the extent they discuss the role of psychosis or severe personality pathology in the development of gender diversity at all it is only to deny that either might prevent transition. , , , The WPATH standards acknowledge the small evidence base on differentiating psychotic from non-psychotic aetiologies of gender diversity, comprised entirely of case reports. Their main reference on the topic noted that of 19 previously published cases 16 had been judged psychotic in the absence of gender dysphoria, with 4 of these nonetheless treated with hormones or surgery and suffering harm as a result. Another review indicated that up to 6% of patients with gender dysphoria had a comorbid psychotic disorder, and listed a number of case studies where antipsychotic treatment was associated with a reduction or resolution of gender dysphoria. Despite this, the WPATH standards appear more concerned that comorbid psychosis might prevent gender diverse patients from accessing the GAMOC than that patients with psychosis might be harmed by the affirmation of psychotic beliefs. Close reading of GAMOC guidelines reveals a fatal deficiency. The guidelines assert that the experience of a gender identity that is different from biological sex is in all cases a healthy variant of normal, , but they do nothing to explain the nature or variance of the experience of gender identity, what it means for gender identity to be different from biological sex, or healthy and pathological variations. This complete failure to describe the phenomenology and psychopathology of gender diversity makes it impossible for the guidelines to meaningfully describe what gender diversity is, or to demonstrate that it does not involve pathology.
Gender identity is a concept describing a type of human experience. It can only be understood by applying the clinical skills of psychiatry with knowledge of phenomenology. As Schreber illustrates, it is certain that pathology causes some cases of gender diversity. Differentiating between healthy and pathological gender diversity, or, more likely, gauging the relative contribution of healthy and pathological processes originating within or in the environment of each patient, can only be achieved by the comparison of an individual’s patterns of behaviour with patterns of normal and pathological development. Phenomenology and psychopathology are core competencies of psychiatric practice, and of no other medical specialty, yet the GAMOC guidelines are designed to exclude psychiatric skills and knowledge. As should be clear from the foregoing discussion, the reason is that the GAMOC’s core clinical principle of unquestioning gender-affirmation, and the core assumption on which it relies – that gender diversity by definition is never caused by endogenous pathology – are both incompatible with competent and ethical psychiatric practice. Thus, it is misleading to think of the GAMOC guidelines as primarily clinical documents. In place of medical diagnosis, they assert a political right designed to expand the boundaries of personal liberty: the right to define a gender identity. In the current formulation of the GAMOC, this is an absolute right with no fixed definition and no constraints. , Self-defined gender identity does not have to be coherent, persistent, or intelligible to a healthcare provider or the average citizen. Traditional guidelines, such as the RANZCP guidelines for the treatment of mood disorder, outline a process of clinical reasoning which matches diagnoses to treatments informed by patient preferences based on risk-benefit analyses. GAMOC guidelines abandon the clinical discipline of diagnosis and make treatment contingent upon the unconstrained subjective experiences of children and potentially disturbed adults. This is unethical, because modern medicine relies upon accurate diagnosis and evidence-based clinical reasoning to ensure that treatment is likely to help and not harm patients. The depathologization movement raises homosexuality as a model of the potential social goods and lack of harms that can be achieved by eliminating a stigmatising diagnosis. However, as Meyer points out, homosexuality was only redefined after a debate where ‘we as a society and as scientists agree [on what] are abnormal behaviours, cognitions, and emotions’ allowing for the emergence of ‘a scientific and social consensus’ (p675). No such debate has been started, and no such consensus yet exists for gender identity. Gender-affirming care is fundamentally incompatible with competent, ethical medical practice. It predicates a class of experiences which diverge from those of the vast majority of human beings, but refuses to describe normal experience or the patterns of divergence. It assumes there are no pathological aetiologies of gender diversity and protects this assumption by forbidding the assessment of pathology in individual patients, and by forbidding the evaluation of treatment outcomes by RCTs.
While the RANZCP initially endorsed the GAMOC, in Position Statement 103 (PS103), it removed this endorsement without explanation, indicating that while some patients prefer affirmation, the evidence about the benefits and harms of providing or withholding GAMOC does not justify its recommendations. PS103 does not provide any evidence or rationale for its statement that ‘Being Trans or Gender Diverse does not represent a mental health condition’. In essence, the RANZCP advises psychiatrists to be aware that the GAMOC exists, but to provide appropriate patient-centred, evidence-based psychiatric care for mental health conditions as if it did not. This appears to be a pragmatic compromise that allows PS103 to avoid a more critical position on the GAMOC by limiting its scope to the treatment of mental illness. Apart from the untested and otherwise undeveloped assertion that gender diversity ‘does not represent a mental health condition’, PS103 is entirely consistent with the arguments made above. Although it is clear that this compromise balances the concerns of different stakeholders, the medicolegal implications for psychiatrists and their patients may be too important to long defer a conclusive position on the aetiological role of mental illness in gender diversity. For example, the lack of evidence for the GAMOC has led one insurer to restrict reimbursement for private practitioners treating gender dysphoria. In addition, the courts have relied upon medical college positions to assume that GAMOC is the accepted standard of care for gender diversity in Australia. Given these stakes, the RANZCP should either provide the evidence and rationale for the position that mental illness plays no aetiological role in gender diversity, or acknowledge that it does play a role in some or all cases and facilitate the phenomenological and psychopathological understanding necessary for safe and ethical treatment.
In the absence of models of the phenomenology and psychopathology of gender diversity, it is impossible to meaningfully judge what proportion of cases involves pathology or assess the role of pathology in individual patients. Unquestioning gender-affirming care is therefore unable to exclude the possibility that it is reinforcing the pathologies of some, most, or all of its patients. This is unethical, and it is the responsibility of psychiatrists to ensure that no patients are harmed by this dangerous model of care.
|
Lectin histochemistry of posterior lingual glands of developing rats | 30a60b3e-26a4-4ced-ae5b-a3950ed85306 | 10293267 | Anatomy[mh] | The oral cavity is part of the digestive system whose main function is to aid food intake. The salivary glands produce and secrete saliva, which keeps the oral cavity moist. Lingual gland’s saliva is an important growth factor for taste receptor cells . The tongue has two minor salivary glands, the anterior lingual and post lingual . The anterior lingual gland, Blandin-Nühn's gland, is a mixed gland located on the inferior surface of the tongue apex , . In humans, Weber's glands open onto the tonsils of the tongue , , , whereas in rats, Weber’s glands open onto the dorsal epithelium at the back of the tongue, and are involved in food mass formation and swallowing , . Weber's glands are muciparous; however, the presence of serous cells has been suggested in humans and rats , . The von Ebner’s glands, which open at the base of the sulcus of the circumvallate papilla and papillae foliate, are serous , , and secrete saliva and wash out the taste pits of taste buds in the papillary sulcus epithelium, thereby renewing and maintaining taste receptor function , – . They also produce tongue lipase, which hydrolyzes triacylglycerols in the acidic gastric lumen and aids the first step in dietary fat digestion , . In rats, the development of the minor salivary glands occurs in the late embryonic period , whereas the development of major salivary glands begins in the parotid gland on embryonic day 14 , , , in the submandibular gland on embryonic day 13 , and in the sublingual gland approximately on embryonic day 18 . The posterior lingual glands are believed to be involved in the renewal of taste receptors through salivation and washing of the taste pits of the taste buds with saliva. Lectins are often used as markers for epithelial and mesenchymal cells to identify specific cell populations because of their ability to bind specifically to glycohydrate epitopes on the cell membrane. By searching for the binding mode of lectins, the localization of glycoconjugates in different tissues and the characteristics of cellular glycans in each tissue can be elucidated. Lectin histochemical studies have been reported on the distribution of lectins in various species. In summary, the previous reports indicate that serous cells of salivary glands show binding to Ulex europeus agglutinin-I (UEAI), Soybean Glycine maximus agglutinin (SBA), Peanut ( Arachis hypogaea ) agglutinin (PNA) and Wheat germ ( Triticum vulgaris ) agglutinin (WGA), while mucous cells tend to show binding to Ricinus communis agglutinin (RCA) and Horse gram ( Dolichos biflorus ) agglutinin (DBA) – . From the 21 lectins used in our laboratory and potentially binding to salivary glands, we found that SBA, PNA, and WGA bind to serous cells and DBA binds to mucous cells in adult rats. However, although the localization and site differences of these lectin-binding patterns have been clarified, the development of the rat posterior lingual glands and changes in lectin-binding patterns have not been clarified. In this study, we examined the development of the rat posterior lingual glands and lectin-binding patterns using SBA, PNA, WGA, and DBA as indices, with the aim of clarifying the relationship between the development of the posterior lingual gland and the renewal of taste reception among its functions. The following results were obtained. Weber’s glands and von Ebner glands were fully developed by postnatal day 21. Mucous cells of the Weber’s glands were shown to bind DBA in adult rats, and SBA, WGA, PNA and DBA in the early developmental stage. Serous cells of von Ebner’s glands were shown to bind SBA, WGA, and PNA in adult rats, and DBA bound to the cellular membrane during early development but disappeared during development. Matured taste buds were found on the circumvallate papilla on postnatal day 1. The present findings indicate that the maturation of posterior lingual glands is closely associated with changes in food habit, and that the Weber’s gland functions as a serous gland in the early postnatal stage, while the von Ebner’s gland is not yet mature. The lectin-binding properties suggest the presence of Galβ (1,3) > Galβ(1,4) > Gal, αGalNAc > αGal > βGalNAc, and NeuAc > (GalNAc) 2–3 >>>GlcNAc sugar residues in serous cells and GalNAcα(1,3) sugar residues in mucous cells (Table ). The results obtained provide data for the functional characterization of the rat posterior lingual gland.
Hematoxylin and Eosin (H&E) staining In adult rats, von Ebner's glands were found to be located just below the circumvallate papilla, separated on both sides by the lingual septum, clumped within the muscularis, and with the opening at the base of the circumvallate papilla sulcus (Fig. a). The Weber's glands were located laterally and posteriorly to the von Ebner’s glands, opening directly onto the dorsal surface of the tongue (Fig. a). On embryonic day 18, the sulcus epithelium of the circumvallate papilla had not yet been inserted into the mesenchyme. Serial sections showed clusters of epithelial cells in the lateral portion of the posterior tongue (Fig. b) which were continuous with the surface layer of the tongue epithelium. On embryonic day 20, we observed an infiltration of the sulcus epithelium of the circumvallate papilla into the mesenchyme, with no evident formation of the von Ebner’s glands. The Weber's glands were also found in the mesenchyme in high number (Fig. c) with numerous clusters of epithelial cells present in the mesenchyme at 1 day of age (Fig. d), and their number and size increased further at 3 days of age (Fig. e). Thereafter, the cell population of epithelial cells in the mesenchyme increased over time, and both the Weber’s and von Ebner’s glands were fully developed by 28 days of age (Fig. f). The clusters of epithelial cells observed at 18 days of age were formed by cells with darker cytoplasm (Fig. a). On embryonic day 20, the number of cells forming these cell masses increased, and the cytoplasm of these cells were similarly dark (Fig. b). No apparent formation of terminal cells of the serous glands at the tip of the sulcus epithelium of the circumvallate papilla, which were plunging into the mesenchyme on embryonic day 18 was observed (Fig. c). At 1 day of age, the terminal part of Weber's glands showed characteristics of mucous cells, with some cells having a light-toned cytoplasm as observed in adult rats, and others having dark-toned cytoplasm (Fig. d). Duct and epithelial mass were observed from the base of the papillae sulcus, some of which showed serous cell morphology and the formation of von Ebner’s glands (Fig. e). Some of the Weber’s glands in rats aged 1 day showed flat nuclei in the clear-toned cytoplasm, and the number of these cells increased at 3 days of age. Serous demilunes were also observed (Fig. f). von Ebner’s glands showed tubular arrangements of terminal cells with round nuclei on the basal side of the dark-toned cytoplasm (Fig. g). At 7 days of age, Weber's glands showed mucous cells, numerous serous cells, and a serous demilunes, whereas von Ebner’s glands demonstrated advanced branching of the terminal part of the gland . Furthermore, Weber's glands showed a marked increase in mucous cells with age. At 21–28 days of age, the observations were similar to those noted in adult rats, and the cytoplasm of the terminal cells were transparent and filled with mucus secretions. The nuclei were oval and located basolaterally. Weber's glands contained mucous cells and serous cells with dark-toned cytoplasm (Fig. h). The von Ebner’s glands showed increased terminal areas and branching, with triangular terminal cells surrounding the gland lumen, as seen in typical exocrine glands. In addition, round nuclei were located slightly basolateral to the basophilic cytoplasm (Fig. i). In the epithelium of the circumvallate papilla sulcus, taste bud primordia were present on embryonic day 18; however, no taste pits were observed (Fig. a). On postnatal day 1, taste buds with taste pits were observed (Fig. b), and thereafter the number of taste buds with taste pits increased (Fig. c). Lectin histochemistry Among the lectins that have been reported to be used in salivary glands, in this study, we tested their binding in serous cells and mucous cells in the posterior lingual gland of adult rats. Three or more tissue fragments from each sample were validated with 21 different lectins. The binding of the lectins was classified into four levels: strong (3), moderate intensity (2), weak intensity (1), and unreactive (0). Different binding intensities were denoted by minimum and maximum values. The lectin that bind stongly to serous cells and unreactively to mucous cells are PNA, SBA, and WGA, which were used as markers for serous cells. The lectin that binds stongly to mucus cells and binds unreactively to mucus cells was DBA, which was designated as a marker for mucus cells. These 4 lectins were excellent markers for serous and mucous cells in adult animals, and were used as indicators of mature glandular cells and for serous and mucous cells during rat development (Table ). Abbreviations: Con A, ( Canavalia ensiformis ) agglutinin; SBA, Soybean ( Glycine maximus ) agglutinin; WGA, Wheat germ ( Triticum vulgaris ) agglutinin; DBA, Horse gram ( Dolichos biflorus ) agglutinin; UEA-I, ( Ulex europaeus ) agglutinin-I; RCA-I, ( Ricinus communis ) agglutinin-I; PNA, peanut ( Arachis hypogaea ) agglutinin; GS-I, ( Griffonia simplicifolia ) Lectin I; pea, ( Pisum sativum ); LCA, ( Lens culinaris ) agglutinin; PHA-E, ( Phaseolus vulgaris ) erythroagglutinin; PHA-L, ( Phaseolus vulgaris ) leucoagglutinin; SJA, ( Sophora japonicum ) agglutinin, sWGA, Succinyl ( Triticum vulgaris ) agglutinin; GS-II , ( Griffonia simplicifolia ) Lectin II; DSA, Thorn apple ( Datura stramonium ) agglutinin; ECL ( Erythrina cristagalli ) Lectin; Jacalin, ( Artocarpus integrifolia ); LEL, ( Lycopersicon esculentum ) Lectin; STL, ( Solanum tuberosum ) Lectin; VVA, ( Vicia villosa ) agglutinin. Lectin binding in the Weber’s glands SBA binding to the cellular membrane was observed, albeit weakly, in cell masses on embryonic day 18 (Fig. a), and some SBA binding was noted in the cytoplasm on postnatal day 1 (Fig. b). At postnatal day7, SBA bound to the cytoplasm and cellular membrane, and at postnatal day14, SBA showed a similar binding pattern and bound to the serous demilunes (Fig. c). However, by postnatal day 21, the binding observed in the cytoplasm and cellular membrane was minimal (Fig. d). In the postnatal period, SBA binding was also strong in saliva. WGA bound to the cellular membrane of cells forming clusters on postnatal day 18, but not to the cytoplasm (Fig. e). WGA binding was observed in the cytoplasm, and from postnatal day 3, it was seen in the cell membrane (Fig. f); however, by postnatal day 7, both cytoplasmic and cell membrane WGA binding had disappeared. It was observed in serous cells (Fig. g). On postnatal day 14, WGA binding was observed in the serous demilunes, but not on the cellular membrane (Fig. h). DBA exhibited strong binding to the cellular membrane on embryonic day 18 (Fig. i) and to the cytoplasm on postnatal day 1 (Fig. j). Thereafter, DBA bound to both the cellular membrane and cytoplasm; however, in adult rats, DBA binding to the cellular membrane decreased (Fig. k, l). PNA bound to the cellular membrane from embryonic day 18 (Fig. m) and to the cytoplasm from postnatal day 3, and binding was observed until postnatal day 14 (Fig. n, o); however, by postnatal day 21, PNA binding to the cellular membrane was absent (Fig. p). Lectin binding in the von Ebner’s glands From postnatal day 1, when the terminal cell primordium of von Ebner’s glands was recognized, SBA binding was observed at the cellular membrane (Fig. a), and the binding remained constant with age (Fig. b–d). Binding was also observed in the cytoplasm from postnatal day14 (Fig. c). Adult rats showed SBA binding in the cytoplasm and on the cellular membrane (Fig. d). WGA exhibited almost no binding at postnatal day 1 (Fig. e); however, at postnatal day 5, 10, WGA binding to the cellular membrane was strong (Fig. f, g). In adult rats, WGA binding was also observed in the cytoplasm (Fig. h). DBA binding was observed on the cellular membrane at postnatal day 1 (Fig. i), and the same binding pattern was observed until postnatal day14 (Fig. j, k). However, after postnatal day 21, no DBA binding was observed in either the cytoplasm or cellular membrane (Fig. l). PNA binding was identified on the cellular membrane at postnatal day 1 (Fig. m), which grew stronger with age (Fig. n, o). PNA binding was also observed in the cytoplasm of adult rats (Fig. p).
In adult rats, von Ebner's glands were found to be located just below the circumvallate papilla, separated on both sides by the lingual septum, clumped within the muscularis, and with the opening at the base of the circumvallate papilla sulcus (Fig. a). The Weber's glands were located laterally and posteriorly to the von Ebner’s glands, opening directly onto the dorsal surface of the tongue (Fig. a). On embryonic day 18, the sulcus epithelium of the circumvallate papilla had not yet been inserted into the mesenchyme. Serial sections showed clusters of epithelial cells in the lateral portion of the posterior tongue (Fig. b) which were continuous with the surface layer of the tongue epithelium. On embryonic day 20, we observed an infiltration of the sulcus epithelium of the circumvallate papilla into the mesenchyme, with no evident formation of the von Ebner’s glands. The Weber's glands were also found in the mesenchyme in high number (Fig. c) with numerous clusters of epithelial cells present in the mesenchyme at 1 day of age (Fig. d), and their number and size increased further at 3 days of age (Fig. e). Thereafter, the cell population of epithelial cells in the mesenchyme increased over time, and both the Weber’s and von Ebner’s glands were fully developed by 28 days of age (Fig. f). The clusters of epithelial cells observed at 18 days of age were formed by cells with darker cytoplasm (Fig. a). On embryonic day 20, the number of cells forming these cell masses increased, and the cytoplasm of these cells were similarly dark (Fig. b). No apparent formation of terminal cells of the serous glands at the tip of the sulcus epithelium of the circumvallate papilla, which were plunging into the mesenchyme on embryonic day 18 was observed (Fig. c). At 1 day of age, the terminal part of Weber's glands showed characteristics of mucous cells, with some cells having a light-toned cytoplasm as observed in adult rats, and others having dark-toned cytoplasm (Fig. d). Duct and epithelial mass were observed from the base of the papillae sulcus, some of which showed serous cell morphology and the formation of von Ebner’s glands (Fig. e). Some of the Weber’s glands in rats aged 1 day showed flat nuclei in the clear-toned cytoplasm, and the number of these cells increased at 3 days of age. Serous demilunes were also observed (Fig. f). von Ebner’s glands showed tubular arrangements of terminal cells with round nuclei on the basal side of the dark-toned cytoplasm (Fig. g). At 7 days of age, Weber's glands showed mucous cells, numerous serous cells, and a serous demilunes, whereas von Ebner’s glands demonstrated advanced branching of the terminal part of the gland . Furthermore, Weber's glands showed a marked increase in mucous cells with age. At 21–28 days of age, the observations were similar to those noted in adult rats, and the cytoplasm of the terminal cells were transparent and filled with mucus secretions. The nuclei were oval and located basolaterally. Weber's glands contained mucous cells and serous cells with dark-toned cytoplasm (Fig. h). The von Ebner’s glands showed increased terminal areas and branching, with triangular terminal cells surrounding the gland lumen, as seen in typical exocrine glands. In addition, round nuclei were located slightly basolateral to the basophilic cytoplasm (Fig. i). In the epithelium of the circumvallate papilla sulcus, taste bud primordia were present on embryonic day 18; however, no taste pits were observed (Fig. a). On postnatal day 1, taste buds with taste pits were observed (Fig. b), and thereafter the number of taste buds with taste pits increased (Fig. c).
Among the lectins that have been reported to be used in salivary glands, in this study, we tested their binding in serous cells and mucous cells in the posterior lingual gland of adult rats. Three or more tissue fragments from each sample were validated with 21 different lectins. The binding of the lectins was classified into four levels: strong (3), moderate intensity (2), weak intensity (1), and unreactive (0). Different binding intensities were denoted by minimum and maximum values. The lectin that bind stongly to serous cells and unreactively to mucous cells are PNA, SBA, and WGA, which were used as markers for serous cells. The lectin that binds stongly to mucus cells and binds unreactively to mucus cells was DBA, which was designated as a marker for mucus cells. These 4 lectins were excellent markers for serous and mucous cells in adult animals, and were used as indicators of mature glandular cells and for serous and mucous cells during rat development (Table ). Abbreviations: Con A, ( Canavalia ensiformis ) agglutinin; SBA, Soybean ( Glycine maximus ) agglutinin; WGA, Wheat germ ( Triticum vulgaris ) agglutinin; DBA, Horse gram ( Dolichos biflorus ) agglutinin; UEA-I, ( Ulex europaeus ) agglutinin-I; RCA-I, ( Ricinus communis ) agglutinin-I; PNA, peanut ( Arachis hypogaea ) agglutinin; GS-I, ( Griffonia simplicifolia ) Lectin I; pea, ( Pisum sativum ); LCA, ( Lens culinaris ) agglutinin; PHA-E, ( Phaseolus vulgaris ) erythroagglutinin; PHA-L, ( Phaseolus vulgaris ) leucoagglutinin; SJA, ( Sophora japonicum ) agglutinin, sWGA, Succinyl ( Triticum vulgaris ) agglutinin; GS-II , ( Griffonia simplicifolia ) Lectin II; DSA, Thorn apple ( Datura stramonium ) agglutinin; ECL ( Erythrina cristagalli ) Lectin; Jacalin, ( Artocarpus integrifolia ); LEL, ( Lycopersicon esculentum ) Lectin; STL, ( Solanum tuberosum ) Lectin; VVA, ( Vicia villosa ) agglutinin.
SBA binding to the cellular membrane was observed, albeit weakly, in cell masses on embryonic day 18 (Fig. a), and some SBA binding was noted in the cytoplasm on postnatal day 1 (Fig. b). At postnatal day7, SBA bound to the cytoplasm and cellular membrane, and at postnatal day14, SBA showed a similar binding pattern and bound to the serous demilunes (Fig. c). However, by postnatal day 21, the binding observed in the cytoplasm and cellular membrane was minimal (Fig. d). In the postnatal period, SBA binding was also strong in saliva. WGA bound to the cellular membrane of cells forming clusters on postnatal day 18, but not to the cytoplasm (Fig. e). WGA binding was observed in the cytoplasm, and from postnatal day 3, it was seen in the cell membrane (Fig. f); however, by postnatal day 7, both cytoplasmic and cell membrane WGA binding had disappeared. It was observed in serous cells (Fig. g). On postnatal day 14, WGA binding was observed in the serous demilunes, but not on the cellular membrane (Fig. h). DBA exhibited strong binding to the cellular membrane on embryonic day 18 (Fig. i) and to the cytoplasm on postnatal day 1 (Fig. j). Thereafter, DBA bound to both the cellular membrane and cytoplasm; however, in adult rats, DBA binding to the cellular membrane decreased (Fig. k, l). PNA bound to the cellular membrane from embryonic day 18 (Fig. m) and to the cytoplasm from postnatal day 3, and binding was observed until postnatal day 14 (Fig. n, o); however, by postnatal day 21, PNA binding to the cellular membrane was absent (Fig. p).
From postnatal day 1, when the terminal cell primordium of von Ebner’s glands was recognized, SBA binding was observed at the cellular membrane (Fig. a), and the binding remained constant with age (Fig. b–d). Binding was also observed in the cytoplasm from postnatal day14 (Fig. c). Adult rats showed SBA binding in the cytoplasm and on the cellular membrane (Fig. d). WGA exhibited almost no binding at postnatal day 1 (Fig. e); however, at postnatal day 5, 10, WGA binding to the cellular membrane was strong (Fig. f, g). In adult rats, WGA binding was also observed in the cytoplasm (Fig. h). DBA binding was observed on the cellular membrane at postnatal day 1 (Fig. i), and the same binding pattern was observed until postnatal day14 (Fig. j, k). However, after postnatal day 21, no DBA binding was observed in either the cytoplasm or cellular membrane (Fig. l). PNA binding was identified on the cellular membrane at postnatal day 1 (Fig. m), which grew stronger with age (Fig. n, o). PNA binding was also observed in the cytoplasm of adult rats (Fig. p).
Here, the development of Weber's and von Ebner's glands, which are posterior lingual glands around the circumvallate papilla in rats, was searched histochemically and by lectin histochemistry. In rodents, 4 types of papillae exist on the dorsal surface of the tongue: fungiform, circumvallate, foliate, and filiform papillae . In humans, the papillae vallate are circular bulges with 8–9 papillae . However, in mice and rats, a single circumvallate papilla exists on the posterior midline of the dorsal surface of the tongue. This circumvallate papilla contains taste buds , . In addition, the von Ebner’s glands open onto the base of the fornix and lobulated papillary sulcus, and the Weber's glands are present laterally and posteriorly to the von Ebner's glands. The Weber's glands are mucous glands , ; however, the presence of serous cells has been suggested in humans, rats, and mice , . Here, the Weber's glands contained mucous cells and fewer serous and serous semilunar cells. These findings are consistent with previous findings in rats , , humans , , and other animals , . In the present study, on embryonic day 18, the epithelial cells of the mesenchyme in the posterior part of the tongue were continuous with the tongue epithelium and could be considered the Weber's gland primordium. These cells showed serous cell morphology. Hamosh et al . reported that the development of the Weber's glands preceded that of the von Ebner’s glands, and secretory granules were present by embryonic day 20. These findings are in close agreement with those of the present study. Regarding the development of the palatine glands, which are considered to be muciparous like Weber's glands, Shinzato et al. . reported that in rats, epithelial incision began on embryonic day 17, terminal cells were formed on embryonic day 18, and mucous cells appeared on embryonic day 21. This change is almost identical to the development of the Weber’s glands demonstrated in the present study. Here, von Ebner’s glands were not formed on embryonic day 18; however, from embryonic day 20, the epithelium of the circumvallate papilla sulcus started infiltrating into the mesenchyme. Hamosh et al . showed that in Sprague Dawley rats, the epithelial growth from the foliate and circumvallate papillae on embryonic day 19–20 was noted as the von Ebner’s glands began to develop, and no terminal formation or production of secretory granules were observed until 3–4 days after birth. The results of the present study are in close agreement with that of these findings. Analysis of sugar residues related to lectin histochemistry is underway in various fields , – . The lectins are often used as markers identify specific cell populations because they bind specifically to glycohydrate epitopes on the cell membrane. By searching for the binding mode of lectins, the localization of glycoconjugates in different tissues and the characteristics of cellular glycans in each tissue can be elucidated. We found that SBA, PNA, and WGA bind to serous cells and DBA binds to mucous cells in adult rats. The results showed that only DBA was found to bind to Weber's gland mucosa cells in adult rats, while all four lectins, SBA, PNA, WGA and DBA, bind to the cells early in development, and all but DBA disappear during development. On the other hand, serous cells of von Ebner's glands bound SBA, WGA and PNA in adult rats, but all four lectins, SBA, PNA, WGA and DBA, bound to the cytoplasm during the developmental stage, and DBA was lost during development. The lectin-binding mode of the developing rats was similar to that of adult rats at postnatal day 21. In addition, mature taste buds were observed in the circumvallate papilla on postnatal day 1. These results indicate that the maturation of the posterior lingual glands is closely related to changes in eating habits, that they are already accompanied by taste receptors in the early postnatal period, but that the Weber's glands, which are mucous glands, function as serous glands and compensate for their function during the period when the von Ebner’s glands are immature. Lectin-binding properties indicate that serous cells. The lectin-binding properties suggested the presence of Galβ (1,3) > Galβ(1,4) > Gal, αGalNAc > αGal > βGalNAc, NeuAc > (GalNAc) 2–3 >>>GlcNAc sugar residues in the serous cells and GalNAcα(1,3) sugar residues in mucous cells. Histochemically, Weber's glands, which are mucous glands, appeared as a cell mass and had the morphology of serous cells at embryonic day 18. At the same time, PNA, SBA, WGA, and DBA showed binding properties to the cellular membrane in lectin. This suggests that the sugar residues are Galβ (1,3) > Galβ(1,4) > Gal, α-D-GalNAc;β-D-GalNAc, NeuAc > (GalNAc) 2–3 >>>GlcNAc , and GalNAcα(1,3). As the Weber's gland matured, mucous cells appeared at postnatal day1, coinciding with this period of SBA and DBA binding. At this time, the cytoplasm of the mucus cells contained sugar residues of αGalNAc > αGal > βGalNAc and GalNAcα(1,3), and a little later, on postnatal day3, binding was observed to PNA and WGA, suggesting the presence of Galβ (1,3) > Galβ(1,4) > Gal and NeuAc > (GalNAc) 2–3 >>> GlcNAc in addition to these The presence of Galβ (1,3) > Galβ(1,4) > Gal and Β-D-GlcNAc;NeuNAc was suggested. Furthermore, WGA disappears on postnatal day 7 and PNA and SBA stop binding on postnatal day 14, suggesting that NeuAc > (GalNAc) 2–3 >>>GlcNAc disappears on postnatal day 7, Galβ (1,3) > Galβ(1,4) > Gal andαGalNAc > αGal > βGalNAc disappear on postnatal day 14, and only GalNAcα(1,3) residue may be present, respectively. Thus, it is suggested that the localization of sugar residues in the cell may change during development. The loss of lectin binding coincided with the change in food intake from a liquid to a solid diet. The salivary glands of rats aged 12–14 weeks and found that SBA, DBA, and PNA showed strong binding to serous cells of the posterior lingual gland, and SBA and DBA showed moderate binding to mucous cells. Further, in the serous semilunar cells found in mucous cells, the lectin-binding pattern was similar to that of serous cells. The authors also reported that SBA, DBA, and PNA showed moderate binding in serous cells, while the parotid gland, which is also serous in nature, showed weak binding in conduits . These findings for DBA and PNA are consistent with those of the present study; however, the results regarding SBA are different. WGA also highlighted reactions in many of the serous terminal cells of the posterior lingual gland, as well as in mucous cells. Furthermore, PNA reacted with most of the serous terminal cells and only with serous cells in areas with a mixture of mucous and serous cells, but not in the conduit epithelium of the mucous glands. These findings on lectin binding differ in part from the results of the present study. Lectins continue to bind to the same carbohydrates regardless of species, but the differences in binding properties may be due to differences in the carbohydrates present. The differences in binding patterns of lectin in methodology may also account for the discrepancy. The methodological difference is that some lectins are inactivated by phosphate complexed with Ca2 + , which causes excessive darkening of the tissue fragments and makes them extremely difficult to observe, so PBS was used for lectin staining in this study. Here, terminal cells in the Weber's glands on embryonic day 18 and the von Ebner's glands on postnatal day 1 bound both serous cells and mucus cells before the lectin-binding pattern became identical to that of adult animals. This suggests that the Weber’s glands in late embryonic stages initially assume serous properties before forming mucous cells. It is possible that both the von Ebner’s and Weber’s glands function as mixed glands for some time after the onset of posterior lingual gland formation in the late embryonic period. In the Weber's glands in rats, undifferentiated cells appear first and form serous cells that differentiate into intermediate-type cells, then into mucous cells. In the present study, we propose that the change in lectin-binding properties observed in the Weber's glands corresponds with this phenomenon. In our study, we found that the posterior lingual gland in rats aged 21–28 days showed the same lectin-binding pattern as that of adult animals. In serous cells, αGalNAc > αGal > βGalNAc residues appeared at postnatal day14, suggesting that the salivary glands were also mature at this time, postnatal days 21–28, when the individuals matured, and Galβ (1,3) > Galβ(1,4) > Gal and NeuAc > (GalNAc) 2–3 >>>GlcNAc may be present. The timing of the appearance of these sugar residues may express the maturity of the developmental stages. Mucus cells may also be salivating and functioning on postnatal day1, when GalNAcα(1,3) residues are present. Furthermore, rats were weaned at postnatal day 21 and kept on solid feed thereafter. Thus, the change in diet may have caused a change in the properties of salivary gland cells. After postnatal day 10, the incisors erupt, and the rats can ingest solid food. During this period, the von Ebner’s glands secrete serous saliva and Weber's glands secrete mucous saliva. Mucous saliva is involved in food mass formation, and such changes in feeding behavior are thought to be closely related to changes in salivary components, i.e., the function of salivary secretory cells. Future studies should investigate these changes in weaning time and feed. The von Ebner’s glands secrete serous saliva, which cleans the taste pits of the taste buds in the sulcus epithelium of the papillae foliate and circumvallate papilla thereby maintaining and renewing taste receptors , . Here, morphologically mature taste buds with taste pits were observed in the sulcus epithelium of the circumvallate papilla on postnatal day 1, and the number of these buds increased thereafter. The appearance of mature postnatal taste buds in the epithelium of the papillary sulcus is consistent with the findings of previous reports , . By postnatal day 3, The von Ebner's glands opening to the floor of the papillary sulcus was not mature; however, the Weber's glands opening to the mucosa behind the tongue showed morphological and histochemical characteristics of serous cells, suggesting that they secrete serous saliva. Therefore, until the von Ebner’s glands mature, Weber's glands are thought to renew taste receptors by cleaning the taste pits of the taste buds in the papillary sulcus epithelium, compensating for the function that the von Ebner’s glands perform. Similar changes may occur in humans, the same mammal species. Not only do the secretory products of the von Ebner’s glands influence the taste response, but also the glyco-chemical properties of each taste bud , . We identified cases in which lectins bound to the cellular membrane and cytoplasm. The binding of lectins on the cellular membrane is thought to be due to the sugar chains that make up the cellular membrane. In contrast, lectin-binding in the cytoplasm is thought to be due to the binding of sugars to proteins formed intracellularly, which are modified by intracellular organelles. Here, we did not examine the intracellular aspects of lectins; however, future studies should examine this.
The development of the Weber's and von Ebner’s glands from the late embryonic stages of the rat were examined using lectin histochemistry using H&E staining, SBA, WGA, DBA, and PNA. We identified that the Weber's glands were a mass of serous cells on embryonic day 18, which formed mucous cells on postnatal day 3, and the number of mucous cells gradually increased with age, showing similar morphology to that of adult rats on postnatal day 21. Further, the von Ebner’s glands began to form around embryonic day 20; however, serous cells were not observed until postnatal day 3. Additionally, in the posterior lingual glands of adult rats, SBA, WGA, and PNA were found in serous cells, and DBA in mucous cells (Fig. ). During early development, the Weber's and von Ebner’s glands showed both serous and mucous cell lectin-binding patterns, which were different from the lectin-binding patterns observed in adult rats. These findings suggest that the Weber's and von Ebner’s glands may function as mixed glands for a short period after the onset of posterior lingual gland formation during the postnatal period. Since developing rats showed the same lectin-binding pattern as adult rats from postnatal day 21–28, we suggest that their saliva secretion changed as their diet changed from liquid to solid.
Experimental animals At least three Sprague–Dawley rats of each age (embryonic day 18 and 20, and postnatal days 1, 3, 5, 7, 10, 14, 21, 28, 43, 56, and 63) were used. The day on which the vaginal plug was identified was defined as embryonic day 0, and the day of birth as postnatal day 0. Animals were weaned at postnatal day 21. For animals in the fetal period, pregnant rats were laparotomized under deep anesthesia administered intraperitoneally, and the fetuses were harvested and dissected. The specimens were fixed using immersion fixation. After birth, rats were placed under deep anesthesia administered intraperitoneally. The chest was opened, a catheter was inserted into the apex of the animal's heart and advanced to the base of the vena cava, an incision was made in the right ear, and the animal was immersed in 0.02 M phosphate-buffered saline (PBS; pH 7.4). The head was perfused and fixed in 0.1 M phosphate buffer (pH 7.4) containing 4% paraformaldehyde, and the tongue was removed and immersed in the same fixative solution. The rats were fed solid feed ad libitum. All animal experiments were approved by the Institutional Animal Care and Use Committee of Osaka University Graduate School of Dentistry and complied with the guidelines for the care and use of laboratory animals at Osaka University (Approval No. 19–024-0). All animal experiments complied with the ARRIVE guidelines. Hematoxylin & Eosin staining The posterior part of the tongue, including the circumvallate papilla, was paraffin-embedded and serial sections of 10 μm thickness were prepared. The paraffin sections were stained with hematoxylin and eosin (H&E) stain, and the morphological changes in the circumvallate papilla, Weber's gland, and von Ebner’s gland were observed under a light microscope. Lectin histochemistry The posterior part of the tongue, including the circumvallate papilla, was embedded in optimal cutting temperature compound (Sakura Finetek Co., Tokyo, Japan), and 10 μm thick frozen sections were prepared and affixed to MAS–coated glass slides (Matsunami Glass Ind., Ltd., Osaka, Japan). The sections were washed with 0.02 M PBS, treated with 3% hydrogen peroxide-PBS for 30 min to inactivate endogenous peroxidase, washed with 0.02 M PBS, and incubated with biotin-labeled. Con A, SBA, WGA, DBA, UEA-I, RCA-I, PNA, GS-I, pea , LCA, PHA-E, PHA-L, SJA, sWGA, GS-II, DSA, ECL, Jacalin, LEL, STL and VVA (0.5 μg/ml; Vector, Burlingame, CA, USA) in a wet box for 16 h at room temperature. The cells were then washed with 0.02 M PBS and treated with Avidin–Biotin Complex (Vector) for 90 min. We searched for the above 21 lectins in 3 sections per adult rat with N = 7 or greater. We performed lectin histochemical searches on the posterior lingual glands of rats. In the posterior lingual gland of rats, except for 4 lectins, some lectins showed binding to both serous and mucous cells, some did not show binding to both cells, and some rats showed unstable binding depending on the individual rats (Table ). Their lectins bind strongly to serous cells and unreactively to mucus cells. PNA, SBA and WGA, were used as markers for serous cells. The lectin that binds strongly to mucus cells and binds unreactively to serous cells was DBA, which was used as a marker for mucus cells in this study. Sections were washed with 0.02 M PBS and further washed with 0.05 M Tris–HCl-buffered saline (TBS), pH 7.6, followed by 0.04% 3,3-diaminobenzidine (Sigma-Aldrich Co., Tokyo, Japan), and horseradish peroxidase activity was visualized using 0.05 M TBS containing 0.003% hydrogen peroxide water, and sensitized with 0.1% nickel ammonium sulfate. All reactions were performed at room temperature. After the reactions, the cells were contrast stained using methylene blue, dehydrated in ascending ethanol series, permeabilized with Lemosol, sealed in Permount ® (Fisher Scientific, NJ), and observed under an optical microscope – .
At least three Sprague–Dawley rats of each age (embryonic day 18 and 20, and postnatal days 1, 3, 5, 7, 10, 14, 21, 28, 43, 56, and 63) were used. The day on which the vaginal plug was identified was defined as embryonic day 0, and the day of birth as postnatal day 0. Animals were weaned at postnatal day 21. For animals in the fetal period, pregnant rats were laparotomized under deep anesthesia administered intraperitoneally, and the fetuses were harvested and dissected. The specimens were fixed using immersion fixation. After birth, rats were placed under deep anesthesia administered intraperitoneally. The chest was opened, a catheter was inserted into the apex of the animal's heart and advanced to the base of the vena cava, an incision was made in the right ear, and the animal was immersed in 0.02 M phosphate-buffered saline (PBS; pH 7.4). The head was perfused and fixed in 0.1 M phosphate buffer (pH 7.4) containing 4% paraformaldehyde, and the tongue was removed and immersed in the same fixative solution. The rats were fed solid feed ad libitum. All animal experiments were approved by the Institutional Animal Care and Use Committee of Osaka University Graduate School of Dentistry and complied with the guidelines for the care and use of laboratory animals at Osaka University (Approval No. 19–024-0). All animal experiments complied with the ARRIVE guidelines.
The posterior part of the tongue, including the circumvallate papilla, was paraffin-embedded and serial sections of 10 μm thickness were prepared. The paraffin sections were stained with hematoxylin and eosin (H&E) stain, and the morphological changes in the circumvallate papilla, Weber's gland, and von Ebner’s gland were observed under a light microscope.
The posterior part of the tongue, including the circumvallate papilla, was embedded in optimal cutting temperature compound (Sakura Finetek Co., Tokyo, Japan), and 10 μm thick frozen sections were prepared and affixed to MAS–coated glass slides (Matsunami Glass Ind., Ltd., Osaka, Japan). The sections were washed with 0.02 M PBS, treated with 3% hydrogen peroxide-PBS for 30 min to inactivate endogenous peroxidase, washed with 0.02 M PBS, and incubated with biotin-labeled. Con A, SBA, WGA, DBA, UEA-I, RCA-I, PNA, GS-I, pea , LCA, PHA-E, PHA-L, SJA, sWGA, GS-II, DSA, ECL, Jacalin, LEL, STL and VVA (0.5 μg/ml; Vector, Burlingame, CA, USA) in a wet box for 16 h at room temperature. The cells were then washed with 0.02 M PBS and treated with Avidin–Biotin Complex (Vector) for 90 min. We searched for the above 21 lectins in 3 sections per adult rat with N = 7 or greater. We performed lectin histochemical searches on the posterior lingual glands of rats. In the posterior lingual gland of rats, except for 4 lectins, some lectins showed binding to both serous and mucous cells, some did not show binding to both cells, and some rats showed unstable binding depending on the individual rats (Table ). Their lectins bind strongly to serous cells and unreactively to mucus cells. PNA, SBA and WGA, were used as markers for serous cells. The lectin that binds strongly to mucus cells and binds unreactively to serous cells was DBA, which was used as a marker for mucus cells in this study. Sections were washed with 0.02 M PBS and further washed with 0.05 M Tris–HCl-buffered saline (TBS), pH 7.6, followed by 0.04% 3,3-diaminobenzidine (Sigma-Aldrich Co., Tokyo, Japan), and horseradish peroxidase activity was visualized using 0.05 M TBS containing 0.003% hydrogen peroxide water, and sensitized with 0.1% nickel ammonium sulfate. All reactions were performed at room temperature. After the reactions, the cells were contrast stained using methylene blue, dehydrated in ascending ethanol series, permeabilized with Lemosol, sealed in Permount ® (Fisher Scientific, NJ), and observed under an optical microscope – .
|
Druggable genome-wide Mendelian randomization identifies therapeutic targets for metabolic dysfunction-associated steatotic liver disease | 78b44c5c-982c-4b8d-bf1e-bd42f2ea8473 | 11938603 | Pathologic Processes[mh] | With the increasing prevalence of obesity, metabolic dysfunction-associated steatotic liver disease (MASLD) has become a major global health concern, affecting approximately 25% of adults . MASLD is characterized by excessive triglyceride accumulation in the liver alongside at least one cardiometabolic risk factor. Some studies suggest that evidence from non-alcoholic fatty liver disease (NAFLD) research is applicable to MASLD, and the terms NAFLD and MASLD can be used interchangeably [ – ]. Therefore, this study adhered to these classification principles in data extraction and concept descriptions. MASLD involves a range of histologically classified pathologies, including simple steatosis, metabolic dysfunction-associated steatohepatitis (MASH), and cirrhosis . MASH, a late-stage form of MASLD, is expected to become a predominant indication for liver transplantation. Despite its severity, no approved drug therapies for MASLD exist, and the few investigated therapies have shown limited efficacy . Advancing research on the cellular and molecular pathogenesis of MASLD is essential for identifying therapeutic targets and developing targeted treatments. Genetic variants associated with MASLD and MASH have been identified through genome-wide association studies (GWAS) and candidate gene approaches . These findings further highlight the important role of genetic factors in the risk of developing MASLD and in disease progression . With the increasing popularity of GWAS, Mendelian randomization (MR) has been shown to be effective in investigating the etiology of this disease and in predicting the efficacy of clinical medications through simulating randomized controlled trials . In MR analyses of drug target genes, the use of mRNA, proteins, or other downstream biomarkers as exposure factors is a robust analytical approach that is more closely aligned with the causal pathway than other exposure phenotypes. Commonly, cis-expression quantitative trait loci (cis-eQTL) within the genomic regions of drug target genes are employed as proxies, which are regulators that influence gene expression . This study clarified mechanisms underlying MASLD and identified potential therapeutic targets for drug development by examining the association between protein-coding genes and MASLD. Using two independent MASLD GWAS datasets, the eQTL identified in blood were combined. MR analyses were conducted to identify potential drug targets that might slow MASLD progression. The associations between representative druggable genes and MASLD risk were examined, along with the associations between these genes and 16 metabolism-related disease risk factors and seven common disease characteristics. Although several similar studies have been conducted, this study offers advantages owing to its large genome-wide association study (GWAS) sample size, broad ethnic representation, and validation of population samples through RNA sequencing [ – ].
This study was a secondary analysis of publicly available data. Informed consent was obtained from all participants following the original GWAS protocol, and all ethical approvals for GWAS were obtained from the original study authors. The experimental verification part of our study was approved by the Institutional Review Board of Tianjin Medical University General Hospital (Tianjin, China) and conducted following the Declaration of Helsinki norms. Identification of cis-eQTL data associated with druggable genes (cis-eQTL of blood and liver tissue from genotype-tissue expression [GTEx]) A total of 4,302 druggable genes, annotated using the HUGO Gene Nomenclature Committee (HGNC) database, were identified on autosomal chromosomes . These included 1,375 protein therapy targets in clinical development, 2,281 proteins related to members of key drug target families, and 646 proteins associated with drug–target interactions. Given that cis-eQTLs are more closely associated with genes relevant to drug discovery, statistically significant cis-eQTLs were obtained from the eQTLGen Consortium to minimize bias (false discovery rate < 0.05, 1 Mb per probe). Consequently, 31,684 human peripheral blood eQTLs were analyzed . A total of 4,302 genetic tools for druggable targets were generated through selecting cis-eQTLs within 100 kb of the genomic location of each gene, ultimately identifying eQTLs for 2,664 druggable genes. Using the same method, the drug target gene cis-eQTLs of liver tissue were also obtained from GTEx v10.0 ( https://gtexportal.org/home/releaseInfoPage ) to validate the significance of the gene results after the blood eQTL MR analysis. Outcome data A genome-wide meta-analysis was conducted on four cohorts (United Kingdom [UK] Biobank, Estonian Biobank, Electronic Medical Records and Genomics (eMERGE), and (Finnish Genetic) FinnGen ) of participants of European ancestry with electronic health records for MASLD. The analysis included the eMERGE and FinnGen cohorts, an updated MASLD GWAS from the UK Biobank (2558 cases; 395,241 controls), and a newly performed GWAS from the Estonia Biobank (4119 cases; 190120 controls). In total, 8,434 MASLD cases and 770,180 controls were included in these cohorts . A separate sample cohort was used for external validation (Table ). The study was a genomics-specific case-control study in the UK Biobank with a definitive diagnosis of MASLD based on the diagnostic criteria recommended in recent consensus guidelines . GWAS analyses were performed on 4,761 MASLD cases and 373,227 healthy controls without MASLD who were subjected to sensitivity analyses, excluding other secondary liver pathologies that coexisted, and adjusting for body mass index (BMI) and alcohol intake. Metabolism-related diseases Sixteen metabolism-related disease risk factors were selected, including lipid profiles (total cholesterol [TC], triglyceride [TG]), high-density lipoproteins [HDL-C], low-density lipoproteins [LDL-C], apolipoproteins A1 and B (ApoA1, ApoB), and lipoprotein a (Lp(a)) . Additionally, blood pressure metrics, including systolic blood pressure, diastolic blood pressure, were included . Three blood glucose indicators were considered: fasting glucose, fasting insulin, and glycated hemoglobin . Four anthropometric characteristics were also analyzed (BMI, waist circumference, hip circumference, and waist-to-hip ratio) . Furthermore, seven metabolic syndrome-related conditions, such as essential hypertension , obesity , hyper-triglyceridemia , type 2 diabetes mellitus , myocardial infarction , and ischemic stroke, were included in the analysis . The GWAS database, operated by the Medical Research Council–Integrative Epidemiology Unit (MRC-IEU) consortium, was used as the source of initial exposure data for the Atherosclerotic Heart Disease Study (Table ) . MR and colocalization (cis-eQTL of blood and liver tissue from GTEx) A two-sample MR R package was employed to analyze the MR data. A series of criteria were applied to exclude low-quality instrumental variables . First, weak single nucleotide polymorphisms (SNPs) were excluded by calculating the F-statistic (F-statistic < 10). Subsequently, SNPs that did not exhibit linkage disequilibrium with independent conditions (r² < 0.1, based on the 1000 Genomes Europe reference panel) were selected as instrumental variables. Additionally, genes with greater MASLD trait exposure were excluded through Steiger filtering, a heterogeneity filtering method to filter SNPs whose coefficient of determination (R 2 ) (R-squared, a statistical measure that represents the proportion of the variance for a dependent variable that is explained by an independent variable) with the exposure is not substantially greater than with the outcome. In the primary analyses, MR estimates for each SNP were calculated using the Wald ratio method. The SNP estimates were meta-analyzed using inverse variance weighting (IVW), MR-Egger, and weighted median models . MR-Egger regression introduces a nuisance parameter to account for directional pleiotropy, assesses whether genetic variants exhibit pleiotropic effects on the outcome, and compares the observed distance of all the variants to the regression line with the expected distance under the null hypothesis of no horizontal pleiotropy . Bonferroni correction was used to assess significance thresholds for multiple test exposures. A P -value < 1.90 × 10 5 ( P = 0.05/2644) was defined as a statistically significant difference, and validation of these significantly different targets was subsequently repeated in another independent cohort. Using the same MR process, drug target gene cis-eQTLs of liver tissue from GTEx were also used to validate the significance of the gene results after the blood eQTLs MR analysis. Significant MR results common to both independent cohorts with MASLD risk were identified using a coloc R package. For the eQTL dataset, a priori probabilities of 1 × 10 4 for cis-eQTL (H1) and MASLD association were assumed, along with 1 × 10 5 for a single variant affecting both traits (H4) . Genes with significant co-localization ( a posteriori probability PH4 > 0.80) were considered as potential drug therapy targets. Subsequently, the association between these targets and 23 metabolic syndrome-related traits was investigated using MR analysis to assess potential safety concerns and alternative indications. Summary data-based MR (SMR) The primary analysis applied involved MR analysis of blood eQTL and diseases. Given that MASLD is liver-specific, prior validation was conducted using liver eQTLs from GTEx in addition to an MR analysis of the disease. To ensure the robustness of study results, a dual SMR analysis was conducted with liver eQTLs from two versions of the GTEx database (v10.0 and v8.0) in conjunction with disease outcomes. Patients From July 2023 to March 2024, patients with obesity who underwent sleeve gastrectomy at the Tianjin Medical University General Hospital received liver biopsies to assess steatosis, inflammation, hepatocyte ballooning, and fibrosis, using a semi-quantitative method of the nonalcoholic steatohepatitis activity score (NAS) . Patients were categorized as having simple steatosis (NAS 1–2, without ballooning or fibrosis) or MASH (NAS ≥ 5 or NAS 3–4 with fibrosis). The prevalence rate for MASH within the cohort under investigation was 42.12%. Liver and blood samples were collected for RNA sequencing and stored at − 80 °C for subsequent analysis. RNA sequencing and data analysis Overall, 41 patients with obesity who underwent metabolic surgery with intraoperative liver biopsies were selected for RNA sequencing by Shanghai OE Biotech. RNA extraction and library preparation methods are detailed in Supplementary File . Libraries were sequenced on the BGI sequencing platform (BGI T7), generating 150 bp paired-end reads. Trimmomatic, a multithread command tool, removed adapters and low-quality reads, with clean reads mapped to genes using the program HISAT2 . Read counts per gene were obtained with HTSEQCOUNT , with fragments per kilobase per million mapped fragments (FPKM) values calculated using CUFFLINKS . Differentially expressed genes (DEGs) between groups were determined using truncated P -values < 0.05 and fold change > 1.5 or < 0.67 using DESeq2 with P -values corrected for multiple testing using the corresponding R package function for Benjamini–Hochberg method. Gene expression patterns were analyzed using hierarchical clustering. Histological analysis Liver tissues were fixed with 4% paraformaldehyde, paraffin-embedded, and sectioned at 5 μm. Tissue sections were then stained with hematoxylin and eosin (HE) and Masson’s stain and evaluated using an orthogonal light microscope (Nikon, NIKON ECLIPSE E100). Immunohistochemical staining was performed on at least six liver samples from each group to detect protein expression of CD33 (Proteintech Group, 17425-1-AP) and MFGE8 (Proteintech Group, 25951-1-AP). Results were interpreted under a white light microscope (Nikon Instruments Ltd. E100). Using the AIPATHWELL analysis software, the same tan nuclei were uniformly selected to identify positive cells in all sections, while blue nuclei were selected as other cells. Each section was analyzed to obtain the number of positive and total cells. The percentage of positive cells (number of positive cells/total number of cells * 100) was calculated as the positivity rate (%). Quantitative polymerase chain reaction Total RNA was isolated from liver using a mirVana RNA Isolation Kit (AM1561), and its yield and integrity were assessed using a NanoDrop 2000 spectrophotometer and ethidium bromide-stained agarose gel electrophoresis, respectively. RNA was converted to cDNA with the TransScript All-in-One First-Strand cDNA Synthesis SuperMIX for qPCR (quantitative polymerase chain reaction). qPCR testing was conducted on a LightCycler ® 480 II using PerfectStart Green qPCR SuperMix, with amplification conditions consisting of 94 °C for 30 s, followed by 45 cycles of 94 °C for 5 s and 60 °C for 30 s. Product specificity was confirmed by melting curve analysis from 60 °C to 97 °C. mRNA expression levels were normalized to GAPDH and calculated with the 2 −ΔΔCt method. Primers were designed and synthesized by Shanghai Ouyi Biomedical Technology Co. and Beijing Kengke Xinye Biotechnology Co, respectively (Table ). Statistical analysis Statistical data are represented as mean and standard deviation values. Group differences were assessed using one-way analysis of variance (ANOVA) followed by Tukey’s post-hoc test for pairwise comparisons. When data did not meet the assumptions of parametric tests, a Kruskal–Wallis analysis test was employed to assess differences across more than two groups. P -values < 0.05 were considered statistically significant.
A total of 4,302 druggable genes, annotated using the HUGO Gene Nomenclature Committee (HGNC) database, were identified on autosomal chromosomes . These included 1,375 protein therapy targets in clinical development, 2,281 proteins related to members of key drug target families, and 646 proteins associated with drug–target interactions. Given that cis-eQTLs are more closely associated with genes relevant to drug discovery, statistically significant cis-eQTLs were obtained from the eQTLGen Consortium to minimize bias (false discovery rate < 0.05, 1 Mb per probe). Consequently, 31,684 human peripheral blood eQTLs were analyzed . A total of 4,302 genetic tools for druggable targets were generated through selecting cis-eQTLs within 100 kb of the genomic location of each gene, ultimately identifying eQTLs for 2,664 druggable genes. Using the same method, the drug target gene cis-eQTLs of liver tissue were also obtained from GTEx v10.0 ( https://gtexportal.org/home/releaseInfoPage ) to validate the significance of the gene results after the blood eQTL MR analysis.
A genome-wide meta-analysis was conducted on four cohorts (United Kingdom [UK] Biobank, Estonian Biobank, Electronic Medical Records and Genomics (eMERGE), and (Finnish Genetic) FinnGen ) of participants of European ancestry with electronic health records for MASLD. The analysis included the eMERGE and FinnGen cohorts, an updated MASLD GWAS from the UK Biobank (2558 cases; 395,241 controls), and a newly performed GWAS from the Estonia Biobank (4119 cases; 190120 controls). In total, 8,434 MASLD cases and 770,180 controls were included in these cohorts . A separate sample cohort was used for external validation (Table ). The study was a genomics-specific case-control study in the UK Biobank with a definitive diagnosis of MASLD based on the diagnostic criteria recommended in recent consensus guidelines . GWAS analyses were performed on 4,761 MASLD cases and 373,227 healthy controls without MASLD who were subjected to sensitivity analyses, excluding other secondary liver pathologies that coexisted, and adjusting for body mass index (BMI) and alcohol intake.
Sixteen metabolism-related disease risk factors were selected, including lipid profiles (total cholesterol [TC], triglyceride [TG]), high-density lipoproteins [HDL-C], low-density lipoproteins [LDL-C], apolipoproteins A1 and B (ApoA1, ApoB), and lipoprotein a (Lp(a)) . Additionally, blood pressure metrics, including systolic blood pressure, diastolic blood pressure, were included . Three blood glucose indicators were considered: fasting glucose, fasting insulin, and glycated hemoglobin . Four anthropometric characteristics were also analyzed (BMI, waist circumference, hip circumference, and waist-to-hip ratio) . Furthermore, seven metabolic syndrome-related conditions, such as essential hypertension , obesity , hyper-triglyceridemia , type 2 diabetes mellitus , myocardial infarction , and ischemic stroke, were included in the analysis . The GWAS database, operated by the Medical Research Council–Integrative Epidemiology Unit (MRC-IEU) consortium, was used as the source of initial exposure data for the Atherosclerotic Heart Disease Study (Table ) .
A two-sample MR R package was employed to analyze the MR data. A series of criteria were applied to exclude low-quality instrumental variables . First, weak single nucleotide polymorphisms (SNPs) were excluded by calculating the F-statistic (F-statistic < 10). Subsequently, SNPs that did not exhibit linkage disequilibrium with independent conditions (r² < 0.1, based on the 1000 Genomes Europe reference panel) were selected as instrumental variables. Additionally, genes with greater MASLD trait exposure were excluded through Steiger filtering, a heterogeneity filtering method to filter SNPs whose coefficient of determination (R 2 ) (R-squared, a statistical measure that represents the proportion of the variance for a dependent variable that is explained by an independent variable) with the exposure is not substantially greater than with the outcome. In the primary analyses, MR estimates for each SNP were calculated using the Wald ratio method. The SNP estimates were meta-analyzed using inverse variance weighting (IVW), MR-Egger, and weighted median models . MR-Egger regression introduces a nuisance parameter to account for directional pleiotropy, assesses whether genetic variants exhibit pleiotropic effects on the outcome, and compares the observed distance of all the variants to the regression line with the expected distance under the null hypothesis of no horizontal pleiotropy . Bonferroni correction was used to assess significance thresholds for multiple test exposures. A P -value < 1.90 × 10 5 ( P = 0.05/2644) was defined as a statistically significant difference, and validation of these significantly different targets was subsequently repeated in another independent cohort. Using the same MR process, drug target gene cis-eQTLs of liver tissue from GTEx were also used to validate the significance of the gene results after the blood eQTLs MR analysis. Significant MR results common to both independent cohorts with MASLD risk were identified using a coloc R package. For the eQTL dataset, a priori probabilities of 1 × 10 4 for cis-eQTL (H1) and MASLD association were assumed, along with 1 × 10 5 for a single variant affecting both traits (H4) . Genes with significant co-localization ( a posteriori probability PH4 > 0.80) were considered as potential drug therapy targets. Subsequently, the association between these targets and 23 metabolic syndrome-related traits was investigated using MR analysis to assess potential safety concerns and alternative indications.
The primary analysis applied involved MR analysis of blood eQTL and diseases. Given that MASLD is liver-specific, prior validation was conducted using liver eQTLs from GTEx in addition to an MR analysis of the disease. To ensure the robustness of study results, a dual SMR analysis was conducted with liver eQTLs from two versions of the GTEx database (v10.0 and v8.0) in conjunction with disease outcomes.
From July 2023 to March 2024, patients with obesity who underwent sleeve gastrectomy at the Tianjin Medical University General Hospital received liver biopsies to assess steatosis, inflammation, hepatocyte ballooning, and fibrosis, using a semi-quantitative method of the nonalcoholic steatohepatitis activity score (NAS) . Patients were categorized as having simple steatosis (NAS 1–2, without ballooning or fibrosis) or MASH (NAS ≥ 5 or NAS 3–4 with fibrosis). The prevalence rate for MASH within the cohort under investigation was 42.12%. Liver and blood samples were collected for RNA sequencing and stored at − 80 °C for subsequent analysis.
Overall, 41 patients with obesity who underwent metabolic surgery with intraoperative liver biopsies were selected for RNA sequencing by Shanghai OE Biotech. RNA extraction and library preparation methods are detailed in Supplementary File . Libraries were sequenced on the BGI sequencing platform (BGI T7), generating 150 bp paired-end reads. Trimmomatic, a multithread command tool, removed adapters and low-quality reads, with clean reads mapped to genes using the program HISAT2 . Read counts per gene were obtained with HTSEQCOUNT , with fragments per kilobase per million mapped fragments (FPKM) values calculated using CUFFLINKS . Differentially expressed genes (DEGs) between groups were determined using truncated P -values < 0.05 and fold change > 1.5 or < 0.67 using DESeq2 with P -values corrected for multiple testing using the corresponding R package function for Benjamini–Hochberg method. Gene expression patterns were analyzed using hierarchical clustering.
Liver tissues were fixed with 4% paraformaldehyde, paraffin-embedded, and sectioned at 5 μm. Tissue sections were then stained with hematoxylin and eosin (HE) and Masson’s stain and evaluated using an orthogonal light microscope (Nikon, NIKON ECLIPSE E100). Immunohistochemical staining was performed on at least six liver samples from each group to detect protein expression of CD33 (Proteintech Group, 17425-1-AP) and MFGE8 (Proteintech Group, 25951-1-AP). Results were interpreted under a white light microscope (Nikon Instruments Ltd. E100). Using the AIPATHWELL analysis software, the same tan nuclei were uniformly selected to identify positive cells in all sections, while blue nuclei were selected as other cells. Each section was analyzed to obtain the number of positive and total cells. The percentage of positive cells (number of positive cells/total number of cells * 100) was calculated as the positivity rate (%).
Total RNA was isolated from liver using a mirVana RNA Isolation Kit (AM1561), and its yield and integrity were assessed using a NanoDrop 2000 spectrophotometer and ethidium bromide-stained agarose gel electrophoresis, respectively. RNA was converted to cDNA with the TransScript All-in-One First-Strand cDNA Synthesis SuperMIX for qPCR (quantitative polymerase chain reaction). qPCR testing was conducted on a LightCycler ® 480 II using PerfectStart Green qPCR SuperMix, with amplification conditions consisting of 94 °C for 30 s, followed by 45 cycles of 94 °C for 5 s and 60 °C for 30 s. Product specificity was confirmed by melting curve analysis from 60 °C to 97 °C. mRNA expression levels were normalized to GAPDH and calculated with the 2 −ΔΔCt method. Primers were designed and synthesized by Shanghai Ouyi Biomedical Technology Co. and Beijing Kengke Xinye Biotechnology Co, respectively (Table ).
Statistical data are represented as mean and standard deviation values. Group differences were assessed using one-way analysis of variance (ANOVA) followed by Tukey’s post-hoc test for pairwise comparisons. When data did not meet the assumptions of parametric tests, a Kruskal–Wallis analysis test was employed to assess differences across more than two groups. P -values < 0.05 were considered statistically significant.
Study design Figure shows the study process, while Table outlines the data sources. Initially, we screened 4,302 human protein-coding genes to evaluate their potential as drug targets. Independent cis-eQTL variants of human blood were then selected conditionally, and two-sample MR was performed to assess the association between drug-target mRNA expression and MASLD risk. MR results were simultaneously adjusted multiple times and validated in a second cohort. Colocalization analyses were conducted to test whether the MR results were affected by different causal variants owing to chain imbalance. Additionally, the drug target gene cis-eQTL of liver tissue from GTEx was used to conduct MR and SMR analysis to validate the significance of gene results of the blood eQTL MR analysis. Safety and alternative indications for the final targets were also evaluated via MR. Finally, RNA-seq was performed on liver biopsy samples, with immunohistochemistry and qPCR experimental methods used to validate the results of RNA-seq and colocalization. Discovery analysis A total of 2,644 druggable genes were identified by screening cis-eQTL data from the eQTLGen consortium. Subsequently, two-sample MR analyses were conducted on European MASLD patient data. In the discovery cohort, IVW meta-analysis was used in conjunction with effect estimates for each instrumental variable, with analysis performed on 8,434 patients and 770,180 controls from the meta-analysis of four cohorts. Ultimately, following a series of tests, it was determined that the predicted gene expression of four genes ( MFGE8 , CD33 , LILRA3 , and TEK ) was associated with an increased risk of developing MASLD ( P < 1.90E − 5 [IVW], 0.05 Bonferroni-corrected 2,644 drug targets, Table –8). Replication analysis Effect estimates for the four genes identified in the discovery phase were validated using the UK Biobank cohort ( N = 373,227). The validation results for two of the targets ( MFGE8 and CD33 ) were statistically significant ( P < 0.05 [IVW], Table , Table ), while the direction of effect remained consistent. Colocalization analysis To minimize the risk of linkage disequilibrium (LD) confounding MR results, gene colocalization analyses were conducted to assess the probability of sharing causal genetic variation in SNPs associated with MASLD and eQTL. Two genes exhibited strong evidence of colocalization (PP.H4 > 0.8). Specifically, colocalization of CD33 and MFGE8 with MASLD suggests their potential as drug target genes. The results indicated that MFGE8 and MASLD might share a causal variant within the MFGE8 locus (PP.H4 = 0.85, Fig. ), whereas CD33 in blood was highlighted as a candidate gene for MASLD risk (PP.H4 = 0.93, Fig. ). Consequently, MR and colocalization analyses identified two potential drug target genes, providing evidence of a shared genetic effect between eQTL and MASLD risk (Table ). Liver transcriptomics and pathological findings RNA-seq generated a total of 285.42 G of clean data. The effective data volume for each sample ranged from 6.69 to 7.08 G, with Q30 base distribution between 95.85 and 97.27% and an average GC content of 45.35%. By aligning the reads to the reference genome, the genomic alignment for each sample was obtained, with an alignment rate of 98.82–99.12%. Based on the alignment results, expression of protein-coding gene were analyzed, and differential screening was conducted based on the expression of protein-coding genes in different samples, with a total of one differential grouping (Tables – ). Principal component analysis revealed different liver transcriptome profiles between patients in the non-MASH and MASH groups (Figure ). Meanwhile, sample-sample distance clustering analysis showed that samples in the non-MASH group were more distinct from those in the MASH group (Fig. ). A total of 1,422 genes were significantly differentially expressed in the non-MASH group compared with the MASH group, with 695 genes upregulated and 727 genes downregulated. H&E and Masson staining images showed significant macrovesicular steatosis, hepatocyte ballooning degeneration, and inflammatory infiltration in the livers of patients with MASH (Fig. ). CD33 The MR analysis revealed a significant correlation between increased CD33 expression and a heightened risk of MASLD (odds ratio [OR] = 1.17, 95% confidence interval [CI]: 1.10, 1.25, P = 1.39 × 10 −6 ). Consequently, antagonists targeting CD33 are emerging as a promising therapeutic approach to mitigate the risk of MASLD. However, it is imperative to consider potential adverse effects and indications for use during the development of novel pharmacological agents. Consequently, an evaluation was performed to ascertain the causal relationship between CD33 gene inhibitors and 16 potentially modifiable risk factors, in addition to seven metabolism-related diseases. The evidence was inconclusive regarding the association between proxied gene inhibitors of CD33 and lipid metabolism markers, blood pressure, and glucose-related markers ( P > 2.17 × 10 −3 [IVW], 0.05/23 results, Fig. ). However, weak correlations were observed between proxied gene inhibitors of CD33 and Lp(a) ( P = 0.041 [IVW]). For metabolic syndrome-related diseases, genetically predicted CD33 inhibition was significantly negatively associated with atherosclerotic heart disease (OR = 0.997, 95% CI: 0.996–0.999, P = 1.67 × 10 −3 [IVW]) (Fig. ). Weak correlations were presented with essential hypertension (OR = 0.999, 95% CI: 0.998–1.000, P = 1.55 × 10 −2 [IVW]). To explore the possible relationship between CD33 and the risk of MASLD, gene and protein levels in liver tissues of patients with MASLD were examined. The results of liver transcriptomics showed that the mRNA of CD33 in liver tissues of patients in the MASH group was significantly higher than those of patients in the non-MASH group. Immunohistochemical staining showed that the percentage of CD33-positive cells was significantly higher in patients in the MASH group than in those in the non-MASH group. The results showed that the expression of CD33 was significantly higher in patients with MASH (Figs. and ). At the single-cell level, CD33 is expressed in immune cells (myeloid cells, lymphocytes, and erythrocytes) and endothelial cells, with some expression in hepatocytes ( https://singlecell.broadinstitute.org/singlecell ) (Figure ). MFGE8 MFGE8 is another druggable gene screened for the presence of a significance threshold using colocalization and hepatic transcriptomics analysis. MR analysis showed a correlation between reduced MFGE8 expression and increased risk of MASLD (OR = 0.89, 95% CI 0.85–0.94, P = 2.15 × 10 −6 ). Therefore, MFGE8 plays an inhibitory role in the development of MASLD. Significant negative associations were found between the surrogate gene agonists on MFGE8 and lipid indices of triglycerides ( P = 3.86 × 10 −3 [IVW]), ApoB ( P = 1.15 × 10 −4 [IVW]), and a positive association with HDL ( P = 9.64 × 10 −4 [IVW]). No significant associations were observed between surrogate gene agonists of MFGE8 and blood pressure, glucose, and weight-related markers ( P > 2.17 × 10 −3 [IVW], 0.05/23 results). Meanwhile, MR analyses of metabolic diseases showed that MFGE8 activation did not significantly increase the risk of metabolic diseases. The observed correlations between surrogate gene agonists of MFGE8 and atherosclerotic heart disease were weak ( P = 1.37 × 10 −2 [IVW] Fig. ). However, after Bonferroni correction, this association was no longer statistically significant. Furthermore, the RNA and protein expression levels of MFGE8 in human liver tissues were examined. The findings indicated that the RNA expression level of MFGE8 was lower in patients with MASH, suggesting that the level of MFGE8 is negatively correlated with the risk of MASH (Fig. ). Immunohistochemical staining showed that the percentage of MFGE8-positive cells was significantly higher in patients in the non-MASH group than in those in the MASH group. The results showed that the expression of MFGE8 was significantly lower in patients with MASH (Fig. ). At the single-cell level, MFGE8 is expressed in endothelial, epithelial, and hepatocyte cells ( https://singlecell.broadinstitute.org/singlecell ) (Figure ). MR results of eQTL of liver tissue from GTEx MR results of eQTL of liver tissue from GTEx (v10.0) were as follows: CD33 (OR = 1.195, 95% CI: 1.12–1.27, P = 1.05 × 10 −8 [IVW]); and MFGE8 (OR = 1.02, 95% CI: 0.997–1.03, P = 9.38 × 10 −2 [IVW]). The OR for MFGE8 was 1.02 (95% CI: 0.997–1.03), with a P -value of 0.0938. This suggests a non-significant 2% increase in disease risk per MFGE8 copy, as the CI includes one and the P -value is > 0.05, indicating no significant association with disease risk (Table ). SMR analysis findings for liver eQTLs from GTEx (v8.0 and v10.0) GTEx (v8.0) results are as follows: CD33 (OR = 1.077, 95% CI: 0.967–1.187, P = 5.20 × 10 −2 ) and MFGE8 (OR = 0.966, 95% CI: 0.847–1.084, P = 7.34 × 10 −2 ) (Table ). CD33 is relatively significant before FDR adjustment, and the direction is consistent with the results of the main analysis of the discovery set. However, after FDR adjustment, both are no longer significant. GTEx (v10.0) results are as follows: CD33 (OR = 1.847, 95% CI: 1.598–2.096, P = 3.53 × 10 −2 ); MFGE8: (OR = 1.185, 95% CI: 1.058–1.313, P = 2.55 × 10 −1 ). CD33 is relatively significant before FDR adjustment. However, the direction of MFGE8 is not consistent with the results of the main analysis of the discovery set. Moreover, after FDR adjustment, both were no longer significant (Supplement Table ). Sensitivity analysis results Sensitivity analysis results of MR results for eQTL of blood showed that CD33 and MFGE8 (Heterogeneity, steiger_direction, and MR presso tests) were robust (Supplement Table –7). In the liver tissue, CD33 remained robust, while MFGE8 exhibited some heterogeneity (Supplement Table ). The HEIDI test from SMR analysis showed that results of both the target genes were robust (Supplement Table ).
Figure shows the study process, while Table outlines the data sources. Initially, we screened 4,302 human protein-coding genes to evaluate their potential as drug targets. Independent cis-eQTL variants of human blood were then selected conditionally, and two-sample MR was performed to assess the association between drug-target mRNA expression and MASLD risk. MR results were simultaneously adjusted multiple times and validated in a second cohort. Colocalization analyses were conducted to test whether the MR results were affected by different causal variants owing to chain imbalance. Additionally, the drug target gene cis-eQTL of liver tissue from GTEx was used to conduct MR and SMR analysis to validate the significance of gene results of the blood eQTL MR analysis. Safety and alternative indications for the final targets were also evaluated via MR. Finally, RNA-seq was performed on liver biopsy samples, with immunohistochemistry and qPCR experimental methods used to validate the results of RNA-seq and colocalization.
A total of 2,644 druggable genes were identified by screening cis-eQTL data from the eQTLGen consortium. Subsequently, two-sample MR analyses were conducted on European MASLD patient data. In the discovery cohort, IVW meta-analysis was used in conjunction with effect estimates for each instrumental variable, with analysis performed on 8,434 patients and 770,180 controls from the meta-analysis of four cohorts. Ultimately, following a series of tests, it was determined that the predicted gene expression of four genes ( MFGE8 , CD33 , LILRA3 , and TEK ) was associated with an increased risk of developing MASLD ( P < 1.90E − 5 [IVW], 0.05 Bonferroni-corrected 2,644 drug targets, Table –8).
Effect estimates for the four genes identified in the discovery phase were validated using the UK Biobank cohort ( N = 373,227). The validation results for two of the targets ( MFGE8 and CD33 ) were statistically significant ( P < 0.05 [IVW], Table , Table ), while the direction of effect remained consistent.
To minimize the risk of linkage disequilibrium (LD) confounding MR results, gene colocalization analyses were conducted to assess the probability of sharing causal genetic variation in SNPs associated with MASLD and eQTL. Two genes exhibited strong evidence of colocalization (PP.H4 > 0.8). Specifically, colocalization of CD33 and MFGE8 with MASLD suggests their potential as drug target genes. The results indicated that MFGE8 and MASLD might share a causal variant within the MFGE8 locus (PP.H4 = 0.85, Fig. ), whereas CD33 in blood was highlighted as a candidate gene for MASLD risk (PP.H4 = 0.93, Fig. ). Consequently, MR and colocalization analyses identified two potential drug target genes, providing evidence of a shared genetic effect between eQTL and MASLD risk (Table ).
RNA-seq generated a total of 285.42 G of clean data. The effective data volume for each sample ranged from 6.69 to 7.08 G, with Q30 base distribution between 95.85 and 97.27% and an average GC content of 45.35%. By aligning the reads to the reference genome, the genomic alignment for each sample was obtained, with an alignment rate of 98.82–99.12%. Based on the alignment results, expression of protein-coding gene were analyzed, and differential screening was conducted based on the expression of protein-coding genes in different samples, with a total of one differential grouping (Tables – ). Principal component analysis revealed different liver transcriptome profiles between patients in the non-MASH and MASH groups (Figure ). Meanwhile, sample-sample distance clustering analysis showed that samples in the non-MASH group were more distinct from those in the MASH group (Fig. ). A total of 1,422 genes were significantly differentially expressed in the non-MASH group compared with the MASH group, with 695 genes upregulated and 727 genes downregulated. H&E and Masson staining images showed significant macrovesicular steatosis, hepatocyte ballooning degeneration, and inflammatory infiltration in the livers of patients with MASH (Fig. ).
The MR analysis revealed a significant correlation between increased CD33 expression and a heightened risk of MASLD (odds ratio [OR] = 1.17, 95% confidence interval [CI]: 1.10, 1.25, P = 1.39 × 10 −6 ). Consequently, antagonists targeting CD33 are emerging as a promising therapeutic approach to mitigate the risk of MASLD. However, it is imperative to consider potential adverse effects and indications for use during the development of novel pharmacological agents. Consequently, an evaluation was performed to ascertain the causal relationship between CD33 gene inhibitors and 16 potentially modifiable risk factors, in addition to seven metabolism-related diseases. The evidence was inconclusive regarding the association between proxied gene inhibitors of CD33 and lipid metabolism markers, blood pressure, and glucose-related markers ( P > 2.17 × 10 −3 [IVW], 0.05/23 results, Fig. ). However, weak correlations were observed between proxied gene inhibitors of CD33 and Lp(a) ( P = 0.041 [IVW]). For metabolic syndrome-related diseases, genetically predicted CD33 inhibition was significantly negatively associated with atherosclerotic heart disease (OR = 0.997, 95% CI: 0.996–0.999, P = 1.67 × 10 −3 [IVW]) (Fig. ). Weak correlations were presented with essential hypertension (OR = 0.999, 95% CI: 0.998–1.000, P = 1.55 × 10 −2 [IVW]). To explore the possible relationship between CD33 and the risk of MASLD, gene and protein levels in liver tissues of patients with MASLD were examined. The results of liver transcriptomics showed that the mRNA of CD33 in liver tissues of patients in the MASH group was significantly higher than those of patients in the non-MASH group. Immunohistochemical staining showed that the percentage of CD33-positive cells was significantly higher in patients in the MASH group than in those in the non-MASH group. The results showed that the expression of CD33 was significantly higher in patients with MASH (Figs. and ). At the single-cell level, CD33 is expressed in immune cells (myeloid cells, lymphocytes, and erythrocytes) and endothelial cells, with some expression in hepatocytes ( https://singlecell.broadinstitute.org/singlecell ) (Figure ).
MFGE8 is another druggable gene screened for the presence of a significance threshold using colocalization and hepatic transcriptomics analysis. MR analysis showed a correlation between reduced MFGE8 expression and increased risk of MASLD (OR = 0.89, 95% CI 0.85–0.94, P = 2.15 × 10 −6 ). Therefore, MFGE8 plays an inhibitory role in the development of MASLD. Significant negative associations were found between the surrogate gene agonists on MFGE8 and lipid indices of triglycerides ( P = 3.86 × 10 −3 [IVW]), ApoB ( P = 1.15 × 10 −4 [IVW]), and a positive association with HDL ( P = 9.64 × 10 −4 [IVW]). No significant associations were observed between surrogate gene agonists of MFGE8 and blood pressure, glucose, and weight-related markers ( P > 2.17 × 10 −3 [IVW], 0.05/23 results). Meanwhile, MR analyses of metabolic diseases showed that MFGE8 activation did not significantly increase the risk of metabolic diseases. The observed correlations between surrogate gene agonists of MFGE8 and atherosclerotic heart disease were weak ( P = 1.37 × 10 −2 [IVW] Fig. ). However, after Bonferroni correction, this association was no longer statistically significant. Furthermore, the RNA and protein expression levels of MFGE8 in human liver tissues were examined. The findings indicated that the RNA expression level of MFGE8 was lower in patients with MASH, suggesting that the level of MFGE8 is negatively correlated with the risk of MASH (Fig. ). Immunohistochemical staining showed that the percentage of MFGE8-positive cells was significantly higher in patients in the non-MASH group than in those in the MASH group. The results showed that the expression of MFGE8 was significantly lower in patients with MASH (Fig. ). At the single-cell level, MFGE8 is expressed in endothelial, epithelial, and hepatocyte cells ( https://singlecell.broadinstitute.org/singlecell ) (Figure ).
MR results of eQTL of liver tissue from GTEx (v10.0) were as follows: CD33 (OR = 1.195, 95% CI: 1.12–1.27, P = 1.05 × 10 −8 [IVW]); and MFGE8 (OR = 1.02, 95% CI: 0.997–1.03, P = 9.38 × 10 −2 [IVW]). The OR for MFGE8 was 1.02 (95% CI: 0.997–1.03), with a P -value of 0.0938. This suggests a non-significant 2% increase in disease risk per MFGE8 copy, as the CI includes one and the P -value is > 0.05, indicating no significant association with disease risk (Table ).
GTEx (v8.0) results are as follows: CD33 (OR = 1.077, 95% CI: 0.967–1.187, P = 5.20 × 10 −2 ) and MFGE8 (OR = 0.966, 95% CI: 0.847–1.084, P = 7.34 × 10 −2 ) (Table ). CD33 is relatively significant before FDR adjustment, and the direction is consistent with the results of the main analysis of the discovery set. However, after FDR adjustment, both are no longer significant. GTEx (v10.0) results are as follows: CD33 (OR = 1.847, 95% CI: 1.598–2.096, P = 3.53 × 10 −2 ); MFGE8: (OR = 1.185, 95% CI: 1.058–1.313, P = 2.55 × 10 −1 ). CD33 is relatively significant before FDR adjustment. However, the direction of MFGE8 is not consistent with the results of the main analysis of the discovery set. Moreover, after FDR adjustment, both were no longer significant (Supplement Table ).
Sensitivity analysis results of MR results for eQTL of blood showed that CD33 and MFGE8 (Heterogeneity, steiger_direction, and MR presso tests) were robust (Supplement Table –7). In the liver tissue, CD33 remained robust, while MFGE8 exhibited some heterogeneity (Supplement Table ). The HEIDI test from SMR analysis showed that results of both the target genes were robust (Supplement Table ).
With the rising prevalence of obesity and its associated metabolic sequelae, MASLD has emerged as the predominant chronic liver condition worldwide . The etiopathogenesis of MASLD is multifaceted and not completely elucidated, contributing to a lack of effective pharmacotherapies. Therefore, the identification of pathogenic targets for MASLD is critical for the advancement of tailored treatment approaches. In this extensive MR study comprising 4,761 individuals with MASLD and 373,227 controls, GWAS data, pharmacogenomic information, and gene expression profiles were investigated. The study findings indicate that inhibition of CD33 gene function and activation of MGFE8 are inversely associated with the risk of MASLD. Furthermore, it was observed that the genetically predicted inhibition of CD33 is significantly inversely correlated with atherosclerotic heart disease. CD33, a member of the sialic acid-binding immunoglobulin-like lectin (Siglec) family, is a cell surface protein that mediates cell-cell interactions and immune regulation. As a transmembrane protein, CD33 is expressed across hematopoietic and phagocytic cells, such as microglia, dendritic cells, monocytes, macrophages, granulomonocyte precursors, and hematopoietic progenitor cells . However, its precise associations with diseases is not fully elucidated. CD33 is highly expressed in leukemia blasts and myeloid leukemia initiating cells, yet it is not present in primitive stem cells and multipotent progenitors. While expressed in common myeloid precursors, CD33 levels are reduced in mature granulocytes and circulating macrophages, making it a promising target for immunotherapy in acute myeloid leukemia (AML) . CD33 is expressed on mature myeloid cells and hematopoietic progenitors, necessitating caution to prevent non-tumor toxic effects when developing CD33 antibody-based therapies. Mylotarg (Gemtuzumab Ozogamicin, GO) is an antibody-drug conjugate (ADC) that consists of an anti- CD33 IgG4 monoclonal antibody linked to calicheamicin. It targets the CD33 antigen on AML cells, is internalized, and releases calicheamicin to eradicate the leukemia cells, extending survival in certain patients . Currently, researchers are developing more effective CD33 -targeted therapeutic agents, such as antibodies against different epitopes of CD33 . The CD33 gene, expressed on brain microglial cells and linked to Alzheimer’s disease (AD) risk, regulates processes such as cytokine release and immune cell growth. CD33 might facilitate AD onset through hindering microglial clearance of β-amyloid . Research suggests AD starts with lysosomal autophagy dysfunction in neurons, leading to β-amyloid deposits . Currently, no CD33 -related drugs for the treatment of AD exist. However, the complexity of CD33 -expressing isoforms and safety concerns related to cytotoxicity have limited the ability to effectively target and use the drug. Some studies indicate that CD33 is an immune checkpoint receptor for HBV-induced immune tolerance . The multifaceted role of CD33 in various health and disease states underscores its importance in immune regulation. Regarding MASLD, a notable aspect of CD33 ’s function is its association with myeloid-derived suppressor cells (MDSCs), which serve as a key regulator of hepatic lipid metabolism. MDSC counts increase with lipid accumulation, and those derived from polymorphonuclear cells in the livers of obese mice have been shown to enhance liver lipid deposition through secreting pro-inflammatory factors (e.g., S100A9), disrupting normal liver lipid metabolism . In humans, both polymorphonuclear cell-derived MDSCs (PMN-MDSCs) and monocyte-derived MDSCs (M-MDSCs) express CD33 . It is inferred that the increased expression of CD33 in MASLD might be related to the increase in myeloid-derived suppressor cells owing to long-term chronic inflammation. A preprint study has identified four novel genetic targets for MASLD, including CD33 , by analyzing 2,941 plasma proteins from 43,978 individuals in the UK Biobank. Integrating human genetics, transcriptomics, liver imaging, and biopsy proteomics, the research aligns with our results. It indicates that the association of CD33 to MASLD is mediated through body weight, with pathway analysis implicating CD33 in the regulation of interleukin-1beta production . The precise role of CD33 in the pathogenesis of MASLD warrants further exploration. Despite the existence of CD33 -targeted therapies, the relationship between CD33 and MASLD has not been fully investigated. The findings propose CD33 as a conceivable therapeutic target for MASLD intervention. MFGE8 was first discovered in the milk fat globules of mouse mammary epithelial cells and functions as a secreted cellular protein. MFGE8 is an anti-inflammatory factor and is involved in a variety of physiological processes, including fertilization , angiogenesis , innate immunity , and tumorigenesis . Studies have shown that MFGE8 expression is associated with myocardial hypertrophy, coronary atherosclerosis , and anti-inflammatory effects such as post-operative injury , and sepsis . MFGE8 is an endogenous inhibitor of inflammation-induced IL-1β production and inhibits macrophage-induced necrotic cell and ATP-dependent IL-1β production by inducing the interaction of integrin β3 and P2 × 7 receptors in cells. Zhang Lei et al. reported that MFGE8 expression is decreased in patients with MASLD and in mouse models. During the process of lipid accumulation in hepatocytes, MFGE8 was shown to negatively regulate inflammation and lipid accumulation through binding to apoptosis signal-regulating kinase 1 (ASK1) and inhibiting its activation. The MR analysis suggests a causal association between MFGE8 agonists and lipid metabolism markers (TG, HDL, apo B). Under metabolic stress, a decrease in MFGE8 promotes ASK1-dependent inflammation, suggesting that MFGE8 may serve as a potential therapeutic target for preventing the progression of MASLD, which is consistent with our study findings. MFG-E8 participates in a series of inflammatory processes mediated by the liver-spleen axis. Some studies showed that tingible body macrophages in the germinal centers of the spleen and lymph nodes strongly express MFG-E8, and that the MFG-E8–/– mice developed splenomegaly, with the formation of numerous germinal centers . Tarantino G reported all process is in agreement with the well-ascertained chronic low-grade inflammation, characteristics of NAFLD, key-process mediated by the liver spleen axis . In this process, there are also some similar targets such as S1009, nd S100A8 and S100A9 heterodimers, along with S100A9 homodimers, were formed after an inflammatory challenge in protein extracts from the spleen .
The complexity of MASLD pathogenesis, which involves various pathophysiological processes and multiple factors across various tissues and organs, presents a significant challenge in identifying suitable pharmacological intervention targets. This study focused on druggable genes to improve the efficacy, safety, and success rates of drug development for MASLD, identifying CD33 and MFGE8 as potential targets. Additionally, this population-based research suggests that these proteins could be crucial in the development of future therapeutic targets for MASLD. MR analysis was also conducted to address potential safety concerns and explore alternative indications, which are crucial for their clinical application. While this study offers valuable insights, it has some limitations. A population overlap was observed between the validation and discovery sets, slightly weakening the validation effect. However, MR and SMR analysis of the liver tissue from GTEx was performed. The relationship between trans-eQTLs and disease was not analyzed, although cis-eQTLs comprise an integral part of quantitative loci that regulate gene expression. The challenge in conducting MR analysis to fully mitigate the potential influence of horizontal pleiotropy should be acknowledged, as well as potential residual pleiotropy and the inability to fully capture gene-environment interactions. The investigation of side effects was confined to outcomes related to metabolic disorders, necessitating future research to broaden this scope and address potential systemic adverse effects. The study’s sample population in the MR section was restricted to individuals of European descent, limiting the generalizability to other populations. Sensitivity analysis showed that only the result for MFGE8 among all MR results for eQTLs of liver tissue (GTEx) exhibited some heterogeneity, likely due to the small GTEx sample size.
Given the lack of effective pharmacological interventions for MASLD, developing novel therapeutics targeting CD33 and MFGE8 is critical. Future research is needed to determine whether CD33 inhibitors and MFGE8 activators produce consistent effects in preclinical and clinical settings, and whether the expression regulation of the two targets would produce the expected clinical results through the application of computer simulation and molecular design technology, as well as their fundamental biological pathways.
Below is the link to the electronic supplementary material. Supplementary Material 1 : Supplementary file S1 RNA extraction and library preparation. Supplementary Material 2 : Fig. S1 . Liver transcriptome principal component analysis. Supplementary Material 3 : Fig. S2 . Single cell sequencing analysis. Supplementary Material 4 : Table S1 . Data sources for MASLD in this study. Supplementary Material 5 : Table S2. Data sources for the analysis. Supplementary Material 6 : Table S3. Primer list. Supplementary Material 7 : Table S4. Supplementary MR results. Supplementary Material 8 : Table S5. Summary of the results of genetic variants in significant genes. Supplementary Material 9 : Table S6. Heterogeneity test results. Supplementary Material 10 : Table S7. Horizontal pleiotropy test results. Supplementary Material 11 : Table S8. Steiger test results. Supplementary Material 12 : Table S9 . External validation MR result. Supplementary Material 13 : Table S10. Colocalization results. Supplementary Material 14 : Table S11. Population characteristics and comparison between subjects with and without MASH. Supplementary Material 15 : Table S12. Transcriptomics raw data. Supplementary Material 16 : Table S13. MR results of liver eQTL. Supplementary Material 17 : Table S14. SMR analysis results of liver eQTL.
|
Current Insights on Bioactive Molecules, Antioxidant, Anti-Inflammatory, and Other Pharmacological Activities of Cinnamomum camphora Linn | 536cd45d-47b4-44be-8e51-01f8a4480b0c | 9568346 | Pharmacology[mh] | The WHO stated that around 80% global population uses different types of traditional medicine to treat many diseases . Traditional medicine is the knowledge, skills, and procedures that indigenous peoples and other cultures have used for a long time in order to preserve health and avoid illness . One of the oldest forms of medicine is Unani medicine, which originated in Greek and is based on seven essential factors responsible for the maintenance of health and imbalance any one of them can lead to disease or even to death. The great physicians, Avicenna and Galan, stated that the primary elements contribute to the formation of things in nature . The intermixture of primary elements forms temperament. The temperament indicates the state of equilibrium to the number of elements and the ratio of the particular compound and different combination responses to its specific nature. Hence, any changes in the quality or quantity of humor alter the equilibrium and disturb the normal temperament. Proper physiological functions are maintained by the homeostasis of temperament. Simple temperamental imbalances cause intrinsic power (immunity) to fight back and maintain body normal levels. If the body's temperament, functions are impaired owing to a change in humor, diet therapy, pharmacotherapy, regimental therapy, or surgery may be required, depending on the situation . Temperament ( Mizaj ) and humor, which is the principal concept in this system and disturbances in the quality and quantity of humor, cause numerous conditions. The four main modes of therapy available in Unani medicine are regimental diet therapy, pharmacotherapy, and surgery. Correction through pharmacotherapy and regimental therapy helps to maintain homeostasis of the humor . First-degree drugs are safer, temperamental quality is less, and second-degree drugs are also safe and have strong temperamental quality but no toxic effect. Third-degree temperamental medicine drugs are strong, the toxic effect may manifest prominently, and fourth-degree drugs are excessively strong and toxic. The drug is moderate ( Mutadil ) in temperament; it has no toxic, no temperamental quality, and the activity is only limited medicinal effect. Commonly, cold temperamental drugs are suitable for hot temperamental individuals and may produce effects on cold temperamental individuals in inappropriate doses. Cinnamomum camphora (L.) is a traditional Unani medicinal plant with a third-degree cold and dry temperament and hence useful in hot temperamental individuals . C. camphora (L.) is a renowned Unani medicinal herb applied for several disease conditions in Unani as well as other traditional medicines. Camphor is a terpene ketone derived from C camphora wood or synthetically produced from turpentine. White, yellow, brown, and blue camphor oil are the four different fractions of camphor oil . Camphor is obtained by distillation with water from the wood of trees or plants and purified by sublimation, and it occurs in translucent white crystals . Since the ancient era in the Unani traditional system of medicine, C. camphora has been using its ethnomedicinal properties like antiseptic, analgesic, and rubefacient properties. Camphor has been use for very long time in various traditional systems of medicine such as Ayurveda, Unani, Siddha, and Chinese. It has been used in Unani medicine mainly in respiratory disorders (Amrāz-i-Riyah) , gastrointestinal ( Amrāz-i-Me‘da wa Am'a ), integument disease (Amrāz-i-Jild ), eye diseases (Amrāz-i-Ayn) , and nervine and cerebral disorders (Amrāz-i- damaghi wa a'sabi ) especially in hot conditions for headache, strengthening senses and brain , bilious diarrhea , inflammation of the liver , and useful in bladder and kidney inflammation . Furthermore, externally, it is used for various ailments such as eye diseases, ear pain, joint, muscular pain, chest congestion, and headache applications such as ear drops or gargling with or without other suitable drugs . An overdose may result in systemic toxicity. Signs of intoxication include gastrointestinal pain, emesis, agitation, tremors, and convulsions, which are followed by CNS depression marked by apnea and coma . Furthermore, it is an important ingredient of Arq Ajeeb used as a prophylactic medicine for COVID-19 as per AYUSH guidelines. This comprehensive appraisal is to familiarize the reader towards the extensive, well recognized, and broad applications of camphor both in Unani and contemporary applications. To collect information on C. camphora for its temperament ( Mizaj ), adverse effects ( Mudir ), corrective ( Muslih ), substitute ( Badal ), ethnomedicinal properties ( Afa'l ), Unani compound formulations, and ethnomedicinal therapeutic uses, a literature survey of traditional Unani texts was conducted. Additionally, full-text paper and thorough search of electronic databases such as PubMed, Scopus, Google Scholar, and Research Gate were conducted to gather all accessible information on phytochemical, physicochemical, and pharmacological investigations relevant to C. camphora . All relevant articles are written in English up to 2022. The search occurred between August 2020 and May 2021. The keywords used were as follows: “ C. camphora ,” “chemical component,” “Unani Medicine,” “ Kafoor ,” “preclinical studies,” “clinical trial,” “phytochemical,” “adverse effect,” “toxicity,” and “traditional.” Chemical structure images were taken by PubChem. Standard Unani Medical Terminology of WHO was reviewed to define the suitable Unani terminologies. The scientific name and synonyms were authenticated and reproduced using The Plant List ( http://www.theplantlist.org ). A total of 464 papers and 21 books were retrieved, 386 were excluded, and we included research and review papers from the electronic database . Twenty-one included Unani classical manuscripts, and herbal pharmacopeial texts were consulted, including the incorporation of Urdu translation of the traditional textbooks such as Makhzan al-Mufridat , Al Jami ul Mufradat Al Advia Wal Aghzia (1197-1248 AD), Muhit-i-A'zam (1806–1902 AD), Khazainul Adwiya , (19th century), and Bustan ul-Mufridat , Kitab ul Mansuri (850-925 AD). We collected data from traditional, classical Unani and herbal pharmacopoeia literature and up-to-date reviews and research to address a traditional and contemporary overview of the application of C. camphora in various ailments. We conducted this review to report most of the information on C. camphora therapeutic and traditional uses for several diseases. In addition, we also included a comparison between therapeutic traditional uses and its current research to prove its ethnomedicinal properties. Furthermore, the mechanism of natural bioactive molecules isolated from C. camphora was also highlighted. Preclinical and clinical trials were also reviewed to prove the effect of C. camphora in various diseases. Because no previous publications have incorporated this type of information in the review article, this review aims to overview and analyze the taxonomy, distribution, macroscopic description of the plant, various ethnomedicinal properties, therapeutic Unani applications, natural bioactive molecules isolated from different parts of the plant, present mechanism of action of natural bioactive molecules, comparison of therapeutic Unani applications proven currently by preclinical and clinical studies, gap, and future recommendation. Hence, the following are the primary contributions of this study: Traditional Unani overview of C. camphora plants such as temperament, ethnomedicinal properties, therapeutic applications in various disease conditions, adverse effects, corrective, substitute, and Unani compound formulations containing C. camphora with dose and therapeutic applications Various natural bioactive molecules separated from various parts of the C. camphora plant with their structures and their mechanisms of action depiction such as analgesic, anti-inflammatory, antioxidant, and anti-allergic We also highlighted current research carried out in vitro, in vivo, and silico pharmacological studies such as antioxidant, anti-inflammatory, anti-allergic, antibacterial, antifungal activity, anxiolytic and antidepressant, analgesic, anti-hyperlipidemic, antifertility, hepatoprotective, antifertility, wound healing, prostaglandin synthesis inhibition, and oestrogenic activities We included preclinical and clinical studies of the main active biomolecules to report the significant use of C. camphora in day-to-day life. It also has a prophylactic effect against the SAR-CoV virus as per the study, and it is one of the main ingredients of Arq Ajeeb compound formulation used as a prophylactic inhaler to prevent COVID-19 infection This review also contributes toward comparative therapeutic evaluation, research gaps, future recommendations, and conclusions C. camphora The vernacular name summarized in . Camphor tree is a shrub or an evergreen tree belonging to the family Lauraceae, Laurales order, genus Cinnamomum, and species camphora . Over 250-300 species of the genus are distributed globally . Twenty-six species are found in India, and approximately 40 species are commonly used for medical conditions. Leaves and stem bark are the sources of medicinal activity. Cinnamomum zeylanicum , C. camphora , C. burmannii , C. cassia , C. tamala , and C. verum species are rice sources of aromatic oil . Camphor ( Kafoor ) is an exudate of a camphor tree as per the description in Unani literature. Moisture or liquid comes out from cracks or incision in the tree and freezes out as rust. It is also expelled through the hole of the tree , or the wood of that tree is chopped and soaked and heated in the water causing sublimation . It is clear crystal white with a strong smell . Camphor is found naturally or artificially synthesized. The natural camphor is D-camphor, whereas the synthetic one is L-camphor . Authentic Unani texts described numerous varieties of camphor. The best type is Kaisuri followed by Riyahi. Kaisuri is found in the city of Qaisr on the island of Tarindib , so the name has been given Kafuri Kaisur . There are three different kinds of camphor: Formosa camphor , Borneo or Barus camphor, and blumea or Ngai camphor. Borneo camphor is high priced, and it is naturally formed in the stems of Dryobalanops camphor grown in Dutch Sumatra and sinks in water. This is considered the best type. Borneo Camphor is high priced, and it is naturally formed in the stems of Dryobalanops camphor grown in Dutch Sumatra and sinks in water, and the third type is Blumea or Ngai camphor. Its opposite, frequently three-nerved, long petiolate, oblong or ovate, 5- to 12.5-cm long, and 2.5 to 5-cm broad leaves are usually three-nerved. Its flowers are tiny and hermaphrodite or produced via polygamy or abortion. Typically, females are larger and have few components. Nine stamens are there, unless they are aborted. The ovary is sessile, free from the perianth, style is narrow, the stigma is discoid, and the style is narrow or obscurely 3-lobed . The fruit is a berry with spreading, somewhat expanded perianth, completely or partially deciduous segments, and less frequently persistent seeds . 4.1. Temperament (Mizaj) Temperament is one of the unique features and fundamental principles of the Unani system of medicine . All medicinal substance, plants, animals, and minerals have their temperament. The temperament of drugs is used as a tool to assess the actions and toxicological properties of Unani drugs. The medications were divided into four groups based on their innate nature: hot ( Hārr ), cold ( Bārid ), moist ( Ratb ), and dry ( Yābis ) in terms of their effect on a moderate human body and four degrees 1, 2, 3, or 4 in terms of increasing intensity of action. As a result, different scholars have claimed that its temperament is variable . The temperament of camphor is cold and dry in the third degree , cold second, and dry in the third-degree . 4.2. Ethnomedicinal Properties (Afa'l) Camphor, administered orally, has several pharmacological properties such as expectorant (Munaffith-i-Balgham) ; stimulant (Muharrik) ; brain and heart tonic (Muqawwi-i-Qalb wa Dim a ¯ gh) ; exhilarant of brain and heart (Mufarrih al- Qalb wa Dim a ¯ gh) ; antipyretic (D a ¯ fi-i-Humm a ¯ ) ; hemostatic ( H ˙ a ¯ bis-i-Dam); anti-pyretic for tubercular infection ( H ˙ umm a ¯ Diqq) (a form of fever that gradually depletes the body's fluids and weakens its organs, resulting in weight loss) ; antispasmodic (D a ¯ fi?-i-Tashannuj) ; astringent (Q a ¯ bi d ˙ ) ; disinfectant (Mani'-i-'Ufunat) ; anaphrodisiac (Duf-i-Bah- ; constipation (Qabz) ; and carminative, reflux expectorant, which stimulate the heart and respiratory system and analgesic and sedative to the nervous system . Additionally, the overview of ethnomedicinal properties, therapeutic applications in Unani medicine, and pharmacological activities are mentioned in ( https://www.ayurtimes.com/cinnamomum-camphora/3/6/2022 and https://plant.ces.ncsu.edu/plants/cinnamomum-caomphora/ ). 4.2.1. Topical Application Topical application of camphor has antiseptic and massages externally and initially; it has stimulant rubefacient (Muhammir) and then anesthetic (Mukhaddir) and analgesic (Musakkin-i-Alam) ethnomedicinal properties . 4.3. Therapeutic Application as per Unani System of Medicine As per Unani system of medicine, Camphor is useful in respiratory, gastrointestinal, musculoskeletal, integumentary, oral cavity, ear condition, eye diseases, and other general conditions . Temperament is one of the unique features and fundamental principles of the Unani system of medicine . All medicinal substance, plants, animals, and minerals have their temperament. The temperament of drugs is used as a tool to assess the actions and toxicological properties of Unani drugs. The medications were divided into four groups based on their innate nature: hot ( Hārr ), cold ( Bārid ), moist ( Ratb ), and dry ( Yābis ) in terms of their effect on a moderate human body and four degrees 1, 2, 3, or 4 in terms of increasing intensity of action. As a result, different scholars have claimed that its temperament is variable . The temperament of camphor is cold and dry in the third degree , cold second, and dry in the third-degree . Camphor, administered orally, has several pharmacological properties such as expectorant (Munaffith-i-Balgham) ; stimulant (Muharrik) ; brain and heart tonic (Muqawwi-i-Qalb wa Dim a ¯ gh) ; exhilarant of brain and heart (Mufarrih al- Qalb wa Dim a ¯ gh) ; antipyretic (D a ¯ fi-i-Humm a ¯ ) ; hemostatic ( H ˙ a ¯ bis-i-Dam); anti-pyretic for tubercular infection ( H ˙ umm a ¯ Diqq) (a form of fever that gradually depletes the body's fluids and weakens its organs, resulting in weight loss) ; antispasmodic (D a ¯ fi?-i-Tashannuj) ; astringent (Q a ¯ bi d ˙ ) ; disinfectant (Mani'-i-'Ufunat) ; anaphrodisiac (Duf-i-Bah- ; constipation (Qabz) ; and carminative, reflux expectorant, which stimulate the heart and respiratory system and analgesic and sedative to the nervous system . Additionally, the overview of ethnomedicinal properties, therapeutic applications in Unani medicine, and pharmacological activities are mentioned in ( https://www.ayurtimes.com/cinnamomum-camphora/3/6/2022 and https://plant.ces.ncsu.edu/plants/cinnamomum-caomphora/ ). 4.2.1. Topical Application Topical application of camphor has antiseptic and massages externally and initially; it has stimulant rubefacient (Muhammir) and then anesthetic (Mukhaddir) and analgesic (Musakkin-i-Alam) ethnomedicinal properties . Topical application of camphor has antiseptic and massages externally and initially; it has stimulant rubefacient (Muhammir) and then anesthetic (Mukhaddir) and analgesic (Musakkin-i-Alam) ethnomedicinal properties . As per Unani system of medicine, Camphor is useful in respiratory, gastrointestinal, musculoskeletal, integumentary, oral cavity, ear condition, eye diseases, and other general conditions . C. camphora in the Traditional and Contemporary Era Unani classical texts mention doses range from minimum to 182 up to 250 mg to 550 mg. According to another opinion 7 gm/week , the maximum dose is 364 mg to 728 mg. More than 8.75 gm reduces sexual power or may cause death . The minimum dose is 182 to 364 mg and can be given for strengthening the patient . The lethal dose in adult humans is 5 to 20 g. One teaspoon of camphorated oil (~1 mL of camphor) was lethal to 16 and 19-month-old children . Unani scholars stated that overdose or misuse of this drug may adversely affect cold temperamental people ( Barid Mizaj) , stagnation of sperm ( Munjamid Mani ), and weakness in individual's stomach that reduces sexual power and sperm quality and forms kidney stone . Camphor is quickly absorbed by the mucous membranes, skin, and gastrointestinal tract in liquid form. Symptoms may appear 5–90 minutes after consumption. The rate of absorption is heavily reliant on the existence of food and other compounds . In humans, intoxication signs are abdominal distress, emesis, tremors, excitement, seizures, and CNS depression characterized by apnea and coma . The traditional Unani medicine discusses various corrective agents ( Muslih ) and forms of simple and compound drugs that can use according to the condition internally or externally to manage or minimize the adverse effects of camphor. As camphor is cold temperament and to combat it coldness, hot and aromatic herbs such as oil of Viola odorata L. ( Rogan-i-Banafsha ), oil of Narcissus tazetta Linn ( Rogan-i-Nargis ), Ambar , and Castoreum ( Jundabedastar ) are used as corrective. Oil of Iris ensata Thunb ( Roghan-i-Sosan ), a flower of Viola odorata L. ( Gul-i-Banafsha ) , and a confection of Rosa damascene act as a corrective in a condition such formation of renal stone caused by camphor use . Narcissus tazetta L. ( Banafsha ), Nelumbo nucifera Gaertn ( Niloufer ), Crocus sativus , a confection of Rosa damescene ( Gulkand ), Ambar , and Musk are drugs that act as corrective in headache caused by use of camphor . Badal ) The idea of drug substitution ( Abdāl-i-Adwiya ) is a significant criterion of Unani pharmacotherapy . In Unani Medicine, replacing the main drug with a substitute having the same or closest pharmacological action with the first desired drug and a substitute can be chosen depending on the situation. Need for drug substitutes, Al-Razi surmised that “frequently, all drugs needed for treatment are not easily available everywhere. As a result, if a physician is uninformed of the replacements that must be used in place of the principal drug, the medical profession's objectivity will be compromised” . The great physician, Avicenna states that a substitute can be used “When the initially intended medicine is unavailable” . In case of non-availability of Camphor, Barbarea vulgaris ( Tabashir Sufaida ) or Pterocarpus santalinus ( Sandal ) or fossil resin of Pinus succinifera ( Kahruba ) can use as a substitute. Murakkabat ) Compound medicines are pharmaceuticals that contain two or more herbs as ingredients in a variety of dose formulations and administration routes. Topical preparations include ointment, lotions, and fine powders for ocular use. Oral preparations include pills, tablets, powder, and semisolid confection forms. Several Unani compound formulations as per pharmacopoeia preparation possessing different ethnomedicinal properties, therapeutic applications, and dosage forms with doses acting on different body systems have been described in detail in . C. camphora and their Mechanism of Action Seventy-four compounds were discovered in leaf, branch, wood, and root chromatograms of C. camphora tissues . Phytochemical components of C. camphora are phenolics, flavonoids such as tannins (2.09%), saponins, alkaloids (3.85%), and carbohydrates . The bioactive compound of C. camphora oil identifies the relevant analgesic effects β -caryophyllene, α -caryophyllene, germacrene D, bicyclogermacrene, unidentified, nerolidol, spathulenol, and unidentified (E)- α -atlantone . Another study identified 96 various compounds in the essential oils by two-dimensional gas chromatography such as methyl isobutyl ketone, pinene ( α and β ), α -thujene, camphene, sabinene, α -phellandrene, hexanal, 3-hexanal-1-ol, 1-hexaol, and sabinene . The major bioactive molecules of C. camphora are camphor, linalool, safrole, and cineole. The major bioactive molecules in different parts of the plant are mentioned in and . The pathophysiology of respiratory diseases (sinusitis, asthma, bronchitis, and COPD) is mucociliary dysfunction, inflammation-induced edema, and hypersecretion of goblet cells that probably plays an important role. Cineole natural substance from camphor has pharmacological properties that are known to reduce inflammation, secretion of goblet cells decreases, the ciliary beat frequency is sped up, and bronchodilatory and mucolytic properties hence help in the drainage of sinuses and other respiratory organs . Consequently, cineole would be therapeutically beneficial for asthma and bronchitis patients based on its proven broncho-dilating and anti-inflammatory effects. Other studies also have proven the effect on cineole . A new elucidation for the analgesic application of camphor is the combination of transient receptor potential A1 (TRPA1) inhibition and desensitization of TRPV1. Camphor activates and then desensitizes TRPV1, thereby having an analgesic action. Linalool bioactive molecules showed pain reduction in mouse models such as inflammatory pain, acetic acid-induced writhing response, and the hot plate test. The likely mechanism perhaps is related to its regulation of NMDA receptors and suppression of pro-inflammatory cytokines . In a clinical investigation, topical borneol treatment dramatically reduced pain compared to placebo. Furthermore, an in vivo study in mice that exhibited TRPM8 channels may perhaps be the molecular target of borneol . C. camphora natural bioactive molecule menthol after topical application (skin, mucous membrane, and oral and nasal cavities) also activates TRPA1, s highly sensitive menthol receptor that contributes in counterirritants and analgesic activities. This suggests the involvement of different kinetics of channels and fast desensitization due to these sensory effects of menthol [46]. Clinical investigations and contemporary medical experiments in migraine and vascular headache demonstrated the crucial role of nitric oxide (NO). NO and nitric oxide synthase (NOS) inhibitors can significantly reduce the severity, frequency, intensity, and accompanying symptoms of migraine attacks. A protein complex known as nuclear factor-kappa β(NF-B) regulates DNA transcription, cytokine synthesis, and cell viability. NF- κ B induces the expression of inducible nitric oxide synthase (iNOS), and overexpression of iNOS can catalyze L-arginine to yield nitric oxide (NO). Proinflammatory factors are also activated by NF- κ B and then cause a neurogenic inflammation reaction, which sensitizes the pain center and initiates headache. In the pathogenesis of migraine, NF- κ B has a significant mediating role. The essential oil of camphor leaves in the mouse model showed noteworthy analgesic action on migraine by inhibiting the nuclear factor-kappa B (NF- κ B)/inducible nitric oxide synthase (iNOS) pathway and reduced neurogenic inflammation. Essential oil of C. camphor perhaps inhibits the iNOS and expression of NF- κ B and therefore decreases NO production and neurogenic inflammatory response . Therefore, it could treat migraine. The main analgesic compounds recognized in the camphor's essential oil camphor leaves were nerolidol and (E)- α -atlantone . Citronellol, which affects cyclooxygenase (COX) 1 and 2, is the enzymes involved in the production of prostaglandins from arachidonic acid and decreases the production of inflammatory mediators which is related to its ability to reduce cell migration and paw edema . Markel et al. investigated the influence of camphor on the expression of oestrogenic genes. They discussed how the UV filter 4-methyl benzylidene camphor (4-MBC) is oestrogenic and interferes with the thyroid axis. They discovered that, in rats, exposure to 4-MBC altered the mRNA levels of ER-alpha, progesterone receptor (PR), preproenkephalin (PPE), and insulin-like growth factor-I (IGF-I) in the brain in a sex- and region-specific manner. Methanolic extract of C. camphora contains anti-inflammatory mechanisms that limit NO and PGE2 synthesis in LPS/IFN-activated macrophages and prevent the generation of TNF IL-6 and IL-1 from RAW264.7 cell . The proanthocyanidins (PAs) in the leaves of C. camphora inhibit tyrosinase monophenolase and hence have been proven to have anti-tyrosinase activity. The PAs also showed strong antioxidant capacity with the ferric reducing antioxidant power (FRAP), scavenging 2,2-diphenyl-1-picrylhydrazyl (DPPH) and 1,2′-azino-bis (3-ethylbenzthiazoline-6-sulphonic acid) (ABTS) assays . The preclinical studies of the main natural active biomolecules are summarized in . summarizes the preclinical and clinical studies of the compound formulation of camphor. 9.1. In Vitro Pharmacological Properties Numerous in vitro experimental studies show that antioxidant, antimicrobial, anti-inflammatory, and miscellaneous activities have been demonstrated by numerous research on C. camphora . 9.1.1. Antioxidant Activity By interacting with biological components within the cell, the oxidation process damages cells, resulting in a variety of illnesses and chronic diseases like cancer and cardiovascular conditions. Additionally, oxidation changes the nutritional value and safety of food by producing secondary reaction products . Oxidative stress is a condition when antioxidant levels are low. Antioxidant activity of polyphenols, which is influenced by their polyphenolic structure, has the effect of removing free radicals and improving antioxidant activity . Due to an excess of reactive oxygen species (ROS), oxidative stress develops when the body's antioxidant system becomes depleted. Due to this, an increase in the concentration of free radicals inside cells is the root cause of many chronic disorders such as nonalcoholic fatty liver, type 2 diabetics, neurological conditions, and reproductive-related problems . Reactive oxygen species (ROS) at baseline levels are necessary for basic physiological activities. Ageing, obesity, type 2 diabetes mellitus (T2DM), depression, and neurodegeneration are all conditions that are significantly impacted by oxidative stress . Camphor has antioxidant, hepatoprotective, antidepressant, estrogenic, and anti-inflammatory qualities in addition to being an antioxidant that prevents oxidative damage and neutralizes free radicals. Camphor's antioxidant capabilities may lessen tissue damage and oxidative stress. In scavenging DPPH, ABTS, and ferric reducing antioxidant power (FRAP) assays, the phytochemical proanthocyanidins (PAs) from leaves and branches of C. camphora displayed significant antioxidant activity . The antioxidant activity of hexane, chloroform, and ethanol extracts was determined using the DPPH (2,2-diphenyl-1-picrylhydrazyl) technique on dried camphor leaves. Another study evaluated the antioxidant activity of leaves of C. camphora in three different solvents and was tested by using the DPPH method, and hot extraction (Soxhlet) and cold extraction (maceration) methods were applied for the presence of components in the camphor leaves. The antioxidant activity of ethanol extracts was higher than that of other extracts. These findings show that camphor leaves, which have significant antioxidant quality, are excellent for pharmaceutical composition. Linalool, nerolidol, and borneol are the phenolic compounds extracted from the ethanolic extract. The hot extraction method by using ethanol solvent can extract antioxidant and mineral content against camphor leaves (see ). Liu et al. established the flavonoids extracted from C. camphora leaves' in vitro antioxidative capacity. Both the ferric reducing antioxidant power assay and the 1,1-diphenyl-2-picrylhydrazyl (DPPH) free radical scavenging assay demonstrated a dose-dependent increase in antioxidant activity in the flavonoids, which is outstanding compared to commercial antioxidants . 9.1.2. Antimicrobial Numerous disorders in the body are brought on by pathogens, which are inhibited by antimicrobial agents, which also stop the establishment of microbial colonies. One of the biggest problems in human health is the over and improper use of antibiotics. Additionally, the rapid spread of microorganisms that are resistant to antibiotics is concerning. The herbal potential of C. camphora is recognized to serve as an antibiotic, antiviral, and antifungal. Wang et al. described that there is evidence that C. camphora essential oil has therapeutic properties like antibacterial effects. The obtained MICs and MBCs verified the clinical strains' significant susceptibility to CCEO. The development of E. coli biofilms is intimately associated to prolonged E. coli infection and can lead to antibiotic resistance. E. coli was significantly destroyed by CCEO, and the E. coli biofilm was also effectively destroyed. C. camphora essential oil (CCEO) was active against E. coli in suspension and biofilms, two states that are common in living organisms. Escherichia coli, one of the most frequent microbial pathogens, is mainly responsible for biofilm-associated opportunistic illnesses like diarrhea, endometritis, and mastitis . In another in vitro study of camphor ethanolic extract has been showed antibacterial action against Escherichia coli , Staphylococcus aureus , and Pseudomonas aeruginosa . Poudel et al. analyzed in vitro study of C. camphora essential oil against five Gram-positive bacteria, Streptococcuspyogenes , Propionibacteriumacnes , Bacilluscereus , Staphylococcusepidermidis , and Staphylococcusaureus , and two Gram-negative bacteria, Pseudomonas aeruginosa and Serratia marcescens , showing its antibacterial properties . The phytochemical compounds found in C. Camphora have a wide variety of antibacterial properties against various pathogens. Leaf, branch, and wood essential oil were tested again using seven strains of fungi, Aspergillus niger , Aspergillus fumigatus , Candida albicans , Microsporum canis , Trichophyton mentagrophytes , Microsporum gypseum , and Trichophyton rubrum . Serratia marcescens responded favorably to the wood essential oil's antimicrobial properties. Camphor, 1,8-cineole, -terpineol, and safrole were the main ingredients in the wood oil; hence, the reported activity of the wood oil against S. marcescens may be the result of synergism between these and other constituents. Only a small amount of action was seen against S. marcescens by camphor, 1,8-cineole, -terpineol, and safrole. According to one study, camphor and 1,8-cineole work together to have a synergistic antibacterial effect . All fungi were cultured on yeast malt, and studies showed good antifungal activity against Aspergillus niger and Aspergillus fumigatus , while the leaf essential oil showed good antifungal activity comparatively to other parts of the plant . Kulzam is a well-known Unani liquid composition used to cure several ailments such as cough, colds, and sore throats . The major components identified in Kulzam were camphor, menthol, etc. The ingredients of the formulation are Sat-i-Pudina , Sat-i-Ajwain , camphor , Roghan baid majnun , Roghan-i-darchini , Roghan-i-zaitun , and Roghan-i-laung . Kulzam demonstrated a significant effect on all tested microorganisms at both 100 and 150 (micro) levels of the undiluted formulation (test sample), and at 150 (micro) level, it inhibited growth more than the standard. Furthermore, it highlights the fact that gram-negative microorganisms are more vulnerable to inhibitory activity than gram-positive ones. Comparing the formulation to that of standard clotrimazole, it showed a very significant zone of inhibition against the fungus Candida albicans and Aspergillus fumigatus . In an in vitro investigation, essential oil from C. camphora leaves, flowers, and twigs showed antifungal action again 7 strains including Aspergillus clavatus , Aspergillus niger , Chaetomium globosum , Cladosporium cladosporioides , Myrothecium verrucaria , Penicillium citrinum , and Trichoderma viride in 1000 μ g/ml concentration . In addition, when compared to other sections of the plant, the leaf oil exhibited the best antifungal efficacy . Five locally gathered plant species' fresh leaves were hydro distilled using Clevenger's apparatus to separate the essential oils and stored in a glass jar. Using the poisoned food approach on a potato dextrose agar medium, the oils were evaluated for resistance to Aspergillus flavus at 5000 ppm. Only the oil of C. camphora showed absolute fungitoxicity against the test fungus among the five essential oils examined . Studies showed camphor oil possesses mycostatic application against Aspergillus flavu s . According to Karashima et al. , CHO cell showed induce expression of TRPA1, 0.5 mg/ml tetracycline was added to the culture medium, and cells were used 5–24 h after induction and menthol used as test drug . C. camphora constituent menthol activates TRPA1 and inhibits it in mouse neuron in in vitro study, suggesting the involvement of different kinetics of channel and fast desensitization due to these sensory effects of menthol, a widely used additive in counterirritants and analgesic activity. 9.1.3. Anti-Inflammatory and Prostaglandin Synthesis Inhibition Inflammation is a healing process that is triggered by pathogen, toxins, and radiations. These factors set off the immune system and cause inflammatory reactions in the organs of the host, which may result in cell death and/or illness. Unani traditional medicine is potentially useful for the treatment of inflammation-related diseases, such as rheumatism, bronchitis, asthma, COPD, acute non-purulent rhinosinusitis, dermatitis, neurodegenerative diseases, and muscle pains. There are well-known anti-inflammatory compounds that have been extracted from plants and evaluated in human clinical trials. Cinelol, cineole, citronellal, and camphor make up the majority of them . Numerous investigations have revealed that C. camphora has an anti-inflammatory activity in vitro. An in vitro investigation of C. camphora leaf extract indicated that it reduced the generation of inflammatory chemokines. Its leaves had a significant impact on 2,4-dinitrochlorobenzene-induced atrophic dermatitis in mice. An in vitro investigation of C. camphora leaf ethanolic extract indicated that it reduced the generation of inflammatory chemokines. Its leaves had a significant impact on 2,4-dinitrochlorobenzene-induced atrophic dermatitis in mice. Lee et al. examined the inhibitory impact of CCex on IFN- (10 ng/mL) stimulated HaCaT keratinocytes' ability to produce the inflammatory chemokine (MDC). The outcomes demonstrated that CCex inhibited MDC formation by IFN- in a concentration-dependent manner. The MeOH extract of C. Camphora inhibited prostaglandin E2 (PGE2) production in LPS/IFN-activated macrophages by up to 70%. To further understand C. camphora's anti-inflammatory activity, researchers looked at macrophage-mediated inflammatory events like cytokine production, NO release, PGE2 release, functional activation of adhesion molecules, and oxidative stress. It can have a strong immunomodulatory influence on numerous inflammatory responses at the transcriptional level, according to the findings of the study . Methanolic extract of C. camphora contains anti-inflammatory mechanisms that limit NO and PGE2 synthesis in LPS/IFN-activated macrophages and prevent the generation of IL-1, IL-6, and TNF- from RAW264.7 cells . By interacting with biological components within the cell, the oxidation process damages cells, resulting in a variety of illnesses and chronic diseases like cancer and cardiovascular conditions. Additionally, oxidation changes the nutritional value and safety of food by producing secondary reaction products . The effectiveness of EOC derived from leaves in treating allergic inflammation, such as atopic dermatitis, was described by Kang . The extract significantly reduced inflammation in low-calcium, high-temperature human adult keratinocytes and improved 2,4-dinitrochlorobenzene-induced atopic dermatitis in mice. These results will make it easier to create EOC as a novel, all-natural treatment for inflammatory skin disorders . 9.1.4. Anti-Hyperlipidemic Activity Camphor compound was examined in rats with experimental dyslipoproteinemia for its pharmacotherapeutic efficacy, antioxidant, and anticoagulant action. The positive results of the study allowed this substance to be recommended for the prevention of atherosclerotic damage to the vascular endothelium and the prevention of thrombogenesis . 9.1.5. Antifertility Activity In order to understand the impact of camphor as a male local contraceptive, in-vitro effect of camphor on human sperm vialbility and motility was examined. A decrease in sperm motility and viability in an in vitro investigation where camphor was used indicates that fertilization efficacy is reduced. Camphor may work as a contraceptive effect. The sperm motility and viability decrease are probably because of a fall in fructose levels or denaturation of protein and cholesterol, which are the energy sources for sperm motility . 9.2. In Vivo Pharmacological Studies 9.2.1. Wound Healing Activity Camphor, a potent wound healing and ant wrinkle drug, reduced MMP1 expression but increased collagen and elastin expression in UV-exposed mouse skin after 4 weeks of therapy. Camphor might prevent the loss of elastin and help it recover after UV-induced damage to retain skin suppleness . It also decreased the depths of the epidermis and subcutaneous fat layer in UV-exposed mouse skin. The ethyl acetate soluble fraction of an ethanolic extract of C. camphora leaves in Wister rats showed improvement in wound healing and increased wound contraction due to enhance and accelerated activity of fibroblast and epithelial cell migration to the wound site and early dermal and epidermal regeneration. Furthermore, the treated group also showed a considerable increase in collagen content . 9.2.2. Anti-Testosterone Activity Jugular vein samples were taken for hormonal analysis from Awassi lambs and rams fed C. camphora at a dose of 20 mg/kg/animal, and semen samples were collected from the animals using artificial vagina in the control group. The study found that the testosterone hormone concentration in the treatment group was much lower than in the control group, which could be attributed to camphor's oestrogenic impact, which reduces testosterone hormone levels. Camphor may suppress catecholamine secretion by inhibiting nicotine acetylcholine receptors, which has an influence on male sexual behavior and reproductively via its effect on blood testosterone levels and/or the sympathetic nervous system. During the second and fourth weeks of the experimental study, mass activity in the camphor group was significantly ( P < 0.05) lower than in the control group, whereas the individual sperm motility percentage showed no significant differences between the camphor and control groups throughout the entire experimental period, i.e., over the final three weeks of the trial, the camphor group displayed lower levels testosterone . 9.2.3. Oestrogenic Effect Maerkel et al. evaluated the estrogenic effect on the brain and reproductive organs both prenatal and postnatal exposure to UV filter 4-methylbenzylidene camphor (4-MBC) in rats. Following pre- and postnatal exposure to the UV filter 4-MBC, the current study found alterations in the expression level and estrogen sensitivity of target genes as well as in the steroid receptor coactivator SRC-1 in sexually dimorphic brain areas of adult rat offspring. They discussed how the UV filter 4-methyl benzylidene camphor (4-MBC) is oestrogenic and interferes with the thyroid axis. They found that 4-MBC exposure changed the mRNA levels of ER-alpha, progesterone receptor (PR), preproenkephalin (PPE), and insulin-like growth factor-I (IGF-I) at the brain level in rats in a sex- and region-specific manner . 9.2.4. Anti-Allergic Activity Edema, a dysfunctional skin barrier, and the invasion of several inflammatory cell types are the hallmarks of allergic skin inflammation, such as atopic dermatitis (AD). In vivo , C. camphora leaves (100 mg/kg) improved atopic dermatitis symptoms by lowering serum immunoglobulin E levels, reducing lymph node thickness and length, decreasing ear edema, and lowering the number of inflammatory cells infiltrating the ears. Atrophic dermatitis is an allergic inflammatory disorder that can be treated with the leaves of C. camphora . IFN- γ , an important mediator of immunity and inflammation, induces the Janus tyrosine kinases- signal transducer and activator of transcription (JAK-STAT) signal pathway. The investigators reported that in skin inflammation lesions, leaf extract inhibited macrophage-derived chemokine (MDC/CCL22) production via the downregulation of (STAT) 1 and extracellular signal-regulated kinase 1/2 (ERK1/2) pathways and, hence, improved several symptoms (ear edema and lymph node size) change in blood parameter (serum IgE) and histological changes in mice with allergic dermatitis. By administering DNCB to mice, we established experimental AD in order to research the effects of CCex on AD in vivo. IgE levels are correlated with the severity of AD and are linked to defective skin barrier, making IgE a key therapeutic target for AD. When compared to the induction group in this investigation, the CCex-treated group had considerably lower serum IgE levels ( p 0.001). Comparing the cutaneous edema in the CCex-treated mice to that in the induction mice on day 29, the difference was significant ( p 0.001). Additionally, the CCex-treated group showed considerably less epidermal thickness and inflammatory cell infiltration than the induction group . 9.2.5. Anxiolytic and Antidepressant Activity Antidepressant and anxiolytic medications are used to treat depression and anxiety. Albino mice weighing 18-30 gm were used for the study CCO given three different doses 250 mg/kg CCO orally, 500 mg/kg, and 750 mg/kg CCO orally in each group. and imipramine15mg/kg given intra peritoneal as a standard control. C. camphora oil (CCO) showed significant anti-anxiety and antidepressant effects compared to the control group in rat models. Numerous monoterpenoid compounds in the essential oil of C. camphora are confirmed by phytochemical studies. β -thujone, β -pinene, linalool, and limonene are monoterpenoids that are testified to have antidepressant applications. Furthermore, recent research has proposed the antidepressant action of β -pinene, which increases dopamine level and inhibits MAO activity in rabbits. Additionally, few studies showed that numerous biologically active molecules, including monoterpenoids, are potent inhibitors of MAO-A and MAO-B . 9.2.6. Antioxidant Activity As previously mentioned, C. camphora seed kernel oil increased the concentrations of superoxide dismutase and catalase in diet-induced rats, which consequently boosted antioxidant activity and reduced malondialdehyde concentration (a biomarker of lipid peroxidation and oxidative stress) . 9.2.7. Analgesic Activity C. camphora leaf essential oils showed a significant analgesic effect against nitroglycerin-induced experimental migraine in mice models and inhibited the nuclear factor-kappa Beta, inducible nitric oxide synthase, and nitric oxide pathway . 9.2.8. Hepatoprotective Activity According to Johari et al. , camphor powder solution given female rats by intraperitoneal injection for 14 days showed hepatoprotective activity in the treatment of a deferent type of liver conditions. On the liver enzymes, it is proven to have a stimulating impact. However, the researchers recommended that camphor use in a higher dosage uninterruptedly probably leads to a substantial increase in the concentration of liver enzymes . Numerous in vitro experimental studies show that antioxidant, antimicrobial, anti-inflammatory, and miscellaneous activities have been demonstrated by numerous research on C. camphora . 9.1.1. Antioxidant Activity By interacting with biological components within the cell, the oxidation process damages cells, resulting in a variety of illnesses and chronic diseases like cancer and cardiovascular conditions. Additionally, oxidation changes the nutritional value and safety of food by producing secondary reaction products . Oxidative stress is a condition when antioxidant levels are low. Antioxidant activity of polyphenols, which is influenced by their polyphenolic structure, has the effect of removing free radicals and improving antioxidant activity . Due to an excess of reactive oxygen species (ROS), oxidative stress develops when the body's antioxidant system becomes depleted. Due to this, an increase in the concentration of free radicals inside cells is the root cause of many chronic disorders such as nonalcoholic fatty liver, type 2 diabetics, neurological conditions, and reproductive-related problems . Reactive oxygen species (ROS) at baseline levels are necessary for basic physiological activities. Ageing, obesity, type 2 diabetes mellitus (T2DM), depression, and neurodegeneration are all conditions that are significantly impacted by oxidative stress . Camphor has antioxidant, hepatoprotective, antidepressant, estrogenic, and anti-inflammatory qualities in addition to being an antioxidant that prevents oxidative damage and neutralizes free radicals. Camphor's antioxidant capabilities may lessen tissue damage and oxidative stress. In scavenging DPPH, ABTS, and ferric reducing antioxidant power (FRAP) assays, the phytochemical proanthocyanidins (PAs) from leaves and branches of C. camphora displayed significant antioxidant activity . The antioxidant activity of hexane, chloroform, and ethanol extracts was determined using the DPPH (2,2-diphenyl-1-picrylhydrazyl) technique on dried camphor leaves. Another study evaluated the antioxidant activity of leaves of C. camphora in three different solvents and was tested by using the DPPH method, and hot extraction (Soxhlet) and cold extraction (maceration) methods were applied for the presence of components in the camphor leaves. The antioxidant activity of ethanol extracts was higher than that of other extracts. These findings show that camphor leaves, which have significant antioxidant quality, are excellent for pharmaceutical composition. Linalool, nerolidol, and borneol are the phenolic compounds extracted from the ethanolic extract. The hot extraction method by using ethanol solvent can extract antioxidant and mineral content against camphor leaves (see ). Liu et al. established the flavonoids extracted from C. camphora leaves' in vitro antioxidative capacity. Both the ferric reducing antioxidant power assay and the 1,1-diphenyl-2-picrylhydrazyl (DPPH) free radical scavenging assay demonstrated a dose-dependent increase in antioxidant activity in the flavonoids, which is outstanding compared to commercial antioxidants . 9.1.2. Antimicrobial Numerous disorders in the body are brought on by pathogens, which are inhibited by antimicrobial agents, which also stop the establishment of microbial colonies. One of the biggest problems in human health is the over and improper use of antibiotics. Additionally, the rapid spread of microorganisms that are resistant to antibiotics is concerning. The herbal potential of C. camphora is recognized to serve as an antibiotic, antiviral, and antifungal. Wang et al. described that there is evidence that C. camphora essential oil has therapeutic properties like antibacterial effects. The obtained MICs and MBCs verified the clinical strains' significant susceptibility to CCEO. The development of E. coli biofilms is intimately associated to prolonged E. coli infection and can lead to antibiotic resistance. E. coli was significantly destroyed by CCEO, and the E. coli biofilm was also effectively destroyed. C. camphora essential oil (CCEO) was active against E. coli in suspension and biofilms, two states that are common in living organisms. Escherichia coli, one of the most frequent microbial pathogens, is mainly responsible for biofilm-associated opportunistic illnesses like diarrhea, endometritis, and mastitis . In another in vitro study of camphor ethanolic extract has been showed antibacterial action against Escherichia coli , Staphylococcus aureus , and Pseudomonas aeruginosa . Poudel et al. analyzed in vitro study of C. camphora essential oil against five Gram-positive bacteria, Streptococcuspyogenes , Propionibacteriumacnes , Bacilluscereus , Staphylococcusepidermidis , and Staphylococcusaureus , and two Gram-negative bacteria, Pseudomonas aeruginosa and Serratia marcescens , showing its antibacterial properties . The phytochemical compounds found in C. Camphora have a wide variety of antibacterial properties against various pathogens. Leaf, branch, and wood essential oil were tested again using seven strains of fungi, Aspergillus niger , Aspergillus fumigatus , Candida albicans , Microsporum canis , Trichophyton mentagrophytes , Microsporum gypseum , and Trichophyton rubrum . Serratia marcescens responded favorably to the wood essential oil's antimicrobial properties. Camphor, 1,8-cineole, -terpineol, and safrole were the main ingredients in the wood oil; hence, the reported activity of the wood oil against S. marcescens may be the result of synergism between these and other constituents. Only a small amount of action was seen against S. marcescens by camphor, 1,8-cineole, -terpineol, and safrole. According to one study, camphor and 1,8-cineole work together to have a synergistic antibacterial effect . All fungi were cultured on yeast malt, and studies showed good antifungal activity against Aspergillus niger and Aspergillus fumigatus , while the leaf essential oil showed good antifungal activity comparatively to other parts of the plant . Kulzam is a well-known Unani liquid composition used to cure several ailments such as cough, colds, and sore throats . The major components identified in Kulzam were camphor, menthol, etc. The ingredients of the formulation are Sat-i-Pudina , Sat-i-Ajwain , camphor , Roghan baid majnun , Roghan-i-darchini , Roghan-i-zaitun , and Roghan-i-laung . Kulzam demonstrated a significant effect on all tested microorganisms at both 100 and 150 (micro) levels of the undiluted formulation (test sample), and at 150 (micro) level, it inhibited growth more than the standard. Furthermore, it highlights the fact that gram-negative microorganisms are more vulnerable to inhibitory activity than gram-positive ones. Comparing the formulation to that of standard clotrimazole, it showed a very significant zone of inhibition against the fungus Candida albicans and Aspergillus fumigatus . In an in vitro investigation, essential oil from C. camphora leaves, flowers, and twigs showed antifungal action again 7 strains including Aspergillus clavatus , Aspergillus niger , Chaetomium globosum , Cladosporium cladosporioides , Myrothecium verrucaria , Penicillium citrinum , and Trichoderma viride in 1000 μ g/ml concentration . In addition, when compared to other sections of the plant, the leaf oil exhibited the best antifungal efficacy . Five locally gathered plant species' fresh leaves were hydro distilled using Clevenger's apparatus to separate the essential oils and stored in a glass jar. Using the poisoned food approach on a potato dextrose agar medium, the oils were evaluated for resistance to Aspergillus flavus at 5000 ppm. Only the oil of C. camphora showed absolute fungitoxicity against the test fungus among the five essential oils examined . Studies showed camphor oil possesses mycostatic application against Aspergillus flavu s . According to Karashima et al. , CHO cell showed induce expression of TRPA1, 0.5 mg/ml tetracycline was added to the culture medium, and cells were used 5–24 h after induction and menthol used as test drug . C. camphora constituent menthol activates TRPA1 and inhibits it in mouse neuron in in vitro study, suggesting the involvement of different kinetics of channel and fast desensitization due to these sensory effects of menthol, a widely used additive in counterirritants and analgesic activity. 9.1.3. Anti-Inflammatory and Prostaglandin Synthesis Inhibition Inflammation is a healing process that is triggered by pathogen, toxins, and radiations. These factors set off the immune system and cause inflammatory reactions in the organs of the host, which may result in cell death and/or illness. Unani traditional medicine is potentially useful for the treatment of inflammation-related diseases, such as rheumatism, bronchitis, asthma, COPD, acute non-purulent rhinosinusitis, dermatitis, neurodegenerative diseases, and muscle pains. There are well-known anti-inflammatory compounds that have been extracted from plants and evaluated in human clinical trials. Cinelol, cineole, citronellal, and camphor make up the majority of them . Numerous investigations have revealed that C. camphora has an anti-inflammatory activity in vitro. An in vitro investigation of C. camphora leaf extract indicated that it reduced the generation of inflammatory chemokines. Its leaves had a significant impact on 2,4-dinitrochlorobenzene-induced atrophic dermatitis in mice. An in vitro investigation of C. camphora leaf ethanolic extract indicated that it reduced the generation of inflammatory chemokines. Its leaves had a significant impact on 2,4-dinitrochlorobenzene-induced atrophic dermatitis in mice. Lee et al. examined the inhibitory impact of CCex on IFN- (10 ng/mL) stimulated HaCaT keratinocytes' ability to produce the inflammatory chemokine (MDC). The outcomes demonstrated that CCex inhibited MDC formation by IFN- in a concentration-dependent manner. The MeOH extract of C. Camphora inhibited prostaglandin E2 (PGE2) production in LPS/IFN-activated macrophages by up to 70%. To further understand C. camphora's anti-inflammatory activity, researchers looked at macrophage-mediated inflammatory events like cytokine production, NO release, PGE2 release, functional activation of adhesion molecules, and oxidative stress. It can have a strong immunomodulatory influence on numerous inflammatory responses at the transcriptional level, according to the findings of the study . Methanolic extract of C. camphora contains anti-inflammatory mechanisms that limit NO and PGE2 synthesis in LPS/IFN-activated macrophages and prevent the generation of IL-1, IL-6, and TNF- from RAW264.7 cells . By interacting with biological components within the cell, the oxidation process damages cells, resulting in a variety of illnesses and chronic diseases like cancer and cardiovascular conditions. Additionally, oxidation changes the nutritional value and safety of food by producing secondary reaction products . The effectiveness of EOC derived from leaves in treating allergic inflammation, such as atopic dermatitis, was described by Kang . The extract significantly reduced inflammation in low-calcium, high-temperature human adult keratinocytes and improved 2,4-dinitrochlorobenzene-induced atopic dermatitis in mice. These results will make it easier to create EOC as a novel, all-natural treatment for inflammatory skin disorders . 9.1.4. Anti-Hyperlipidemic Activity Camphor compound was examined in rats with experimental dyslipoproteinemia for its pharmacotherapeutic efficacy, antioxidant, and anticoagulant action. The positive results of the study allowed this substance to be recommended for the prevention of atherosclerotic damage to the vascular endothelium and the prevention of thrombogenesis . 9.1.5. Antifertility Activity In order to understand the impact of camphor as a male local contraceptive, in-vitro effect of camphor on human sperm vialbility and motility was examined. A decrease in sperm motility and viability in an in vitro investigation where camphor was used indicates that fertilization efficacy is reduced. Camphor may work as a contraceptive effect. The sperm motility and viability decrease are probably because of a fall in fructose levels or denaturation of protein and cholesterol, which are the energy sources for sperm motility . By interacting with biological components within the cell, the oxidation process damages cells, resulting in a variety of illnesses and chronic diseases like cancer and cardiovascular conditions. Additionally, oxidation changes the nutritional value and safety of food by producing secondary reaction products . Oxidative stress is a condition when antioxidant levels are low. Antioxidant activity of polyphenols, which is influenced by their polyphenolic structure, has the effect of removing free radicals and improving antioxidant activity . Due to an excess of reactive oxygen species (ROS), oxidative stress develops when the body's antioxidant system becomes depleted. Due to this, an increase in the concentration of free radicals inside cells is the root cause of many chronic disorders such as nonalcoholic fatty liver, type 2 diabetics, neurological conditions, and reproductive-related problems . Reactive oxygen species (ROS) at baseline levels are necessary for basic physiological activities. Ageing, obesity, type 2 diabetes mellitus (T2DM), depression, and neurodegeneration are all conditions that are significantly impacted by oxidative stress . Camphor has antioxidant, hepatoprotective, antidepressant, estrogenic, and anti-inflammatory qualities in addition to being an antioxidant that prevents oxidative damage and neutralizes free radicals. Camphor's antioxidant capabilities may lessen tissue damage and oxidative stress. In scavenging DPPH, ABTS, and ferric reducing antioxidant power (FRAP) assays, the phytochemical proanthocyanidins (PAs) from leaves and branches of C. camphora displayed significant antioxidant activity . The antioxidant activity of hexane, chloroform, and ethanol extracts was determined using the DPPH (2,2-diphenyl-1-picrylhydrazyl) technique on dried camphor leaves. Another study evaluated the antioxidant activity of leaves of C. camphora in three different solvents and was tested by using the DPPH method, and hot extraction (Soxhlet) and cold extraction (maceration) methods were applied for the presence of components in the camphor leaves. The antioxidant activity of ethanol extracts was higher than that of other extracts. These findings show that camphor leaves, which have significant antioxidant quality, are excellent for pharmaceutical composition. Linalool, nerolidol, and borneol are the phenolic compounds extracted from the ethanolic extract. The hot extraction method by using ethanol solvent can extract antioxidant and mineral content against camphor leaves (see ). Liu et al. established the flavonoids extracted from C. camphora leaves' in vitro antioxidative capacity. Both the ferric reducing antioxidant power assay and the 1,1-diphenyl-2-picrylhydrazyl (DPPH) free radical scavenging assay demonstrated a dose-dependent increase in antioxidant activity in the flavonoids, which is outstanding compared to commercial antioxidants . Numerous disorders in the body are brought on by pathogens, which are inhibited by antimicrobial agents, which also stop the establishment of microbial colonies. One of the biggest problems in human health is the over and improper use of antibiotics. Additionally, the rapid spread of microorganisms that are resistant to antibiotics is concerning. The herbal potential of C. camphora is recognized to serve as an antibiotic, antiviral, and antifungal. Wang et al. described that there is evidence that C. camphora essential oil has therapeutic properties like antibacterial effects. The obtained MICs and MBCs verified the clinical strains' significant susceptibility to CCEO. The development of E. coli biofilms is intimately associated to prolonged E. coli infection and can lead to antibiotic resistance. E. coli was significantly destroyed by CCEO, and the E. coli biofilm was also effectively destroyed. C. camphora essential oil (CCEO) was active against E. coli in suspension and biofilms, two states that are common in living organisms. Escherichia coli, one of the most frequent microbial pathogens, is mainly responsible for biofilm-associated opportunistic illnesses like diarrhea, endometritis, and mastitis . In another in vitro study of camphor ethanolic extract has been showed antibacterial action against Escherichia coli , Staphylococcus aureus , and Pseudomonas aeruginosa . Poudel et al. analyzed in vitro study of C. camphora essential oil against five Gram-positive bacteria, Streptococcuspyogenes , Propionibacteriumacnes , Bacilluscereus , Staphylococcusepidermidis , and Staphylococcusaureus , and two Gram-negative bacteria, Pseudomonas aeruginosa and Serratia marcescens , showing its antibacterial properties . The phytochemical compounds found in C. Camphora have a wide variety of antibacterial properties against various pathogens. Leaf, branch, and wood essential oil were tested again using seven strains of fungi, Aspergillus niger , Aspergillus fumigatus , Candida albicans , Microsporum canis , Trichophyton mentagrophytes , Microsporum gypseum , and Trichophyton rubrum . Serratia marcescens responded favorably to the wood essential oil's antimicrobial properties. Camphor, 1,8-cineole, -terpineol, and safrole were the main ingredients in the wood oil; hence, the reported activity of the wood oil against S. marcescens may be the result of synergism between these and other constituents. Only a small amount of action was seen against S. marcescens by camphor, 1,8-cineole, -terpineol, and safrole. According to one study, camphor and 1,8-cineole work together to have a synergistic antibacterial effect . All fungi were cultured on yeast malt, and studies showed good antifungal activity against Aspergillus niger and Aspergillus fumigatus , while the leaf essential oil showed good antifungal activity comparatively to other parts of the plant . Kulzam is a well-known Unani liquid composition used to cure several ailments such as cough, colds, and sore throats . The major components identified in Kulzam were camphor, menthol, etc. The ingredients of the formulation are Sat-i-Pudina , Sat-i-Ajwain , camphor , Roghan baid majnun , Roghan-i-darchini , Roghan-i-zaitun , and Roghan-i-laung . Kulzam demonstrated a significant effect on all tested microorganisms at both 100 and 150 (micro) levels of the undiluted formulation (test sample), and at 150 (micro) level, it inhibited growth more than the standard. Furthermore, it highlights the fact that gram-negative microorganisms are more vulnerable to inhibitory activity than gram-positive ones. Comparing the formulation to that of standard clotrimazole, it showed a very significant zone of inhibition against the fungus Candida albicans and Aspergillus fumigatus . In an in vitro investigation, essential oil from C. camphora leaves, flowers, and twigs showed antifungal action again 7 strains including Aspergillus clavatus , Aspergillus niger , Chaetomium globosum , Cladosporium cladosporioides , Myrothecium verrucaria , Penicillium citrinum , and Trichoderma viride in 1000 μ g/ml concentration . In addition, when compared to other sections of the plant, the leaf oil exhibited the best antifungal efficacy . Five locally gathered plant species' fresh leaves were hydro distilled using Clevenger's apparatus to separate the essential oils and stored in a glass jar. Using the poisoned food approach on a potato dextrose agar medium, the oils were evaluated for resistance to Aspergillus flavus at 5000 ppm. Only the oil of C. camphora showed absolute fungitoxicity against the test fungus among the five essential oils examined . Studies showed camphor oil possesses mycostatic application against Aspergillus flavu s . According to Karashima et al. , CHO cell showed induce expression of TRPA1, 0.5 mg/ml tetracycline was added to the culture medium, and cells were used 5–24 h after induction and menthol used as test drug . C. camphora constituent menthol activates TRPA1 and inhibits it in mouse neuron in in vitro study, suggesting the involvement of different kinetics of channel and fast desensitization due to these sensory effects of menthol, a widely used additive in counterirritants and analgesic activity. Inflammation is a healing process that is triggered by pathogen, toxins, and radiations. These factors set off the immune system and cause inflammatory reactions in the organs of the host, which may result in cell death and/or illness. Unani traditional medicine is potentially useful for the treatment of inflammation-related diseases, such as rheumatism, bronchitis, asthma, COPD, acute non-purulent rhinosinusitis, dermatitis, neurodegenerative diseases, and muscle pains. There are well-known anti-inflammatory compounds that have been extracted from plants and evaluated in human clinical trials. Cinelol, cineole, citronellal, and camphor make up the majority of them . Numerous investigations have revealed that C. camphora has an anti-inflammatory activity in vitro. An in vitro investigation of C. camphora leaf extract indicated that it reduced the generation of inflammatory chemokines. Its leaves had a significant impact on 2,4-dinitrochlorobenzene-induced atrophic dermatitis in mice. An in vitro investigation of C. camphora leaf ethanolic extract indicated that it reduced the generation of inflammatory chemokines. Its leaves had a significant impact on 2,4-dinitrochlorobenzene-induced atrophic dermatitis in mice. Lee et al. examined the inhibitory impact of CCex on IFN- (10 ng/mL) stimulated HaCaT keratinocytes' ability to produce the inflammatory chemokine (MDC). The outcomes demonstrated that CCex inhibited MDC formation by IFN- in a concentration-dependent manner. The MeOH extract of C. Camphora inhibited prostaglandin E2 (PGE2) production in LPS/IFN-activated macrophages by up to 70%. To further understand C. camphora's anti-inflammatory activity, researchers looked at macrophage-mediated inflammatory events like cytokine production, NO release, PGE2 release, functional activation of adhesion molecules, and oxidative stress. It can have a strong immunomodulatory influence on numerous inflammatory responses at the transcriptional level, according to the findings of the study . Methanolic extract of C. camphora contains anti-inflammatory mechanisms that limit NO and PGE2 synthesis in LPS/IFN-activated macrophages and prevent the generation of IL-1, IL-6, and TNF- from RAW264.7 cells . By interacting with biological components within the cell, the oxidation process damages cells, resulting in a variety of illnesses and chronic diseases like cancer and cardiovascular conditions. Additionally, oxidation changes the nutritional value and safety of food by producing secondary reaction products . The effectiveness of EOC derived from leaves in treating allergic inflammation, such as atopic dermatitis, was described by Kang . The extract significantly reduced inflammation in low-calcium, high-temperature human adult keratinocytes and improved 2,4-dinitrochlorobenzene-induced atopic dermatitis in mice. These results will make it easier to create EOC as a novel, all-natural treatment for inflammatory skin disorders . Camphor compound was examined in rats with experimental dyslipoproteinemia for its pharmacotherapeutic efficacy, antioxidant, and anticoagulant action. The positive results of the study allowed this substance to be recommended for the prevention of atherosclerotic damage to the vascular endothelium and the prevention of thrombogenesis . In order to understand the impact of camphor as a male local contraceptive, in-vitro effect of camphor on human sperm vialbility and motility was examined. A decrease in sperm motility and viability in an in vitro investigation where camphor was used indicates that fertilization efficacy is reduced. Camphor may work as a contraceptive effect. The sperm motility and viability decrease are probably because of a fall in fructose levels or denaturation of protein and cholesterol, which are the energy sources for sperm motility . 9.2.1. Wound Healing Activity Camphor, a potent wound healing and ant wrinkle drug, reduced MMP1 expression but increased collagen and elastin expression in UV-exposed mouse skin after 4 weeks of therapy. Camphor might prevent the loss of elastin and help it recover after UV-induced damage to retain skin suppleness . It also decreased the depths of the epidermis and subcutaneous fat layer in UV-exposed mouse skin. The ethyl acetate soluble fraction of an ethanolic extract of C. camphora leaves in Wister rats showed improvement in wound healing and increased wound contraction due to enhance and accelerated activity of fibroblast and epithelial cell migration to the wound site and early dermal and epidermal regeneration. Furthermore, the treated group also showed a considerable increase in collagen content . 9.2.2. Anti-Testosterone Activity Jugular vein samples were taken for hormonal analysis from Awassi lambs and rams fed C. camphora at a dose of 20 mg/kg/animal, and semen samples were collected from the animals using artificial vagina in the control group. The study found that the testosterone hormone concentration in the treatment group was much lower than in the control group, which could be attributed to camphor's oestrogenic impact, which reduces testosterone hormone levels. Camphor may suppress catecholamine secretion by inhibiting nicotine acetylcholine receptors, which has an influence on male sexual behavior and reproductively via its effect on blood testosterone levels and/or the sympathetic nervous system. During the second and fourth weeks of the experimental study, mass activity in the camphor group was significantly ( P < 0.05) lower than in the control group, whereas the individual sperm motility percentage showed no significant differences between the camphor and control groups throughout the entire experimental period, i.e., over the final three weeks of the trial, the camphor group displayed lower levels testosterone . 9.2.3. Oestrogenic Effect Maerkel et al. evaluated the estrogenic effect on the brain and reproductive organs both prenatal and postnatal exposure to UV filter 4-methylbenzylidene camphor (4-MBC) in rats. Following pre- and postnatal exposure to the UV filter 4-MBC, the current study found alterations in the expression level and estrogen sensitivity of target genes as well as in the steroid receptor coactivator SRC-1 in sexually dimorphic brain areas of adult rat offspring. They discussed how the UV filter 4-methyl benzylidene camphor (4-MBC) is oestrogenic and interferes with the thyroid axis. They found that 4-MBC exposure changed the mRNA levels of ER-alpha, progesterone receptor (PR), preproenkephalin (PPE), and insulin-like growth factor-I (IGF-I) at the brain level in rats in a sex- and region-specific manner . 9.2.4. Anti-Allergic Activity Edema, a dysfunctional skin barrier, and the invasion of several inflammatory cell types are the hallmarks of allergic skin inflammation, such as atopic dermatitis (AD). In vivo , C. camphora leaves (100 mg/kg) improved atopic dermatitis symptoms by lowering serum immunoglobulin E levels, reducing lymph node thickness and length, decreasing ear edema, and lowering the number of inflammatory cells infiltrating the ears. Atrophic dermatitis is an allergic inflammatory disorder that can be treated with the leaves of C. camphora . IFN- γ , an important mediator of immunity and inflammation, induces the Janus tyrosine kinases- signal transducer and activator of transcription (JAK-STAT) signal pathway. The investigators reported that in skin inflammation lesions, leaf extract inhibited macrophage-derived chemokine (MDC/CCL22) production via the downregulation of (STAT) 1 and extracellular signal-regulated kinase 1/2 (ERK1/2) pathways and, hence, improved several symptoms (ear edema and lymph node size) change in blood parameter (serum IgE) and histological changes in mice with allergic dermatitis. By administering DNCB to mice, we established experimental AD in order to research the effects of CCex on AD in vivo. IgE levels are correlated with the severity of AD and are linked to defective skin barrier, making IgE a key therapeutic target for AD. When compared to the induction group in this investigation, the CCex-treated group had considerably lower serum IgE levels ( p 0.001). Comparing the cutaneous edema in the CCex-treated mice to that in the induction mice on day 29, the difference was significant ( p 0.001). Additionally, the CCex-treated group showed considerably less epidermal thickness and inflammatory cell infiltration than the induction group . 9.2.5. Anxiolytic and Antidepressant Activity Antidepressant and anxiolytic medications are used to treat depression and anxiety. Albino mice weighing 18-30 gm were used for the study CCO given three different doses 250 mg/kg CCO orally, 500 mg/kg, and 750 mg/kg CCO orally in each group. and imipramine15mg/kg given intra peritoneal as a standard control. C. camphora oil (CCO) showed significant anti-anxiety and antidepressant effects compared to the control group in rat models. Numerous monoterpenoid compounds in the essential oil of C. camphora are confirmed by phytochemical studies. β -thujone, β -pinene, linalool, and limonene are monoterpenoids that are testified to have antidepressant applications. Furthermore, recent research has proposed the antidepressant action of β -pinene, which increases dopamine level and inhibits MAO activity in rabbits. Additionally, few studies showed that numerous biologically active molecules, including monoterpenoids, are potent inhibitors of MAO-A and MAO-B . 9.2.6. Antioxidant Activity As previously mentioned, C. camphora seed kernel oil increased the concentrations of superoxide dismutase and catalase in diet-induced rats, which consequently boosted antioxidant activity and reduced malondialdehyde concentration (a biomarker of lipid peroxidation and oxidative stress) . 9.2.7. Analgesic Activity C. camphora leaf essential oils showed a significant analgesic effect against nitroglycerin-induced experimental migraine in mice models and inhibited the nuclear factor-kappa Beta, inducible nitric oxide synthase, and nitric oxide pathway . 9.2.8. Hepatoprotective Activity According to Johari et al. , camphor powder solution given female rats by intraperitoneal injection for 14 days showed hepatoprotective activity in the treatment of a deferent type of liver conditions. On the liver enzymes, it is proven to have a stimulating impact. However, the researchers recommended that camphor use in a higher dosage uninterruptedly probably leads to a substantial increase in the concentration of liver enzymes . Camphor, a potent wound healing and ant wrinkle drug, reduced MMP1 expression but increased collagen and elastin expression in UV-exposed mouse skin after 4 weeks of therapy. Camphor might prevent the loss of elastin and help it recover after UV-induced damage to retain skin suppleness . It also decreased the depths of the epidermis and subcutaneous fat layer in UV-exposed mouse skin. The ethyl acetate soluble fraction of an ethanolic extract of C. camphora leaves in Wister rats showed improvement in wound healing and increased wound contraction due to enhance and accelerated activity of fibroblast and epithelial cell migration to the wound site and early dermal and epidermal regeneration. Furthermore, the treated group also showed a considerable increase in collagen content . Jugular vein samples were taken for hormonal analysis from Awassi lambs and rams fed C. camphora at a dose of 20 mg/kg/animal, and semen samples were collected from the animals using artificial vagina in the control group. The study found that the testosterone hormone concentration in the treatment group was much lower than in the control group, which could be attributed to camphor's oestrogenic impact, which reduces testosterone hormone levels. Camphor may suppress catecholamine secretion by inhibiting nicotine acetylcholine receptors, which has an influence on male sexual behavior and reproductively via its effect on blood testosterone levels and/or the sympathetic nervous system. During the second and fourth weeks of the experimental study, mass activity in the camphor group was significantly ( P < 0.05) lower than in the control group, whereas the individual sperm motility percentage showed no significant differences between the camphor and control groups throughout the entire experimental period, i.e., over the final three weeks of the trial, the camphor group displayed lower levels testosterone . Maerkel et al. evaluated the estrogenic effect on the brain and reproductive organs both prenatal and postnatal exposure to UV filter 4-methylbenzylidene camphor (4-MBC) in rats. Following pre- and postnatal exposure to the UV filter 4-MBC, the current study found alterations in the expression level and estrogen sensitivity of target genes as well as in the steroid receptor coactivator SRC-1 in sexually dimorphic brain areas of adult rat offspring. They discussed how the UV filter 4-methyl benzylidene camphor (4-MBC) is oestrogenic and interferes with the thyroid axis. They found that 4-MBC exposure changed the mRNA levels of ER-alpha, progesterone receptor (PR), preproenkephalin (PPE), and insulin-like growth factor-I (IGF-I) at the brain level in rats in a sex- and region-specific manner . Edema, a dysfunctional skin barrier, and the invasion of several inflammatory cell types are the hallmarks of allergic skin inflammation, such as atopic dermatitis (AD). In vivo , C. camphora leaves (100 mg/kg) improved atopic dermatitis symptoms by lowering serum immunoglobulin E levels, reducing lymph node thickness and length, decreasing ear edema, and lowering the number of inflammatory cells infiltrating the ears. Atrophic dermatitis is an allergic inflammatory disorder that can be treated with the leaves of C. camphora . IFN- γ , an important mediator of immunity and inflammation, induces the Janus tyrosine kinases- signal transducer and activator of transcription (JAK-STAT) signal pathway. The investigators reported that in skin inflammation lesions, leaf extract inhibited macrophage-derived chemokine (MDC/CCL22) production via the downregulation of (STAT) 1 and extracellular signal-regulated kinase 1/2 (ERK1/2) pathways and, hence, improved several symptoms (ear edema and lymph node size) change in blood parameter (serum IgE) and histological changes in mice with allergic dermatitis. By administering DNCB to mice, we established experimental AD in order to research the effects of CCex on AD in vivo. IgE levels are correlated with the severity of AD and are linked to defective skin barrier, making IgE a key therapeutic target for AD. When compared to the induction group in this investigation, the CCex-treated group had considerably lower serum IgE levels ( p 0.001). Comparing the cutaneous edema in the CCex-treated mice to that in the induction mice on day 29, the difference was significant ( p 0.001). Additionally, the CCex-treated group showed considerably less epidermal thickness and inflammatory cell infiltration than the induction group . Antidepressant and anxiolytic medications are used to treat depression and anxiety. Albino mice weighing 18-30 gm were used for the study CCO given three different doses 250 mg/kg CCO orally, 500 mg/kg, and 750 mg/kg CCO orally in each group. and imipramine15mg/kg given intra peritoneal as a standard control. C. camphora oil (CCO) showed significant anti-anxiety and antidepressant effects compared to the control group in rat models. Numerous monoterpenoid compounds in the essential oil of C. camphora are confirmed by phytochemical studies. β -thujone, β -pinene, linalool, and limonene are monoterpenoids that are testified to have antidepressant applications. Furthermore, recent research has proposed the antidepressant action of β -pinene, which increases dopamine level and inhibits MAO activity in rabbits. Additionally, few studies showed that numerous biologically active molecules, including monoterpenoids, are potent inhibitors of MAO-A and MAO-B . As previously mentioned, C. camphora seed kernel oil increased the concentrations of superoxide dismutase and catalase in diet-induced rats, which consequently boosted antioxidant activity and reduced malondialdehyde concentration (a biomarker of lipid peroxidation and oxidative stress) . C. camphora leaf essential oils showed a significant analgesic effect against nitroglycerin-induced experimental migraine in mice models and inhibited the nuclear factor-kappa Beta, inducible nitric oxide synthase, and nitric oxide pathway . According to Johari et al. , camphor powder solution given female rats by intraperitoneal injection for 14 days showed hepatoprotective activity in the treatment of a deferent type of liver conditions. On the liver enzymes, it is proven to have a stimulating impact. However, the researchers recommended that camphor use in a higher dosage uninterruptedly probably leads to a substantial increase in the concentration of liver enzymes . C. camphora essential oil from seeds, twigs, and leaves showed robust contact toxicity against cotton aphids with median lethal concentration (LC50) values of 146.78, 274.99, and 245.79, mg/L after 48 h of treatment, respectively . Camphor is quickly absorbed by the mucous membranes, skin, and gastrointestinal tract in liquid form. Symptoms may appear 5–90 minutes after consumption . In humans, indications of intoxication include nausea, vomiting, trembling, and convulsions, which are followed by CNS depression characterized by apnea and coma . Unlike petroleum products, camphor is a botanical hydrocarbon, very inexpensive, and can be easily cultivated without any shortages. Therefore, camphor is an exceptional carbon source for the production of high purity, high yield, and high efficiency . According to Unani physicians, seven factors are responsible for the maintenance of health, and loss of any one of these can lead to disease or even death. Dietotherapy and pharmacotherapy are mainly used to maintain the equilibrium of humors to maintain health and treat disease conditions. All single drugs have specific and many ethnopharmacological properties according to their active principles and temperament. C. camphora (L.) is a traditional Unani medicinal plant with a third-degree cold and dry temperament and hence useful in hot temperamental individuals used since ancient times. Nowadays, natural and artificial camphor is also used for medicinal conditions and commercial purposes. A review of Unani and other conventional literature realized that C. camphora has prophylactic and several pharmacological properties for treating medical conditions and strengthening mental and physical properties and it is effective in treating respiratory conditions, musculoskeletal, gastrointestinal, oral, eye, integumentary, and general conditions. Unani physicians evaluated ethno-pharmacological properties, usage, patient temperament, and disease condition when prescribing medications and then selected single pharmaceuticals with correctives ( Muslih ) to reduce undesirable or unwanted effects. Furthermore, to combat the adverse effect of camphor on cold temperamental people , they advised camphor with hot temperament and fragrance herbs such as Zafran , Amber , and Misk . Camphor's distinctive aroma has led to its widespread use in ointments and inhalants, particularly as a remedy to treat respiratory ailments. Unani physicians stated that C. camphora is commonly used in respiratory conditions such as acute and chronic cough, fever, common cold, lung ulcers, pleurisy, pneumonia, coryza, and catarrh in various forms as a single drug or with another herbal, mineral, or animal origin drug as a compound formula as it possesses expectorant, antipyretics, deobstruent, and mucolytic properties . C. camphora has been a useful remedy in symptoms of COVID-19 and inhibits SARS CoV-2 spike glycoprotein CoV . Furthermore, one of the ingredients of many compounds, Unani formulation, is useful in viral infections and respiratory diseases . In vitro, in vivo, and clinical, recent studies have proven its antibacterial , anti-inflammatory , analgesics , and immunomodulatory properties. Hence, it is beneficial in conditions such as acute non-purulent rhinosinusitis, asthma, COPD, bronchitis, and the inhibitory effect of SARS-CoV . It is used as a nasal decongestant and a cough suppressant . Anti-asthmatic, cough suppressant, mucolytic, bronchodilator control, airway mucus hypersecretion, and anti-asthmatic qualities are found in camphor, cineole, linalool, and safrole . Compound drugs of C. camphora and external medication used for the suitable medicament in musculoskeletal conditions such as arthritis and muscular pain with ethnomedicinal properties such as stimulant, rubefacient, anastatic, and analgesics . Camphor is a major active ingredient in liniments and balms used as a topical analgesic, and it is a natural compound . Lee et al. in their in vitro study confirmed anti-inflammatory and antioxidant properties of camphor. They hypothesize that the anti-inflammatory actions of C. camphora is perhaps due to the modulation of cytokine, NO, and PGE(2) production. TRPM8, TRPV3, and TRPV1 (transient receptor potential) channels are activated, and TRPA1 is inhibited by camphor leading to excitation, warm sensation, and desensitization of sensory nerves, itch, relieving pain, and irritation in the applied area . We have designed the word cloud in the current study based on the closest terms used in this study . It would be helpful for the researchers, scientists, doctors, students, and academicians to work in this area. Previously, many researchers are used word cloud in the field of stress , motor imagery , augmented reality , blockchain technology , and premenstrual syndrome with oxidative stress . In addition, we also designed the network visualization based on previous published studies mentioned in . This network visualization may also be helpful to the researchers, scientists, doctors, students, and academicians to work in this area. Previously, many researchers are used network visualization in the field of premenstrual syndrome, insomnia , bruxism , motor imagery, augmented reality, stress , and blockchain technology. Camphor is a cardiac and nervine tonic in proper dose and dosage forms. It acts as an exhilarant, tonic, stimulant innate faculty of soul and nerves. The Unani formulation such as Jauhar-i-Kafoor is used for inhalation as nervine and exhilarant for vital organs and useful in convulsion. Jauhar-i-Kafoor is also useful as cardiac tonic for conditions such as a syncope, weakness of heart functions, and palpitation. Overdose or misused of camphor can cause adverse effects such as abdominal discomfort, tremor, apnea, coma, and CNS depression. Hence, a proper dose stimulates and enhances nervine and cardiac activity, while an overdose can disrupt the system. However, more research is required to confirm. According to the findings, a proper dose stimulates and enhances nervine and cardiac activity, while an overdose can disrupt the system. However, more research is required to confirm. Studies have shown that C. camphora or some of its components might find some applications in the future for the treatment of memory disorders or for improving brain functions in patients . Strength includes authenticity and research articles showed that C. camphora has therapeutic applications in various diseases, proven in few recent preclinical and clinical studies. However, other ethnopharmacological activities with eye applications and gastrointestinal and cardiovascular diseases require additional modern pharmacological interpretations to explicate its basic mechanism. The review showed that isolated compounds, extract dose range, route of administration, and dose frequency are clarified in the existing recent research. However, more preclinical and clinical trials are recommended to explore its therapeutic applications in other diseases. Furthermore, it has been normally combined with other single drugs in Unani medicine; hence, drug interactions should be researched further in conventional therapies. There are no studies that have revealed interactions between conventional medicine and camphor, and additional research is needed to determine the safety and efficacy of camphor. Nonetheless, prospective clinical studies are recommended to validate the herb's healing effect. In humans, indications of intoxication include gastrointestinal like nausea, vomiting, and nervous system trembling and convulsions, which are followed by CNS depression characterized by apnea and coma. The comprehensive review concludes that it is a drug that has been effectively used in Unani medicine for centuries to treat several ailments, particularly respiratory, nervous, musculoskeletal, and eye disorders. Camphor has been related to a variety of biological activities, including antibacterial, antifungal, anti-inflammatory, analgesics, antitussive, and antioxidant. However, bioactivity was often determined using an essential oil rich in camphor rather than pure camphor, and clinical studies of pure camphor have not been found. Moreover, sparse preclinical and clinical researches are available. However, more studies are needed to explore the pharmacological activities. This review has demonstrated that camphor is a medication with many wide ranges of applications. |
Using a person-centered approach in clinical care for patients with complex chronic conditions: Perspectives from healthcare professionals caring for Veterans with COPD in the U.S. Veterans Health Administration’s Whole Health System of Care | 21915d78-470b-4c10-9936-5e76e7b3deba | 10289382 | Patient-Centered Care[mh] | The Veterans Health Administration (VHA) is a subdivision of the United States Department of Veterans Affairs (VA) that serves eligible military Veterans and their family members. For over a decade, VHA, the largest nationally integrated healthcare system in the U.S., has been undergoing an unprecedented transformation to a Whole Health (WH) System of Care that promotes Veterans’ health and well-being in a person-centered manner. VHA’s WH system seeks to empower Veterans to collaborate with their care team as equal partners, encourage the use of self-care skills, and promote access to both conventional and complementary approaches to disease treatment and prevention . The central element of the WH System implementation is WH Clinical Care–that is, primary and specialty care services offered in line with a WH approach, as opposed to the older disease-centered approach. WH Clinical Care encompasses four principles for approaching clinical encounters: (1) the state of the Veteran’s health and well-being is assessed in a comprehensive, whole-person way; (2) what matters most to the Veteran–their meaning or purpose in life–is elicited early and prioritized throughout the encounter; (3) the Veteran is supported in setting and pursuing realistic health and well-being goals that are related to what matters most to them; and (4) the Veteran is equipped with knowledge, information, and resources in support of these goals, which may include connecting the Veteran with other WH System of Care offerings–e.g., peer-facilitated groups or complementary and integrative health (CIH) providers . VHA’s WH Clinical Care is fully aligned with the principles of person-centered care. We define person-centered care as an approach to care that considers each individual who comes into contact with the healthcare system as a person with a unique life context and aligns the care with the individual’s preferences, goals, and priorities [ – ]. In other words, patients are viewed as more than the diseases from which they suffer. Indeed, WH Clinical Care incorporates and scales up many of the key principles that the person-centered care movement has been advocating for many decades but that are not always explicitly captured in brief definitions–a holistic view of health and illness as affected by multiple factors; empowerment of the patient to engage in their own care (sometimes described as patient-directed care); and integration of non-medical and supportive services, to name a few [ , – ]. In the United States, the urgency of attending to the health and well-being of the whole person–not just to diseases of specific organs or body systems–has recently been highlighted in a new strategic plan issued by the National Center for Complementary and Integrative Health (NCCIH) of the National Institutes of Health (NIH) . Calls for making person-centered care a matter of U.S. policy have been issued for decades, including, notably, a recent proposal by the National Academies of Sciences, Engineering, and Medicine to scale VHA’s Whole Health model at the national level [ , , ]. Similar developments are taking place globally [ – ]. In this context, it is more critical than ever that VHA’s experiences of building a WH System of Care are thoroughly described and analyzed, so as to allow policymakers, clinicians, and patient advocates in the U.S. and beyond to learn from VHA’s lessons. As a national-level, publicly funded health system, VHA’s experiences are relevant beyond the U.S. healthcare landscape. Prior research shows that, despite the advances already made, implementing WH Clinical Care has been an enormous challenge for VHA, and the slowest component of the implementation so far . Efforts to increase clinicians’ use of WH Clinical Care have had varying success [ , , ], and many have argued that having a system that supports this approach to care is critical . Much of the existing literature, however, has focused on the high-level organizational challenges of implementation as perceived by frontline clinicians and leaders . It is essential to better understand frontline healthcare professionals’ perspectives on the meaning, rationale and practicality providing WH Clinical Care, because their conceptualizations have an immediate bearing on the success of implementation efforts. While VHA envisions WH Clinical Care as applicable to all patient populations, it is a particularly natural fit for patients with complex chronic conditions. Such individuals face a range of functional and psychosocial challenges that WH Clinical Care is well-positioned to address, given its person-centered focus on supporting patients in pursuing a holistic vision of health and well-being in line with their values, goals, and preferences . Chronic obstructive pulmonary disease (COPD) is a case in point. COPD is a complex, systemic condition that affects and is affected by all areas of the individual’s physical, psychological, and social well-being. Breathlessness, a common symptom of COPD, causes tremendous distress to patients and caregivers , and acute exacerbations of breathlessness may result in emergency room visits or costly hospital stays . Individuals living with COPD also experience decreased everyday activity , social isolation [ – ], depression and anxiety [ – ], feelings of guilt and self-blame , and distress about the future [ – ]. A growing chorus of voices in the academic, clinician, and patient advocate communities has been calling for implementation of a person-centered approach wherein the clinical team empowers and supports the individual with COPD to pursue a better quality of life and a sustained sense of well-being, in line with this individual’s values, goals, and preferences [ – ]. Despite evidence in support of this patient-centered approach to COPD care [ – ] and its alignment with VHA’s WH Clinical Care model, such care is not widely practiced—either in the VHA or beyond. In this paper, we sought to understand how healthcare professionals across disciplines at VHA view the value of and the facilitators and barriers to providing WH Clinical Care to Veterans with COPD, as a lens to understand their receptivity to WH Clinical Care in general.
We conducted a qualitative study to understand factors affecting the uptake of the Whole Health Clinical Care in outpatient care for Veterans with COPD. Between October 2020 and November 2021, we recruited staff members at a large urban VA Medical Center (VAMC) that includes both tertiary inpatient care and an extensive selection of outpatient services to participate in semi-structured interviews. We sought to interview healthcare workers across diverse services and disciplines that typically play an important role in managing outpatient care of Veterans with COPD, including primary care, pulmonary medicine, and palliative care. In alignment with the WH vision wherein whole-person care incorporates broader psychosocial well-being, we also interviewed staff in mental healthcare services (which are increasingly recognized as a key element of comprehensive care for patients with COPD), the chaplain service (as spiritual/existential concerns are well-documented in the population of interest), and the site’s Whole Health service, which is responsible for promoting and supporting the implementation of the Whole Health approach locally. As part of the larger study, we also interviewed Veterans with COPD. We chose not to integrate an analysis of their perspectives into the current manuscript as the emphases and range of topics covered in their interviews were so different as to merit a separate manuscript (currently in preparation). The study was approved by the VA Bedford Institutional Review Board (IRB), as well as by the IRB of the VA site where the data for this study were collected. The interview guide ( ) was developed iteratively with input from co-authors and drawing on the Consolidated Framework for Implementation Research (CFIR) for guidance on the content and structure. CFIR was chosen for its comprehensive, multi-level conceptualization of implementation determinants, such as perceptions of the intervention (e.g., WH Clinical Care), the inner setting (e.g., specific VAMC and VHA as a whole), outer setting (e.g., the nation-wide push for patient-centered care), characteristics of individuals involved (e.g., healthcare professionals for the purposes of this study) and the process of implementation (e.g., engagement of VAMC in the WH System of Care implementation). Participants were invited to participate in the study via e-mail; e-mail addresses were obtained from internal VA address books and public VA provider directories. Interviews took place over the phone or Microsoft Teams, depending on the participants’ preference. Prior to each interview, the interviewer (first author) went over the study description and participants’ rights. Verbal informed consent was obtained prior to participation. All interviews were recorded with participants’ permission. After each interview, the interviewer took notes to capture the content and initial analytical reflections. Interview data were analyzed using qualitative content analysis . The first author coded the transcripts in the qualitative data analysis software ATLAS.ti using inductive codes (derived from the data). The inductive codes were then organized into larger categories corresponding to the four elements of WH Clinical Care described above. Throughout the coding and analysis process, the first author used memos to record insights about similarities and differences across the data set, as well as emerging concepts. Drawing on post-interview notes, memos, and coded data, the first author generated an initial set of themes, each of them pertinent to the four elements of WH Clinical Care. These initial themes were refined with the help of the senior author, who also reviewed the larger dataset to ensure that the themes adequately reflect the patterns in the data. The naming and content of the themes were finalized by incorporating several rounds of oral and written input of other authors.
We interviewed 25 individuals from a wide range of disciplines and services. The majority of participants were physicians (48%) (see for more details). Our overarching finding is that participants had lingering questions/concerns regarding each of the WH Clinical Care elements–namely, (1) whole-person approach to assessment; (2) eliciting what matters most; (3) supporting the Veteran in setting and pursuing personally meaningful and realistic health and well-being goals; and (4) equipping the Veteran with knowledge, information, and resources in support of these goals. The four themes identified in participants’ discussions of WH Clinical Care, each of them associated with the specific element of WH Clinical Care, are presented in , with their brief names intended to capture the question/tension in point. 1. Assessing the Veteran’s life context: How much is too much? When asked to describe their typical appointments with Veterans who have COPD, many participants described their efforts to understand the Veteran’s life context, especially during the first appointment–a practice that broadly aligns with the WH Clinical Care model. Incorporation of life context in clinical conversations, however, varied in scope and focus. In some cases, the details of the descriptions fully embody WH principles. For example, a primary care provider who was very familiar and comfortable with the WH terminology described a comprehensive approach wherein the full scope of the Veteran’s social support needs is explored, beyond the narrow focus on the disease: “…I use a Whole Health template to take kind of like a more extensive social history than I might otherwise. And what I like about it, is that… it’s, like, really a fuller account of human flourishing and so, you know, asking about things like what’s most important to you in life and social stressors and social support. <…> I wanna know where they live, of course. I wanna make sure they have reliable transportation. <…> …sometimes, you know, there are a lot of veterans who aren’t working or maybe… who are medically retired because of service related-injuries or whatnot, and… there’s a lot of support in terms of like… supportive work therapy or whatever” (PRIMARY-PHYS-04). This quote stands in contrast with one from a different clinician, a pulmonologist who was both sympathetic toward WH and self-admittedly not very knowledgeable about it: “…I don’t necessarily use the terminology of Whole Health care, but in general, my style in trying to take care of patients is to try to connect with them on a personal level and… <find> out what’s going on in their lives, and often that… may influence what’s going on with their respiratory symptoms and their COPD. The issue of smoking cessation is completely wrapped up in their lifestyle and… what stressors they’re encountering. So… it’s pretty routine for me to ask about their environment in terms of… what could be affecting their breathing. So, the presence of animals in the house, dust, mold, the condition of the house, etc. You know, smokers in the house whether it’s the patient themselves or family members so I get a… rough idea what’s going on <in> the house or… wherever they’re living or if they’re unstable in their housing situation” (PULM-PHYS-01). This interviewee, like the previous one, recounted a comprehensive assessment of the Veteran’s life context, which is broadly aligned with WH principles, yet they placed emphases in a notably different way. Specifically, all of the questions described concern factors in the Veteran’s environment that may affect their respiratory symptoms. Such assessment would result in a more circumscribed picture than the “fuller account of human flourishing” described by the participant quoted earlier. A related, yet distinct point of tension came up in the context of assessing for mental health and socioeconomic difficulties. A primary care provider expressed discomfort with the idea of probing into this area, perceiving it as difficult to address within a physician’s scope of practice: “…some questions you don’t want to ask because you can’t address them. You know, it is very difficult for me as an MD physician. If… something is wrong with their housing… all I do really is, ‘oh, let me have you talk to a social worker.’ You know… sometimes you don’t want to ask the question if you don’t want to know the answer, or at least know how to answer the question for them or how to actually help them” (PRIMARY-PHYS-02). Several participants reported a similar uneasiness around inquiring about spiritual or existential concerns that are sometimes present in patients with COPD, such as shame, guilt, and fear of death, during routine appointments. For example, in response to the interviewer question about whether such topics come up in the appointment, a primary care provider answered: “Yeah, but we really don’t have enough time for that. …this would require a separate visit, you know, because you open up a door that you have to close it very fast and that would be not appropriate to do. Obviously, if somebody’s coming with some concerns or symptoms, and we can use large part of the visit for discussion for that particular aspect of their life, that’s a different thing but not during the routine visits. So… we usually don’t go–if there is a sense that there is a need for that, we have enabled, you know, help from our Behavioral Health” (PRIMARY-PHYS-01). A notable counterpoint to the routine avoidance of this sphere of life/well-being came from an interview with a WH Coach (a WH service staff member whose role involves delivering health and well-being coaching to Veterans), who modeled a thoughtful approach to framing the conversation about spiritual concerns: “…what’s the best practice for me… is to kinda initiate that conversation by explaining to them that for some this might be about your religion, but for others it’s about a connection… to something outside of yourself. <…> So… for Veterans who are not religious, it doesn’t have to feel awkward to them because it’s not just about religion. You know, it’s really about your soul, it’s about connecting. And once you…have that conversation, I think it’s much better received than, you know, to just come in and say, ‘We’re going to talk about spirit and soul,’ and the first thing you do is roll out the Chaplain. <…> So, although we have Chaplains available at the VA to kinda come in and speak to Spirit and Soul, you know, I think anyone can do it because, again, it’s really that connection that the Veterans resonate with and connect with” (WH-04). In sum, while there was a buy-in into the need to understand the Veteran’s context, there were disagreements as to which elements of this context are deemed as relevant and appropriate to assess. 2. Understanding what really matters: To what end? A major element of WH Clinical Care–and person-centered care, more broadly–is the emphasis on understanding “what really matters” to the patient. In this paradigm, “what really matters” refers to the deepest motivating force running through the Veteran’s life, the Veteran’s innermost goals and priorities. Inquiring about whether participants ask about what really matters to Veterans as part of their practice has produced a variety of responses–from descriptions that would not be out of place in WH training materials to accounts that expressed unease or confusion about this idea. Specifically, we identified three types of perspectives that are described in detail below: Using the “what really matters” question to take the Veteran’s lead Using the question to gain leverage in the service of the clinician’s agenda Challenge in effectively using the question 2.1. Taking the Veteran’s lead On one end of the spectrum are participants who appear to have fully internalized the imperative to elicit the Veteran’s life priorities–they ask what really matters in order to take the Veteran’s lead. For example, a pulmonologist involved in the pulmonary rehabilitation program described how the program staff seek to understand what the desired level of everyday functioning would be for the Veteran, in order to help each Veteran realize their unique personal vision: “…yeah, we <ask> them, ‘what’s important to you?’–you know… ‘what <does> a good quality of life mean to you’ and I think <if they say> ‘I’m totally happy sitting on the couch. I just don’t wanna be short of breath when I’m sitting on the couch.’ Then, okay, that’s very different from someone who tells me… ‘I used to being able to… go on walks with my grandchildren and I can’t go out anymore.’ …So, yeah, it’s… what do they value in their lives, what do they want you to do, and is there a way that we can help them achieve that” (PR-PHYS-01). This passage is exemplary as it expresses an understanding that what may be important to one Veteran is not important to another–an issue that is at the heart of the debate around the extent to which the patient’s goals and preferences should be prioritized by the clinician. A similar approach was expressed in all of our interviews with palliative care providers, who saw their role as supporting Veterans in doing what really matters to them: “…thinking about the veterans I’ve cared for… over the last year with COPD, many of them… are just so advanced by the time they get to us that they’re not able to get out garden and drive and do all the things that used to matter most. So, then it’s what matters most now . And a lot of it is spending time with family. …like, this… Veteran who was in our unit just temporarily, and his wife was having this prolonged admission, he… really, really wanted to go home. <…> …what mattered most to him was being back in his home environment… so we discharged him home. So, you know, I think as people progress what I found is the things that are—matter most to them are very—more practical, more tangible. <….> So, the <what> matters most can be more kind of vague and sometimes it’s very concrete and so… even taking care of these practical things goes a long way” (PALL-PHYS-03). 2.2. Leveraging what really matters A contrasting approach involved strategically using the information about what matters to the Veteran in service of the clinician’s agenda. For example, one primary care provider recounted routinely using WH “techniques” to encourage lifestyle change in Veterans with chronic pain and diabetes, but not those with COPD. In the quote below, this clinician shares that their decision to participate in the interview was motivated by the desire to learn more about using a WH approach with Veterans with COPD: “I… use Whole Health a lot… and I tend to use it more, like, in pain management; I <also> used it a lot <in> lifestyle change around, like, diabetes, of getting involved in exercise and working on your diet and kind of having a Whole Health coach meet up with you periodically <…> …it was one reason I… was interested in talking to you is ‘cause it is not one that I had thought about as far as, like, inhaler adherence or even… getting beyond just smoking cessation…” (PRIMARY-NP-01). This participant is clearly familiar with and enthusiastic about WH language and tools. At the same time, when this interviewee contemplates the prospect of asking Veterans with COPD about what matters most to them, they see it, in a somewhat narrow manner, as a tool to promote inhaler adherence and smoking cessation, rather than as an invitation for an open-ended exploration of the Veterans’ own priorities which might or might not fully align with these best practices. This interviewee’s approach thus displays a second understanding of the rationale for the “what really matters” question–to leverage and support Veteran’s own motivation to make healthy behavioral changes. In other words, the clinician asks about what matters most to the Veteran in order to facilitate Veteran’s engagement in the goals that the clinician has already designated as important (“lifestyle change”). This understanding was also present in the words of another participant, a psychologist in mental health care services, who was knowledgeable about WH yet also described using the question about what is important as a way to promote treatment adherence, thus subordinating “what really matters” to a biomedical rationale: “I think it’s a great question because what it’s really speaking to is… what’s important to the patient, and what’s important to the patient is going to drive their health behaviors and their choices. They may not recognize that, but that’s where your leverage is. If they say what’s important to me is, I don’t know, walking with my grandchildren or… something like that, then you’re going to be able to help with their motivation and their treatment adherence through this value of what’s important to them” (MH-PSYCH-01). 2.3. Challenge in effectively using the question Finally, some participants found it difficult to effectively inquire about what really matters to the Veteran during a clinical encounter. This concern was most fully demonstrated in an interview with a primary care provider that is worth quoting at length. First, the interviewee explicitly takes issue with the narrative of WH trainings and reports abandoning the question of “what really matters” after receiving excessively vague answers that were difficult to relate to the rest of the clinical conversation: “…it never works the way that it has been portrayed to work in the training, and I actually find the Veterans are… as… confused by the question as we are, you know… they will always say kind of my health, my family and you get much beyond that… and then the connection to how do I connect that to your COPD is just not there. I try to force them to have the connection there anyway in a way that doesn’t feel natural and doesn’t help the conversation… Or it’s just hard for me to connect COPD to that thing that maybe they there are not able to do as well” (PRIMARY-PHYS-02). The same participant then endorsed the more scripted/concrete elements of WH Clinical Care: “<I> actually like the more action-oriented parts of Whole Health … like, ‘okay, great, it sounds like Tai Chi might be good for you,’ and I’m not gonna try to make an artificial connection to why. <…> …there’s the poster that we put up that has the different circles that show all those different components and I have been surprised. There are Veterans that will see that and say, ‘oh yeah, that’s a different way of thinking about it.’ And I haven’t engaged with that framework, those more kind of domain-specific questions as much as might be recommended.…there is value I can see in putting more meat on the bone of that kind of single question, what is most important to you. When you break it down into the different circles of health that might help them, that might help that conversation go better and to be honest it is not a part of my routine practice. <…> …maybe I just need to be shown the light, shown some examples where it really, really made the difference and I just haven’t had that experience yet” (PRIMARY-PHYS-02). The difficulty with making the question about “what really matters” work was acknowledged by other participants. Two primary reasons for this challenge were offered. One was that the question was poorly contextualized; i.e., little guidance was provided for how to incorporate it into a routine appointment. “I think we need to learn how it can be useful as a tool. I don’t think it’s valuable if it’s just a question we ask everybody for the sake of asking the question because honestly when you ask most patients what’s important to you they’re gonna say their family. …so I—we just need to know, okay, glad that I asked you that, but, like, what do I do with that information? How is that helpful to me?” (PRIMARY-PHYS-03). “I think what maybe gets lost is, like, what do you do with that information and… what does it mean. <…> I mean, I think it’s a weird question. …I love it and I think it’s a weird question to ask because it has no context around it, right? …I’m not surprised at all that providers are, like, ‘what do I do with this?’ <…> …could it be part of… a psychosocial needs assessment that isn’t just… plopped in the middle of an intake about other things, …in a more dedicated… 10- or 15-minute conversation that is more focused on psychosocial stuff? And, like, you could put this ‘what matters most question’ in that… and you need to give providers… options of what to do next that isn’t too burdensome, and, like, what do you do with that information?” (MH-PSYCH-01). Another explanation, interestingly, pointed to the challenges of asking Veterans who may not be prone to reflection to share something as complex and intensely personal as what matters most to them: “…a lot of Veterans don’t know the answer to the question which is surprising and interesting because those are some of the people who end up in my office ‘cause… they’ve lost their sense of what’s important or they don’t feel like they can… live a life that is important anymore. <…> So, I think it may … just not be registering with in their lives. Like, who, like, they may have never been one of those people who stepped back and, like, explored this bigger question, especially male older adults. Like, I think… a lot of them are not super-emotional-minded or… even that insightful… …they don’t know what to do with those questions. But some of them do, it just depends” (MH-PSYCH-01). 3. Goal setting and personal health planning: Whose job is it, anyway? In the WH Clinical Care model, the Veteran and their provider or clinical team are encouraged to collaborate on setting shared goals based on what really matters to the Veteran. These goals are then, ideally, documented in the Veteran’s chart to facilitate information sharing across clinical team members, as well as to enable goal follow-up and adjustment. Although participants were open to and in some cases experienced with setting shared goals with Veterans, several raised concerns about feasibility, and there were differing perspectives on the appropriate scope of and distribution of responsibility for goal setting and health planning. Several interviewees described how they deliberate with the Veteran on the best course of action in light of the Veteran’s priorities and preferences as a routine part of their practice, although they did not always use WH-specific language such as SMART goals. For example, a pulmonologist specializing in sleep issues shared the following: “…often, I’ll have patients who say, well, ‘I really want to lose some weight.’ <…> And… so we’ll go that route instead of positive airway pressure. And I say, you know, ‘let’s reassess in a few months and see who you’re doing with that.’ There are other patients who really want to choose the CPAP, and even though they fail at it for months and even years sometimes, they don’t want to consider a different… treatment option. …And sometimes, they just <do not>… want treatment at all.” (PULM-PHYS-02). Although the concept of setting goals was relatively familiar, the interviewees were more concerned about the feasibility of conducting personal health planning as a comprehensive and longitudinal process. For example, a pulmonologist said the following in response to a question whether they had seen or added to a Veteran’s personal health plan: “No. [Laughing] I have not. It’s possible that I have seen it… and I was not–to be honest–you know, motivated to go look at it because we spend so much time in front of the computer that I try to do what I need to do to get the computer work done. So, I have not put together a… personal health plan for a patient. That would be, you know… an example of a perception of a specialist deferring that to the primary care physician because I think it’s wrapped up in their… management of their hypertension, management of their diet, their hyperlipidemia, their diabetes and so forth beyond just my recommendation that they walk or that they exercise that they, you know, whatever I can do to enroll them in either MOVE! Program or <pulmonary> rehab” (PULM-PHYS-01). In other words, while this participant was open to the idea of making a few recommendations or placing referrals, they saw health planning as more appropriate for a primary care provider to take on. Another pulmonologist expressed a similar opinion, albeit in a more forceful fashion. What is notable about the passage below is that this participant sees personal health planning as misaligned with their role in the context of regular outpatient care for Veterans with COPD yet endorses it as their dominant approach in the context of severe, end-of-life COPD: “It’s ridiculous. <…> I’m sorry, I’m being completely honest. …I’m a specialist so I am not the primary care doctor… …when my patient is coming to see me, they… want me to address their lung issues, their COPD issues… <….> Let me amend it this way. So, there are patients that I see in my clinic for COPD that are maybe severe end-stage who are at the tail end of their life where… I’m managing their health status in general. <…> …for these patients… my framework shifts, okay? I’m still the specialist but, yeah, absolutely, in that situation the framework of understanding, you know, what is more important to the patient, what are their goal <so> that framework applies, sure” (PULM-PHYS-03). This notion of personal health planning as lying outside of the pulmonologist’s regular role with regards to providing care to Veterans with COPD was challenged by a primary care provider: “I think that if you’re an expert what an expert does is understand all angles of a problem and be able to comment on it all, and use it all and integrate it all… I think it’s easy for people to say no, not me, no, not my specialty, but… who else’s specialty should it be, if not the expert’s on all of the things that contribute to COPD and all of the things that are available for COPD treatment?” (PRIMARY-PHYS-03). The issue of appropriate division of labor with regards to personal health planning came up in the interviewees in the WH service, as well. Several interviewees whose role involved clinician education felt that it would be appropriate for a provider to set a few shared goals during the appointment yet rely on the WH service to follow up with the Veteran on goal progress: “I think the peers and the coaches are the answer to providers who say, ‘I don’t have time for this.’ You know, the providers can introduce whole health and then can send veterans toward…the coaches to do much more in-depth work over time around Whole Health… so… a coach <can> meet with a veteran one-on-one to continue to talk about goals and values and what’s important. <…> So, I see… the coaches as absolutely being able to kind of expand the reach of Whole Health because they do have the time. That’s their job description. So, taking some of the burden off the providers who don’t have the time, right, to talk every week about Whole Health to individual patients” (WH-06). Offering a counterpoint, a WH Coach we interviewed spoke to the importance of a personal relationship with the clinician–in other words, instead of the one-directional approach (the coach takes over where the clinician leaves off), the coach and the clinician engage in a give and take dynamic, deliberating together over how to empower the Veteran: “I meet with different providers all the time because they call me and say, “Hey… I have this situation.’ You know, ‘I’m trying this with the guy.’ You know, ‘I’d like you to give it a shot or have a conversation with him.’ And oftentimes… because of the relationships that we have, you know, this Veteran’s getting that wrap-around support, not only from the Health Coach, but that provider’s buying into it, and that provider’s also, you know, starting to use some of the language and changing the conversation, as they say, around healthcare” (WH-04). 4. Equipping Veterans with resources: How to navigate all these options? Most interviewees felt that discussing or providing Veterans with COPD with referrals to a broad range of services that may support their well-being was worthwhile, yet some challenges were also raised. To obtain a comprehensive picture, we inquired about services that are usually seen as falling under the WH umbrella, such as complementary and integrative health (CIH) and WH coaches, but also other services, including palliative care, mental health care, and chaplaincy. Two types of barriers to referrals came up in the interviews: (1) barriers related to the interviewees’ knowledge and attitudes about service referrals and (2) barriers related to logistics. For this manuscript, we only focus on the former as the latter are highly granular and specific to VHA and the specific VAMC setting. Generally, participants were open to the idea of connecting Veterans with COPD to services that support their well-being holistically. In a few cases, however, interviewees said that they had not previously considered the possibility and perhaps lacked the knowledge of how these services would help patients with COPD, but would now be interested in doing so in future: “You know, I honestly haven’t recommended it specifically for Veterans with COPD, but I would imagine for our… less severe cases…. it could be really helpful as far as maybe like some pneumonia prevention and getting some movement, getting up and moving around a little bit and helping their immune system with some blood flow ‘cause they aren’t really vigorous exercises, especially Tai Chi… <…> So, I’m sure that more severe COPD is probably—that’s inappropriate, too late, but for our earlier diagnoses, more well-controlled, where definitely exercise is still important, I think that, you know, it’d be interesting to see if it could help prevent pneumonia” (PRIMARY-NP-01). For example, none of the participants interviewed were aware of or previously considered connecting Veterans with COPD with a chaplain outside of the inpatient and/or end-of-life context: “I’ve never thought to bring it up in the outpatient setting… …I do attend on the inpatient medical wards and I know, sometimes I’m kind of struck by, ‘oh, the chaplain came by and saw the patient today and left a note,’ and it just kind of it stands out as something like, it’s just like a totally different world than what I practice. So, to be honest I’ve never considered bringing that into outpatient care. And, you know, I don’t think a Veteran has ever asked me to and I refuse, I don’t think I would ever do that, it’s just never occurred to me and it’s never occurred to me to suggest it if I detect that there was maybe spiritual need or maybe even spiritual distress that was in play” (PRIMARY-PHYS-02). One of the chaplains interviewed attributed this phenomenon to the inadequate awareness of what spiritual care is or how it is relevant to the Veteran’s team’s ability to provide whole-person care at all stages of life: “There is huge poverty of understanding. And a lot of denialism that happens. People know what they know, and they are less inclined to learn what they don’t know. You know, people will trivialize it with spiritual care because they don’t understand it. Maybe because they… figure for themselves, ‘if I don’t have any use for it, you don’t have any use for it.’ But……the least you can do when you don’t understand is to learn. <…> …because our target is the same. …we’re here to serve one individual and that’s the Veteran. If we understand that this Veteran has different dimensions and have been in several different places, then we cannot just compartmentalize ’em and say, ‘oh no, I’m just a psychiatrist.’ <…> So part of my goal is… that we can communicate… much more fluently… and we can tackle it from… different… specialty areas but knowing that… we are here to serve one individual” (CHAPL-01). Some participants also felt that Veterans with COPD themselves may see well-being offerings like CIH as unrealistic to pursue given the severity of their condition and/or logistical barriers, despite the obvious relevance of these offerings for improving their quality of life: “A lot of them have very little breath. <….> At rest, they’re just totally fine, but they’re sick and tired of sitting still. They don’t want to be at rest anymore. They remember when they could walk three miles a day, and they want to do that. They remember when they could wash the dishes without becoming winded. They remember when they could prepare dinner. They can’t stand to look out the window and see their wife shoveling the leaves because they want to do it. This is the pain of COPD. <…> But… they don’t like to leave the house, some of them. <…> Some of them have a car, some of them don’t. Some of them need to take their oxygen with them everywhere they go, and they’re afraid the tank will run out, you know. So, there’re a lot of barriers… and a lot of them are old, and they’re going to say, ‘Ach, what do I need with meditation?’ Or, like, ‘Yoga, that’s for young people.’ You know, I haven’t done a good job of selling it. I think that most of them could benefit, is the truth” (PALL-PHYS-01). In such scenarios, conveying the benefits and relevance of such services requires knowledge and tact on the part of the provider. One of the WH Coaches interviewed gave an example of displaying such tact: “So, I kind of explain what it’s like to be in the class. …I’ll make some recommendations of, ‘You have COPD. Avoid the hot Yoga; that is not going to be something that is in your wheelhouse to do right now.’ And then I’ll explain what it is and the benefits to them, as well, of, you know, typically people with COPD are older. Typically, they have some form of mobility issue. So, explaining that, you know, ‘This is going to help stretch you out a little bit. It’s going to work on strengthening some of your core. They do a lot of breathing during it, which can, of course, have a positive impact on your overall breathing.’ With the Tai Chi, I go into explaining, you know, ‘It has a lot of help with balance,’ ‘cause, again, something I’ve noticed is a lot of patients with COPD are also having some form of unrelated balance issues. So, again, explaining, ‘You know, this can help your balance. It can help your posture, which if you’re standing up straighter, your lungs are going to be a little bit bigger; might help you with some of the breathing’” (WH-02). A final barrier to placing referrals to services for holistically supporting Veterans’ well-being that was commonly mentioned was lack of awareness that these offerings are available: “So, I haven’t really thought about it… although I’m a yogi myself… …I just don’t necessarily think about referring…. I don’t have a problem with it… it never really occurred to me because I think I just wasn’t aware of it. So, I think that I probably am not aware of most of the offerings that would be available” (PULM-PHYS-01).
When asked to describe their typical appointments with Veterans who have COPD, many participants described their efforts to understand the Veteran’s life context, especially during the first appointment–a practice that broadly aligns with the WH Clinical Care model. Incorporation of life context in clinical conversations, however, varied in scope and focus. In some cases, the details of the descriptions fully embody WH principles. For example, a primary care provider who was very familiar and comfortable with the WH terminology described a comprehensive approach wherein the full scope of the Veteran’s social support needs is explored, beyond the narrow focus on the disease: “…I use a Whole Health template to take kind of like a more extensive social history than I might otherwise. And what I like about it, is that… it’s, like, really a fuller account of human flourishing and so, you know, asking about things like what’s most important to you in life and social stressors and social support. <…> I wanna know where they live, of course. I wanna make sure they have reliable transportation. <…> …sometimes, you know, there are a lot of veterans who aren’t working or maybe… who are medically retired because of service related-injuries or whatnot, and… there’s a lot of support in terms of like… supportive work therapy or whatever” (PRIMARY-PHYS-04). This quote stands in contrast with one from a different clinician, a pulmonologist who was both sympathetic toward WH and self-admittedly not very knowledgeable about it: “…I don’t necessarily use the terminology of Whole Health care, but in general, my style in trying to take care of patients is to try to connect with them on a personal level and… <find> out what’s going on in their lives, and often that… may influence what’s going on with their respiratory symptoms and their COPD. The issue of smoking cessation is completely wrapped up in their lifestyle and… what stressors they’re encountering. So… it’s pretty routine for me to ask about their environment in terms of… what could be affecting their breathing. So, the presence of animals in the house, dust, mold, the condition of the house, etc. You know, smokers in the house whether it’s the patient themselves or family members so I get a… rough idea what’s going on <in> the house or… wherever they’re living or if they’re unstable in their housing situation” (PULM-PHYS-01). This interviewee, like the previous one, recounted a comprehensive assessment of the Veteran’s life context, which is broadly aligned with WH principles, yet they placed emphases in a notably different way. Specifically, all of the questions described concern factors in the Veteran’s environment that may affect their respiratory symptoms. Such assessment would result in a more circumscribed picture than the “fuller account of human flourishing” described by the participant quoted earlier. A related, yet distinct point of tension came up in the context of assessing for mental health and socioeconomic difficulties. A primary care provider expressed discomfort with the idea of probing into this area, perceiving it as difficult to address within a physician’s scope of practice: “…some questions you don’t want to ask because you can’t address them. You know, it is very difficult for me as an MD physician. If… something is wrong with their housing… all I do really is, ‘oh, let me have you talk to a social worker.’ You know… sometimes you don’t want to ask the question if you don’t want to know the answer, or at least know how to answer the question for them or how to actually help them” (PRIMARY-PHYS-02). Several participants reported a similar uneasiness around inquiring about spiritual or existential concerns that are sometimes present in patients with COPD, such as shame, guilt, and fear of death, during routine appointments. For example, in response to the interviewer question about whether such topics come up in the appointment, a primary care provider answered: “Yeah, but we really don’t have enough time for that. …this would require a separate visit, you know, because you open up a door that you have to close it very fast and that would be not appropriate to do. Obviously, if somebody’s coming with some concerns or symptoms, and we can use large part of the visit for discussion for that particular aspect of their life, that’s a different thing but not during the routine visits. So… we usually don’t go–if there is a sense that there is a need for that, we have enabled, you know, help from our Behavioral Health” (PRIMARY-PHYS-01). A notable counterpoint to the routine avoidance of this sphere of life/well-being came from an interview with a WH Coach (a WH service staff member whose role involves delivering health and well-being coaching to Veterans), who modeled a thoughtful approach to framing the conversation about spiritual concerns: “…what’s the best practice for me… is to kinda initiate that conversation by explaining to them that for some this might be about your religion, but for others it’s about a connection… to something outside of yourself. <…> So… for Veterans who are not religious, it doesn’t have to feel awkward to them because it’s not just about religion. You know, it’s really about your soul, it’s about connecting. And once you…have that conversation, I think it’s much better received than, you know, to just come in and say, ‘We’re going to talk about spirit and soul,’ and the first thing you do is roll out the Chaplain. <…> So, although we have Chaplains available at the VA to kinda come in and speak to Spirit and Soul, you know, I think anyone can do it because, again, it’s really that connection that the Veterans resonate with and connect with” (WH-04). In sum, while there was a buy-in into the need to understand the Veteran’s context, there were disagreements as to which elements of this context are deemed as relevant and appropriate to assess.
A major element of WH Clinical Care–and person-centered care, more broadly–is the emphasis on understanding “what really matters” to the patient. In this paradigm, “what really matters” refers to the deepest motivating force running through the Veteran’s life, the Veteran’s innermost goals and priorities. Inquiring about whether participants ask about what really matters to Veterans as part of their practice has produced a variety of responses–from descriptions that would not be out of place in WH training materials to accounts that expressed unease or confusion about this idea. Specifically, we identified three types of perspectives that are described in detail below: Using the “what really matters” question to take the Veteran’s lead Using the question to gain leverage in the service of the clinician’s agenda Challenge in effectively using the question
On one end of the spectrum are participants who appear to have fully internalized the imperative to elicit the Veteran’s life priorities–they ask what really matters in order to take the Veteran’s lead. For example, a pulmonologist involved in the pulmonary rehabilitation program described how the program staff seek to understand what the desired level of everyday functioning would be for the Veteran, in order to help each Veteran realize their unique personal vision: “…yeah, we <ask> them, ‘what’s important to you?’–you know… ‘what <does> a good quality of life mean to you’ and I think <if they say> ‘I’m totally happy sitting on the couch. I just don’t wanna be short of breath when I’m sitting on the couch.’ Then, okay, that’s very different from someone who tells me… ‘I used to being able to… go on walks with my grandchildren and I can’t go out anymore.’ …So, yeah, it’s… what do they value in their lives, what do they want you to do, and is there a way that we can help them achieve that” (PR-PHYS-01). This passage is exemplary as it expresses an understanding that what may be important to one Veteran is not important to another–an issue that is at the heart of the debate around the extent to which the patient’s goals and preferences should be prioritized by the clinician. A similar approach was expressed in all of our interviews with palliative care providers, who saw their role as supporting Veterans in doing what really matters to them: “…thinking about the veterans I’ve cared for… over the last year with COPD, many of them… are just so advanced by the time they get to us that they’re not able to get out garden and drive and do all the things that used to matter most. So, then it’s what matters most now . And a lot of it is spending time with family. …like, this… Veteran who was in our unit just temporarily, and his wife was having this prolonged admission, he… really, really wanted to go home. <…> …what mattered most to him was being back in his home environment… so we discharged him home. So, you know, I think as people progress what I found is the things that are—matter most to them are very—more practical, more tangible. <….> So, the <what> matters most can be more kind of vague and sometimes it’s very concrete and so… even taking care of these practical things goes a long way” (PALL-PHYS-03).
A contrasting approach involved strategically using the information about what matters to the Veteran in service of the clinician’s agenda. For example, one primary care provider recounted routinely using WH “techniques” to encourage lifestyle change in Veterans with chronic pain and diabetes, but not those with COPD. In the quote below, this clinician shares that their decision to participate in the interview was motivated by the desire to learn more about using a WH approach with Veterans with COPD: “I… use Whole Health a lot… and I tend to use it more, like, in pain management; I <also> used it a lot <in> lifestyle change around, like, diabetes, of getting involved in exercise and working on your diet and kind of having a Whole Health coach meet up with you periodically <…> …it was one reason I… was interested in talking to you is ‘cause it is not one that I had thought about as far as, like, inhaler adherence or even… getting beyond just smoking cessation…” (PRIMARY-NP-01). This participant is clearly familiar with and enthusiastic about WH language and tools. At the same time, when this interviewee contemplates the prospect of asking Veterans with COPD about what matters most to them, they see it, in a somewhat narrow manner, as a tool to promote inhaler adherence and smoking cessation, rather than as an invitation for an open-ended exploration of the Veterans’ own priorities which might or might not fully align with these best practices. This interviewee’s approach thus displays a second understanding of the rationale for the “what really matters” question–to leverage and support Veteran’s own motivation to make healthy behavioral changes. In other words, the clinician asks about what matters most to the Veteran in order to facilitate Veteran’s engagement in the goals that the clinician has already designated as important (“lifestyle change”). This understanding was also present in the words of another participant, a psychologist in mental health care services, who was knowledgeable about WH yet also described using the question about what is important as a way to promote treatment adherence, thus subordinating “what really matters” to a biomedical rationale: “I think it’s a great question because what it’s really speaking to is… what’s important to the patient, and what’s important to the patient is going to drive their health behaviors and their choices. They may not recognize that, but that’s where your leverage is. If they say what’s important to me is, I don’t know, walking with my grandchildren or… something like that, then you’re going to be able to help with their motivation and their treatment adherence through this value of what’s important to them” (MH-PSYCH-01).
Finally, some participants found it difficult to effectively inquire about what really matters to the Veteran during a clinical encounter. This concern was most fully demonstrated in an interview with a primary care provider that is worth quoting at length. First, the interviewee explicitly takes issue with the narrative of WH trainings and reports abandoning the question of “what really matters” after receiving excessively vague answers that were difficult to relate to the rest of the clinical conversation: “…it never works the way that it has been portrayed to work in the training, and I actually find the Veterans are… as… confused by the question as we are, you know… they will always say kind of my health, my family and you get much beyond that… and then the connection to how do I connect that to your COPD is just not there. I try to force them to have the connection there anyway in a way that doesn’t feel natural and doesn’t help the conversation… Or it’s just hard for me to connect COPD to that thing that maybe they there are not able to do as well” (PRIMARY-PHYS-02). The same participant then endorsed the more scripted/concrete elements of WH Clinical Care: “<I> actually like the more action-oriented parts of Whole Health … like, ‘okay, great, it sounds like Tai Chi might be good for you,’ and I’m not gonna try to make an artificial connection to why. <…> …there’s the poster that we put up that has the different circles that show all those different components and I have been surprised. There are Veterans that will see that and say, ‘oh yeah, that’s a different way of thinking about it.’ And I haven’t engaged with that framework, those more kind of domain-specific questions as much as might be recommended.…there is value I can see in putting more meat on the bone of that kind of single question, what is most important to you. When you break it down into the different circles of health that might help them, that might help that conversation go better and to be honest it is not a part of my routine practice. <…> …maybe I just need to be shown the light, shown some examples where it really, really made the difference and I just haven’t had that experience yet” (PRIMARY-PHYS-02). The difficulty with making the question about “what really matters” work was acknowledged by other participants. Two primary reasons for this challenge were offered. One was that the question was poorly contextualized; i.e., little guidance was provided for how to incorporate it into a routine appointment. “I think we need to learn how it can be useful as a tool. I don’t think it’s valuable if it’s just a question we ask everybody for the sake of asking the question because honestly when you ask most patients what’s important to you they’re gonna say their family. …so I—we just need to know, okay, glad that I asked you that, but, like, what do I do with that information? How is that helpful to me?” (PRIMARY-PHYS-03). “I think what maybe gets lost is, like, what do you do with that information and… what does it mean. <…> I mean, I think it’s a weird question. …I love it and I think it’s a weird question to ask because it has no context around it, right? …I’m not surprised at all that providers are, like, ‘what do I do with this?’ <…> …could it be part of… a psychosocial needs assessment that isn’t just… plopped in the middle of an intake about other things, …in a more dedicated… 10- or 15-minute conversation that is more focused on psychosocial stuff? And, like, you could put this ‘what matters most question’ in that… and you need to give providers… options of what to do next that isn’t too burdensome, and, like, what do you do with that information?” (MH-PSYCH-01). Another explanation, interestingly, pointed to the challenges of asking Veterans who may not be prone to reflection to share something as complex and intensely personal as what matters most to them: “…a lot of Veterans don’t know the answer to the question which is surprising and interesting because those are some of the people who end up in my office ‘cause… they’ve lost their sense of what’s important or they don’t feel like they can… live a life that is important anymore. <…> So, I think it may … just not be registering with in their lives. Like, who, like, they may have never been one of those people who stepped back and, like, explored this bigger question, especially male older adults. Like, I think… a lot of them are not super-emotional-minded or… even that insightful… …they don’t know what to do with those questions. But some of them do, it just depends” (MH-PSYCH-01).
In the WH Clinical Care model, the Veteran and their provider or clinical team are encouraged to collaborate on setting shared goals based on what really matters to the Veteran. These goals are then, ideally, documented in the Veteran’s chart to facilitate information sharing across clinical team members, as well as to enable goal follow-up and adjustment. Although participants were open to and in some cases experienced with setting shared goals with Veterans, several raised concerns about feasibility, and there were differing perspectives on the appropriate scope of and distribution of responsibility for goal setting and health planning. Several interviewees described how they deliberate with the Veteran on the best course of action in light of the Veteran’s priorities and preferences as a routine part of their practice, although they did not always use WH-specific language such as SMART goals. For example, a pulmonologist specializing in sleep issues shared the following: “…often, I’ll have patients who say, well, ‘I really want to lose some weight.’ <…> And… so we’ll go that route instead of positive airway pressure. And I say, you know, ‘let’s reassess in a few months and see who you’re doing with that.’ There are other patients who really want to choose the CPAP, and even though they fail at it for months and even years sometimes, they don’t want to consider a different… treatment option. …And sometimes, they just <do not>… want treatment at all.” (PULM-PHYS-02). Although the concept of setting goals was relatively familiar, the interviewees were more concerned about the feasibility of conducting personal health planning as a comprehensive and longitudinal process. For example, a pulmonologist said the following in response to a question whether they had seen or added to a Veteran’s personal health plan: “No. [Laughing] I have not. It’s possible that I have seen it… and I was not–to be honest–you know, motivated to go look at it because we spend so much time in front of the computer that I try to do what I need to do to get the computer work done. So, I have not put together a… personal health plan for a patient. That would be, you know… an example of a perception of a specialist deferring that to the primary care physician because I think it’s wrapped up in their… management of their hypertension, management of their diet, their hyperlipidemia, their diabetes and so forth beyond just my recommendation that they walk or that they exercise that they, you know, whatever I can do to enroll them in either MOVE! Program or <pulmonary> rehab” (PULM-PHYS-01). In other words, while this participant was open to the idea of making a few recommendations or placing referrals, they saw health planning as more appropriate for a primary care provider to take on. Another pulmonologist expressed a similar opinion, albeit in a more forceful fashion. What is notable about the passage below is that this participant sees personal health planning as misaligned with their role in the context of regular outpatient care for Veterans with COPD yet endorses it as their dominant approach in the context of severe, end-of-life COPD: “It’s ridiculous. <…> I’m sorry, I’m being completely honest. …I’m a specialist so I am not the primary care doctor… …when my patient is coming to see me, they… want me to address their lung issues, their COPD issues… <….> Let me amend it this way. So, there are patients that I see in my clinic for COPD that are maybe severe end-stage who are at the tail end of their life where… I’m managing their health status in general. <…> …for these patients… my framework shifts, okay? I’m still the specialist but, yeah, absolutely, in that situation the framework of understanding, you know, what is more important to the patient, what are their goal <so> that framework applies, sure” (PULM-PHYS-03). This notion of personal health planning as lying outside of the pulmonologist’s regular role with regards to providing care to Veterans with COPD was challenged by a primary care provider: “I think that if you’re an expert what an expert does is understand all angles of a problem and be able to comment on it all, and use it all and integrate it all… I think it’s easy for people to say no, not me, no, not my specialty, but… who else’s specialty should it be, if not the expert’s on all of the things that contribute to COPD and all of the things that are available for COPD treatment?” (PRIMARY-PHYS-03). The issue of appropriate division of labor with regards to personal health planning came up in the interviewees in the WH service, as well. Several interviewees whose role involved clinician education felt that it would be appropriate for a provider to set a few shared goals during the appointment yet rely on the WH service to follow up with the Veteran on goal progress: “I think the peers and the coaches are the answer to providers who say, ‘I don’t have time for this.’ You know, the providers can introduce whole health and then can send veterans toward…the coaches to do much more in-depth work over time around Whole Health… so… a coach <can> meet with a veteran one-on-one to continue to talk about goals and values and what’s important. <…> So, I see… the coaches as absolutely being able to kind of expand the reach of Whole Health because they do have the time. That’s their job description. So, taking some of the burden off the providers who don’t have the time, right, to talk every week about Whole Health to individual patients” (WH-06). Offering a counterpoint, a WH Coach we interviewed spoke to the importance of a personal relationship with the clinician–in other words, instead of the one-directional approach (the coach takes over where the clinician leaves off), the coach and the clinician engage in a give and take dynamic, deliberating together over how to empower the Veteran: “I meet with different providers all the time because they call me and say, “Hey… I have this situation.’ You know, ‘I’m trying this with the guy.’ You know, ‘I’d like you to give it a shot or have a conversation with him.’ And oftentimes… because of the relationships that we have, you know, this Veteran’s getting that wrap-around support, not only from the Health Coach, but that provider’s buying into it, and that provider’s also, you know, starting to use some of the language and changing the conversation, as they say, around healthcare” (WH-04).
Most interviewees felt that discussing or providing Veterans with COPD with referrals to a broad range of services that may support their well-being was worthwhile, yet some challenges were also raised. To obtain a comprehensive picture, we inquired about services that are usually seen as falling under the WH umbrella, such as complementary and integrative health (CIH) and WH coaches, but also other services, including palliative care, mental health care, and chaplaincy. Two types of barriers to referrals came up in the interviews: (1) barriers related to the interviewees’ knowledge and attitudes about service referrals and (2) barriers related to logistics. For this manuscript, we only focus on the former as the latter are highly granular and specific to VHA and the specific VAMC setting. Generally, participants were open to the idea of connecting Veterans with COPD to services that support their well-being holistically. In a few cases, however, interviewees said that they had not previously considered the possibility and perhaps lacked the knowledge of how these services would help patients with COPD, but would now be interested in doing so in future: “You know, I honestly haven’t recommended it specifically for Veterans with COPD, but I would imagine for our… less severe cases…. it could be really helpful as far as maybe like some pneumonia prevention and getting some movement, getting up and moving around a little bit and helping their immune system with some blood flow ‘cause they aren’t really vigorous exercises, especially Tai Chi… <…> So, I’m sure that more severe COPD is probably—that’s inappropriate, too late, but for our earlier diagnoses, more well-controlled, where definitely exercise is still important, I think that, you know, it’d be interesting to see if it could help prevent pneumonia” (PRIMARY-NP-01). For example, none of the participants interviewed were aware of or previously considered connecting Veterans with COPD with a chaplain outside of the inpatient and/or end-of-life context: “I’ve never thought to bring it up in the outpatient setting… …I do attend on the inpatient medical wards and I know, sometimes I’m kind of struck by, ‘oh, the chaplain came by and saw the patient today and left a note,’ and it just kind of it stands out as something like, it’s just like a totally different world than what I practice. So, to be honest I’ve never considered bringing that into outpatient care. And, you know, I don’t think a Veteran has ever asked me to and I refuse, I don’t think I would ever do that, it’s just never occurred to me and it’s never occurred to me to suggest it if I detect that there was maybe spiritual need or maybe even spiritual distress that was in play” (PRIMARY-PHYS-02). One of the chaplains interviewed attributed this phenomenon to the inadequate awareness of what spiritual care is or how it is relevant to the Veteran’s team’s ability to provide whole-person care at all stages of life: “There is huge poverty of understanding. And a lot of denialism that happens. People know what they know, and they are less inclined to learn what they don’t know. You know, people will trivialize it with spiritual care because they don’t understand it. Maybe because they… figure for themselves, ‘if I don’t have any use for it, you don’t have any use for it.’ But……the least you can do when you don’t understand is to learn. <…> …because our target is the same. …we’re here to serve one individual and that’s the Veteran. If we understand that this Veteran has different dimensions and have been in several different places, then we cannot just compartmentalize ’em and say, ‘oh no, I’m just a psychiatrist.’ <…> So part of my goal is… that we can communicate… much more fluently… and we can tackle it from… different… specialty areas but knowing that… we are here to serve one individual” (CHAPL-01). Some participants also felt that Veterans with COPD themselves may see well-being offerings like CIH as unrealistic to pursue given the severity of their condition and/or logistical barriers, despite the obvious relevance of these offerings for improving their quality of life: “A lot of them have very little breath. <….> At rest, they’re just totally fine, but they’re sick and tired of sitting still. They don’t want to be at rest anymore. They remember when they could walk three miles a day, and they want to do that. They remember when they could wash the dishes without becoming winded. They remember when they could prepare dinner. They can’t stand to look out the window and see their wife shoveling the leaves because they want to do it. This is the pain of COPD. <…> But… they don’t like to leave the house, some of them. <…> Some of them have a car, some of them don’t. Some of them need to take their oxygen with them everywhere they go, and they’re afraid the tank will run out, you know. So, there’re a lot of barriers… and a lot of them are old, and they’re going to say, ‘Ach, what do I need with meditation?’ Or, like, ‘Yoga, that’s for young people.’ You know, I haven’t done a good job of selling it. I think that most of them could benefit, is the truth” (PALL-PHYS-01). In such scenarios, conveying the benefits and relevance of such services requires knowledge and tact on the part of the provider. One of the WH Coaches interviewed gave an example of displaying such tact: “So, I kind of explain what it’s like to be in the class. …I’ll make some recommendations of, ‘You have COPD. Avoid the hot Yoga; that is not going to be something that is in your wheelhouse to do right now.’ And then I’ll explain what it is and the benefits to them, as well, of, you know, typically people with COPD are older. Typically, they have some form of mobility issue. So, explaining that, you know, ‘This is going to help stretch you out a little bit. It’s going to work on strengthening some of your core. They do a lot of breathing during it, which can, of course, have a positive impact on your overall breathing.’ With the Tai Chi, I go into explaining, you know, ‘It has a lot of help with balance,’ ‘cause, again, something I’ve noticed is a lot of patients with COPD are also having some form of unrelated balance issues. So, again, explaining, ‘You know, this can help your balance. It can help your posture, which if you’re standing up straighter, your lungs are going to be a little bit bigger; might help you with some of the breathing’” (WH-02). A final barrier to placing referrals to services for holistically supporting Veterans’ well-being that was commonly mentioned was lack of awareness that these offerings are available: “So, I haven’t really thought about it… although I’m a yogi myself… …I just don’t necessarily think about referring…. I don’t have a problem with it… it never really occurred to me because I think I just wasn’t aware of it. So, I think that I probably am not aware of most of the offerings that would be available” (PULM-PHYS-01).
In this paper, we explored the multidisciplinary perspectives of VHA healthcare professionals on using WH Clinical Care with Veterans with COPD as a lens for investigating their attitudes and experiences with WH Clinical Care for patients who have complex chronic conditions, more broadly. We found that our interviewees experienced tensions and uncertainties across all four elements of WH Clinical Care: from person-centered assessment to understanding what really matters to Veterans, collaborative goal setting, and equipping Veterans with resources to holistically support their health and wellbeing. These challenges reflect a complex interplay of constructs across all domains of our initial analytical framework, CFIR–from the inherent characteristics of WH Clinical Care as an intervention (its complexity and adaptability) to the features of both the inner setting (especially the quality of professional networks and communication, but also organizational culture and implementation climate) and the outer setting (particularly patient needs as understood by our interviewees), alike. Individual characteristics of our participants, and especially their knowledge and beliefs about WH, also undoubtedly shaped their perspectives. Our work carries several important implications that are relevant not only to VHA’s ongoing efforts to infuse a WH approach throughout its system of care, but also to other organizations and systems that are concerned with implementing person-centered models of care–both with regards to patients with complex chronic conditions like COPD and on the system level, as a whole. We elaborate on these general lessons below, focusing on three areas in particular: (1) the paradox of selective implementation; (2) the uncertainties around division of labor and coordination; and (3) the challenge of navigating the inherent tensions between the ethos of person-centered care, on one hand, and the logics of biomedical rationality and economic expediency, on the other. Our first finding of interest is that our participants felt affinity with some elements of WH Clinical Care but not others (for example, a physician may be open to placing a referral to a CIH provider but not to engaging in personal health planning around what really matters to the patient). In other words, instead of perceiving WH Clinical Care as a single paradigm to be followed in its entirety, healthcare professionals may draw piecemeal on specific tools and concepts. This selective approach to WH Clinical Care is not necessarily at odds with VHA’s own vision. The WH Clinical Care implementation efforts to date have been built on the premise that many clinicians may already be utilizing at least some person-centered care principles in their practice, even if unwittingly or inconsistently so. Indeed, the WH Implementation Guide reads, “Often you may discover that a person or team’s practice already includes elements of Whole Health Clinical Care . Meet clinicians where they are, connect Whole Health with the work they are already doing or are required to do, and partner with them to establish goals to help them move towards a fully transformed Whole Health Clinical Care approach” (emphasis ours). The phenomenon of selective uptake of WH Clinical Care can be viewed through two contrasting, yet complementary lenses. On one hand, it presents an opportunity: Not treating WH Clinical Care–or person-centered care, more broadly–as an all-or-nothing proposition may empower clinicians to incorporate those elements of the approach that are most aligned with their established interests and strengths into practice. Paradoxically, however, this selective uptake of person-centered care principles and practices may create a challenge of its own. In implementation science terms, it is the challenge of balancing adaptability vs. fidelity: successful implementation initiatives must tailor an intervention to fit the context while also preserving the intervention’s core, indispensable characteristics, yet the boundary between the core and the periphery may be far from obvious [ , , ]. In the case of complex, multi-level interventions such as WH Clinical Care or any other person-centered care initiative, this ambiguity is all the more pronounced. The biomedical paradigm is strongly entrenched in healthcare settings. Therefore, there is a risk that clinicians may end up using person-centered care practices sporadically and without fully embracing the spirit of person-centered care, which would defeat the very purpose of implementation. Any educational and outreach efforts undertaken to support the implementation of WH Clinical Care–and person-centered care, more broadly–must not only support healthcare professionals in incorporating select person-centered practices and tools into their established approach, but also encourage them to critically rethink the approach itself so as to promote a true paradigm shift from the biomedical model to a person-centered one. i Individualized coaching or audit and feedback interventions , used iteratively and over a period of time, may be well-suited for this purpose, but larger culture change interventions are also crucial. Our second finding with broader implications concerns the uncertainty about an appropriate division of labor between various types of healthcare professionals in providing WH Clinical Care. Some of our participants seemed to think that the responsibility for the entirety of WH Clinical Care lies on the physician’s–particularly, primary care physician’s–shoulders, which was, understandably, perceived as burdensome. Others perceived WH as mainly the responsibility of other individuals–PACT nurses, WH Coaches, other WH service employees (e.g., CIH instructors, WH educators)–and reported little to no integration of a WH approach into their practice. This finding speaks to the importance of a coordinated approach across multiple disciplines and services in implementing WH Clinical Care. Using the language of normalization process theory , an approach that seeks to describe how interventions get embedded into the everyday fabric of practices within an organization, WH implementation efforts are yet to achieve a sufficient degree of either relational integration (seamless incorporation of the innovation–in this case, WH–into the existing networks of professional relationships between individuals and groups) or contextual integration (integration of the innovation into the existing organizational context, including structures and practices). This challenge is hardly unique to the VHA setting. Many systems within and outside the U.S. are grappling with the imperative of scaling up person-centered care principles beyond the patient-clinician dyadic encounter–i.e., making care coordination itself person-centered [ – ]. In VHA and beyond, overcoming this challenge would require an intentional effort to leverage teams and collaborative relationships more broadly within and across clinical settings. Approaches such as relationship-centered care and relational coordination may be well-poised to help guide or at least inform such efforts. In general, however, while care coordination and/or integration are enshrined in some patient-centered and person-centered care frameworks, there is yet insufficient guidance around the best practices for coordinating care in a way that is consistent with what really matters to the patient, and there are still numerous opportunities for greater conceptual clarity that can be addressed by future research . Finally, our work highlights both the importance and challenges of explicitly attending to rhetoric and discourse of person-centered care in general and the WH model in particular . Specifically, our findings highlight an important tension at the heart of the WH initiative. In our interviews, we uncovered two perspectives. One embraces the ethical imperative of structuring care around patients’ values, goals, and needs, whatever these may be. The other, however, sees WH Clinical Care as valuable only to the extent that it helps achieve a predetermined objective –Veterans’ compliance with treatment and engagement in healthy behaviors, more broadly. These contrasting views, we posit, derive from the lingering tension in the discursive framing that VA’s Office of Patient-Centered Care & Cultural Transformation (OPCC&CT) has adopted in its advocacy for WH–a tension that can be seen in broader discussions around person-centered care beyond the VHA context. WH promotional and educational materials tend to emphasize that the use of WH Clinical Care by healthcare providers would both create a meaningful, supportive environment for the Veteran and result in downstream outcomes that are desirable from the healthcare system standpoint and underpin healthcare providers’ performance metrics (i.e., improvement in clinically meaningful outcomes, reduced care utilization, lower cost burden). It is understandable that the case for person-centered care may need to be supported with arguments that are compelling from a biomedical and economic standpoint. However, we argue that efforts to implement person-centered care–in the VHA, as well as beyond–can benefit from a more attentive and intentional approach to discourse and framing. Research has shown that patient-centered and person-centered care language can be coopted and leveraged in the service of a disease-centered and provider-centric agenda . Outreach and education efforts ought to spend more time unpacking the tension between the biomedical and the person-centered models. For example, training programs could incorporate working through and debriefing on s scenarios wherein the ethos of person-centered care clashes with the logics of biomedical expediency and economic rationality instead of almost exclusively dwelling on situations when the two are aligned. Vignettes and case studies could explore how to navigate situations when the patient’s values, goals, and priorities turn out to not fully align with the clinician’s vision for what may be best for the patient. The training could then emphasize that a truly person-centered approach requires that the clinician and patient build on a foundation of trust and explore how clinical expertise can be leveraged in the service of the patient’s goals. Such vignettes could even show how the areas of misalignment or disagreement may shrink over time, as long as this possibility is not overstated in excessively idealistic terms. On a more fundamental level, however, such educational efforts would not resolve the underlying conflict between the overt endorsement of person-centered care and the organizational priorities and incentives that may be directly at odds with person-centered care principles.. It appears that institutional person-centered care implementation efforts are plagued by a contradiction: they are targeting institutions whose entrenched assumptions and routines reflect a biomedical (disease-centered) paradigm, a directive rather than partnership-oriented approach to the patient-clinician relationship, and a preoccupation with minimizing costs and maximizing efficiency above all else. The tension between the ethos of person-centered care and organizational structures and priorities is, once again, not unique to the VHA setting, let alone to person-centered care for patients with COPD, and has been noted in such diverse contexts as home-based care for older adults, care for individuals with kidney failure, diabetes self-management programs, and others [ – ]. When structures and norms that are at odds with person-centered care are still in place and unquestioned, implementation efforts may succeed in transforming healthcare workers’ beliefs about the value of being person-centered and even some of their practices, yet the gulf “between knowing and doing person-centeredness,” as Franklin and colleagues put it , would remain. No simple solutions here can be offered beyond making a larger observation that as VHA and other systems in the United States and beyond progress in their person-centered care implementation efforts, these tensions may become even more overt and profound, stimulating difficult yet important conversations. The main limitation of our study is that we collected our data at a single–albeit large and influential–health system in the United States, the Veterans Health Administration. The VHA possesses organizational characteristics that make it stand out in the U.S. healthcare landscape (e.g., national-level integration, focus on military Veterans and their families/caregivers, concern with providing comprehensive care with robust mental health and social support components). Additionally, the VHA is embedded in the U.S. sociocultural and political-economic context, which differs in significant ways from healthcare organizations and health systems in other regions of the world. However, we argue that the overall implications of our study are broadly relevant and transcend both the VHA and U.S. context. The importance of going beyond the clinician-patient dyad to leverage team-based and inter-service care coordination has been broadly recognized and incorporated into numerous existing conceptual frameworks of person-centered and patient-centered care [ , , ]. The difficulty of integrating person-centered/holistic approaches into largely biomedical/disease-centered systems has also been described in various settings worldwide , and may indeed be one of the foundational challenges at the heart of person-centered care implementation. Finally, the tension between the ethos of person-centered care and the logics of economic efficiency is hardly a VHA- or U.S.-specific phenomenon. Indeed, health systems in other parts of the world, including Europe and the U.K., have been under an increasing pressure from austerity policies . The other key limitation of the study is that we sought to explore healthcare professionals’ perspectives on the use of WH Clinical Care with patients living with complex chronic conditions through the lens of a specific condition–COPD. This focus has enabled us to obtain rich, example-specific perspectives. At the same time, however, we were not able to explore all aspects of WH Clinical Care for patients with complex chronic conditions due to the enormity of the topic. While we maintain that the key insights shared by our interviewees are broad and not COPD-specific, it is possible that we would have heard about different challenges and different understandings of WH Clinical Care, had we chosen to select a different condition or several conditions as the lens for our inquiry.
We conducted a qualitative research study to understand the perspectives of healthcare professionals in the largest nationally integrated health system in the U.S., the Veterans Health Administration, on practicing Whole Health Clinical Care with Veterans with complex chronic conditions, using their experiences of providing care to Veterans with COPD in particular as the guiding lens for our inquiry. We identified questions and concerns that participants had regarding each of the four components of WH Clinical Care. We further explored the broader relevance of our findings, including the implications of the phenomenon of selective uptake of person-centered care elements by healthcare professionals for education and outreach efforts, the importance for a team-based approach for sustainable person-centered care implementation, and the imperative of grappling with the lingering tension between the ethos of person-centered care and the logics of biomedical rationality and economic expediency. Our work can serve as a template for future research efforts focused on investigating person-centered care experiences of healthcare professionals in different care settings and services.
S1 Appendix Interview guides. (DOCX) Click here for additional data file.
|
Workforce Trends Among Canadian Medical Oncologists and Medical Oncology Trainees over Two Decades | defb79da-1fd5-4c41-afe1-db2dc512dfcb | 11854901 | Internal Medicine[mh] | Cancer is the leading cause of death in Canada, accounting for 27% of all national deaths at the time of the last national census in 2023 . Additionally, cancer rates have risen annually in Canada, with 239,100 cases projected for the year 2023 . The rising burden of cancer within Canada is of interest to health system decision makers. Global projections predict a rise in both the incidence of cancer as well as the demand for systemic therapy in upcoming years . Medical oncologists are physicians who manage the care of cancer patients and oversee the delivery of systemic therapy and, as a whole, they can be described as the MO workforce. Supply and demand projections from America, Europe, Australia, and Spain suggest that the demand for oncologic services may outstrip the supply if adequate workforce growth is not maintained . Thus, from a policy perspective, it is vital that trends in the MO workforce within Canada are summarized such that prior policy can be evaluated and future policy directed in order to ensure an adequate supply of MOs to service the country’s cancer system. Although workforce trends for radiation oncologists in Canada have been published , no readily available summary of the Canadian MO workforce has been performed. An annual headcount from the Canadian Medical Association (CMA) provides an overview of the MO workforce nationally in a given year but with limited examination of longitudinal trends and regional data and, additionally, there is no information about demographics trends of MO trainees (residents and fellows) over time, which is key to informing future modeling of workforce supply. Recent work predicting future MO workforce requirements within Canada focused on the number of MOs in the country as a whole without acknowledging that there are likely regional differences in MO supply and, thus, possible regional shortfalls, which would need to be addressed for optimal cancer care . Importantly, this work did not include any analysis on the supply of trainees into the MO system, which directly affects the future workforce. Therefore, we aim to collate available data on the MO workforce in Canada and to perform a descriptive analysis of this workforce over a 25-year span. Here, we describe the changes in the composition of the existing Canadian MO workforce and the supply of future MOs (trainees) nationally and by region between 1994 and 2020. A secondary objective was to characterize MO workforce metrics relative to annual cancer diagnoses over time as a surrogate of cancer burden relative to provider coverage. 2.1. Study Design This study is a descriptive analysis of pre-existing databases. Because no hypothesis testing was conducted and the purpose of this analysis was to serve as a summary of the available data, no statistical analysis was performed. 2.2. Databases for MO Workforce Information on practicing MO was obtained from both the Canadian Medical Association—Physician Data Centre (CMA-PDC; 1994–2019) and the Canadian Institute for Health information—Scott’s Medical Database (CIHI-SMDB; 1994–2020) . Details of each database used are shown in . Data from each database were manually extracted by one reviewer (AF). MO workforce data were extracted for the years 1994–2020 because these years included the greatest overlap between the CIHI-SMDB database (1968–2020) and the CMA-PDC database (1994–2019). Each database reports similar demographic data including the number of MO practicing in each province, as well as their gender and age. For categories where both databases contained identical information such as the number of MO, the data were averaged. The age distribution of MO was categorized using different cutoffs in the CIHI-SMDB versus CMA-PDC databases; thus, data could not be aggregated and were instead reported separately for each database. 2.3. Database for MO Trainees Information on MO trainees (1994–2020) was derived from the Canadian Post M.D Education Registry (CAPER) database . The CAPER database collects data submitted by the post-graduate programs of all 17 Canadian medical schools and includes two-year residents as well as fellows undergoing additional training after graduation from medical oncology residency programs. The 1994–1999 reports were obtained from CAPER and added to the CAPER subspecialty reports for 2000–2019, which are available online. Our analysis excludes visa-sponsored trainees who are required to return to their home country after completion of training. The data are tabulated per academic year (July–June of the following year). 2.4. Regional Analyses Regional analyses were performed according to the official regions of Canada . Canada is divided into the following regions: (1) West Coast (British Columbia); (2) Prairie Provinces (Alberta, Saskatchewan, and Manitoba); (3) Central Canada (Ontario and Quebec); and (4) Atlantic Canada (Newfoundland, Nova Scotia, New Brunswick, and Prince Edward Island). As Central Canada contains the two most populous Canadian provinces (Quebec, and Ontario), separate analysis were performed for each province. The Northern Territories were not included in the analysis since there are no MO or training programs in these regions. 2.5. Cancer Incidence National and regional cancer incidence data for 1994 to 2020 were obtained from the Canadian Cancer Registry database with Statistics Canada. Projected cancer incidence data were derived using CANPROJ projection software . CANPROJ is a validated projection modeling package in R software (version 4.1.0, R Foundation for Statistical Computing, Vienna, Austria and RStudio version 1.4.1717 RStudio Inc., Boston, MA, USA) that uses trends in historical data to select the best-fit model for projected years based on a decision algorithm comprised of six age–period–cohort models . Since cancer incidence data for the province of Quebec were not available from 2011 onwards, Quebec-specific incidence counts were projected from 2011 to 2020 versus 2020 only for all other regions . Annual ratios of incident cancer cases per medical oncologist were calculated using the number of actual or projected incident cancer cases in Canadians 20–90 years old divided by the number of practicing medical oncologists from 1994 to 2020. Cancer incidence in Canadians 0–19 years old was excluded because the care of cancer patients who are under 18 years old are managed by pediatric oncologists. Both national and regional incident cancer cases per medical oncologist ratios were generated. This study is a descriptive analysis of pre-existing databases. Because no hypothesis testing was conducted and the purpose of this analysis was to serve as a summary of the available data, no statistical analysis was performed. Information on practicing MO was obtained from both the Canadian Medical Association—Physician Data Centre (CMA-PDC; 1994–2019) and the Canadian Institute for Health information—Scott’s Medical Database (CIHI-SMDB; 1994–2020) . Details of each database used are shown in . Data from each database were manually extracted by one reviewer (AF). MO workforce data were extracted for the years 1994–2020 because these years included the greatest overlap between the CIHI-SMDB database (1968–2020) and the CMA-PDC database (1994–2019). Each database reports similar demographic data including the number of MO practicing in each province, as well as their gender and age. For categories where both databases contained identical information such as the number of MO, the data were averaged. The age distribution of MO was categorized using different cutoffs in the CIHI-SMDB versus CMA-PDC databases; thus, data could not be aggregated and were instead reported separately for each database. Information on MO trainees (1994–2020) was derived from the Canadian Post M.D Education Registry (CAPER) database . The CAPER database collects data submitted by the post-graduate programs of all 17 Canadian medical schools and includes two-year residents as well as fellows undergoing additional training after graduation from medical oncology residency programs. The 1994–1999 reports were obtained from CAPER and added to the CAPER subspecialty reports for 2000–2019, which are available online. Our analysis excludes visa-sponsored trainees who are required to return to their home country after completion of training. The data are tabulated per academic year (July–June of the following year). Regional analyses were performed according to the official regions of Canada . Canada is divided into the following regions: (1) West Coast (British Columbia); (2) Prairie Provinces (Alberta, Saskatchewan, and Manitoba); (3) Central Canada (Ontario and Quebec); and (4) Atlantic Canada (Newfoundland, Nova Scotia, New Brunswick, and Prince Edward Island). As Central Canada contains the two most populous Canadian provinces (Quebec, and Ontario), separate analysis were performed for each province. The Northern Territories were not included in the analysis since there are no MO or training programs in these regions. National and regional cancer incidence data for 1994 to 2020 were obtained from the Canadian Cancer Registry database with Statistics Canada. Projected cancer incidence data were derived using CANPROJ projection software . CANPROJ is a validated projection modeling package in R software (version 4.1.0, R Foundation for Statistical Computing, Vienna, Austria and RStudio version 1.4.1717 RStudio Inc., Boston, MA, USA) that uses trends in historical data to select the best-fit model for projected years based on a decision algorithm comprised of six age–period–cohort models . Since cancer incidence data for the province of Quebec were not available from 2011 onwards, Quebec-specific incidence counts were projected from 2011 to 2020 versus 2020 only for all other regions . Annual ratios of incident cancer cases per medical oncologist were calculated using the number of actual or projected incident cancer cases in Canadians 20–90 years old divided by the number of practicing medical oncologists from 1994 to 2020. Cancer incidence in Canadians 0–19 years old was excluded because the care of cancer patients who are under 18 years old are managed by pediatric oncologists. Both national and regional incident cancer cases per medical oncologist ratios were generated. 3.1. National Demographic Trends for MO Between 1994 and 2020, the number of medical oncologists in Canada rose from 161 in 1994 to 642 in 2020, an absolute increase of 481 MOs (+299%) ( a). The average annual growth rate of the MO workforce over the period was 5.7% per year, with a maximum increase of 39.9% between 1994 and 1995 ( a). The minimum rate of growth of 1% occurred between 2017 and 2018. Over the study period, the number of incident cancer cases in Canada rose by 88% from 120,255 in 1994 to 225,800 in 2020. The incident cancer cases per medical oncologist ratio nationally fell from 749:1 to 352:1 ( b). The proportion of female oncologists in Canada rose from 24.9% in 1994 to 46.0% in 2020 ( c). The proportion of Canadian MO attaining their medical degree in Canada was stable over time, with 77% and 82% of MO attaining their MD in Canada in 1994 and 2020, respectively ( d). The average age of the MO workforce in Canada rose from 44 years old in 1994 to 48 in 2020. ( a). The proportion of oncologists in each age category over time as estimated by each database are depicted in b (CIHI-SMDB) and c (CMA-PDC). In 1994 and 2020, the CIHI proportion of oncologists <39 in age was similar at 33% and 32%, respectively; however, the proportion between age 40 and 49 fell from 44% to 28%, while the proportion between 50 and 64 rose from 22% to 29%. Only 2% of oncologists were over 65 in 1994 and, by 2020, this number had risen to 11%. The CMA-PDC age data show a similar trend, with the proportion of oncologists aged 55 years old or greater rising from 15% in 1994 to 31% in 2019 ( c). 3.2. Regional Demographic Trends for MO The number of MO increased across all regions of Canada over the study period ( a). The largest proportional increase was seen in the Atlantic provinces, with 6 MO in 1994 compared to 38 in 2020 (+533%), followed by the West Coast (17 to 105:+518%), Prairie Provinces (15 to 91:+507%), Ontario (64 to 219:+242%), and Quebec (59 to 189:+220%). The increase in the number of MOs in all regions was higher over the first half of the study interval (1994–2007) versus the latter half (2008–2019) ( a). Between 1994 and 2020, the ratio of incident cancer cases per MO provider fell across all regions ( b). In 1994, the Atlantic and Prairie Provinces had higher incident cases per MO ratios than the other regions, with ratios of 1836:1 and 1257:1, respectively, compared to 909:1 on the West Coast, 682:1 in Ontario, and 543:1 in Quebec. In 2020, the highest ratio was in Atlantic Canada (447:1), followed by Ontario (411:1), the Prairie Provinces (377:1), Quebec (300:1), and the West Coast (261:1). Atlantic Canada is the only region that experienced repeated peaks in this ratio, with specific maximums seen in 1998 (1154:1) and 2006 (980:1) after sustained periods where the ratio fell consistently ( b). According to CIHI-SMDB for the year 2020, Atlantic Canada has the lowest proportion of female MO (34%), followed by the Prairie Provinces (35%), Ontario (46%), Quebec (50%), and the West Coast (52%) ( c). The proportion of female MO providers increased in all regions between 1994 and 2020. Regional variation in the age composition of oncologists according to CIHI-SMB is shown in for all regions. In 1994, the average age of an MO was highest in Quebec (48 years) compared with the West Coast (43), Atlantic Canada (42), the Prairie Provinces (42), and Ontario (40), which had the lowest average age. In contrast, in 2020, the average age of MO was the highest in Ontario (49 years), followed by Atlantic Canada (49 years), Quebec (48 years), the Prairie Provinces (47 years), and the West Coast (46 years). 3.3. Medical Oncologist Trainees The number of MO trainees in Canada rose from 34 in the 1994–1995 academic year to 99 in the 2019–2020 academic year, representing a 191% increase ( a). These trainees were trained at nine universities in four provinces (British Columbia, Alberta, Ontario, and Quebec). With time, MO training programs emerged at the University of Manitoba (Manitoba) in 1993/94, University of Laval (Quebec) in 1995/96, Dalhousie University (Nova Scotia) in 2003/04, Queens University (Ontario) in 2006/07, University of Sherbrooke (Quebec) in 2010/11, and Memorial University (Newfoundland and Labrador) in 2015/16, increasing training capacity to 15 residency programs in seven provinces. No MO training programs exist in Saskatchewan, New Brunswick, Prince Edward Island, and the Canadian Territories. The proportion of MO trainees by gender is shown in b. Female trainees outnumbered male trainees for all years except for 1994–1995 and 2006–2008 ( b). In 1994, 66% of trainees were male compared to 42% in 2019, corresponding to a shift towards consistently more female MO trainees throughout the study period. The regional supply of trainees through time is shown in c. The raw number of trainees has risen across all regions, but the proportional increase varies by region. The proportion of total Canadian MO trainees supplied by each province is shown in d. Only 1 trainee per year came from Quebec in 1994 (3% of total trainees) but the number increased to 18 (18% of total trainees) in 2019, a rise of 1700%. Atlantic Canada trained four MO trainees (4% of total) in 2019 compared with none in 1994. While the proportion of total MO trainees trained in Ontario has fallen between 1994 (56%; 20 trainees) and 2019 (49%; 49 trainees) the absolute number has risen. Similarly, the proportion of MO trainees trained on the West Coast has also fallen between 1994 (15%; 5 trainees) and 2019 (12%; 12 trainees). The Prairie Provinces have maintained the same portion of national trainees between 1994 (15%; 5 trainees) and 2019 (16%; 16 trainees) while increasing the absolute number of trainees. Between 1994 and 2020, the number of medical oncologists in Canada rose from 161 in 1994 to 642 in 2020, an absolute increase of 481 MOs (+299%) ( a). The average annual growth rate of the MO workforce over the period was 5.7% per year, with a maximum increase of 39.9% between 1994 and 1995 ( a). The minimum rate of growth of 1% occurred between 2017 and 2018. Over the study period, the number of incident cancer cases in Canada rose by 88% from 120,255 in 1994 to 225,800 in 2020. The incident cancer cases per medical oncologist ratio nationally fell from 749:1 to 352:1 ( b). The proportion of female oncologists in Canada rose from 24.9% in 1994 to 46.0% in 2020 ( c). The proportion of Canadian MO attaining their medical degree in Canada was stable over time, with 77% and 82% of MO attaining their MD in Canada in 1994 and 2020, respectively ( d). The average age of the MO workforce in Canada rose from 44 years old in 1994 to 48 in 2020. ( a). The proportion of oncologists in each age category over time as estimated by each database are depicted in b (CIHI-SMDB) and c (CMA-PDC). In 1994 and 2020, the CIHI proportion of oncologists <39 in age was similar at 33% and 32%, respectively; however, the proportion between age 40 and 49 fell from 44% to 28%, while the proportion between 50 and 64 rose from 22% to 29%. Only 2% of oncologists were over 65 in 1994 and, by 2020, this number had risen to 11%. The CMA-PDC age data show a similar trend, with the proportion of oncologists aged 55 years old or greater rising from 15% in 1994 to 31% in 2019 ( c). The number of MO increased across all regions of Canada over the study period ( a). The largest proportional increase was seen in the Atlantic provinces, with 6 MO in 1994 compared to 38 in 2020 (+533%), followed by the West Coast (17 to 105:+518%), Prairie Provinces (15 to 91:+507%), Ontario (64 to 219:+242%), and Quebec (59 to 189:+220%). The increase in the number of MOs in all regions was higher over the first half of the study interval (1994–2007) versus the latter half (2008–2019) ( a). Between 1994 and 2020, the ratio of incident cancer cases per MO provider fell across all regions ( b). In 1994, the Atlantic and Prairie Provinces had higher incident cases per MO ratios than the other regions, with ratios of 1836:1 and 1257:1, respectively, compared to 909:1 on the West Coast, 682:1 in Ontario, and 543:1 in Quebec. In 2020, the highest ratio was in Atlantic Canada (447:1), followed by Ontario (411:1), the Prairie Provinces (377:1), Quebec (300:1), and the West Coast (261:1). Atlantic Canada is the only region that experienced repeated peaks in this ratio, with specific maximums seen in 1998 (1154:1) and 2006 (980:1) after sustained periods where the ratio fell consistently ( b). According to CIHI-SMDB for the year 2020, Atlantic Canada has the lowest proportion of female MO (34%), followed by the Prairie Provinces (35%), Ontario (46%), Quebec (50%), and the West Coast (52%) ( c). The proportion of female MO providers increased in all regions between 1994 and 2020. Regional variation in the age composition of oncologists according to CIHI-SMB is shown in for all regions. In 1994, the average age of an MO was highest in Quebec (48 years) compared with the West Coast (43), Atlantic Canada (42), the Prairie Provinces (42), and Ontario (40), which had the lowest average age. In contrast, in 2020, the average age of MO was the highest in Ontario (49 years), followed by Atlantic Canada (49 years), Quebec (48 years), the Prairie Provinces (47 years), and the West Coast (46 years). The number of MO trainees in Canada rose from 34 in the 1994–1995 academic year to 99 in the 2019–2020 academic year, representing a 191% increase ( a). These trainees were trained at nine universities in four provinces (British Columbia, Alberta, Ontario, and Quebec). With time, MO training programs emerged at the University of Manitoba (Manitoba) in 1993/94, University of Laval (Quebec) in 1995/96, Dalhousie University (Nova Scotia) in 2003/04, Queens University (Ontario) in 2006/07, University of Sherbrooke (Quebec) in 2010/11, and Memorial University (Newfoundland and Labrador) in 2015/16, increasing training capacity to 15 residency programs in seven provinces. No MO training programs exist in Saskatchewan, New Brunswick, Prince Edward Island, and the Canadian Territories. The proportion of MO trainees by gender is shown in b. Female trainees outnumbered male trainees for all years except for 1994–1995 and 2006–2008 ( b). In 1994, 66% of trainees were male compared to 42% in 2019, corresponding to a shift towards consistently more female MO trainees throughout the study period. The regional supply of trainees through time is shown in c. The raw number of trainees has risen across all regions, but the proportional increase varies by region. The proportion of total Canadian MO trainees supplied by each province is shown in d. Only 1 trainee per year came from Quebec in 1994 (3% of total trainees) but the number increased to 18 (18% of total trainees) in 2019, a rise of 1700%. Atlantic Canada trained four MO trainees (4% of total) in 2019 compared with none in 1994. While the proportion of total MO trainees trained in Ontario has fallen between 1994 (56%; 20 trainees) and 2019 (49%; 49 trainees) the absolute number has risen. Similarly, the proportion of MO trainees trained on the West Coast has also fallen between 1994 (15%; 5 trainees) and 2019 (12%; 12 trainees). The Prairie Provinces have maintained the same portion of national trainees between 1994 (15%; 5 trainees) and 2019 (16%; 16 trainees) while increasing the absolute number of trainees. This is the first report describing temporal trends in the Canadian MO workforce, including the MO trainee cohort, by age, gender, and region over a 25-year period. Several important findings have emerged. Between 1994 and 2019, the Canadian MO workforce has increased by 298%, a pace exceeding the rise in cancer incidence. Regional differences in the supply of medical oncologists relative to regional cancer incidence were evident, with Atlantic Canada persistently having the lowest MO supply relative to its annual cancer incidence. Over time, the average age of Canadian MO has increased and, in Ontario, a larger proportion of MO are nearing retirement age than other provinces. The rise in the number of incident cancer cases relative to MO providers has resulted in a lower ratio of annual incident cancers per MO provider over time, from 749:1 in 1994 to 352:1 in 2020. These data could be explained by either the appropriate expansion of the MO workforce to accommodate increased demand or initial undersupply of MO relative to demand, followed by a supply correction. To determine whether Canada at present has an undersupply of oncologists relative to its cancer incidence, we compared Canada’s incident cancer/MO ratio to that of other health systems across similar developed countries . Canada’s ratio of incident cancer per MO provider remains higher than the United States, Australia, and most European comparators . Although the number of MO in Canada has increased with time, comparative data suggest a relative undersupply persists. What is perhaps most concerning is that, although Australia has a higher supply of MO compared with our estimates, a recent study from Australia suggests that even their workforce is inadequate to meet current oncologic demand based on Australian providers seeing higher than expected new consult numbers per year . Empiric evidence of Canadian MO undersupply is shown by recent reporting, suggesting significant delays in receiving chemotherapy in both the West Coast and Prairie regions of Canada . Characterizing growth in the MO workforce may not accurately quantify workload per MO. Estimates suggest that 57% of all incident cancer will require systemic therapy; however, this remains the most used workforce metric across the literature; thus, this is the metric we examined . MO workload is influenced by complex factors, including referral rates, cancer incidence rates in each jurisdiction, treatment complexity, potential inpatient care for treatment toxicity or complications, and participation in clinical trial or medical education activities . A recent Canadian projection model has suggested that, even if the number of Canadian MO increases with time relative to cancer incidence, the number of systemic therapy starts per oncologist is still projected to rise and patients are receiving more lines of treatment than in the past . Therefore, cancer incidence and consultation workload are imperfect surrogates for provider workload and more robust metrics are needed. This is further supported by the fact that waitlists on the west coast are currently far in excess of acceptable standards for Canada; yet, according to our study, they have the most favorable incidence to MO ratio and, thus, we would expect to see the lowest wait time in this region . Therefore, updating the Canadian Oncology Utilization study, completed in 2002, would be of value, as many of the assumptions from this study are unlikely to retain validity 20 years later . Comparing our findings with Canadian Radiation Oncology (RO) workforce data , there were no notable peaks in the ratio of cancer incidence per provider among MO as there were in RO, which rose to a peak ratio of 1196:1 in 2003/2006. Drivers of this peak were multifactorial but may have included large swings in the number of Canadian RO trainees coinciding with a large number of retirements in Quebec and an exodus of trainees due to poor job market conditions between 1996 and 2001 . This underscores the importance of a stable trainee supply if we are to maintain a stable oncology workforce and avoid the undersupply issues seen in radiation oncology. Regionally, Atlantic Canada appears to have a relative undersupply of oncologists compared with the rest of the country. In 2020, Atlantic Canada had the highest ratio of annual incident cancer cases per MO provider at 447:1 and the lowest ratio with 261:1 in BC. This trend parallels the RO literature, with Atlantic Canada having a higher proportion of the nation’s cancer cases than they have of the countries ROs . The reasons for this are not entirely clear but warrant further exploration. Aside from Quebec, there is a trend towards an older demographic, with a larger proportion of MO now in the age bracket >50 years of age across Canada. For example, in Ontario, 44% of MO are now >50 years old, with 14% now over the age of 65 years compared with 6% and none in 1994, respectively. As more MOs reach the typical retirement age of 65 , increased turnover and replacements will be required to maintain stability in the workforce. In Quebec, the average age of MO rose rapidly in 1994 when many practicing hematologists recertified as medical oncologists, which, together with low trainee numbers, led to the Quebec MO population being older than the rest of Canada. More recently, the proportion of Quebec MOs < 50 years of age has started to rise , as the proportion of Canadian MO trainees from Quebec has risen from 2/26 (8%) in 1994 to 18/99 (18%) in 2019. These data support the assertion that workforce numbers can potentially be modulated through increases in the number of trainees in each region. We observed a trend towards gender parity over time in the MO population. As of 2020, 46% of medical oncologists were female compared with only 25% in 1994. However, this proportion varied by region, with only 34% and 35% of MO in Atlantic Canada and the Prairies, respectively, being female in 2020. Our findings mirror a sample of Australian MO from 2018, which showed that 49% of survey respondents were female . Prior qualitative evidence suggests the emergence of a perception that female MO are becoming more common and that the specialty itself is changing to accommodate the needs of gender balance . Specifics as to what this accommodation means are not given in this paper. Between 1997 and 2001, there was an apparent contracture in MO residents, with only 39, 37, 31, 37, and 42 residents in each year, respectively. Ontario alone had 17 MO residents in 1996, but this fell to 13, 10, 17, 14, and 16 for the ensuing five years. Although the number of Canadian MOs expanded during this time, the population of MOs in Ontario declined slightly, hitting a trough in the year 2000. A similar trend is observed in Quebec between 1994 and 2008, where trainee numbers are proportionately lower compared with the rest of Canada. The stagnant MO trainee numbers may reflect healthcare cuts in the mid-1990s across multiple jurisdictions, including Quebec . Anecdotally, our colleagues in Quebec have endorsed a period ~1998–2008 where medical school enrollment was actively cut, leading to a subsequent decline in subspecialty training. Following reversal of these spending cuts, there was both a rapid increase in the proportion of trainees from Quebec after 2012 and the slope of the MO supply curve thereafter ( and ). Several policy implications arise from our study. First, Canada appears to have a relative undersupply of medical oncologists relative to many other developed health systems. Thus, supply side measures to increase the number of MO would be beneficial. Theoretically increasing the number of oncology training slots should help to increase the net supply of oncologists in Canada. However, a look at the 2025 Canadian Residents Matching Service (CaRMs) subspecialty match report suggests that there were unfilled MO residency slots in all of our described regions . Addressing this supply side shortfall will require a better understanding of the underlying causes for the shortfall and strategies targeting the specific issues identified. For instance, the number of domestic trainees remains low, policies to facilitate recruitment of more foreign trained MOs may help address wait times. Our study has several limitations. Most importantly, the numbers reported by the two major databases used are not entirely concordant from year to year. Thus, the exact number of active MO in the country is not precisely known, which means these data should not be used for granular policy decisions such as local hiring of MO. Additionally, both databases may overestimate the number of practicing MO in a region. For example, the number of departures from the MO workforce are not reliably captured by either database; counting oncologists who have retired or left the country in our sample could over-estimate the supply of MO in Canada for a given year. Additionally, neither database captures the full-time equivalent (FTE) an individual MO works nor part-time employment, making it challenging to ascertain the full capacity of the workforce. Furthermore, the main metric of annual incident cancer per MO provider is only a crude estimate of the relationship between supply and demand. This metric only takes into account new cancer cases but does not account for cancer prevalence, referral rates, surveillance for cancer recurrence or progression, treatment complexity, and risk of complications, all of which contribute to the clinical workload in practice. Finally, we did not report the number of dedicated fellowship positions per region. Theoretically, provinces with more fellowship slots may ultimately retain a higher proportion of trainees than provinces whose residents transition to fellowship in other provinces, further exacerbating regional differences in MO supply. To improve workforce modelling accuracy in the future, an integrated database which includes both individual oncologist workload metrics (annual consults, follow-up assessments and treated patients) as well as region-specific cancer incidence alongside the number of oncologists in each region would be extremely useful. Additionally, continuing this work in 5-year windowing would allow for ongoing monitoring of oncology workforce capacity and inform policy pertaining to health human resource recruitment and funding. Our study shows that the number of medical oncologists in Canada has increased over a 25-year period, despite which the ratio of incident cancer/MO is still higher in Canada compared with other developed countries. Furthermore, our MO population is aging, which may have significant implications for the MO supply in the future. Since the number of trainees produced by a region directly influences the number of practicing MO, more MO trainees in underserved regions such as Atlantic Canada may help address undersupply in those regions of Canada. Future studies should also consider examining alternate models of care, such as general practitioners in oncology (GPOs), nurse practitioners (NPs), and/or pharmacists, that have been increasingly adopted over time and may help address MO workload and geographic disparities. |
Implementation of a patient-centered remote wound monitoring system for management of diabetic foot ulcers | a9e86a78-1a53-4663-923a-f90d55be9336 | 10244728 | Patient-Centered Care[mh] | Chronic lower extremity wounds are a major source of global morbidity, disability, and healthcare utilization ( – ). Diabetic foot ulcers (DFU) represent an increasingly common and difficult to treat subset of lower extremity wounds ( , ). In the United States, diabetes affects 37.3 million persons, of whom 19% to 33% will develop a DFU during their lifetime ( , ). Complications of DFU are common and morbid, including up to a 60% occurrence of diabetic foot infection and a 15% to 20% risk of subsequent lower extremity amputation ( ). Both incident DFU and poor healing disproportionately affect socioeconomically vulnerable populations, persons with complex medical needs, and/or persons with limited access to high-quality wound care ( ). In-person multidisciplinary diabetic foot and wound management is standard of care for the treatment of DFU ( , ). However, the model of multidisciplinary care typically requires frequent in-person wound assessments, which may not be achievable for patients due to numerous barriers. Patients with DFU and their caregivers consistently identify time constraints (e.g., difficulty finding available appointment times, conflicts with occupational and care-giving responsibilities), financial insecurity, mobility deficits, and lack of access to safe transportation as barriers to accessing treatment ( , ). Remote wound care offers a potential approach to overcoming these barriers. In response to the COVID-19 pandemic, the use of telemedicine has expanded exponentially in the United States ( ) Telemedicine strategies have been applied to the management of DFU with mixed results ( ). Patients and physicians have expressed enthusiasm for remote wound monitoring solutions, but most current systems rely on trained healthcare providers (e.g., home care nurses) or non-expert clinicians in the home for execution ( , ). There is a paucity of data on the feasibility, compliance, and outcomes of a remote wound monitoring system that relies on patients and their caregivers to perform their own wound scans. The Minuteful for Wound Digital Management System (Healthy.io, Tel Aviv, Israel) is a novel, remote wound monitoring system that captures wound measurements and analyzes tissue distribution in real-time through use of a smartphone application. Use of this digital management system by clinicians has been shown to be successful in non-US healthcare settings such as England ( ), but a newer patient-facing version of the technology has recently been developed. We conducted a pilot study of patients with DFU to assess patient engagement, reliability, and satisfaction with the Minuteful for Wound Digital Management System.
Patient population We enrolled 25 patients who presented to the Johns Hopkins Hospital multidisciplinary diabetic limb preservation clinic with an active DFU between July 1 and November 30, 2022. Patients were considered for enrollment in the study if they were proficient in English, ≥18 years of age, had an active DFU, had completed any planned revascularization and/or wound debridement procedures, and were willing and able to use a smartphone to capture weekly wound scans for an 8-week study period. For patients with multiple wounds, the largest wound that was accessible for imaging was designated to be monitored using the device throughout the study. Patients were excluded from the study if the wound was too large to capture in a single wound scan, if the wound was in a location that was not accessible to the patient or their caregiver, or if they were unable to operate the smartphone application. Patients who wished to participate but did not have access to a smartphone were loaned a smartphone with the app pre-installed for the duration of their participation in the study. The Johns Hopkins University Institutional Review Board approved the study, and all patients provided written informed consent to participate. Minuteful for wound digital management system The Minuteful for Wound Digital Management System (Healthy.io, Tel Aviv Israel) consists of dedicated calibration markers (stickers), a smartphone application (Minuteful for Wound app), and a web-based Portal (Minuteful for Wound Portal) that turns any smart mobile device into a wound care management tool ( ). The use of calibration markers helps the application identify the wound area and controls for different lighting conditions and camera types. The Minuteful for Wound app guides patients through the process of collecting clinical data and capturing scans of their wound using the embedded smartphone camera. The captured scan is transferred to a cloud-based server, where a set of distinct algorithms is used to analyze and translate it into a set of measurements for each wound. The measurements are securely displayed in the cloud-based Minuteful for Wound Portal to assist healthcare professionals in managing and monitoring the wound healing process ( ). Study protocol Following informed consent, the patients and their caregivers were instructed how to download and log into the Minuteful for Wound app on their smartphone. The primary user (patient or caregiver) was then provided with a box of calibration stickers specifically for use with the Minuteful for Wound app and taught how to apply the stickers and scan the wound. The primary user was given the opportunity to ask questions and practice scanning the wound and, once they were proficient, completed the first scan for upload in the clinic. Written information about the study and use of the application, including a user manual and a brochure, were also provided to the patient and their caregiver. Once trained to use the app, the primary user was asked to obtain weekly at-home wound scans during regular dressing changes. Users were asked to capture a minimum of one wound scan per week to allow for flexibility in scanning, but were encouraged to capture scans with each dressing change when possible. The quality of the scans was standardized using in-app boundary conditions, an algorithmic mechanism that enables results to be presented to the clinicians in the Portal. In cases where the environmental conditions did not meet the device’s prerequisites (e.g., not enough motion during the scan or it was blurry, the lighting was too bright or too dark, or the calibration markers did not remain in the camera field for the entire scan), the patient would be prompted by the app to re-perform the scan. All remotely collected assessments were securely transmitted to the HIPAA-compliant Minuteful for Wound Portal for review by the study team. All wound assessments were reviewed by members of the study team (consisting of a vascular surgeon, surgical podiatrist, and general surgery resident) once per week to assess progress. A weekly wound update was then provided to the patient by phone, and patients with concern for clinically stagnating wounds were asked to visit the clinic for an in-person assessment within the next week. Primary users who did not complete a weekly wound scan were called by the study team with a reminder. A Healthy.io engagement team was available to support the primary user in completing remote assessments as needed. A technology support hotline was available from 8am to 6pm EST on Mondays through Fridays to provide live phone support for app use. At the end of the 8-week study period primary users were asked to complete a useability and satisfaction survey to assess ease of use and usefulness of the Minuteful for Wound Digital Management System. The survey was developed using a mixed Likert scale and open-ended question design through iterative processing by the study team ( ). Patient data We captured age, sex, race, ethnicity, insurance status, area deprivation index and comorbidities for each patient through direct review and abstraction of the electronic medical records. Area deprivation index, which is a comprehensive measure of neighborhood socioeconomic deprivation, was calculated using each patient’s complete address and Neighborhood Atlas mapping ( , ). Hypertension was defined as systolic blood pressure >160 mmHg on 3 separate visits or current use of antihypertensive medications. Hyperlipidemia was defined as a cholesterol level >200mg/dL, LDL level >130 mg/dL, or use of cholesterol-lowering medications. Coronary artery disease was defined as a documented history of myocardial infarction or previous coronary revascularization. Congestive heart failure was defined based on a documented diagnosis, echocardiogram findings, or the Framingham criteria ( ). Peripheral artery disease was defined as an ankle-brachial index (ABI) of ≤0.8, toe pressure of <70 mmHg, or a history of a revascularization procedure on the affected limb. Chronic kidney disease was defined as an eGFR of <90 mL/min/1.73m². The wound characteristics for each patient were documented by study staff at the time of study enrollment including wound location, size, vascular studies, Wound, Ischemia and foot Infection (WIfI) score ( ), and previous revascularization and podiatric surgical interventions. WIfI classification was assigned based on post-revascularization and wound debridement characteristics to determine wound stage at the time of enrollment. Study outcomes The primary outcomes were patient engagement and satisfaction with the Minuteful for Wound Digital Management System. Patient engagement was determined by the recorded number of successful scans performed over the study period. ‘Optimally engaged patients’ were defined as patients completing 100% of study scans (equivalent to at least one scan every week). ‘Highly engaged patients’ were defined as patients meeting a threshold of 75% to 99% of study scans (equivalent to at least one scan every other week and a half). ‘Engaged patients’ were defined as patients meeting a predefined threshold of 50% to 74% of study scans (equivalent to at least one scan every other week, which is equivalent to the expected frequency for standard in-person wound care visits). ‘Not engaged patients’ were defined as patients completing 25% to 49% (equivalent to at least one scan per month). Patients who completed <25% of wound scans (i.e. <2 scans over 8 weeks) were considered to be study failures. Patient and caregiver satisfaction with the application was determined by survey responses to questions about ease of use and overall usefulness. A patient was determined to be satisfied if they provided a response of 4 or 5 (“Agree” or “Strongly Agree,” respectively) on the Likert Scale. Secondary outcomes were the proportion of scans that led to a change in patient management, including changes to wound care plan or a change in planned next in-person visit; number of reminder phone calls made to patients; change in wound area from study enrollment to study completion; and the proportion of patients who achieved wound healing. To evaluate the scanning experience of the patients, the number of boundary condition alerts and scan attempts were recorded for each patient assessment. Statistical analysis Baseline patient demographics, comorbidities, wound information, and study outcomes were tabulated and reported using means (standard deviations) or percent (N) as appropriate. Change in wound area over the course of the study was compared using paired t-tests, with P<0.05 denoting statistical significance. Qualitative data collected from open-ended questions on the patient survey were analyzed by two reviewers who used open coding, resolved discrepancies with triangulation, and applied thematic analysis.
We enrolled 25 patients who presented to the Johns Hopkins Hospital multidisciplinary diabetic limb preservation clinic with an active DFU between July 1 and November 30, 2022. Patients were considered for enrollment in the study if they were proficient in English, ≥18 years of age, had an active DFU, had completed any planned revascularization and/or wound debridement procedures, and were willing and able to use a smartphone to capture weekly wound scans for an 8-week study period. For patients with multiple wounds, the largest wound that was accessible for imaging was designated to be monitored using the device throughout the study. Patients were excluded from the study if the wound was too large to capture in a single wound scan, if the wound was in a location that was not accessible to the patient or their caregiver, or if they were unable to operate the smartphone application. Patients who wished to participate but did not have access to a smartphone were loaned a smartphone with the app pre-installed for the duration of their participation in the study. The Johns Hopkins University Institutional Review Board approved the study, and all patients provided written informed consent to participate.
The Minuteful for Wound Digital Management System (Healthy.io, Tel Aviv Israel) consists of dedicated calibration markers (stickers), a smartphone application (Minuteful for Wound app), and a web-based Portal (Minuteful for Wound Portal) that turns any smart mobile device into a wound care management tool ( ). The use of calibration markers helps the application identify the wound area and controls for different lighting conditions and camera types. The Minuteful for Wound app guides patients through the process of collecting clinical data and capturing scans of their wound using the embedded smartphone camera. The captured scan is transferred to a cloud-based server, where a set of distinct algorithms is used to analyze and translate it into a set of measurements for each wound. The measurements are securely displayed in the cloud-based Minuteful for Wound Portal to assist healthcare professionals in managing and monitoring the wound healing process ( ).
Following informed consent, the patients and their caregivers were instructed how to download and log into the Minuteful for Wound app on their smartphone. The primary user (patient or caregiver) was then provided with a box of calibration stickers specifically for use with the Minuteful for Wound app and taught how to apply the stickers and scan the wound. The primary user was given the opportunity to ask questions and practice scanning the wound and, once they were proficient, completed the first scan for upload in the clinic. Written information about the study and use of the application, including a user manual and a brochure, were also provided to the patient and their caregiver. Once trained to use the app, the primary user was asked to obtain weekly at-home wound scans during regular dressing changes. Users were asked to capture a minimum of one wound scan per week to allow for flexibility in scanning, but were encouraged to capture scans with each dressing change when possible. The quality of the scans was standardized using in-app boundary conditions, an algorithmic mechanism that enables results to be presented to the clinicians in the Portal. In cases where the environmental conditions did not meet the device’s prerequisites (e.g., not enough motion during the scan or it was blurry, the lighting was too bright or too dark, or the calibration markers did not remain in the camera field for the entire scan), the patient would be prompted by the app to re-perform the scan. All remotely collected assessments were securely transmitted to the HIPAA-compliant Minuteful for Wound Portal for review by the study team. All wound assessments were reviewed by members of the study team (consisting of a vascular surgeon, surgical podiatrist, and general surgery resident) once per week to assess progress. A weekly wound update was then provided to the patient by phone, and patients with concern for clinically stagnating wounds were asked to visit the clinic for an in-person assessment within the next week. Primary users who did not complete a weekly wound scan were called by the study team with a reminder. A Healthy.io engagement team was available to support the primary user in completing remote assessments as needed. A technology support hotline was available from 8am to 6pm EST on Mondays through Fridays to provide live phone support for app use. At the end of the 8-week study period primary users were asked to complete a useability and satisfaction survey to assess ease of use and usefulness of the Minuteful for Wound Digital Management System. The survey was developed using a mixed Likert scale and open-ended question design through iterative processing by the study team ( ).
We captured age, sex, race, ethnicity, insurance status, area deprivation index and comorbidities for each patient through direct review and abstraction of the electronic medical records. Area deprivation index, which is a comprehensive measure of neighborhood socioeconomic deprivation, was calculated using each patient’s complete address and Neighborhood Atlas mapping ( , ). Hypertension was defined as systolic blood pressure >160 mmHg on 3 separate visits or current use of antihypertensive medications. Hyperlipidemia was defined as a cholesterol level >200mg/dL, LDL level >130 mg/dL, or use of cholesterol-lowering medications. Coronary artery disease was defined as a documented history of myocardial infarction or previous coronary revascularization. Congestive heart failure was defined based on a documented diagnosis, echocardiogram findings, or the Framingham criteria ( ). Peripheral artery disease was defined as an ankle-brachial index (ABI) of ≤0.8, toe pressure of <70 mmHg, or a history of a revascularization procedure on the affected limb. Chronic kidney disease was defined as an eGFR of <90 mL/min/1.73m². The wound characteristics for each patient were documented by study staff at the time of study enrollment including wound location, size, vascular studies, Wound, Ischemia and foot Infection (WIfI) score ( ), and previous revascularization and podiatric surgical interventions. WIfI classification was assigned based on post-revascularization and wound debridement characteristics to determine wound stage at the time of enrollment.
The primary outcomes were patient engagement and satisfaction with the Minuteful for Wound Digital Management System. Patient engagement was determined by the recorded number of successful scans performed over the study period. ‘Optimally engaged patients’ were defined as patients completing 100% of study scans (equivalent to at least one scan every week). ‘Highly engaged patients’ were defined as patients meeting a threshold of 75% to 99% of study scans (equivalent to at least one scan every other week and a half). ‘Engaged patients’ were defined as patients meeting a predefined threshold of 50% to 74% of study scans (equivalent to at least one scan every other week, which is equivalent to the expected frequency for standard in-person wound care visits). ‘Not engaged patients’ were defined as patients completing 25% to 49% (equivalent to at least one scan per month). Patients who completed <25% of wound scans (i.e. <2 scans over 8 weeks) were considered to be study failures. Patient and caregiver satisfaction with the application was determined by survey responses to questions about ease of use and overall usefulness. A patient was determined to be satisfied if they provided a response of 4 or 5 (“Agree” or “Strongly Agree,” respectively) on the Likert Scale. Secondary outcomes were the proportion of scans that led to a change in patient management, including changes to wound care plan or a change in planned next in-person visit; number of reminder phone calls made to patients; change in wound area from study enrollment to study completion; and the proportion of patients who achieved wound healing. To evaluate the scanning experience of the patients, the number of boundary condition alerts and scan attempts were recorded for each patient assessment.
Baseline patient demographics, comorbidities, wound information, and study outcomes were tabulated and reported using means (standard deviations) or percent (N) as appropriate. Change in wound area over the course of the study was compared using paired t-tests, with P<0.05 denoting statistical significance. Qualitative data collected from open-ended questions on the patient survey were analyzed by two reviewers who used open coding, resolved discrepancies with triangulation, and applied thematic analysis.
Patient cohort We enrolled 25 patients over the month study period. Mean age was 65.5 (SD, 13.7) years, 60.0% of our participants were male, and 52.0% self-identified as non-Hispanic Black adults. The most common comorbidities were hypertension (68.0%), chronic kidney disease (68.0%), and peripheral artery disease (56.0%) ( ). There were a wide variety of wound locations ( ). Mean wound area at baseline was 18.0 cm² (SD, 15.2), 36.0% of wounds were severe (WIfI stage 3 or 4), 64.0% of patients had undergone lower extremity revascularization, and 80.0% had undergone surgical debridement prior to enrollment. Home health was involved in the care of 60.0% of patients, and a wide range of wound dressing treatment strategies were used ( ). Study participation The primary user was a caregiver in 60.0% of cases and the patient in 40.0%. Home health was not involved in wound scanning for this feasibility study. Twenty-eight percent of users borrowed a smartphone for the purposes of study participation. Overall, patients submitted a mean number of 5.80 (SD, 5.30) wound scans over the total 8-week study period, equal to a mean of 0.72 (SD, 0.63) wound scans per week. Patients received a mean of 2.28 (SD, 2.25) reminder phone calls to submit wound scans from the study team during the study period ( ). Patients made or received a mean of 6.32 (SD, 5.18) technical calls from the Healthy.io technical team. Clinical study outcomes App engagement was variable, with 20.0% of patients completing 100% of the weekly wound scans (optimally engaged), 28.0% completing 75-99% of weekly wound scans (highly engaged), 12.0% completing 50-74% of weekly wound scans (engaged), and 28.0% completing <25% weekly wounds scans (i.e., study failure) ( ). Study failures were investigated and classified as communication difficulties in 16.0% of patients and lack of caregiver availability for assistance with the scans in 12.0% of patients. Overall, 21/25 (84.0%) patients completed at least one in-home wound scan. Thirty-six percent of patients were advised that they should undergo an early change in their wound management plan at least once during the study based on weekly review of their wound scan by the study team. Treatment changes included a change in wound treatment in 20.0% of patients and initiation of an earlier appointment for in-person clinic evaluation in 16.0% of patients. There were no instances where use of the digital management system resulted in a delayed diagnosis of wound deterioration. At the conclusion of the study, there was a mean decrease in wound area of 7.67 cm 2 (SD, 9.72) per patient (P=0.005), equivalent to a mean decrease of 41.6% (SD, 15.8%). Complete wound healing was achieved in 12.0% of patients (3/25). Patient satisfaction Primary users who completed at least one at-home wound scan during the study period were asked to take part in a post-study survey, with a response rate of 81.0% (17/21). Of the four participants who did not complete the survey, three were unable to be reached by telephone after completion of the study and one declined to participate. Of the survey respondents, 88.2% agreed or strongly agreed that the Minuteful Wound app was easy to use. Overall, 94.1% of patients found the digital wound management system to be useful, with the large majority noting they felt more involved in their wound care, more responsible for their health, and more able to access healthcare services ( ). One patient even stated that they felt “empowered [to be accountable for their health].” Common themes that recurred throughout the patient surveys included appreciation for close wound monitoring without the need for travel to the clinic. However, many respondents expressed room for improvement with “more instantaneous feedback.” Despite the asynchronous design of the app use and study team evaluation, 100% of patients felt confident that the information they sent to the study team was received. Usage data outcomes Among the 179 wound scans attempted by patients and/or their caregiver, 145 (81.0%) were free of boundary condition violations and successfully submitted to the Portal via the app. Eleven patients did not receive any boundary condition notifications during the study period, meaning they produced satisfactory and clinically valid wound documentation on all first wound scan attempts throughout the study period. Of the 34 scans that did not meet the prerequisite boundary conditions, 19 (55.9%) were rectified after a single user notification. The boundary conditions that were violated for these patients were much more common during the beginning of the study and decreased over the course of 8 weeks ( ).
We enrolled 25 patients over the month study period. Mean age was 65.5 (SD, 13.7) years, 60.0% of our participants were male, and 52.0% self-identified as non-Hispanic Black adults. The most common comorbidities were hypertension (68.0%), chronic kidney disease (68.0%), and peripheral artery disease (56.0%) ( ). There were a wide variety of wound locations ( ). Mean wound area at baseline was 18.0 cm² (SD, 15.2), 36.0% of wounds were severe (WIfI stage 3 or 4), 64.0% of patients had undergone lower extremity revascularization, and 80.0% had undergone surgical debridement prior to enrollment. Home health was involved in the care of 60.0% of patients, and a wide range of wound dressing treatment strategies were used ( ).
The primary user was a caregiver in 60.0% of cases and the patient in 40.0%. Home health was not involved in wound scanning for this feasibility study. Twenty-eight percent of users borrowed a smartphone for the purposes of study participation. Overall, patients submitted a mean number of 5.80 (SD, 5.30) wound scans over the total 8-week study period, equal to a mean of 0.72 (SD, 0.63) wound scans per week. Patients received a mean of 2.28 (SD, 2.25) reminder phone calls to submit wound scans from the study team during the study period ( ). Patients made or received a mean of 6.32 (SD, 5.18) technical calls from the Healthy.io technical team.
App engagement was variable, with 20.0% of patients completing 100% of the weekly wound scans (optimally engaged), 28.0% completing 75-99% of weekly wound scans (highly engaged), 12.0% completing 50-74% of weekly wound scans (engaged), and 28.0% completing <25% weekly wounds scans (i.e., study failure) ( ). Study failures were investigated and classified as communication difficulties in 16.0% of patients and lack of caregiver availability for assistance with the scans in 12.0% of patients. Overall, 21/25 (84.0%) patients completed at least one in-home wound scan. Thirty-six percent of patients were advised that they should undergo an early change in their wound management plan at least once during the study based on weekly review of their wound scan by the study team. Treatment changes included a change in wound treatment in 20.0% of patients and initiation of an earlier appointment for in-person clinic evaluation in 16.0% of patients. There were no instances where use of the digital management system resulted in a delayed diagnosis of wound deterioration. At the conclusion of the study, there was a mean decrease in wound area of 7.67 cm 2 (SD, 9.72) per patient (P=0.005), equivalent to a mean decrease of 41.6% (SD, 15.8%). Complete wound healing was achieved in 12.0% of patients (3/25).
Primary users who completed at least one at-home wound scan during the study period were asked to take part in a post-study survey, with a response rate of 81.0% (17/21). Of the four participants who did not complete the survey, three were unable to be reached by telephone after completion of the study and one declined to participate. Of the survey respondents, 88.2% agreed or strongly agreed that the Minuteful Wound app was easy to use. Overall, 94.1% of patients found the digital wound management system to be useful, with the large majority noting they felt more involved in their wound care, more responsible for their health, and more able to access healthcare services ( ). One patient even stated that they felt “empowered [to be accountable for their health].” Common themes that recurred throughout the patient surveys included appreciation for close wound monitoring without the need for travel to the clinic. However, many respondents expressed room for improvement with “more instantaneous feedback.” Despite the asynchronous design of the app use and study team evaluation, 100% of patients felt confident that the information they sent to the study team was received.
Among the 179 wound scans attempted by patients and/or their caregiver, 145 (81.0%) were free of boundary condition violations and successfully submitted to the Portal via the app. Eleven patients did not receive any boundary condition notifications during the study period, meaning they produced satisfactory and clinically valid wound documentation on all first wound scan attempts throughout the study period. Of the 34 scans that did not meet the prerequisite boundary conditions, 19 (55.9%) were rectified after a single user notification. The boundary conditions that were violated for these patients were much more common during the beginning of the study and decreased over the course of 8 weeks ( ).
As our healthcare system has been adapting to the ever-changing climate of the COVID-19 pandemic, telemedicine advancement has become a priority for medical technology companies. We aimed to determine whether a novel digital wound monitoring system could be effectively used by patients and/or their caregivers to provide clinicians with high-quality wound data to guide care. We found that patients and caregivers could successfully learn how to use the Minuteful for Wound smartphone app, and the majority successfully engaged in its use. Study participants found the digital wound management system useful, and more than a third of patients benefited from its use in the form of early treatment modification. A number of software companies have developed smartphone apps to measure and record wound size, including for DFU ( , – ). Remote wound monitoring programs may be helpful in the management of chronic wounds, particularly as a communication tool between patients and their healthcare providers when close follow-up is necessary ( ). However, most of the applications are developed with physicians and nurses being the intended users. A recent study enrolled patients from a rural Veteran’s Affairs wound care clinic in a remote wound telemedicine program and showed excellent wound healing outcomes, but the wound telemedicine was facilitated by a trained telepresenter ( ). Similarly, a meta-analysis of telemedicine versus in-person management of DFU showed similar or possibly improved wound healing, amputation, and mortality outcomes for patients managed via telemedicine ( ). However, all telemedicine studies identified in the meta-analysis involved use of a trained nurse or similar healthcare provider to facilitate the telemedicine communication between the patient and physician. Our study is unique in that it assessed a remote wound monitoring system designed to be patient-facing, where patients and their caregivers had total responsibility for capturing and submitting remote wound scans on a repeated basis. Patient engagement in our study was high compared to prior studies of telemedicine use. In a study of emergency department patients undergoing acute laceration repair or incision and drainage procedures, 58% of patients sent at least one picture of their wound through a Mobile Post-operative Wound Evaluator (mPOWEr) smartphone app ( ). In our study, 84.0% of patients submitted at least one at-home wound scan. However, overall engagement was lower because we defined patient engagement as capturing ≥50% of expected weekly scans. The need for repeated wound scans places a larger burden on the patients and their caregivers than a single wound capture but was designed to simulate the frequency of standard in-person wound monitoring in the clinic. Reminder phone calls from the study team were required approximately twice per patient over the course of the 8-week study. Whether this burden is sustainable for larger numbers of patients is unclear. A prior study also demonstrated that telemedicine costs for managing DFU patients by telemedicine are approximately $2222 USD lower per patient compared to standard in-person monitoring ( ). Primary users in our study reported high rates of satisfaction with the Healthy.io Minuteful for Wound Digital Management System. Patients are generally in favor of remote wound monitoring based on data from prior studies, including studies specific to DFU ( – ). In a scoping review of telemedicine solutions for DFU, four main maps emerged: “A whole human not merely a hole in a human,” “Less of a burden on the family, the community, and the environment,” “Competences and continuity of care are essential for high-quality care” and “The quality and modality of the technology.” Consistent with these concepts, our patients reported less frequent in-person appointments, better continuity of care, and more accountability with care as benefits. We also observed some drawbacks to the technology. Specifically, primary users felt that more instantaneous feedback about the wound would be helpful. Future iterations of the app will involve a 2-way in-app communication tool that will allow patients to receive feedback more synchronously and remove the burden of the weekly phone call. In addition to patients’ desire for more timely wound feedback, we encountered other challenges in our study. While our rate of study completion was high (72.0%), seven patients failed to complete the study. One of the major concerns around the use of remote wound monitoring systems is how they can be utilized by socioeconomically disadvantaged patients. Nearly one third of patients in our study required a borrowed smartphone because they lacked a smartphone with the specifications needed to run the app; most patients in our study had a preexisting smartphone, but many were older models not compatible with this technology. Both smartphone and reliable internet access are barriers to implementation in vulnerable populations ( , ). Our multidisciplinary diabetic limb preservation clinic serves a large number of patients from socially disadvantaged backgrounds, however, patients enrolled in this study resided in less disadvantaged neighborhoods than a typical patient in our clinic ( ). Making remote wound monitoring technology accessible to a wide range of populations will be important for successful adoption moving forward. There are a number of limitations to this study. We enrolled only a small number of patients in this pilot study, and we did not have a control group for comparison. We were not able to evaluate hospital financial data associated with the implementation and continued use of the app, but plan to do so in the future. Finally, our study was not designed or powered to assess wound healing outcomes, and due to the feasibility design we did not attempt to alter patient care based on the wound images provided. However, our findings did lead to the successful initiation of a now ongoing randomized controlled trial comparing use of the Minuteful for Wound Digital Management System compared to standard of care in-person monitoring ( ).
The Healthy.io Minuteful for Wound Digital Management System is a feasible means of remote wound monitoring for use by patients and their caregivers. We were able to show good patient engagement, satisfaction, and usage data in a pilot study design. Our results suggest the feasibility of patient-facing technology for the remote wound app monitoring of diabetic foot ulcers. This study is the impetus for a new randomized controlled trial designed to study wound healing efficacy for remote wound app monitoring vs. standard in-person clinic visits for the treatment of lower extremity wounds, in which we hope to show the barriers that often interfere with in-person follow up visits will no longer interfere with proper wound care.
The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.
The studies involving human participants were reviewed and approved by Johns Hopkins University Institutional Review Board. The patients/participants provided their written informed consent to participate in this study. Written informed consent was obtained from the individual(s) for the publication of any potentially identifiable images or data included in this article.
AK and CH, project conceptualization, patient enrollment, trial coordination, wound scan review, data analysis, manuscript development, and review. SB and EL-A, data analysis, manuscript development, and review. KM, MS, DS, CA, and ES, manuscript development and review. DJ, project conceptualization, data analysis, manuscript development, and review. RS, patient enrollment, wound scan review, manuscript development, and review. JR, patient enrollment and trial coordination. All authors contributed to the article and approved the submitted version.
|
The expression of | 83b7024f-68a4-46b3-b66c-5085d6c84a86 | 11404481 | Anatomy[mh] | Small cell lung carcinoma (SCLC), in principle, results from the inactivation of the common tumour suppressor genes TP53 and Rb1 , and has traditionally been considered relatively genetically homogeneous. Rudin et al reported that SCLCs have been divided into four molecular subtypes (ASCL1, NEUROD1, POU2F3, and YAP1), with heterogeneity at the mRNA level . Each subtype has a different expression profile. For example, ASCL1 and NEUROD1‐dominant SCLCs correspond to high‐neuroendocrine (NE) and high‐TTF‐1 phenotypes, while POU2F3 and YAP1‐dominant SCLCs usually belong to low‐NE and low‐TTF‐1 phenotypes . Another aspect of SCLC heterogeneity is lineage plasticity between NE and non‐NE cells. The transformation of non‐SCLC (NSCLC) into SCLC may result from lineage plasticity (which also includes the transformation of SCLC to NSCLC ). Lineage plasticity, involving flexible phenotypic changes, favours cancer cell survival under adverse conditions such as hypoxia and driver‐targeted therapies and increases the proliferative ability and malignancy of SCLC. Plasticity may be a driving cause of the difficulty in the treatment and poor prognosis of SCLC, making it a hot topic in SCLC research . The question of what pathological findings explain plasticity in SCLC is of significant interest to pathologists; however, limited evidence is available on this subject. Combined SCLC, in which SCLC is combined with a non‐SCLC histotype, stands out as a potential target indicating plasticity in SCLC. Component‐wise analysis using chromosome studies and next‐generation sequencing (NGS) suggested that combined SCLC forms from a single clone. However, due to its rarity and difficulty in accessing specimens, detailed pathological observations have not been fully conducted. YAP1 is a downstream effector of the Hippo pathway and is involved in several biological processes such as cell proliferation, migration, and epithelial‐mesenchymal transition (EMT) . It is up‐regulated in many human malignancies , and its overexpression in NSCLC is considered a poor prognostic factor . In SCLC, YAP1 is involved in cell growth, EMT, drug resistance, and non‐NE expression . In SCLC, four molecular subtypes have been proposed by Rudin et al , including YAP1‐dominant SCLC , although some reports state that YAP1‐expressing SCLC is a cell line‐based concept and not applicable to primary SCLC . As mentioned above, YAP1 can be considered a non‐NE marker in SCLC and may be a candidate for plasticity between NE and non‐NE types. Furthermore, molecular markers and/or transcription factors may be involved in the development and lineage determination of SCLC, but limited information is available on this. This study aimed to carry out detailed pathological observations of combined SCLC, focusing on YAP1 and other transcription factors, to approach pathological findings indicating SCLC plasticity. Sample and histological examination This study was approved by the Ethics Committee of Kobe University Hospital (No. B220045) and was conducted in accordance with the Declaration of Helsinki. A total of 100 consecutive cases of surgically resected SCLCs at Kobe University Hospital between 2000 and 2023 and Nara Medical University between 2016 and 2022 were included. Resection methods included total lung resection ( n = 1), lobectomy ( n = 58), segmentectomy ( n = 3), and partial resection ( n = 38). All patients did not receive pre‐operative treatment, such as chemotherapy or radiotherapy. Fifty‐three percent of patients received post‐operative chemotherapy. Twenty‐one patients received carboplatin and etoposide, 11 patients received cisplatin and etoposide, and two patients received cisplatin and irinotecan. Chemotherapy history was unknown for 19 patients. Two pathologists (NJ and CO) made histopathological diagnoses according to the fifth World Health Organization classification. We evaluated as many H&E slides as possible (average 4.7 tumour‐containing slides), because small amounts of NSCLC components were found in some cases. The determination of SCLC or NSCLC was based on H&E, and component amounts were calculated to the nearest 1%. In this study, the partner elements of combined SCLCs included squamous cell carcinoma (SQCC), adenocarcinoma, large‐cell neuroendocrine carcinoma (LCNEC), unclassifiable NSCLC, as well as sarcoma, so components other than SCLC are referred to as ‘partners’. Combined SCLCs were classified into the following three morphological patterns: (1) A separated morphological pattern, in which SCLC and partner are separated by clear cell‐to‐cell boundaries with intervening stroma; (2) a mosaic morphological pattern, in which the cell‐to‐cell boundary between SCLC and partner is unclear without intervening stroma and they cannot be separated; and (3) coexistence of the two patterns. Immunohistochemistry Formalin‐fixed paraffin‐embedded (FFPE) sections were used for immunohistochemistry (IHC) with a Ventana BenchMark GX (Roche, Switzerland) or BOND‐III (Leica, Deer Park, TX, USA) automated immunostainer. Protein expression levels of p53, Rb1, TTF‐1, p40, synaptophysin, chromogranin A, CD56, INSM1, ASCL1, NEUROD1, POU2F3, and YAP1 were investigated. The IHC protocols are summarised in supplementary material, Table . YAP1 and p40 were assessed on whole slides in all cases, and other antibodies were also assessed on whole slides whenever possible, which was approximately 80% of samples. Cases that could not be studied on whole slides were evaluated using spiral arrays, which are less susceptible to heterogeneity . p40 was used for the purpose of identifying cases containing SQCC and not for statistical analysis. Based on a recent consensus, overexpression, complete absence, and cytoplasmic expression were defined as p53 abnormal patterns . Rb1 was classified into two patterns: retained and total loss (>90% loss). For other antibodies, evaluation was performed using the H ‐score. H ‐score (0–300) was defined as the product of intensity (0, 1 = weak, 2 = moderate, and 3 = strong) and proportion (0–100%) of the expressed tumour cells, according to a previous study . In the combined SCLCs, each component was scored separately, as far as possible. SCLC component scores were used for statistical evaluation. Among ASCL1, NEUROD1, and POU2F3, the highest expressed marker with an H ‐score of 50 or more was defined as the dominant transcription subtypes. Cases in which none of the three were expressed above an H ‐score of 50 were defined as triple‐negative (TN). Pure SCLC that predominantly expressed ASCL1, NEUROD1, and POU2F3 were labelled pure SCLC‐A, pure SCLC‐N, and pure SCLC‐P, respectively. Pure SCLC of the TN type was pure SCLC‐TN. Combined SCLC‐A, SCLC‐N, SCLC‐P, and SCLC‐TN were defined similarly. Next‐generation sequencing NGS was performed on 17 combined SCLCs. DNA was extracted from an 8‐μm‐thick FFPE surgical sample using the DNA Isolation Kit for FFPE Tissue Samples (CELLDATE, Fremont, CA, USA). The quality and concentration of the DNA samples were examined using a NanoDrop (Thermo Scientific, Waltham, MA, USA), a 4200 TapeStation, and a Genomic DNA ScreenTape Assay (Agilent Technologies, Santa Clara, CA, USA). Cases with clear boundaries between the SCLC and partner were separated into their respective components by macrodissection. Targeted NGS was performed on MiniSeq (Illumina, Inc., San Diego, CA, USA) using the commercially available gene panel Ampliseq™ for Illumina Cancer Hotspot Ver2 (Illumina, Inc.). This 106‐bp‐sized gene panel targets 50 cancer‐related gene hotspots, including 207 amplicon primers, and can detect single nucleotide variations and indels. Statistical analyses Statistical analyses of patient characteristics, prognosis, and IHC results were performed for combined SCLC and pure SCLC. Comparisons of variables for patient characteristics and pathological features were made using Mann–Whitney test or Fisher's exact test. Recurrence‐free survival (RFS) was defined as the time from the date of surgery until the date of recurrence or death by any cause. Overall survival (OS) was defined as the time from the date of surgery until death by any cause, or until the last follow‐up visit. RFS and OS were evaluated using the Kaplan–Meier method, and differences in survival curves were assessed using the log‐rank test. Statistical analyses of prognosis were performed using EZR version 1.55 (Saitama Medical Center, Jichi Medical University, Saitama, Japan), a graphical user interface of R (The R Foundation for Statistical Computing, Vienna, Austria). For survival analysis, univariate analyses and multivariate analyses adjusted for additional factors were performed . Statistical significance was set at a p value <0.05. All p values were two‐sided. This study was approved by the Ethics Committee of Kobe University Hospital (No. B220045) and was conducted in accordance with the Declaration of Helsinki. A total of 100 consecutive cases of surgically resected SCLCs at Kobe University Hospital between 2000 and 2023 and Nara Medical University between 2016 and 2022 were included. Resection methods included total lung resection ( n = 1), lobectomy ( n = 58), segmentectomy ( n = 3), and partial resection ( n = 38). All patients did not receive pre‐operative treatment, such as chemotherapy or radiotherapy. Fifty‐three percent of patients received post‐operative chemotherapy. Twenty‐one patients received carboplatin and etoposide, 11 patients received cisplatin and etoposide, and two patients received cisplatin and irinotecan. Chemotherapy history was unknown for 19 patients. Two pathologists (NJ and CO) made histopathological diagnoses according to the fifth World Health Organization classification. We evaluated as many H&E slides as possible (average 4.7 tumour‐containing slides), because small amounts of NSCLC components were found in some cases. The determination of SCLC or NSCLC was based on H&E, and component amounts were calculated to the nearest 1%. In this study, the partner elements of combined SCLCs included squamous cell carcinoma (SQCC), adenocarcinoma, large‐cell neuroendocrine carcinoma (LCNEC), unclassifiable NSCLC, as well as sarcoma, so components other than SCLC are referred to as ‘partners’. Combined SCLCs were classified into the following three morphological patterns: (1) A separated morphological pattern, in which SCLC and partner are separated by clear cell‐to‐cell boundaries with intervening stroma; (2) a mosaic morphological pattern, in which the cell‐to‐cell boundary between SCLC and partner is unclear without intervening stroma and they cannot be separated; and (3) coexistence of the two patterns. Formalin‐fixed paraffin‐embedded (FFPE) sections were used for immunohistochemistry (IHC) with a Ventana BenchMark GX (Roche, Switzerland) or BOND‐III (Leica, Deer Park, TX, USA) automated immunostainer. Protein expression levels of p53, Rb1, TTF‐1, p40, synaptophysin, chromogranin A, CD56, INSM1, ASCL1, NEUROD1, POU2F3, and YAP1 were investigated. The IHC protocols are summarised in supplementary material, Table . YAP1 and p40 were assessed on whole slides in all cases, and other antibodies were also assessed on whole slides whenever possible, which was approximately 80% of samples. Cases that could not be studied on whole slides were evaluated using spiral arrays, which are less susceptible to heterogeneity . p40 was used for the purpose of identifying cases containing SQCC and not for statistical analysis. Based on a recent consensus, overexpression, complete absence, and cytoplasmic expression were defined as p53 abnormal patterns . Rb1 was classified into two patterns: retained and total loss (>90% loss). For other antibodies, evaluation was performed using the H ‐score. H ‐score (0–300) was defined as the product of intensity (0, 1 = weak, 2 = moderate, and 3 = strong) and proportion (0–100%) of the expressed tumour cells, according to a previous study . In the combined SCLCs, each component was scored separately, as far as possible. SCLC component scores were used for statistical evaluation. Among ASCL1, NEUROD1, and POU2F3, the highest expressed marker with an H ‐score of 50 or more was defined as the dominant transcription subtypes. Cases in which none of the three were expressed above an H ‐score of 50 were defined as triple‐negative (TN). Pure SCLC that predominantly expressed ASCL1, NEUROD1, and POU2F3 were labelled pure SCLC‐A, pure SCLC‐N, and pure SCLC‐P, respectively. Pure SCLC of the TN type was pure SCLC‐TN. Combined SCLC‐A, SCLC‐N, SCLC‐P, and SCLC‐TN were defined similarly. NGS was performed on 17 combined SCLCs. DNA was extracted from an 8‐μm‐thick FFPE surgical sample using the DNA Isolation Kit for FFPE Tissue Samples (CELLDATE, Fremont, CA, USA). The quality and concentration of the DNA samples were examined using a NanoDrop (Thermo Scientific, Waltham, MA, USA), a 4200 TapeStation, and a Genomic DNA ScreenTape Assay (Agilent Technologies, Santa Clara, CA, USA). Cases with clear boundaries between the SCLC and partner were separated into their respective components by macrodissection. Targeted NGS was performed on MiniSeq (Illumina, Inc., San Diego, CA, USA) using the commercially available gene panel Ampliseq™ for Illumina Cancer Hotspot Ver2 (Illumina, Inc.). This 106‐bp‐sized gene panel targets 50 cancer‐related gene hotspots, including 207 amplicon primers, and can detect single nucleotide variations and indels. Statistical analyses of patient characteristics, prognosis, and IHC results were performed for combined SCLC and pure SCLC. Comparisons of variables for patient characteristics and pathological features were made using Mann–Whitney test or Fisher's exact test. Recurrence‐free survival (RFS) was defined as the time from the date of surgery until the date of recurrence or death by any cause. Overall survival (OS) was defined as the time from the date of surgery until death by any cause, or until the last follow‐up visit. RFS and OS were evaluated using the Kaplan–Meier method, and differences in survival curves were assessed using the log‐rank test. Statistical analyses of prognosis were performed using EZR version 1.55 (Saitama Medical Center, Jichi Medical University, Saitama, Japan), a graphical user interface of R (The R Foundation for Statistical Computing, Vienna, Austria). For survival analysis, univariate analyses and multivariate analyses adjusted for additional factors were performed . Statistical significance was set at a p value <0.05. All p values were two‐sided. Patient characteristics As shown in Figure , 35 cases were morphologically classified as combined SCLC and 65 as pure SCLC. Of the 100 patients included, 87 were males and 13 were females, with a median age of 72 years (range, 40–91 years). All but one male patient (who had combined SCLC‐N) had a history of smoking. The tumour size in the combined SCLC group was significantly larger than that in the pure SCLC group (median 31 versus 25 mm, p = 0.0009); however, no significant differences in other factors were observed between the two groups. IHC As shown in Figure , NEUROD1 expression was significantly higher in combined SCLCs than in pure SCLCs (median H ‐score 50 versus 0, p = 0.04); however, no significant differences in the expression of other proteins or the composition of dominant transcription types were observed. In 98 cases excluding SCLC‐TN, double‐positive ( H ‐score ≥ 50) transcription marker cases were significantly more common in combined SCLCs, and single‐positive cases were more common in pure SCLCs ( p = 0.019), suggesting that combined SCLCs have stronger plasticity between transcription marker expression (Figure ). YAP1 expression Figure shows the distribution of YAP1 expression ( H ‐score) in the SCLC components of combined and pure SCLCs. YAP1 expression ( H ‐score > 1) was observed in 80% of combined SCLCs and 62% of pure SCLCs. The YAP1‐negative rate in SCLC components was 20% for combined SCLCs and 38% for pure SCLCs. Prognosis In univariate and multivariate analyses for OS, the presence of chemotherapy was a significant favourable prognostic factor, and the presence of interstitial pneumonia was a significant unfavourable prognostic factor (Table ). Combined SCLCs tended to have a poorer prognosis than pure SCLCs regarding OS, but this was not significant (Table ). Pathological features of combined SCLCs Thirty‐five combined SCLCs were morphologically classified as mosaic type (5 cases), coexisting mosaic and separated type (7 cases), or separated type (23 cases) (Figures and ). The median proportions of SCLC components in the total tumours of each type were 90%, 70%, and 40%, respectively (Figure ). The proportion of SCLC in the total tumour was significantly lower in the separated morphological pattern than in the other two patterns ( p = 0.018). In separated morphological patterns, clear boundaries existed between SCLC and the partner (Figure ), which were also highlighted by NEUROD1 and Napsin A (Figure ) or POU2F3 and p40 (Figure ). At high magnification, SCLC and its partners were accompanied by clear intervening stromal components, which could be seen in YAP1 expression. YAP1 was focally expressed in areas of morphological SCLC (data not shown). In the mosaic morphological pattern, partners (all of which were NSCLC of this type) were scattered in a mosaic‐like pattern. Indistinct and ambiguous boundaries were observed between the SCLC and NSCLC partners without intervening stroma. YAP1 showed mosaic‐like expression, almost coinciding with areas that were morphologically NSCLC, but was also weakly expressed in areas that were morphologically SCLC (Figure ). The border between SCLC and NSCLC was ambiguous in both H&E and YAP1 cells, indicating a gradual transition between SCLC and NSCLC. In the adenocarcinoma components of combined SCLC showing mosaic morphological types, Rb1 was lost, as in SCLC components (arrows in Figure ). Pathological features of pure SCLCs focusing on YAP1 expression All pure SCLCs exhibited homogeneous small‐cell cytology. Figure shows four representative examples of ASCL1‐dominant pure SCLCs. In YAP1‐negative cases, YAP1 was completely negative, and ASCL1 was positive (Figure ). In cases with a YAP1 H ‐score of 1–99, YAP1 was expressed in a mosaic pattern with a mutually exclusive expression pattern with ASCL1 (Figure ). In cases with a YAP1 H ‐score of 100 or more, both YAP1 and ASCL1 were usually diffusely positive (Figure ). Importantly, YAP1 was expressed with a uniform small‐cell morphology. Based on the assumption that YAP1 is a non‐NE marker , this suggests a mixture of NE and non‐NE phenotypes, even in morphologically homogeneous SCLC. In YAP1‐positive cases of pure SCLCs, YAP1 was expressed mostly in a mosaic‐like pattern, similar to YAP1 mosaic‐like expression in combined SCLCs. Partners in the combined SCLCs As shown in Figure , most of the combined SCLCs had classical LCNEC, SQCC, and adenocarcinoma as partners. The number of partners in combined SCLC was not always one, with 37% (13/35) of combined SCLC cases having a third partner and three cases of combined SCLC having both well‐ and poorly differentiated adenocarcinomas as partners (cases #1, 22, and 25). As a unique partner, one case (case #30) had mixed adenocarcinoma and SQCC, showing a biphasic pattern with p40‐positive SQCC in the outer layer and adenocarcinoma in the inner layer, similar to tumours recognised as mucoepidermoid carcinoma‐like adenosquamous carcinoma in a previous study (Figure ). Two patients (cases #2 and 9) had NSCLC that could not be determined histopathologically or phenotypically, with a mixed expression of TTF‐1 and p40 by double IHC (Figure ). One case (case #3) included myogenin‐positive rhabdomyosarcoma in addition to SQCC (Figure ). The presence of these lineage varieties as partners in combined SCLC suggests that other plasticities occur simultaneously (between adenocarcinoma and SQCC or carcinoma and sarcoma), along with lineage plasticity between NE and non‐NE components. Characteristics of partners in combined SCLCs by dominant transcription subtypes Among the combined SCLCs, the area occupied by SCLC varied from 1% to 99% (median, 40%). No significant differences were observed in the areas occupied by SCLC among the three dominant transcription subtypes. Interestingly, a relationship was found between dominant transcription subtype and NSCLC partners. The adenocarcinoma area in the combined SCLC‐N group was significantly larger than that in the combined SCLC group (median rate, 25% versus 0%, p = 0.006). Conversely, the SQCC area in the combined SCLC‐P group was significantly larger than that in the combined SCLC group (median rate 40% versus 0%, p = 0.0006). Figure also indicates that adenocarcinoma was abundant in combined SCLC‐N, while SQCC dominates in combined SCLC‐P. In contrast, SCLC‐A was a mix of different NSCLC partners with no discernible trend. Concordance of p53 and Rb1 IHC by each component in the combined SCLCs The expressions of p53 and Rb1 were assessed in each component, including the partner, in mosaic patterns, with 85 samples assessed in all 35 cases. p53 showed almost universal abnormal expression in both SCLC and partner (97%; Figure ), with only three components of one combined SCLC showing wild‐type p53 (case #20). Rb1 loss was found in 94% (33 of 35) of SCLC and 78% (39 of 50) of the partners. The p53 and Rb1 aberration concordance rates for all partners showing the same aberrant expression as in SCLC were 100% (34/34) and 72% (24/33), respectively. Discordant Rb1 aberration patterns were observed in nine cases, where all components of SCLC showed loss of Rb1, while one or more of the partners showed retention of Rb1. All cases showing discordant Rb1 expression were of a separated morphological type (Figure ). NGS in combined SCLC A total of 35 DNA samples from 17 patients were examined using NGS. Among the 15 pairs of combined SCLCs in which component‐by‐component analysis was possible, TP53 abnormalities were found in 73% (11/15) and Rb1 abnormalities were identified in 20% (3/15). The TP53 and Rb1 aberration concordance rates for all partners with the identical aberration as SCLC were 91% (10/11) and 66% (2/3), respectively. In one case (case #23), the site of the TP53 mutation differed between NEC (SCLC and LCNEC) and adenocarcinoma. In case #8, only the SCLC had the Rb1 deletion and FBXW7 mutation, although both the SCLC and SQCC shared the same TP53 missense mutation. EGFR mutations were observed in two cases: one was confirmed in both SCLC and adenocarcinoma (case #20; this patient was the only non‐smoker). Other mutations, such as CDH1, BRAF, ABL1, HER4 , and MLH1 , resulted in the same mutation spots in both SCLC and its partner. The variant allele frequency was generally higher in SCLC than in the partners, possibly due to the hypercellularity of SCLC. As shown in Figure , 35 cases were morphologically classified as combined SCLC and 65 as pure SCLC. Of the 100 patients included, 87 were males and 13 were females, with a median age of 72 years (range, 40–91 years). All but one male patient (who had combined SCLC‐N) had a history of smoking. The tumour size in the combined SCLC group was significantly larger than that in the pure SCLC group (median 31 versus 25 mm, p = 0.0009); however, no significant differences in other factors were observed between the two groups. As shown in Figure , NEUROD1 expression was significantly higher in combined SCLCs than in pure SCLCs (median H ‐score 50 versus 0, p = 0.04); however, no significant differences in the expression of other proteins or the composition of dominant transcription types were observed. In 98 cases excluding SCLC‐TN, double‐positive ( H ‐score ≥ 50) transcription marker cases were significantly more common in combined SCLCs, and single‐positive cases were more common in pure SCLCs ( p = 0.019), suggesting that combined SCLCs have stronger plasticity between transcription marker expression (Figure ). expression Figure shows the distribution of YAP1 expression ( H ‐score) in the SCLC components of combined and pure SCLCs. YAP1 expression ( H ‐score > 1) was observed in 80% of combined SCLCs and 62% of pure SCLCs. The YAP1‐negative rate in SCLC components was 20% for combined SCLCs and 38% for pure SCLCs. In univariate and multivariate analyses for OS, the presence of chemotherapy was a significant favourable prognostic factor, and the presence of interstitial pneumonia was a significant unfavourable prognostic factor (Table ). Combined SCLCs tended to have a poorer prognosis than pure SCLCs regarding OS, but this was not significant (Table ). SCLCs Thirty‐five combined SCLCs were morphologically classified as mosaic type (5 cases), coexisting mosaic and separated type (7 cases), or separated type (23 cases) (Figures and ). The median proportions of SCLC components in the total tumours of each type were 90%, 70%, and 40%, respectively (Figure ). The proportion of SCLC in the total tumour was significantly lower in the separated morphological pattern than in the other two patterns ( p = 0.018). In separated morphological patterns, clear boundaries existed between SCLC and the partner (Figure ), which were also highlighted by NEUROD1 and Napsin A (Figure ) or POU2F3 and p40 (Figure ). At high magnification, SCLC and its partners were accompanied by clear intervening stromal components, which could be seen in YAP1 expression. YAP1 was focally expressed in areas of morphological SCLC (data not shown). In the mosaic morphological pattern, partners (all of which were NSCLC of this type) were scattered in a mosaic‐like pattern. Indistinct and ambiguous boundaries were observed between the SCLC and NSCLC partners without intervening stroma. YAP1 showed mosaic‐like expression, almost coinciding with areas that were morphologically NSCLC, but was also weakly expressed in areas that were morphologically SCLC (Figure ). The border between SCLC and NSCLC was ambiguous in both H&E and YAP1 cells, indicating a gradual transition between SCLC and NSCLC. In the adenocarcinoma components of combined SCLC showing mosaic morphological types, Rb1 was lost, as in SCLC components (arrows in Figure ). SCLCs focusing on YAP1 expression All pure SCLCs exhibited homogeneous small‐cell cytology. Figure shows four representative examples of ASCL1‐dominant pure SCLCs. In YAP1‐negative cases, YAP1 was completely negative, and ASCL1 was positive (Figure ). In cases with a YAP1 H ‐score of 1–99, YAP1 was expressed in a mosaic pattern with a mutually exclusive expression pattern with ASCL1 (Figure ). In cases with a YAP1 H ‐score of 100 or more, both YAP1 and ASCL1 were usually diffusely positive (Figure ). Importantly, YAP1 was expressed with a uniform small‐cell morphology. Based on the assumption that YAP1 is a non‐NE marker , this suggests a mixture of NE and non‐NE phenotypes, even in morphologically homogeneous SCLC. In YAP1‐positive cases of pure SCLCs, YAP1 was expressed mostly in a mosaic‐like pattern, similar to YAP1 mosaic‐like expression in combined SCLCs. SCLCs As shown in Figure , most of the combined SCLCs had classical LCNEC, SQCC, and adenocarcinoma as partners. The number of partners in combined SCLC was not always one, with 37% (13/35) of combined SCLC cases having a third partner and three cases of combined SCLC having both well‐ and poorly differentiated adenocarcinomas as partners (cases #1, 22, and 25). As a unique partner, one case (case #30) had mixed adenocarcinoma and SQCC, showing a biphasic pattern with p40‐positive SQCC in the outer layer and adenocarcinoma in the inner layer, similar to tumours recognised as mucoepidermoid carcinoma‐like adenosquamous carcinoma in a previous study (Figure ). Two patients (cases #2 and 9) had NSCLC that could not be determined histopathologically or phenotypically, with a mixed expression of TTF‐1 and p40 by double IHC (Figure ). One case (case #3) included myogenin‐positive rhabdomyosarcoma in addition to SQCC (Figure ). The presence of these lineage varieties as partners in combined SCLC suggests that other plasticities occur simultaneously (between adenocarcinoma and SQCC or carcinoma and sarcoma), along with lineage plasticity between NE and non‐NE components. SCLCs by dominant transcription subtypes Among the combined SCLCs, the area occupied by SCLC varied from 1% to 99% (median, 40%). No significant differences were observed in the areas occupied by SCLC among the three dominant transcription subtypes. Interestingly, a relationship was found between dominant transcription subtype and NSCLC partners. The adenocarcinoma area in the combined SCLC‐N group was significantly larger than that in the combined SCLC group (median rate, 25% versus 0%, p = 0.006). Conversely, the SQCC area in the combined SCLC‐P group was significantly larger than that in the combined SCLC group (median rate 40% versus 0%, p = 0.0006). Figure also indicates that adenocarcinoma was abundant in combined SCLC‐N, while SQCC dominates in combined SCLC‐P. In contrast, SCLC‐A was a mix of different NSCLC partners with no discernible trend. IHC by each component in the combined SCLCs The expressions of p53 and Rb1 were assessed in each component, including the partner, in mosaic patterns, with 85 samples assessed in all 35 cases. p53 showed almost universal abnormal expression in both SCLC and partner (97%; Figure ), with only three components of one combined SCLC showing wild‐type p53 (case #20). Rb1 loss was found in 94% (33 of 35) of SCLC and 78% (39 of 50) of the partners. The p53 and Rb1 aberration concordance rates for all partners showing the same aberrant expression as in SCLC were 100% (34/34) and 72% (24/33), respectively. Discordant Rb1 aberration patterns were observed in nine cases, where all components of SCLC showed loss of Rb1, while one or more of the partners showed retention of Rb1. All cases showing discordant Rb1 expression were of a separated morphological type (Figure ). in combined SCLC A total of 35 DNA samples from 17 patients were examined using NGS. Among the 15 pairs of combined SCLCs in which component‐by‐component analysis was possible, TP53 abnormalities were found in 73% (11/15) and Rb1 abnormalities were identified in 20% (3/15). The TP53 and Rb1 aberration concordance rates for all partners with the identical aberration as SCLC were 91% (10/11) and 66% (2/3), respectively. In one case (case #23), the site of the TP53 mutation differed between NEC (SCLC and LCNEC) and adenocarcinoma. In case #8, only the SCLC had the Rb1 deletion and FBXW7 mutation, although both the SCLC and SQCC shared the same TP53 missense mutation. EGFR mutations were observed in two cases: one was confirmed in both SCLC and adenocarcinoma (case #20; this patient was the only non‐smoker). Other mutations, such as CDH1, BRAF, ABL1, HER4 , and MLH1 , resulted in the same mutation spots in both SCLC and its partner. The variant allele frequency was generally higher in SCLC than in the partners, possibly due to the hypercellularity of SCLC. In our study, the prevalence of combined SCLCs in resected SCLCs was 35%, whereas previous reports have ranged from 5% to 34% . These discrepancies may stem from variations in the thoroughness of efforts to differentiate between SCLC and NSCLC, requiring cautious interpretation of the results. However, we believe our results are accurate because we performed a careful pathological and morphological evaluation. The tumour size of combined SCLCs was significantly larger than that of pure SCLCs. The presence of NSCLC components with lower proliferative activity may lead to clinical recognition after the tumour has become larger. Detailed morphological observations further highlighted the morphological variations in the combined SCLCs. Notably, in one‐third of the combined SCLC cases, partners were distributed in a mosaic pattern, with five cases consisting of only this mosaic pattern. Combined SCLCs with a mosaic pattern had no obvious intervening stroma between the SCLC and partner. This mosaic morphological pattern appears to indicate the presence of a transitional capacity between SCLC and its partner and may be a pathological finding indicative of SCLC plasticity. Another interesting phenomenon was the identification of unique plasticity between partners, such as carcinoma and sarcoma, or adenocarcinoma and SQCC. Furthermore, two cases of combined SCLCs had NSCLC partners that could not be classified as adenocarcinoma or SQCC. Recently, TTF1/p40 diffuse double‐positive NSCLC showing bilineage differentiation at the same cellular level, which may be due to progenitor cell plasticity, has been reported . Plasticity between adenocarcinoma‐SQCCs has been reported to be involved in treatment resistance . Combined SCLC may resist adverse conditions by exhibiting a variety of plasticity. YAP1 expression was found not only in NSCLC partners but also in SCLC components in 80% of combined SCLCs and 62% of pure SCLCs, with most showing a mosaic pattern. Similar mosaic‐like YAP1 expression has already been suggested in several reports . Importantly, YAP1 was expressed in SCLC components with uniform small‐cell morphology, whether combined or pure. Based on the assumption that YAP1 is a non‐NE marker , this suggests a mixture of NE and non‐NE phenotypes, even in morphological SCLC. The YAP1 mosaic pattern appeared to symbolise fluctuations between SCLC (NE) and the partner (non‐NE), even in morphological SCLC. YAP1 mosaic‐like expression was found in both combined and pure SCLC. Combined SCLC is a more morphologically and phenotypically complex tumour than pure SCLC; however, pure SCLC is also a phenotypically complex tumour with various YAP1 patterns, including mosaic‐like expression. If so, mosaic‐like YAP1 expression in pure SCLC may represent phenotypic plasticity and be reminiscent of latent combined SCLC, implying a sequential pathogenesis between combined and pure SCLC. Combined SCLC had significantly higher expression of NEUROD1, and significantly more cases were double‐positive for the transcription markers than pure SCLC. Recent reports have shown a high proportion of the NEUROD1‐dominant type in combined SCLC , suggesting some influence of transcription markers, especially NEUROD1, on the development of combined SCLC. As another novel insight, an association was found between transcription subtypes and types of partner in combined SCLC. Combined SCLC‐A had various NSCLC as partners, while combined SCLC‐N had a significantly higher proportion of adenocarcinoma components, and combined SCLC‐P had a significantly higher proportion of SQCC components. NEUROD1 participates in the formation of alveolar septa and NE differentiation , while POU2F3 is involved in the generation of tuft cells in close association with basal cells . These findings could also provide evidence supporting the relationship between transcription markers and NSCLC partners in combined SCLC. The cellular origin of SCLC has been postulated to be NE cells or NE stem cells , and studies using mouse models have demonstrated that SCLC arises most efficiently from NE cells but also from a subset of alveolar type 2 cells and basal cells . SCLC is also highly plastic and can assume entirely different NE and non‐NE fates via NOTCH signals . Although de novo SCLC can certainly occur, as supported by in situ SCLC , it is also believed that a significant number of SCLC are of epithelial rather than NE origin, reflecting epithelial‐NE conversion in advanced cases . In this study, component‐wise NGS (for 15 pairs of combined SCLCs) and IHC (for 35 pairs) were performed. By NGS, 10 pairs of combined SCLCs had the identical TP53 mutation, and two pairs had the identical Rb1 abnormality. On IHC, 34 pairs of combined SCLCs exhibited the same abnormal p53 pattern, while 24 pairs showed Rb1 loss. The rate of positive abnormalities in NGS is low due to panel sequencing, and the number of NGS cases is limited. Although strictly proving clonality may be challenging, the presence of the identical TP53 mutation in 10 pairs supports the single clone theory of combined SCLC. Combined SCLC is a symbolic morphological finding of plasticity between NE and non‐NE. It is difficult to determine the direction of tumour differentiation; however, the presence of a certain number of cases with Rb1 /Rb1 abnormalities only in SCLC provides evidence for the possibility that SCLC abnormalities occur in the order TP53 /p53 to Rb1 /Rb1 and the possibility that conversion occurs from NSCLC to SCLC. The current standard treatment for all SCLC needs to be reexamined to determine the best therapy. Similar treatment strategies have been recommended for both combined and pure SCLC. However, responses to tyrosine kinase inhibitors (TKIs) in EGFR ‐mutant SCLC have been reported , and EGFR‐TKIs may be an option, especially in cases with a high proportion of EGFR ‐mutant NSCLC. In combined SCLC, it has been reported that different partners have different vulnerabilities to certain cell death pathways, such as ferroptosis , which may provide a basis for suggesting different treatment strategies depending on the type of partner and also strengthen the relationship between partner and molecular and/or transcription‐based classification. The limitations of this study include selection bias due to limited surgical material. Surgical materials must be selected for detailed observation; however, studies using small samples, including biopsies, need to verify this in the future. A second limitation was the impact of block quality, as older blocks may have decreased immunosignal intensity. The median H ‐score was higher in the most recent FFPE blocks than in the oldest ones for the antibodies stained in the nucleus, but not significantly different (pathological images from old and recent FFPE blocks are shown in supplementary material, Figure ). The use of FFPE blocks that are too old should be avoided but, given the limited number of surgical SCLC cases, these had to be used in this study. At our institution, more attention has been paid to fixation and quality control since 2000, which may minimise the effects of age‐related degeneration. The third limitation of NGS in this study was the hotspot analysis and exon sequencing. For rigorous proof of clonality, it may be necessary to analyse genomic tests with higher coverage. In conclusion, combined SCLC is a more morphologically and phenotypically complex tumour compared with pure SCLC. However, pure SCLC is a phenotypically complex tumour in terms of YAP1 expression. The morphological mosaic pattern and YAP1 mosaic‐like expression may help to visualise the ongoing lineage plasticity. Another point is that NSCLC partners in combined SCLC were differentially characterised by transcription marker expression, with NEUROD1 showing high affinity for adenocarcinoma and POU2F3 for SQCC. These results suggest that transcription factors are involved in specific cell lineages and not only in NE cells. NJ and CO contributed to the conception, study design, interpretation and drafting of the manuscript. TF and MT contributed to sample collection, genetic analysis and revision of the manuscript. SM, YT and YM provided the samples, performed prognostic analysis and revised the manuscript. TI contributed to the IHC and revised the manuscript. All the authors approved the final version of the manuscript. Figure S1. Pathological images of an old block and a recent block Table S1. Details of the immunohistochemical protocols |
Information overload and parental perspectives on information provided to parents/carers of paediatric patients undergoing elective surgical procedures | 6016168c-f67a-4d04-ae7c-c75d14c73173 | 11495572 | Health Literacy[mh] | Health literacy includes people’s knowledge, motivation and competences to access, understand, appraise, and apply health information to make judgments and take decisions concerning healthcare, disease prevention and health promotion . Evidence from the United States of America and Europe suggest that some patients may have inadequate health literacy levels to self-manage their health conditions; this applies both broadly, and specifically to diseases such as heart failure , chronic pain and chronic kidney disease . The consequences of limited health literacy can be further compounded by information overload. In the current information age, patients and families may be exposed to a potentially bewildering array of information sources of varying quality and trustworthiness . As outlined in Obamiro and Lee , this information may lead individuals to not paying attention to vital information, incorrectly processing, or even avoiding information, all of which could have a negative impact on health outcomes. Additionally, a recent extensive study assessing consumer priorities for perioperative medicine found that improved communication with parents rated eighth in their top ten priorities . In response to survey data that suggested three-quarters of adults are overwhelmed by cancer information, Jensen et al. developed and validated an 8-item Cancer Information Overload (CIO) scale to evaluate the associations between cancer information overload and cancer-related behaviours such as cancer screening. They found that patients who reported feeling overloaded with cancer information and who consequently had a higher score on the CIO scale, predicted a lower likelihood of colon cancer screening within an 18-month period. The authors concluded that high CIO undermines health efforts. The authors suggest that healthcare providers and public health professionals could use the CIO scale to develop evidence-based communication strategies to mitigate CIO. Cancer information overload is the most popular topic in studies of consumer health information overload; however, the scale has been adapted for use in diet information, weight management, intensive care next of kin and coronary heart disease . Obamiro and Lee assessed the validity of the 8-item scale for use with Australian atrial fibrillation patients. The authors also validated a modified 5-item version for the same population . Both the original 8-item and reduced 5-item scales were reported to be valid and reliable for measuring health information overload among adult Australians living with atrial fibrillation . They suggested that both the 8-item and reduced 5-item modified CIO scales may be suitably adapted for health information overload in patients with other chronic diseases; however the reduced 5-item modified CIO scale may be preferable if time is limited to collect participant responses or in attempts to mitigate respondent fatigue and improve response quality. Information overload may apply not just to chronic conditions. It has been suggested that information overload could negatively impact patients’ abilities to contribute to their healthcare management . In the paediatric setting, where parents are pivotal in making decisions and managing their children’s health perioperatively, the impact of information overload on parents remains largely unexplored, despite its potentially significant consequences. Parents attending hospital with their children for surgery see an array of health professionals perioperatively who provide information on their child’s condition, treatment, and ongoing management . Additionally, the perioperative period is notably stressful for parents, with heightened levels of anxiety for both parents and child, further burdening the mental load of parents. Therefore, our research team conducted a study among parents and carers of children undergoing any elective surgical procedure at our institution with the objective to investigate information communication and overload.
This study was approved by the Child and Adolescent Health Service Human Research Ethics Committee (RGS000003994) and recognised by The University of Western Australia Human Research Ethics Committee (2021/ET000512). The objectives were to: Assess the usefulness of a modified version of the Cancer Information Overload scale with parents/carers of paediatric patients undergoing any elective surgical procedure; Identify the level of information overload, if any, among such parents/carers; Explore information communication and overload from the parents’ /carers’ perspectives. Survey of participants on day of surgery Participants (parents of children undergoing any elective surgical procedure) were invited to participate through poster in the perioperative area advertising with QR code and/or by members of the research team. A questionnaire was developed in Qualtrics (Utah, USA,) and disseminated to participants from May 2021 to October 2021. At the start of the survey parents were asked to consent to participate in the survey. Demographic information (age, gender, Aboriginal or Torres Strait Islander self-identification, postal code, annual income, highest educational level, and employment status) was recorded by respondents. Participants were also asked how many health professionals their child had seen prior to the current surgery in relation to the surgery/condition, e.g., general practitioner, specialist, child health nurse or dentist. Postcodes were classified as metropolitan (metro), regional, or remote based on the Australian Bureau of Statistics Statistical Geography Standard Correspondences Postcode 2017 to Remoteness Area 2016 . Postcodes classified as being in “Major Cities of Australia” were grouped under “metro”, postcodes classified as “Inner Regional Australia” or “Outer Regional Australia” were grouped under “regional”, and those classified as “Remote Australia” or “Very Remote Australia” were grouped under “remote”. Participants were asked to respond to an Information Overload scale. The original 8-item CIO scale contained four points per item (from strongly disagree with a value of 1, to strongly agree with a value of 4) . Scoring of the original CIO scale was performed by summing the value of each item (scores range from 8 to 32), with higher scores indicating a greater degree of health information overload. For this study, we adapted the 5-item CIO scale (scores range from 5 to 20) and the 8-item atrial fibrillation scale using”caring for your children after surgery” (see ). Given that there is no previous data to enable sample size estimation based on power, we were unable to perform a power calculation. As such, we used conservative estimates, using a proportion of 0.5 for parental information overload, an unknown large population size, a 95% level of confidence and a margin of error of 5%, resulting in a minimum sample size of 384 . Semi-structured qualitative interviews To provide complementary data to the survey, semi-structured qualitative interviews were conducted with a subgroup of parents of patients who had undergone elective surgery and who indicated they would be open to further contact in the survey. Prior to the interviews, the participants gave informed consent to participate in the study. A convenience sample were contacted by telephone by research assistants to organise interviews. When staff was available, all potentially eligible families were contacted. An Interview Guide was used to ensure consistency between interviews; however, questions remained open-ended, allowing parents to explore themes they deemed important. The Interview Guide began with an introduction to the research team, details of what the study involved and outlined the interview process. The targeted number of participants was 20, as this is typically large enough for data saturation to occur . For the present study, data saturation was defined as the point where no new themes could be independently identified by two or more researchers from the research team. The interviews were undertaken by trained, experienced members of the research staff over the phone or via video call at a time that was convenient for the interviewee. Only the interviewee and interviewer were present during the interviews. No repeat interviews were conducted. Field notes were taken by the interviewers, in order to highlight preliminary themes and identify when data saturation had been reached. Participants’ responses were recorded via an audio recorder and transcribed verbatim. An abbreviated health literacy assessment was conducted using the validated 4-item BRIEF health literacy screening tool to provide further demographic context . The demographic and health literacy data collected during the interview were analysed to provide context on the participant group. Data analysis Analysis of survey responses was conducted using R Version 4.2 . Responses to individual items in the modified 5-item information overload scale were described using medians, lower quartiles (LQ) and upper quartiles (UQ). To establish evidence for structural validity, construct validity and reliability, psychometric properties of the scale were explored, with internal consistency assessed by considering Cronbach’s alpha, and the factor structure of the scale established by performing exploratory factor analysis (EFA) with oblique rotation. To identify variables associated with information overload, multiple linear regression analysis was performed for the modified 5-item information overload scale total score. Variables considered in the regression analysis included participants’ demographic characteristics (gender, age group, Aboriginal or Torres Strait Islander self-identification, whether a language other than English is spoken at home, education level, employment status, income, postcode remoteness) and whether or not the respondent or any of their children had previously had a similar surgery. Interaction terms between variables were included based on exploration of interaction plots and cross-tables. Backwards stepwise regression was used to reduce the model to significant covariates, where statistical significance was taken at p<0.05. Model assumptions were verified by inspection of diagnostic plots. The summed score of multiple Likert type items into a Likert scale score can be treated as interval data, and is therefore suitable for analysis by linear regression . As lower education level has previously been found to be significantly associated with greater cancer information overload , we predicted that education levels would be negatively correlated with our modified information overload scale scores. While there are few studies examining the relationship between other demographic characteristics and health information overload, we predicted there will be no significant correlations with our other demographic characteristics (age, gender, postcode, annual income, and employment status). Given the low rate of missing data, regression analysis was performed on complete cases only. Thematic analysis was undertaken on the interview transcripts, using the qualitative data analysis software NVivo® 11. Researchers utilized the Triangulation Framework method of analysis to identify themes . The Framework method was chosen because it is useful for research teams comprising of multiple analysts with varying levels of qualitative research experience, as it provides a structured approach to qualitative analysis . Descriptive (quantitative) statistics of demographic and health literacy information were performed using Microsoft Excel®. The proposed study was guided by various methods of trustworthiness proposed by Lincoln and Guba . Analyst triangulation was used during the coding process to improve the reliability of the data, acknowledging that each researcher brings their own attitudes, values and world views to their interpretation of the interview transcripts. Incorporating multiple researcher viewpoints and skillsets into the analysis mitigates the risk of bias or favouring one particular point of view. Prior to analysis, each researcher also reflected on their pre-existing perspectives, assumptions and expectations of the data in order to identify implicit biases, thus further improving the reliability of the data. Three researchers (EB, MD and SG) firstly read through all interview transcripts to familiarise themselves with the data. One initial transcript was coded by three researchers (EB, MD and SG) and principal investigator KL. Researchers then met to discuss discrepancies in coding and reconcile any differences. Following this discussion, a working analytical framework was developed in consultation with KL and the study investigators. This framework comprised of a table of codes identified from the data with associated definitions. Researchers EB, MD and SG then coded the remaining transcripts based on this framework, adding new codes and definitions as they arose in discussion with the study team. Coded data was combined into a framework matrix showing participant data for each code, including direct and/or summarised participant quotes. Codes were then tentatively grouped together according to emerging themes. Final themes were established by mapping connections within and between participant responses in collaboration with the research team. Results are reported in accordance with COREQ (Consolidated criteria for reporting qualitative research) or STROBE guidelines as appropriate.
Participants (parents of children undergoing any elective surgical procedure) were invited to participate through poster in the perioperative area advertising with QR code and/or by members of the research team. A questionnaire was developed in Qualtrics (Utah, USA,) and disseminated to participants from May 2021 to October 2021. At the start of the survey parents were asked to consent to participate in the survey. Demographic information (age, gender, Aboriginal or Torres Strait Islander self-identification, postal code, annual income, highest educational level, and employment status) was recorded by respondents. Participants were also asked how many health professionals their child had seen prior to the current surgery in relation to the surgery/condition, e.g., general practitioner, specialist, child health nurse or dentist. Postcodes were classified as metropolitan (metro), regional, or remote based on the Australian Bureau of Statistics Statistical Geography Standard Correspondences Postcode 2017 to Remoteness Area 2016 . Postcodes classified as being in “Major Cities of Australia” were grouped under “metro”, postcodes classified as “Inner Regional Australia” or “Outer Regional Australia” were grouped under “regional”, and those classified as “Remote Australia” or “Very Remote Australia” were grouped under “remote”. Participants were asked to respond to an Information Overload scale. The original 8-item CIO scale contained four points per item (from strongly disagree with a value of 1, to strongly agree with a value of 4) . Scoring of the original CIO scale was performed by summing the value of each item (scores range from 8 to 32), with higher scores indicating a greater degree of health information overload. For this study, we adapted the 5-item CIO scale (scores range from 5 to 20) and the 8-item atrial fibrillation scale using”caring for your children after surgery” (see ). Given that there is no previous data to enable sample size estimation based on power, we were unable to perform a power calculation. As such, we used conservative estimates, using a proportion of 0.5 for parental information overload, an unknown large population size, a 95% level of confidence and a margin of error of 5%, resulting in a minimum sample size of 384 .
To provide complementary data to the survey, semi-structured qualitative interviews were conducted with a subgroup of parents of patients who had undergone elective surgery and who indicated they would be open to further contact in the survey. Prior to the interviews, the participants gave informed consent to participate in the study. A convenience sample were contacted by telephone by research assistants to organise interviews. When staff was available, all potentially eligible families were contacted. An Interview Guide was used to ensure consistency between interviews; however, questions remained open-ended, allowing parents to explore themes they deemed important. The Interview Guide began with an introduction to the research team, details of what the study involved and outlined the interview process. The targeted number of participants was 20, as this is typically large enough for data saturation to occur . For the present study, data saturation was defined as the point where no new themes could be independently identified by two or more researchers from the research team. The interviews were undertaken by trained, experienced members of the research staff over the phone or via video call at a time that was convenient for the interviewee. Only the interviewee and interviewer were present during the interviews. No repeat interviews were conducted. Field notes were taken by the interviewers, in order to highlight preliminary themes and identify when data saturation had been reached. Participants’ responses were recorded via an audio recorder and transcribed verbatim. An abbreviated health literacy assessment was conducted using the validated 4-item BRIEF health literacy screening tool to provide further demographic context . The demographic and health literacy data collected during the interview were analysed to provide context on the participant group.
Analysis of survey responses was conducted using R Version 4.2 . Responses to individual items in the modified 5-item information overload scale were described using medians, lower quartiles (LQ) and upper quartiles (UQ). To establish evidence for structural validity, construct validity and reliability, psychometric properties of the scale were explored, with internal consistency assessed by considering Cronbach’s alpha, and the factor structure of the scale established by performing exploratory factor analysis (EFA) with oblique rotation. To identify variables associated with information overload, multiple linear regression analysis was performed for the modified 5-item information overload scale total score. Variables considered in the regression analysis included participants’ demographic characteristics (gender, age group, Aboriginal or Torres Strait Islander self-identification, whether a language other than English is spoken at home, education level, employment status, income, postcode remoteness) and whether or not the respondent or any of their children had previously had a similar surgery. Interaction terms between variables were included based on exploration of interaction plots and cross-tables. Backwards stepwise regression was used to reduce the model to significant covariates, where statistical significance was taken at p<0.05. Model assumptions were verified by inspection of diagnostic plots. The summed score of multiple Likert type items into a Likert scale score can be treated as interval data, and is therefore suitable for analysis by linear regression . As lower education level has previously been found to be significantly associated with greater cancer information overload , we predicted that education levels would be negatively correlated with our modified information overload scale scores. While there are few studies examining the relationship between other demographic characteristics and health information overload, we predicted there will be no significant correlations with our other demographic characteristics (age, gender, postcode, annual income, and employment status). Given the low rate of missing data, regression analysis was performed on complete cases only. Thematic analysis was undertaken on the interview transcripts, using the qualitative data analysis software NVivo® 11. Researchers utilized the Triangulation Framework method of analysis to identify themes . The Framework method was chosen because it is useful for research teams comprising of multiple analysts with varying levels of qualitative research experience, as it provides a structured approach to qualitative analysis . Descriptive (quantitative) statistics of demographic and health literacy information were performed using Microsoft Excel®. The proposed study was guided by various methods of trustworthiness proposed by Lincoln and Guba . Analyst triangulation was used during the coding process to improve the reliability of the data, acknowledging that each researcher brings their own attitudes, values and world views to their interpretation of the interview transcripts. Incorporating multiple researcher viewpoints and skillsets into the analysis mitigates the risk of bias or favouring one particular point of view. Prior to analysis, each researcher also reflected on their pre-existing perspectives, assumptions and expectations of the data in order to identify implicit biases, thus further improving the reliability of the data. Three researchers (EB, MD and SG) firstly read through all interview transcripts to familiarise themselves with the data. One initial transcript was coded by three researchers (EB, MD and SG) and principal investigator KL. Researchers then met to discuss discrepancies in coding and reconcile any differences. Following this discussion, a working analytical framework was developed in consultation with KL and the study investigators. This framework comprised of a table of codes identified from the data with associated definitions. Researchers EB, MD and SG then coded the remaining transcripts based on this framework, adding new codes and definitions as they arose in discussion with the study team. Coded data was combined into a framework matrix showing participant data for each code, including direct and/or summarised participant quotes. Codes were then tentatively grouped together according to emerging themes. Final themes were established by mapping connections within and between participant responses in collaboration with the research team. Results are reported in accordance with COREQ (Consolidated criteria for reporting qualitative research) or STROBE guidelines as appropriate.
Survey of participants on day of surgery A total of 382 respondents consented to the Day of Surgery survey. Of these, 2 participants did not respond to any of the questions following consent, and were subsequently excluded from the analysis, leaving 380 participants (female = 290, 76%) from 126 postcodes. The postcode responses indicated that 304 respondents (80%) were in metropolitan areas, with 71 from regional/remote areas and 5 missing. shows a summary of the participant characteristics. Nineteen respondents (5%) identified as Aboriginal or Torres Strait Islander. More than a quarter of respondents (n = 102, 27%) primarily spoke a language other than English at home. 56% of respondents (n = 211) indicated that neither themselves nor any of their children had undergone a similar surgery in the past, and 53% (n = 202) indicated that neither themselves nor any of their children had undergone multiple surgeries in the past. Information overload The responses to the individual items from the modified 5-item information overload scale are summarised in . Most respondents disagreed or strongly disagreed with the majority of the 5 item statements. Item 3 (“Information about caring for my child after surgery all starts to sound the same after a while”) had the highest rate of agreement, with 29% agreeing or strongly agreeing with the statement. A score of 2 (“disagree”) was most commonly selected for all items except for item 2 (“It has gotten to the point where I don’t even care to hear new information about caring for my child after surgery”), for which a score of 1 (“strongly disagree”) was most commonly selected. The mean (SD) of the total score was 9.2 (2.5), and the median (LQ, UQ) was 10 (7, 11). The maximum total score was 18, and the minimum was 5. Cronbach’s alpha for the modified 5-item information overload scale was 0.78, indicating acceptable internal consistency, evidence for acceptable reliability . Exploration of the factor structure indicated a one- or two-factor solution; however, the individual modified information overload scale items were summed to a one-factor solution to align with the design of the original score. Details of the results of EFA are given in . Regression analysis Given the five participants with missing postcode, the complete-case multiple regression analysis was performed on the remaining 375 participants. Unadjusted regression estimates from single-variable models for total score on each candidate independent variable are reported in Table 1 in . However, these do not reflect the true effect of each variable as this analysis does not account for the effect of confounding variables. Interaction terms for education with Aboriginal or Torres Strait Islander, age group with education, and language other than English (LOTE) with education were included in the initial model based on inspection of interaction plots (Figs 1 to 3 in ). Parameter estimates from the full multivariable model before reduction of covariates are shown in Table 2 in . Following backwards stepwise variable selection, the estimated coefficients and 95% CI from the multiple linear regression for the total modified 5-item information overload scale scores are shown in . Model diagnostics confirmed that the normal linear model assumptions were verified. People who spoke a language other than English (LOTE) had higher total score on average (0.98, p<0.001). Education level was shown to have an effect on information overload. None of the Indigenous Australian participants in this cohort reported having university degrees; however, those who completed high school had higher mean information overload total scores compared to those who had technical/TAFE education or who had not completed high school. Of respondents who were not Aboriginal or Torres Strait Islander, those who had TAFE or university qualifications had lower average information overload total score compared to those who had not completed high school (TAFE -1.21, p = 0.013; undergraduate -1.63, p = 0.002; postgraduate -1.12, p = 0.031), indicating lower information overload. The interaction between Aboriginal or Torres Strait Islander heritage and education level was statistically significant (see ). Aboriginal and Torres Strait Islander people whose highest education was high school completion had significantly higher information overload scores than non-Aboriginal people with the same education level (3.4, p = 0.020). No comparison could be made between Aboriginal or Torres Strait Islander and non-Aboriginal participants with undergraduate or postgraduate university degrees, since none of the Indigenous participants in this cohort reported having university degrees. shows the model-estimated mean score and 95% confidence interval for different levels of education and Aboriginal and/or Torres Strait Islander heritage. Semi-structured qualitative interviews Results are reported in accordance with COREQ (Consolidated criteria for reporting qualitative research) . Interviews and data collection were performed by three researchers with experience in qualitative research (authors EB, SB and MD). The interviewers had no prior relationship with the participants. Participants were aware that the research was about their children’s experience with surgery and how they felt about the amount of information they received. A convenience sample of 24 parents took part in semi-structured interviews chosen from respondents who had indicated a willingness to participate in the interview stage (n = 136 of 382; 35; 35.6%). The interviews were conducted between 20 and 59 days following their child’s operation. Of the parents who were contacted for an interview, three parents declined further participation, 28 were interviewed but 4 of these were emergency surgery parents who were subsequently not analysed as the study was for planned elective surgery only. Two interviews were pending when it was determined that data saturation had been reached and therefore these interviews were not scheduled. The BRIEF health literacy screening tool was used with 23 interviewees scoring 17–20 points, indicating they were able to read and comprehend patient education materials. One interviewee had a marginal score which indicated they may need assistance or may struggle with patient education materials. Thirteen parents reported seeing 2–5 health professionals and nine parents more than five health professionals. summarises the characteristics of interview participants. Theme codes and comments from the qualitative analysis are summarised in . Overall, parents were satisfied with the amount of information that they received. Very rarely did any parent complain of ‘information overload’, and of those who did were referring to the overall stress of the hospital environment/staff communication. For example, “ We were dealing with a sick child that was crying so we couldn’t concentrate on anything that was told to me , I would have forgotten everything so having it in writing was excellent ” (PB198). Parents mainly found communication between nurses/doctors and families was positive and effective. However, for some parents, the communication of verbal information by staff was not at the parent’s comprehension level, with one parent stating “ The actual information is clear but then it gets distorted through their (staff) communication ” (PB70). Another parent felt the manner of information delivery could be improved “ Bit more sensitivity to the format and delivery , as opposed to the information itself would be a good investment of time ” (PB85). Parents were pleased with the written information that was given to them to take home. They reported finding this the most effective way to manage the information after surgery. Some parents suggested having more personalised information would be of benefit. Parents with more experience in the healthcare system (either with themselves or children who had previous surgeries) found it easier to retain the information and had already developed strategies on how to process it. One parent noted, “ I guess unfortunate that I have a lot of experience myself in the hospital , unfortunate that I am already aware of all the information , so it is something that is automatic to me ” (PB70). Unsurprisingly, parents who themselves were medical professionals (nurses, GP etc.) found the process much less overwhelming. Previous experience in the hospital and previous experience with personal health concerns and/or chronic illness seemed to play a significant role in how parents perceive and process the information given to them.
A total of 382 respondents consented to the Day of Surgery survey. Of these, 2 participants did not respond to any of the questions following consent, and were subsequently excluded from the analysis, leaving 380 participants (female = 290, 76%) from 126 postcodes. The postcode responses indicated that 304 respondents (80%) were in metropolitan areas, with 71 from regional/remote areas and 5 missing. shows a summary of the participant characteristics. Nineteen respondents (5%) identified as Aboriginal or Torres Strait Islander. More than a quarter of respondents (n = 102, 27%) primarily spoke a language other than English at home. 56% of respondents (n = 211) indicated that neither themselves nor any of their children had undergone a similar surgery in the past, and 53% (n = 202) indicated that neither themselves nor any of their children had undergone multiple surgeries in the past.
The responses to the individual items from the modified 5-item information overload scale are summarised in . Most respondents disagreed or strongly disagreed with the majority of the 5 item statements. Item 3 (“Information about caring for my child after surgery all starts to sound the same after a while”) had the highest rate of agreement, with 29% agreeing or strongly agreeing with the statement. A score of 2 (“disagree”) was most commonly selected for all items except for item 2 (“It has gotten to the point where I don’t even care to hear new information about caring for my child after surgery”), for which a score of 1 (“strongly disagree”) was most commonly selected. The mean (SD) of the total score was 9.2 (2.5), and the median (LQ, UQ) was 10 (7, 11). The maximum total score was 18, and the minimum was 5. Cronbach’s alpha for the modified 5-item information overload scale was 0.78, indicating acceptable internal consistency, evidence for acceptable reliability . Exploration of the factor structure indicated a one- or two-factor solution; however, the individual modified information overload scale items were summed to a one-factor solution to align with the design of the original score. Details of the results of EFA are given in .
Given the five participants with missing postcode, the complete-case multiple regression analysis was performed on the remaining 375 participants. Unadjusted regression estimates from single-variable models for total score on each candidate independent variable are reported in Table 1 in . However, these do not reflect the true effect of each variable as this analysis does not account for the effect of confounding variables. Interaction terms for education with Aboriginal or Torres Strait Islander, age group with education, and language other than English (LOTE) with education were included in the initial model based on inspection of interaction plots (Figs 1 to 3 in ). Parameter estimates from the full multivariable model before reduction of covariates are shown in Table 2 in . Following backwards stepwise variable selection, the estimated coefficients and 95% CI from the multiple linear regression for the total modified 5-item information overload scale scores are shown in . Model diagnostics confirmed that the normal linear model assumptions were verified. People who spoke a language other than English (LOTE) had higher total score on average (0.98, p<0.001). Education level was shown to have an effect on information overload. None of the Indigenous Australian participants in this cohort reported having university degrees; however, those who completed high school had higher mean information overload total scores compared to those who had technical/TAFE education or who had not completed high school. Of respondents who were not Aboriginal or Torres Strait Islander, those who had TAFE or university qualifications had lower average information overload total score compared to those who had not completed high school (TAFE -1.21, p = 0.013; undergraduate -1.63, p = 0.002; postgraduate -1.12, p = 0.031), indicating lower information overload. The interaction between Aboriginal or Torres Strait Islander heritage and education level was statistically significant (see ). Aboriginal and Torres Strait Islander people whose highest education was high school completion had significantly higher information overload scores than non-Aboriginal people with the same education level (3.4, p = 0.020). No comparison could be made between Aboriginal or Torres Strait Islander and non-Aboriginal participants with undergraduate or postgraduate university degrees, since none of the Indigenous participants in this cohort reported having university degrees. shows the model-estimated mean score and 95% confidence interval for different levels of education and Aboriginal and/or Torres Strait Islander heritage.
Results are reported in accordance with COREQ (Consolidated criteria for reporting qualitative research) . Interviews and data collection were performed by three researchers with experience in qualitative research (authors EB, SB and MD). The interviewers had no prior relationship with the participants. Participants were aware that the research was about their children’s experience with surgery and how they felt about the amount of information they received. A convenience sample of 24 parents took part in semi-structured interviews chosen from respondents who had indicated a willingness to participate in the interview stage (n = 136 of 382; 35; 35.6%). The interviews were conducted between 20 and 59 days following their child’s operation. Of the parents who were contacted for an interview, three parents declined further participation, 28 were interviewed but 4 of these were emergency surgery parents who were subsequently not analysed as the study was for planned elective surgery only. Two interviews were pending when it was determined that data saturation had been reached and therefore these interviews were not scheduled. The BRIEF health literacy screening tool was used with 23 interviewees scoring 17–20 points, indicating they were able to read and comprehend patient education materials. One interviewee had a marginal score which indicated they may need assistance or may struggle with patient education materials. Thirteen parents reported seeing 2–5 health professionals and nine parents more than five health professionals. summarises the characteristics of interview participants. Theme codes and comments from the qualitative analysis are summarised in . Overall, parents were satisfied with the amount of information that they received. Very rarely did any parent complain of ‘information overload’, and of those who did were referring to the overall stress of the hospital environment/staff communication. For example, “ We were dealing with a sick child that was crying so we couldn’t concentrate on anything that was told to me , I would have forgotten everything so having it in writing was excellent ” (PB198). Parents mainly found communication between nurses/doctors and families was positive and effective. However, for some parents, the communication of verbal information by staff was not at the parent’s comprehension level, with one parent stating “ The actual information is clear but then it gets distorted through their (staff) communication ” (PB70). Another parent felt the manner of information delivery could be improved “ Bit more sensitivity to the format and delivery , as opposed to the information itself would be a good investment of time ” (PB85). Parents were pleased with the written information that was given to them to take home. They reported finding this the most effective way to manage the information after surgery. Some parents suggested having more personalised information would be of benefit. Parents with more experience in the healthcare system (either with themselves or children who had previous surgeries) found it easier to retain the information and had already developed strategies on how to process it. One parent noted, “ I guess unfortunate that I have a lot of experience myself in the hospital , unfortunate that I am already aware of all the information , so it is something that is automatic to me ” (PB70). Unsurprisingly, parents who themselves were medical professionals (nurses, GP etc.) found the process much less overwhelming. Previous experience in the hospital and previous experience with personal health concerns and/or chronic illness seemed to play a significant role in how parents perceive and process the information given to them.
Most respondents to the initial questionnaire disagreed or strongly disagreed with most of the individual item statements in the modified 5-item information overload scale, indicating that they did not feel overloaded with information about the care of their children after surgery. The linear regression analysis revealed that the interaction between education level and Indigenous Australian self-identification was statistically significant. We found a mean difference in information overload between Indigenous and non-Indigenous Australians that depended on their highest education level. Higher education levels corresponded to lower information overload in our cohort. In a 2020 scoping review by Khaleel et al. lower education level was associated with health information overload in several studies they reviewed; however, one study had reported a higher level of education to be associated with diet information overload. This diet study had a high proportion of Hispanic/Latino participants and more than 20% had below high school education level. Additionally, many of the participants received nutrition assistance services. Therefore, the scoping review authors conclude that despite this contradictory finding there is an association between education level and information overload. Although the proportion of Aboriginal or Torres Strait Islander participants in the online survey cohort mirrored the general population of Western Australia, none of the Indigenous Australian participants in this cohort reported having university degrees; therefore, we cannot make inferences about differences in information overload score between Indigenous and non-Indigenous Australians with university degrees. We also found that households that spoke a language other than English had higher information overload (0.98, p<0.001). As expected, we found no significant correlations between the total modified 5-item information overload score and the other demographic characteristics; age, gender and postcode. Further annual income and employment status were also uncorrelated with the total modified information overload score. This contrasts with Khaleel et al., who reported that low socioeconomic status is positively associated with information overload based on studies they included in their scoping analysis. Overall, the scale could be applied to our population in the perioperative setting and yielded important information to guide information flow to parents. Thematic analysis of the semi-structured interviews revealed that most parents did not feel overloaded with information. Some parents did report feeling overwhelmed when different staff presented differing or contradictory information, and this was identified as a barrier (conflicting information). This can be closely linked to another barrier around conflicting information between what is being given in the hospital and information that parents have accessed via the internet. Another barrier was poor communication by staff. Parents gave examples of information, which was not presented at the parent’s comprehension level, not being tailored to their child or being delivered too fast. This is important to note since we only interviewed parents with English as a first language, and given we found that information overload was significantly higher in those with a language other than English at home, this barrier would have an even greater impact on such families. When questioned on suggestions to help facilitate communication, most parents in the interviews said it was helpful to have written information to take home; this was also reflected in the follow-up survey responses where 77% indicated a preference for written information, 9% verbal and 14% both verbal and written. Another suggestion from the parent interviews was to have take-home care information which was less generic and more personalised. This study assessed a modified Information Overload scale for use in parents/carers of paediatric patients undergoing any elective surgical procedures and explored information overload among parents/carers through qualitative semi-structured interviews. A scoping review of health information overload among health consumers reported on 22 studies, of which 10 focused on cancer information overload (CIO) with nothing in the perioperative space even though this may be of particular importance given parents/patients are given a lot of information at a stressful time, with the added risks of drug misuse and inadequate postoperative care. Limitations The study was conducted at a single centre. However, our institution is the only paediatric tertiary referral centre in Western Australia, and it serves a wide range of patients from both metropolitan and regional areas. The interviewees were English-speaking parents only due lack of funding for interpretation services for research. It is a limitation of the study that information on health literacy was not collected in the survey and that purposive sampling of those with high overload was not used for the qualitative study sampling. This work did not include patients having emergency surgery where time to receive and digest complex information is shorter and often received only verbally and at time when an individual’s information-overload threshold is likely to be lowered . Our results show that parents of elective surgical parents are not suffering from information overload in general, but they do want more information on immediate and late postoperative recovery. This study also highlights the need to take extra care when communicating with families of vulnerable population groups, including those who are non-native English speakers or who may not be university educated. By doing so, staff can make a meaningful difference to the patient experience and their understanding of the information presented to them. Future work in the elective surgery setting should focus on culturally and linguistically diverse patient groups and the level of information overload for this population, which would necessitate alternate communication strategies. The information overload scale may be useful in such research to identify those with the highest levels of overload. Studying the information overload in parents of children undergoing emergency surgery would also be worthwhile.
The study was conducted at a single centre. However, our institution is the only paediatric tertiary referral centre in Western Australia, and it serves a wide range of patients from both metropolitan and regional areas. The interviewees were English-speaking parents only due lack of funding for interpretation services for research. It is a limitation of the study that information on health literacy was not collected in the survey and that purposive sampling of those with high overload was not used for the qualitative study sampling. This work did not include patients having emergency surgery where time to receive and digest complex information is shorter and often received only verbally and at time when an individual’s information-overload threshold is likely to be lowered . Our results show that parents of elective surgical parents are not suffering from information overload in general, but they do want more information on immediate and late postoperative recovery. This study also highlights the need to take extra care when communicating with families of vulnerable population groups, including those who are non-native English speakers or who may not be university educated. By doing so, staff can make a meaningful difference to the patient experience and their understanding of the information presented to them. Future work in the elective surgery setting should focus on culturally and linguistically diverse patient groups and the level of information overload for this population, which would necessitate alternate communication strategies. The information overload scale may be useful in such research to identify those with the highest levels of overload. Studying the information overload in parents of children undergoing emergency surgery would also be worthwhile.
S1 File POLARBEAR interview guide. (PDF) S2 File Exploratory factor analysis. (PDF) S3 File Additional results figures and tables. (PDF) S4 File Participant questionnaire. (PDF)
|
The effect of music on patient anxiety undergoing bronchoscopy: a randomized controlled trial | 007b4e54-ef14-44fe-a9b2-2717d04bed5f | 11951599 | Surgery[mh] | Bronchoscopy is a fundamental examination for both diagnostic and therapeutic purposes in pneumology. The indications of this unavoidable procedure are still dominated by tumoral pathology .Although considered as a safe method with rare major life-threatening complications, bronchoscopy is often associated with anxiety, stress and discomfort .In fact, these negative emotions are quite common among patients undergoing bronchoscopy as well as other invasive procedures This technique has a variable but considerable effect on the degree of anxiety, which can lead to a more difficult, incomplete procedure, or even to a total refusal of the method by the patient . Adequate and comprehensive information on the procedure, as well as adequate premedication and anesthesia, are necessary but may not be sufficient to reduce the patient's anxiety and discomfort . Indeed, this feeling of stress, fear and discomfort is multi-factorial and depends not only on the education and information provided to the patient but also on the operator's experience and the patient's history . Bronchoscopy itself can also be a source of anxiety for the patient. It is well known that these patients carry a heavy psychological burden due to the symptoms of the underlying disease, the uncertain diagnosis, the fear of the unknown or the unexpected .However, acting on these factors may not be enough. Several pharmacological and non-pharmacological methods can therefore be implemented in order to win this care challenge. Currently, sedation can be used during endoscopic procedures to reduce patient anxiety and discomfort, but with a significant rate of cardiorespiratory side effects and cost. .Several non-pharmacological methods have been suggested for this purpose, such as hypnosis , acupuncture and audio-visual distraction. .Music therapy is a form of therapy that uses music as a therapeutic process. Listening to music appears to be beneficial in reducing anxiety in patients undergoing pleural procedures . it has been suggested as a safe and accessible non-pharmacological means to reduce stress and anxiety in patients, across a number of endoscopic procedures . Its effect has been studied mainly in digestive endoscopy . Very few studies have investigated the effect of music during Bronchoscopy, with conflicting findings. A study conducted in Denmark showed that music during Bronchoscopy reduces anxiety in patients with suspected lung cancer, especially when the music is self-selected . . In contrast, Colt & al concluded that there was no significant effect of music on anxiety during bronchoscopy .Results from RCTs can provide healthcare professionals with evidence-based information to make informed decisions about incorporating music interventions as a non-pharmacological intervention into clinical practice such as bronchoscopy. This can lead to improved patient comfort and satisfaction. The aim of this study was to evaluate the effect of music on anxiety levels among patients undergoing bronchoscopy in the pneumology departments at the university hospital of Kairouan and Sousse in 2023.
Study design This study was a prospective, randomized, investigator-blinded, controlled trial performed from April 2, 2023 to October 2, 2023 at the two pneumology department of Farhat Hached University Hospital and Ibn El Jazzar University hospital in the centre of Tunisia. Participants All inpatients and outpatients at the two study centres, scheduled for bronchoscopy under local sedation, were eligible. Were included, Patients aged ≥ 18years who could understand and give consent. Patients with hearing or memory impairment, consciousness disorders, undergoing psychiatric treatment for anxiety disorders and/or taking anxiolytic or sedative medication were excluded from the study. Participants with a history of major depressive disorder, bipolar disorder, generalized anxiety disorder, panic disorder, obsessive–compulsive disorder, post-traumatic stress disorder, or schizophrenia were excluded. These conditions were excluded to minimize potential confounding factors that could influence anxiety levels and responses to the intervention. Patients were also excluded if bronchoscopy had been interrupted for any reason. Verbal and written informed consents about the bronchoscopy procedure and the research study were obtained from all participants prior to enrolment in the study. All participants had the possibility to ask the investigators about the study protocol. Randomization was carried out using a computer-generated randomization list to have the same number of participants in the two groups. To reduces the likelihood of bias in the assignment of participants to groups, a collaborator not involved in the study allocated the patients into two groups: Experimental for the patients undergoing bronchoscopy with music and control group for patients undergoing bronchoscopy without music. He maintained a confidential register of assignments. Intervention Before entering the endoscopy unit, the investigator completed the first part of the pre-established questionnaire including the characteristics of consenting participants. Once installed in the endoscopy room, the patient receives two Bluetooth earphones according to the randomization process. For the control group, the two headsets played no music. For the experimental group, the headsets played a playlist of preselected traditional Tunisian music. For this group, to ensure blinding to the investigator and the bronchoscopist, a collaborator not participating in the study will adjust the volume so that the music does not interfere with patient communication, before the entrance of the investigator and the bronchoscopist. All patients were told not to disclose the allocation. The earphones were placed on both ears approximately 10 min before the bronchoscopy and were not removed until 10 min after the end of the bronchoscopy. the conditions for performing bronchoscopy were similar in the two study centres. Bronchoscopy was performed using a flexible bronchoscope (Olympus BF-XP290) inserted through the nose. The procedure included performing various diagnostic sampling techniques, such as cytological analysis from bronchial aspirate or bronchoalveolar lavage fluid, as well as bronchial and transbronchial biopsies. All patients received local anaesthesia by gargling 5 ml of 2% lidocaine. Lidocaine gel 2% was also used to lubricate the bronchoscope. The total dose of lidocaine during bronchoscopy was limited to 0.8 mg/kg. .The choice of music was made with a music therapist who selected a 30-min playlist of 6 Tunisian songs that are known for their frequent use of traditional instruments such as the oud, violin, and percussion, which add an authentic touch to the music, reinforcing the patients' cultural link with music. In addition to the instrument used, the choice was made according to the rhythm, the crescendo then decrescendo, and tempo. The Tunisian classical songs selected for this study were carefully chosen for their rhythm, starting with a moderate tempo, then progressing to a musical crescendo, followed by a gradual decrescendo. This progression can be seen as a resonant representation of the patients' emotional experience. The initial rhythm can help to establish a sense of stability, the crescendo can elicit positive emotions, and the decrescendo can provide a sense of gradual relaxation. Outcomes The primary outcome was changes in anxiety level as measured by the STAI-S before and after the bronchoscopy(ΔSTAI). The secondary outcomes were changes in comfort level as measured by VAS, the blood pressure, heart rate and the oxygen saturation (Sat O2), before and after bronchoscopy. The duration of the procedure and side effects (desaturation, tachycardia) during the procedure were also recorded. Anxiety was measured using the State trait anxiety inventory,state subscale (STAI-S) in his Arabic version .This scale includes 20 questions with scores ranging from 20 to 80. The highest score corresponds to the highest degree of anxiety. .Comfort was assessed using a visual analogic scale (VAS), which provides a subjective evaluation of patient satisfaction. It consists of a 10 cm horizontal or vertical line whose two extremities represent the minimum and maximum comfort scores (0: very uncomfortable/not at all comfortable), 10: no discomfort (very comfortable). . Data collection tool The data was collected using a pre-established questionnaire developed based on the literature. This questionnaire consists of 4 parts: the first part is related to general characteristics of the patients (age, gender, education level comorbid disease, and symptoms), the second part is related to vital parameters (HR, BP, and Sat 02) before and after bronchoscopy. The third part is related to the bronchoscopy procedure (duration and side effects during bronchoscopy). The fourth part is related to the evaluation of anxiety and comfort using the STAI-S and a VAS. Statistical analysis All statistical analyses were performed using SPSS (Version 21, SPSS Inc., Chicago, USA). Kolmogorov–Smirnov test was used to determine if the data were normally distributed. The central tendencies of the variables studied were summarized by means ± Standard Deviations (SD) and medians ± Inter Quartile Intervals (IIQ). Chi square test was used to compare categorical variables between the two groups. Student t test or Mann–Whitney U test were used to compare continuous variables as appropriate. Statistical significance was determined at the 0.05 level and confidence intervals (CIs) set at 95%. Sample size determination The sample size was set at 136, estimating a 20% dropout and using Statistical significance at the 0.05 level, a confidence interval (CIs) at 95% probability and a power of 80%. The standard deviation was set at 10.4 for STAI State, and a count of 5 was interpreted as a clinically relevant difference . Ethical considerations The study protocol was approved by the ethics committee of the faculty of medicine of Sousse (Tunisia) and registered in the pan African Clinical Trial Registry, on 3 September 2023, with the trial number PACTR202309620440045.
This study was a prospective, randomized, investigator-blinded, controlled trial performed from April 2, 2023 to October 2, 2023 at the two pneumology department of Farhat Hached University Hospital and Ibn El Jazzar University hospital in the centre of Tunisia.
All inpatients and outpatients at the two study centres, scheduled for bronchoscopy under local sedation, were eligible. Were included, Patients aged ≥ 18years who could understand and give consent. Patients with hearing or memory impairment, consciousness disorders, undergoing psychiatric treatment for anxiety disorders and/or taking anxiolytic or sedative medication were excluded from the study. Participants with a history of major depressive disorder, bipolar disorder, generalized anxiety disorder, panic disorder, obsessive–compulsive disorder, post-traumatic stress disorder, or schizophrenia were excluded. These conditions were excluded to minimize potential confounding factors that could influence anxiety levels and responses to the intervention. Patients were also excluded if bronchoscopy had been interrupted for any reason. Verbal and written informed consents about the bronchoscopy procedure and the research study were obtained from all participants prior to enrolment in the study. All participants had the possibility to ask the investigators about the study protocol. Randomization was carried out using a computer-generated randomization list to have the same number of participants in the two groups. To reduces the likelihood of bias in the assignment of participants to groups, a collaborator not involved in the study allocated the patients into two groups: Experimental for the patients undergoing bronchoscopy with music and control group for patients undergoing bronchoscopy without music. He maintained a confidential register of assignments.
Before entering the endoscopy unit, the investigator completed the first part of the pre-established questionnaire including the characteristics of consenting participants. Once installed in the endoscopy room, the patient receives two Bluetooth earphones according to the randomization process. For the control group, the two headsets played no music. For the experimental group, the headsets played a playlist of preselected traditional Tunisian music. For this group, to ensure blinding to the investigator and the bronchoscopist, a collaborator not participating in the study will adjust the volume so that the music does not interfere with patient communication, before the entrance of the investigator and the bronchoscopist. All patients were told not to disclose the allocation. The earphones were placed on both ears approximately 10 min before the bronchoscopy and were not removed until 10 min after the end of the bronchoscopy. the conditions for performing bronchoscopy were similar in the two study centres. Bronchoscopy was performed using a flexible bronchoscope (Olympus BF-XP290) inserted through the nose. The procedure included performing various diagnostic sampling techniques, such as cytological analysis from bronchial aspirate or bronchoalveolar lavage fluid, as well as bronchial and transbronchial biopsies. All patients received local anaesthesia by gargling 5 ml of 2% lidocaine. Lidocaine gel 2% was also used to lubricate the bronchoscope. The total dose of lidocaine during bronchoscopy was limited to 0.8 mg/kg. .The choice of music was made with a music therapist who selected a 30-min playlist of 6 Tunisian songs that are known for their frequent use of traditional instruments such as the oud, violin, and percussion, which add an authentic touch to the music, reinforcing the patients' cultural link with music. In addition to the instrument used, the choice was made according to the rhythm, the crescendo then decrescendo, and tempo. The Tunisian classical songs selected for this study were carefully chosen for their rhythm, starting with a moderate tempo, then progressing to a musical crescendo, followed by a gradual decrescendo. This progression can be seen as a resonant representation of the patients' emotional experience. The initial rhythm can help to establish a sense of stability, the crescendo can elicit positive emotions, and the decrescendo can provide a sense of gradual relaxation.
The primary outcome was changes in anxiety level as measured by the STAI-S before and after the bronchoscopy(ΔSTAI). The secondary outcomes were changes in comfort level as measured by VAS, the blood pressure, heart rate and the oxygen saturation (Sat O2), before and after bronchoscopy. The duration of the procedure and side effects (desaturation, tachycardia) during the procedure were also recorded. Anxiety was measured using the State trait anxiety inventory,state subscale (STAI-S) in his Arabic version .This scale includes 20 questions with scores ranging from 20 to 80. The highest score corresponds to the highest degree of anxiety. .Comfort was assessed using a visual analogic scale (VAS), which provides a subjective evaluation of patient satisfaction. It consists of a 10 cm horizontal or vertical line whose two extremities represent the minimum and maximum comfort scores (0: very uncomfortable/not at all comfortable), 10: no discomfort (very comfortable). .
The data was collected using a pre-established questionnaire developed based on the literature. This questionnaire consists of 4 parts: the first part is related to general characteristics of the patients (age, gender, education level comorbid disease, and symptoms), the second part is related to vital parameters (HR, BP, and Sat 02) before and after bronchoscopy. The third part is related to the bronchoscopy procedure (duration and side effects during bronchoscopy). The fourth part is related to the evaluation of anxiety and comfort using the STAI-S and a VAS.
All statistical analyses were performed using SPSS (Version 21, SPSS Inc., Chicago, USA). Kolmogorov–Smirnov test was used to determine if the data were normally distributed. The central tendencies of the variables studied were summarized by means ± Standard Deviations (SD) and medians ± Inter Quartile Intervals (IIQ). Chi square test was used to compare categorical variables between the two groups. Student t test or Mann–Whitney U test were used to compare continuous variables as appropriate. Statistical significance was determined at the 0.05 level and confidence intervals (CIs) set at 95%.
The sample size was set at 136, estimating a 20% dropout and using Statistical significance at the 0.05 level, a confidence interval (CIs) at 95% probability and a power of 80%. The standard deviation was set at 10.4 for STAI State, and a count of 5 was interpreted as a clinically relevant difference .
The study protocol was approved by the ethics committee of the faculty of medicine of Sousse (Tunisia) and registered in the pan African Clinical Trial Registry, on 3 September 2023, with the trial number PACTR202309620440045.
From April to October 2023, 146 participants were assessed for eligibility. Three patients refused to participate, and 2 patients were aged < 18 years. Participants with psychiatric disease ( n = 2) and hearing impairment ( n = 3) were excluded. The remaining 136 participants were randomly assigned to the control ( n = 68) and experimental group( n = 68) (Fig. ). The baseline characteristics of the study participants (Table ) There were no statistically significant differences in age, gender and comorbidities between the control and experimental group. More than half of participants had low level of education (illiterate or primary school level) with no difference between the two groups ( p = 0.24). The most common respiratory symptoms were cough, chest pain and dyspnea. The main indication for bronchoscopy in both groups was suspicion of lung cancer, which was found in 58.5% of cases. Comparison of procedure related variables between experimental and control group (Table ) Pre and post bronchoscopy vital signs as blood pressure, heart rate and oxygen saturation, were similar in the two groups. No statistically significant difference was found between the two groups in terms of side effects, including tachycardia and desaturation, during bronchoscopy. The mean duration of bronchoscopy was similar in the control and experimental group. Comparison of variation of vital signs between experimental and control group (Fig. ) There was no difference in the ΔSBP and ΔDBP, measured as difference between constants after and before bronchoscopy, between the control and experimental group. The music didn’t have a significant effect on the mean change of Sat02 and HR. Anxiety and comfort evaluation There were no significant differences between the two groups on the absolute values of STAI before bronchoscopy( p = 0.97), nor after the procedure ( p = 0.34) (Table ). Moreover, the music didn’t have a significant effect when analyzing the ΔSTAI ( p = 0.41) (Fig. ). When analyzing the VAS before and after bronchoscopy, there was no difference between the two groups. (Table ).Interestingly, there was a significant effect of music on change of VAS score ( p = 0.018) (Fig. ). On average, music improved the VAS score by 0.85 points compared with the control group.
) There were no statistically significant differences in age, gender and comorbidities between the control and experimental group. More than half of participants had low level of education (illiterate or primary school level) with no difference between the two groups ( p = 0.24). The most common respiratory symptoms were cough, chest pain and dyspnea. The main indication for bronchoscopy in both groups was suspicion of lung cancer, which was found in 58.5% of cases.
) Pre and post bronchoscopy vital signs as blood pressure, heart rate and oxygen saturation, were similar in the two groups. No statistically significant difference was found between the two groups in terms of side effects, including tachycardia and desaturation, during bronchoscopy. The mean duration of bronchoscopy was similar in the control and experimental group.
) There was no difference in the ΔSBP and ΔDBP, measured as difference between constants after and before bronchoscopy, between the control and experimental group. The music didn’t have a significant effect on the mean change of Sat02 and HR.
There were no significant differences between the two groups on the absolute values of STAI before bronchoscopy( p = 0.97), nor after the procedure ( p = 0.34) (Table ). Moreover, the music didn’t have a significant effect when analyzing the ΔSTAI ( p = 0.41) (Fig. ). When analyzing the VAS before and after bronchoscopy, there was no difference between the two groups. (Table ).Interestingly, there was a significant effect of music on change of VAS score ( p = 0.018) (Fig. ). On average, music improved the VAS score by 0.85 points compared with the control group.
This study that aimed to evaluate the effect of music on anxiety during bronchoscopy, showed a decrease of anxiety level using the VAS, in patient listening to music, when compared to the control group. This result was not found when using the S-STAI (Fig. ). The effect of music on anxiety has been the subject of several studies in various fields including psychiatry , oncology , cardiology and surgery and procedures such as endoscopy . In fact, the use of music during surgery under local anesthetic has been shown to reduce patient’s stress levels by masking unpleasant sounds. . In a previous study, it has been demonstrated that the use of music can reduce pre-operative anxiety during short waiting periods. . Bronchoscopy is a procedure that frequently elicits feelings of anxiety and stress among patients , potentially resulting in diminished satisfaction and compliance. Previous studies using music has failed to reduce anxiety levels among patients undergoing bronchoscopy . In the other hand a previous study reported a positive effect of music on STAI level . A systematic review showed that music during bronchoscopy reduced patients' blood pressure and heart rate, but no conclusive results were obtained on the effect of music on anxiety levels . Several different scales, therefore, were used to evaluate the anxiety outcome. In our study we opted to assess anxiety using two types of indicators: psychological, subjectively measured using the STAI and VAS, and physiological, objectively measured using the HR, SatO2 and BP. The STAI is a very widely known and reliable self-report scale that has been used extensively to assess anxiety. . the State portion of this questionnaire including 20 questions may be more complex for our participants who mainly have a low education level when compared to the VAS. The discrepancy between the results obtained from the STAI and the VAS may partly be explained by their differing levels of complexity and the cognitive effort required to respond accurately. The VAS, being a simpler and more intuitive tool, might better reflect participants' immediate emotional states, particularly for individuals with lower literacy levels. Conversely, the STAI's more detailed structure could lead to challenges in comprehension, potentially influencing the accuracy of responses. Additionally, in some situations, individuals may underestimate their anxiety levels when responding to self-report tools like the STAI. This may be due to self-protective strategies, where participants adopt defensive attitudes that prevent them from fully recognizing or admitting their emotional state, particularly fear. For some patients, expressing discomfort through the VAS might feel more acceptable than explicitly identifying fear through the STAI . These potential limitations underscore the value of using multiple measures to assess anxiety, as they provide complementary insights into the participants' experiences. While the STAI offers a robust framework for anxiety evaluation, the VAS may offer a practical and culturally sensitive alternative for populations with limited literacy levels. Further exploration of these tools' limitations and strengths within this context could help refine future studies in this area. To overcome this anxiety during bronchoscopy, different types of music have been studied like binaural beat audio ,piano improvisation , soft music or even self-selected music .Music therapy uses various musical elements, including melody, timbre, rhythm, harmony and pitch, to enhance physical and psychological well-being . In fact, music acts as a distractor, diverting the patient's attention from negative stimuli to something pleasant and encouraging, thereby improving anxiety. . The underlying theory behind the use of music as an anti-anxiety intervention method is its ability to trigger relaxation by stimulating the autonomic nervous system. . it has been shown that in addition to the intrinsic perception of music, the feeling of well-being is linked to symbolic, iconic and behavioral meanings. . For these reasons, we have opted in our study, for Tunisian classical music, which is characterized by its authentic dimension, manifested in its respect for formal structure, poetic texts, rhythms, melodic modes and Tunisian musical intonation, resulting from technical elements characteristic of the traditional musical language of each region. This heritage reflects an aspect of Tunisian culture rooted in the depths of history. This music is part of the "oriental" music movement and is a synthesis of the Tunisian own cultural heritage and external contributions, mainly from the Andalusian and Oriental traditions. In other words, Tunisian music is music that lives, it's full of feeling and Its music that supports people in their daily lives. Strength and limitations To the best of our knowledge, it’s the first study that involves music therapist in the choice of the music, thus meeting the exact definition of music therapy, as distinct from music medicine, used in the majority of previous studies. Furthermore, double blinding to the investigator and the bronchoscopist is one of the strong points of our study. Moreover, anxiety was assessed using two different subjective scales (STAI and VAS) and objective measurement like oxygen saturation and blood pressure. However, some limitations of the study should be mentioned. The time given to the patient to listen to the music (10 min) before bronchoscopy maybe insufficient to relax the patient. One notable limitation is the duration of the music intervention. This time,may have been insufficient to achieve a meaningful relaxation effect. Longer exposure to music might have allowed patients to engage more deeply with the intervention, potentially enhancing its anxiolytic benefits. Future studies should consider evaluating the optimal duration of music exposure needed to maximize its calming effects in this clinical context. Although the STAI is still the most widely used for assessing anxiety, it may not be suitable for our population, as this questionnaire can be confusing for illiterate patients, who make up the majority of our population. To mitigate this limitation, we have included the use of VAS. By using both the STAI and the VAS, we aimed to provide a more comprehensive assessment of anxiety and to account for potential limitations associated with either measure alone. However, it is important to acknowledge that the VAS also has limitations. While it is less dependent on literacy, it may be susceptible to anchoring effects and may not capture the nuances of anxiety as effectively as a more structured self-report measure. In future research, exploring additional methods for assessing anxiety in populations with low literacy levels, such as behavioral observations or interviews, could provide valuable insights into the most appropriate assessment tools for diverse populations. The use of pre-selected music can reduce the expected positive effect of music. Another important consideration is the cultural specificity of the music used. While traditional Tunisian music was chosen to align with the cultural context of the participants, individual preferences for music genre and style were not assessed or incorporated into the intervention. This may have impacted the degree to which participants connected with the music, potentially influencing the intervention's effectiveness. Future studies should consider a more personalized approach to music selection to better cater to individual preferences while maintaining cultural relevance. Broader cultural factors may have influenced participants' responses to both the music intervention and the self-report tools. For example, cultural norms surrounding emotional expression or familiarity with music as a therapeutic tool might have affected the participants’ receptiveness and reported outcomes. These aspects merit further exploration in future research to better understand their influence on study results. In addition, we didn’t use subjective direct measurement to assess anxiety like blood cortisol. While the absence of sedation in our study allowed us to isolate the effects of music on anxiety, it is important to acknowledge that this may limit the generalizability of our findings to clinical settings where sedation is routinely used. Future studies should investigate the effects of music on anxiety in patients undergoing sedated bronchoscopy procedures. Implication for practice and research Future studies can address these limitations. The positive effect of the music can be enhanced by giving the choice to the patients to select the preferred music and increasing the amount of time the music is played. In addition, the use of subjective measurement like blood cortisol or salivary indices may be an interesting prospect for future research.
To the best of our knowledge, it’s the first study that involves music therapist in the choice of the music, thus meeting the exact definition of music therapy, as distinct from music medicine, used in the majority of previous studies. Furthermore, double blinding to the investigator and the bronchoscopist is one of the strong points of our study. Moreover, anxiety was assessed using two different subjective scales (STAI and VAS) and objective measurement like oxygen saturation and blood pressure. However, some limitations of the study should be mentioned. The time given to the patient to listen to the music (10 min) before bronchoscopy maybe insufficient to relax the patient. One notable limitation is the duration of the music intervention. This time,may have been insufficient to achieve a meaningful relaxation effect. Longer exposure to music might have allowed patients to engage more deeply with the intervention, potentially enhancing its anxiolytic benefits. Future studies should consider evaluating the optimal duration of music exposure needed to maximize its calming effects in this clinical context. Although the STAI is still the most widely used for assessing anxiety, it may not be suitable for our population, as this questionnaire can be confusing for illiterate patients, who make up the majority of our population. To mitigate this limitation, we have included the use of VAS. By using both the STAI and the VAS, we aimed to provide a more comprehensive assessment of anxiety and to account for potential limitations associated with either measure alone. However, it is important to acknowledge that the VAS also has limitations. While it is less dependent on literacy, it may be susceptible to anchoring effects and may not capture the nuances of anxiety as effectively as a more structured self-report measure. In future research, exploring additional methods for assessing anxiety in populations with low literacy levels, such as behavioral observations or interviews, could provide valuable insights into the most appropriate assessment tools for diverse populations. The use of pre-selected music can reduce the expected positive effect of music. Another important consideration is the cultural specificity of the music used. While traditional Tunisian music was chosen to align with the cultural context of the participants, individual preferences for music genre and style were not assessed or incorporated into the intervention. This may have impacted the degree to which participants connected with the music, potentially influencing the intervention's effectiveness. Future studies should consider a more personalized approach to music selection to better cater to individual preferences while maintaining cultural relevance. Broader cultural factors may have influenced participants' responses to both the music intervention and the self-report tools. For example, cultural norms surrounding emotional expression or familiarity with music as a therapeutic tool might have affected the participants’ receptiveness and reported outcomes. These aspects merit further exploration in future research to better understand their influence on study results. In addition, we didn’t use subjective direct measurement to assess anxiety like blood cortisol. While the absence of sedation in our study allowed us to isolate the effects of music on anxiety, it is important to acknowledge that this may limit the generalizability of our findings to clinical settings where sedation is routinely used. Future studies should investigate the effects of music on anxiety in patients undergoing sedated bronchoscopy procedures.
Future studies can address these limitations. The positive effect of the music can be enhanced by giving the choice to the patients to select the preferred music and increasing the amount of time the music is played. In addition, the use of subjective measurement like blood cortisol or salivary indices may be an interesting prospect for future research.
This study that aimed to evaluate the effect of music on anxiety during bronchoscopy, showed an improvement of comfort level using the VAS, in patient listening to music, when compared to the control group. Our study supports the findings that music, as a simple, safe and inexpensive complementary methods can in addition to sedatives and analgesics, improve comfort and reduce anxiety of patients undergoing bronchoscopy.
Supplementary Material 1
|
Recommandations pour les consultations d’ophtalmopédiatrie – épidémie Covid-19 : qui faut-il voir ? Qui peut-on reporter ? | 402a6ce2-1446-4cd4-b9d0-7018b0b180f6 | 7251990 | Ophthalmology[mh] | L’auteur déclare ne pas avoir de liens d’intérêts.
|
Delegation ärztlicher Leistungen an rheumatologische Fachassistenten | 1caa3ef2-d3b5-4cc9-af71-51b84716bb1d | 11147826 | Internal Medicine[mh] | Das Vorliegen einer Komorbidität bei Patienten mit einer rheumatoiden Arthritis (RA) ist häufig. Daten der Kerndokumentation des Deutschen Rheuma-Forschungszentrums Berlin (DRFZ) zeigen, dass je nach Altersstufe bei mindestens 80 % der Patienten eine oder mehrere Komorbiditäten auftreten . Neben einer kardiovaskulären oder pulmonalen Komorbidität spielen auch psychische Erkrankungen eine bedeutende Rolle. Die Prävalenz der Depression im Zusammenhang mit einer RA liegt je nach Studie bei 9,5–41,5 % und das Risiko des Auftretens ist in den ersten 5 Jahren am höchsten . Im Rahmen einer Metaanalyse aus 72 Studien konnte eine Prävalenz von 16,1 % ermittelt werden, wobei diese ebenfalls stark variierte und u. a. von dem jeweiligen Messinstrument abhing . Auch bei der Ermittlung der Prävalenz in einem Früharthritiskollektiv konnte gezeigt werden, dass eine Depression mit im Vergleich zu der Normalbevölkerung signifikant höherer Prävalenz auftritt (16,5 % vs. 9,4 %) . Das Auftreten einer Depression ist mit einer erhöhten Krankheitsaktivität assoziiert, korrespondierend nimmt auch die Schwere des subjektiven Krankheitsempfindens zu, wie z. B. die globale Patienteneinschätzung oder die Anzahl druckschmerzhafter Gelenke. Darüber hinaus kommt es zu niedrigeren Remissionsraten, einem schlechteren Therapieansprechen sowie höheren Krankheitskosten . Nicht nur die Kosten der Erkrankung an sich steigen, sondern auch das Risiko von beruflichen Teilhabeeinschränkungen ist erhöht. Kosten im Zusammenhang mit Arbeitsunfähigkeiten sowie eingeschränkte berufliche Teilhabe, wie z. B. eine Erwerbsminderung, machen mit 39–86 % einen wesentlichen Anteil der gesamten Krankheitskosten aus . Das Auftreten einer Depression stellt dabei den besten Prädiktor für eine Arbeitsunfähigkeit dar . Während die Depression schon seit Längerem als Komorbidität bekannt ist, wurde eine Angststörung erst später als zusätzliches Problem erkannt. Wie bei der Depression zeigt sich mit einem berichteten Auftreten von 21–70 % eine hohe Spannbreite in den Angaben der Literatur . Hier scheint insbesondere die Progredienzangst, also die Angst vor einem erneuten Schub oder Fortschreiten der Erkrankung, von hoher Relevanz für die Patienten zu sein. Im Vordergrund stehen die Angst vor Nebenwirkungen bei einer medikamentösen Langzeitbehandlung, Arbeitsunfähigkeit sowie Verlust der Autonomie . Freier et al. 2019 berichten aus der Früharthritiskohorte eine 3fach erhöhte Angstsymptomatik als in der Normalbevölkerung bereits bei einem Erstbesuch (23,3 % vs. 6,8 %) . Ein ausführliches und regelmäßiges Screening auf Komorbidität ist also hoch relevant: 1. aufgrund des Einflusses auf die rheumatische Grunderkrankung und 2. weil sie häufig ihrerseits behandlungsbedürftig ist. Dies wurde 2016 auch von der europäischen Fachgesellschaft European Alliance of Associations for Rheumatology (EULAR) aufgegriffen . Allerdings besteht in der Rheumatologie nach wie vor eine Unterversorgung, die durch Mangel an Rheumatologen, einer regionalen Ungleichheit und einer Fehlallokation ärztlicher Ressourcen entsteht . Häufig ist in der ärztlichen Sprechstunde keine Zeit mehr für das Screening auf Komorbidität vorhanden. Mittlerweile konnten die Vorteile einer teambasierten Versorgungsform mit Delegation ärztlicher Leistungen an geschulte rheumatologische Fachassistenz (RFA) in mehreren Studien auch in Deutschland aufgezeigt werden . Die rechtlichen Rahmenbedingungen sind abgesteckt, und seit 2021 liegt das Musterfortbildungscurriculum der Bundesärztekammer zur Aufstiegsqualifikation „Medizinische Fachangestellte für Rheumatologie vor“ (s. Überblick ). Die Implementierung einer strukturierten Visite durch die RFA bietet das Potenzial, die Versorgung von Patienten mit chronisch entzündlichen Erkrankungen zu verbessern und gleichzeitig der Unterversorgung zu begegnen. Ziel der vorliegenden Arbeit war es, die Auswirkungen einer teambasierten Versorgungsform auf den Verlauf einer Depression und Ängstlichkeit bei Patienten mit einer seropositiven rheumatoiden Arthritis im Krankheitsschub zu untersuchen.
Die Daten wurden im Rahmen der randomisierten, kontrollierten, pragmatischen multizentrischen Studie: „Effektivität der RFA-Sprechstunde (ERFASS)“ zum Vergleich einer teambasierten Versorgung mit der Regelversorgung erhoben . Eingeschlossen wurden volljährige Patienten mit einer vom Arzt gesicherten Diagnose einer Rheumafaktor- und/oder ACPA-positiven rheumatoiden Arthritis (ICD-10 M05.8) im Krankheitsschub zu entweder Therapiebeginn, Therapieumstellung oder bei Therapieeskalation. Ausgeschlossen wurden Patienten, die absehbar für eine 1‑jährige Nachbeobachtungsdauer nicht zur Verfügung standen, mit schweren Begleiterkrankungen nach Beurteilung des behandelnden Arztes, bei mangelnden Deutschkenntnissen und fehlender Einwilligungsfähigkeit. In dem 12-monatigen Beobachtungszeitraum fanden nach der Baselinevisite 5 Folgevisiten statt (Wochen 6, 12, 24, 36 und 52). Die Visiten in Woche 6, 12 und 36 wurden in der Interventionsgruppe zunächst durch die RFA durchgeführt, mit anschließendem kurzen Arztkontakt. Als primärer Endpunkt wurde die Veränderung des „Disease Activity Score 28“ (DAS28) über 12 Monate auf Nicht-Unterlegenheit untersucht . Die sekundären Endpunkte beinhalten unter anderem das Vorliegen einer depressiven und/oder ängstlichen Symptomatik, gemessen mit der „Hospital Anxiety and Depression Scale“ (HADS) , sowie die Patientenzufriedenheit gemessen mit dem „Zufriedenheit in der ambulanten Versorgung – Qualität aus Patientenperspektive“(ZAP)-Fragebogen . Der ZAP-Fragebogen ist ein standardisiertes Instrument zur Messung der prozessbezogenen Patientenzufriedenheit und besteht insgesamt aus 23 Items, die 4 Dimensionen zugeordnet sind: 8 Items Arzt-Patienten-Interaktion, u. a. Verständnis oder Einfühlungsvermögen, 8 Items Information (z. B. zu Ursachen oder Verlauf der Erkrankung), 4 Items Praxisorganisation (z. B. Wartezeit) sowie fachliche Kompetenz mit 3 Items (u. a. Gründlichkeit und Sorgfalt). Darüber hinaus gibt es 3 Globalfragen, welche die „Zufriedenheit mit dem Arzt insgesamt“, das Vertrauen zum Arzt sowie eine Einschätzung der Behandlungsqualität erfragen, die jeweils auf einer Likert-Skala von 1 bis 4 beantwortet werden. In der Studie wurde die originale Formulierung „Arzt/Ärztin“ durch die Begriffe „Behandler/Behandlerin“ ersetzt. Dazu wurden die Autoren des Fragebogens vorab kontaktiert. Der HADS besteht aus 2 Subskalen mit je 7 Items, die jeweils auf einer Likert-Skala von 1 bis 4 beantwortet werden. Bei dem HADS handelt es sich um einen Fragebogen zur Selbstbeurteilung der Ausprägung ängstlicher und depressiver Symptomatik. Der Cut-off für sowohl die Depressions- als auch die Ängstlichkeitsskala des HADS lag bei ≥ 8 (milde/moderate Symptomatik) und der Schwellenwert für Sicherheitswarnungen bei ≥ 11 (abnormale Symptomatik). Die Fragebögen zu den sekundären Endpunkten wurden zu Baseline, Monat 6 und Monat 12 erhoben. Inhalte der Visiten Interventionsgruppe Für die RFA-Visite wurde ein Zeitfenster von 30 min eingeplant. Die Aufgaben umfassten die vorbereitende Anamnese gemäß einer Checkliste, Bestimmung der Krankheitsaktivität (DAS28), Screening auf Komorbidität sowie Informationen zu Medikamenteneinnahme und unerwünschte Medikamentenwirkungen. Ein weiterer Schwerpunkt der Visite lag in der Erfassung der Lebensumstände, psychischer Belastungen, Befragung zur Arbeitsfähigkeit und Ermittlung eines Bedarfs einer Rehabilitation oder anderer unterstützender Maßnahmen. Im Anschluss an das Gespräch mit dem Rheumatologen fand bei Bedarf eine Schulung zum Verständnis und zur Applikation der neuen Therapie statt. Kontrollgruppe Die Patienten in der Kontrollgruppe erhielten weiterhin die Regelversorgung mit Terminen bei dem behandelnden Rheumatologen alle 3 Monate für die je etwa 15–20 min eingeplant wurden. Zusätzlich fand ein kurzer Treat-to-Target(T2T)-Besuch statt. In beiden Gruppen konnten die Patienten bei Problemen zusätzliche Termine bekommen. Fallzahlkalkulation Basierend auf dem Wilcoxon-Vorzeichen-Rang-Test, wurde die Fallzahlkalkulation für die sekundären Endpunkte durchgeführt. Die Fallzahlkalkulation mit G*Power 3 hat bei einer Drop-out-Rate von 10 %, einem α von 0,025, einer Power von 95 sowie einer Effektgröße von d = 0,4 ergeben, dass mindestens 74 Patienten berücksichtigt werden müssen. Statistische Analysen Statistische Analysen wurden mit der Software IBM SPSS Statistics V.25 (IBM, New York) durchgeführt. Der Shapiro-Wilk-Test wurde verwendet, um auf Normalverteilung zu prüfen. Der t‑Test für abhängige Stichproben bzw. der Wilcoxon-Vorzeichen-Rang-Test und der t‑Test für unabhängige Stichproben bzw. der Mann-Whitney-U-Test wurden angewandt, um die Veränderung über die Zeit sowie die Nicht-Unterlegenheit bei einem Grenzwert von 0,4 zu untersuchen. Diese Effektgröße wird als minimaler klinisch relevanter Unterschied bei typischen „patient reported outcomes“ angesehen . Um anteilige Veränderungen zu betrachten, kam der McNemar-Test zur Anwendung.
Interventionsgruppe Für die RFA-Visite wurde ein Zeitfenster von 30 min eingeplant. Die Aufgaben umfassten die vorbereitende Anamnese gemäß einer Checkliste, Bestimmung der Krankheitsaktivität (DAS28), Screening auf Komorbidität sowie Informationen zu Medikamenteneinnahme und unerwünschte Medikamentenwirkungen. Ein weiterer Schwerpunkt der Visite lag in der Erfassung der Lebensumstände, psychischer Belastungen, Befragung zur Arbeitsfähigkeit und Ermittlung eines Bedarfs einer Rehabilitation oder anderer unterstützender Maßnahmen. Im Anschluss an das Gespräch mit dem Rheumatologen fand bei Bedarf eine Schulung zum Verständnis und zur Applikation der neuen Therapie statt. Kontrollgruppe Die Patienten in der Kontrollgruppe erhielten weiterhin die Regelversorgung mit Terminen bei dem behandelnden Rheumatologen alle 3 Monate für die je etwa 15–20 min eingeplant wurden. Zusätzlich fand ein kurzer Treat-to-Target(T2T)-Besuch statt. In beiden Gruppen konnten die Patienten bei Problemen zusätzliche Termine bekommen.
Für die RFA-Visite wurde ein Zeitfenster von 30 min eingeplant. Die Aufgaben umfassten die vorbereitende Anamnese gemäß einer Checkliste, Bestimmung der Krankheitsaktivität (DAS28), Screening auf Komorbidität sowie Informationen zu Medikamenteneinnahme und unerwünschte Medikamentenwirkungen. Ein weiterer Schwerpunkt der Visite lag in der Erfassung der Lebensumstände, psychischer Belastungen, Befragung zur Arbeitsfähigkeit und Ermittlung eines Bedarfs einer Rehabilitation oder anderer unterstützender Maßnahmen. Im Anschluss an das Gespräch mit dem Rheumatologen fand bei Bedarf eine Schulung zum Verständnis und zur Applikation der neuen Therapie statt.
Die Patienten in der Kontrollgruppe erhielten weiterhin die Regelversorgung mit Terminen bei dem behandelnden Rheumatologen alle 3 Monate für die je etwa 15–20 min eingeplant wurden. Zusätzlich fand ein kurzer Treat-to-Target(T2T)-Besuch statt. In beiden Gruppen konnten die Patienten bei Problemen zusätzliche Termine bekommen.
Basierend auf dem Wilcoxon-Vorzeichen-Rang-Test, wurde die Fallzahlkalkulation für die sekundären Endpunkte durchgeführt. Die Fallzahlkalkulation mit G*Power 3 hat bei einer Drop-out-Rate von 10 %, einem α von 0,025, einer Power von 95 sowie einer Effektgröße von d = 0,4 ergeben, dass mindestens 74 Patienten berücksichtigt werden müssen.
Statistische Analysen wurden mit der Software IBM SPSS Statistics V.25 (IBM, New York) durchgeführt. Der Shapiro-Wilk-Test wurde verwendet, um auf Normalverteilung zu prüfen. Der t‑Test für abhängige Stichproben bzw. der Wilcoxon-Vorzeichen-Rang-Test und der t‑Test für unabhängige Stichproben bzw. der Mann-Whitney-U-Test wurden angewandt, um die Veränderung über die Zeit sowie die Nicht-Unterlegenheit bei einem Grenzwert von 0,4 zu untersuchen. Diese Effektgröße wird als minimaler klinisch relevanter Unterschied bei typischen „patient reported outcomes“ angesehen . Um anteilige Veränderungen zu betrachten, kam der McNemar-Test zur Anwendung.
Bei der ERFASS-Studie handelte es sich um ein Subprojekt des vom Innovationsfonds geförderten Projektes Rheuma-VOR (#01NVF16029). Die Studie wurde mit Zustimmung der zuständigen Ethikkommission der Medizinischen Hochschule Hannover durchgeführt (# 3638-2017), im Einklang mit nationalem Recht sowie gemäß der Deklaration von Helsinki von 1975 (in der aktuellen, überarbeiteten Fassung) durchgeführt. Die Rheuma-Liga Niedersachsen e. V. war in alle Schritte der Studie in Bezug auf die Planung, Auswertung und Diskussion der Ergebnisse involviert. Von allen beteiligten Patienten liegt eine Einverständniserklärung vor. Die Studie wurde beim deutschen Register für klinische Studien registriert (DRKS00013055).
Es wurden 224 Patienten auf die Kontroll- und Interventionsgruppe randomisiert (113, 111). Die Drop-out-Rate betrug 8 % (s. Überblick ). Von 101 (KG) und 97 (IG) Patienten konnte der über alle Zeitpunkte vollständige Datensatz der Hospital Anxiety and Depression Scale ausgewertet werden. Die deskriptiven Ergebnisse zu Baseline sind in Tab. aufgeführt. In Bezug auf die Veränderung der Ängstlichkeit der Patienten konnte eine signifikante Verbesserung des Scores in der Interventionsgruppe nachgewiesen werden, in der Kontrollgruppe jedoch nicht. Zu Baseline gaben in der Interventions- sowie Kontrollgruppe 37 % der Patienten moderate bis ausgeprägte ängstliche Symptome an (entsprechend eines HADS Scores ≥ 8). Zu Monat 6 sank dieser Anteil in beiden Gruppen, und zwar auf 29 % (IG) bzw. 30 % (KG). In der IG ging der Anteil nach 12 Monaten weiter zurück auf 24 % ( p < 0,001), in der KG stieg er wieder auf 38 % an ( p = 0,5) Der Unterschied zwischen den Gruppen ist signifikant ( p = 0,035, Tab. ; Abb. a). Der Verlauf der von den Patienten berichteten depressiven Symptomatik hat sich in beiden Gruppen signifikant verbessert ( p = 0,001), zwischen der Interventions- und Kontrollgruppe allerdings nicht unterschieden ( p = 0,866, Tab. ). In der Kontrollgruppe kam es nach einem initialen Abfall von 38 auf 29 % des Anteils der Patienten mit einer depressiven Symptomatik nach Monat 6 wieder zu einem leichten Anstieg (32 %). In der Interventionsgruppe blieb dahingegen der Anteil zunächst in etwa gleich (26 %, 25 %), nach Monat 6 kam es jedoch zu einem deutlichen Abfall des Anteils der Patienten mit einer depressiven Symptomatik (18 %). Diese Veränderungen sind weder in der gesamten Studienpopulation ( p = 0,092) noch in der Kontrollgruppe ( p = 0,324) oder der Interventionsgruppe ( p = 0,090) signifikant (Abb. b). Die Anteile der Patienten mit einer milden/moderaten Angstsymptomatik bleiben in beiden Gruppen über den Zeitverlauf nahezu unverändert, die Veränderung findet vornehmlich in der Gruppe der Patienten statt, die eine abnormale Angstsymptomatik berichten (Abb. a). Die depressive Symptomatik verändert sich dahingegen in allen 3 Ausprägungen (Abb. b). Hinsichtlich der Veränderung der 3 Globalfragen des ZAP über 12 Monate bezüglich des Vertrauens zu den Behandelnden ( p = 0,775), der Qualität der Behandlung im Allgemeinen ( p = 0,283) sowie der Zufriedenheit mit dem zuletzt besuchten Behandelnden im Allgemeinen ( p = 0,690) wurden keine signifikanten Unterschiede zwischen den Gruppen festgestellt (Tab. ). Bei der Zufriedenheit mit den Behandelnden in Bezug auf die Dimension der Information haben sich die Patienten in der IG signifikant besser informiert gefühlt als in der KG ( p = 0,03, Tab. ).
Im Rahmen der ERFASS-Studie konnte insgesamt ein positiver Effekt einer teambasierten Versorgungsform auf den Verlauf einer patientenberichteten Symptomatik einer Depression und Ängstlichkeit bei Patienten mit einer Rheumafaktor- und/oder ACPA-positiven rheumatoiden Arthritis im Krankheitsschub nachgewiesen werden. Informationen und Wissen können helfen, Angst zu reduzieren, und Selbstregulierung ist dabei ein wichtiger Aspekt im Umgang mit chronischen Erkrankungen . Diese Selbstregulierung kann beigebracht und gelernt werden , und hier liegt ein wichtiger Ansatzpunkt für eine RFA-Visite. Durch die Visite bei der RFA können die Patienten viel über ihre Erkrankung und den Umgang damit lernen. Hilfestellung bei der Applikation der medikamentösen Therapie ist ein wichtiger Punkt, aber auch Informationen dazu, wie eine richtige Ernährung und Bewegung dabei helfen können, die Krankheit zu bewältigen. Dieses erhöhte Wissen und das Erlernen von Fähigkeiten können ihnen eine gewisse Sicherheit geben und dadurch positive Auswirkungen auf die Bewältigung der Angst haben . Dies könnte erklären, warum lediglich die Angst und nicht die Depression signifikant besser geworden ist. Darüber hinaus konnte anhand inflammatorischer Wirkmechanismen gezeigt werden, dass ein signifikanter Zusammenhang zwischen RA und Depressionen besteht . Bei Angststörungen konnten bisher keine direkten Zusammenhänge nachgewiesen werden, jedoch indirekte Zusammenhänge, die zeigen konnten, dass der Einfluss von psychologischem Stress auf Angststörungen durch inflammatorische Zytokine mediiert werden kann . Der psychologische Stress kann durch die verbesserte Selbstwirksamkeit der Patienten gelindert werden, worüber im Verlauf auch die Angststörung reduziert werden kann. Die Behandlung der immunologischen Auslöser der Erkrankung kann auch einen positiven Einfluss auf die Symptome der Depression haben . Da sich die Krankheitsaktivität in beiden Gruppen jedoch nicht signifikant über die Zeit unterscheidet, ist es auch naheliegend, dass sich die depressive Symptomatik ähnlich verändert. In einer von Meisters et al. veröffentlichten Studie wurden Versorgungslücken aus Patienten- und Rheumatologensicht nach EULAR-Kriterien in 35 europäischen Ländern untersucht. Wichtige von den Patienten erwähnte Versorgungslücken sind unter anderem: Erhalt von ausreichenden Informationen bezüglich der Erkrankung und verschiedener Behandlungsmöglichkeiten sowie ein Eingehen auf die Patientenbedürfnissen. Im Rahmen der ERFASS-Studie konnte gezeigt werden, dass sich die Patienten in der teambasierten Versorgungsform signifikant besser informiert gefühlt haben (Abb. ). Im Gegensatz zu der RFA-Visite bietet die reguläre Sprechstunde für eine ausgedehnte Schulung der Patienten keine Zeit, wodurch erklärt werden könnte, warum sich in der Kontrollgruppe die Angst nicht signifikant verbessert. Als zusätzliche Unterstützung und Vertiefung der Interaktion könnte der Einsatz von Digitalen Gesundheitsanwendungen (DiGAs) bei Angst und Depression überlegt werden. Patienten haben so die Möglichkeit, nach Einführung durch die RFA eigenständig zu Hause die Anwendung durchzuführen und bei Bedarf regelmäßig zu besprechen. Auffällig ist, dass die Veränderung zwischen den beiden Gruppen zwischen Monat 6 und Monat 12 geschieht. Eine Erklärung hierfür könnte sein, dass die Patienten bislang in den meisten Fällen noch keine Erfahrung in der zusätzlichen Versorgung durch eine RFA hatten und dem zunächst skeptisch gegenüberstanden. Der Mehrwert der teambasierten Versorgung wurde dann erst im Verlauf erkannt. Unsere Ergebnisse entsprechen internationalen Studien, die ebenfalls zeigen konnten, dass Visiten bei RFAs zu signifikanten Verbesserungen des psychischen Wohlergehens führen können . Die Studie ist mit einigen Limitationen verbunden. Es wurden nur Patienten mit einer seropositiven RA im Krankheitsschub in die Studie eingeschlossen. Eine Übertragbarkeit auf andere Indikationen ist nicht uneingeschränkt möglich. Eine weitere Limitation ist, dass die Randomisierung der Patienten innerhalb der jeweiligen Zentren stattfand, und nicht als einzelne Cluster. Somit kann ein Halo-Effekt nicht grundsätzlich ausgeschlossen werden. Darüber hinaus fand entsprechend den rechtlichen Rahmenbedingungen immer ein Arztkontakt statt. Gemäß Studienprotokoll orientierte sich die Länge des Kontaktes u. a. an den Bedürfnissen des Patienten. In der teambasierten Versorgung steht der Arzt selbstverständlich ebenfalls als Ansprechpartner den Patienten zur Verfügung. Es konnte jedoch eine deutliche Zeitersparnis aufseiten der Ärzte nachgewiesen werden , sodass umfassende Information aufseiten der Rheumatologen eher unwahrscheinlich ist.
Die Einbeziehung einer 30-minütigen RFA-Visite bei ambulanten Wiedervorstellungen von Patienten mit einer RA hat zu einer Reduktion ängstlicher Symptome und zu einem Informationsgewinn bei den Betroffenen geführt. Neben der Zeitersparnis für die ärztlichen Tätigkeiten unterstreicht diese Verbesserung in der Ergebnisqualität die Versorgungsrelevanz der Delegation ärztlicher Leistungen in der Rheumatologie durch RFA.
|
Pharmacogenomic Profiling of Cisplatin-Resistant and -Sensitive Human Osteosarcoma Cell Lines by Multimodal Targeted Next Generation Sequencing | cc0bc570-c108-4512-abd1-ab6393ade396 | 9570120 | Pharmacology[mh] | High-grade osteosarcoma (HGOS), the most common malignant tumor of bone, is treated by surgery and systemic neo-adjuvant multidrug chemotherapy . Cisplatin (CDDP), together with high-dose methotrexate and doxorubicin, is invariably included in standard chemotherapy for this tumor . Pharmacogenetic studies have revealed several single nucleotide polymorphisms (SNPs) of genes belonging to either DNA repair, drug transport, folate metabolism, and detoxification pathways to be associated with therapy-related parameters in HGOS, as survival and drug response, or development of drug-associated toxicity . The general goal of these studies was the identification of genomic variations associated with drug response or adverse toxicities, which may provide useful information to improve treatment efficacy and simultaneously reduce the risk of chemotherapy-related toxicities . In the last decade, pharmacogenomic approaches have been increasingly applied to the study of HGOS providing a series of interesting insights related to genetic polymorphisms, which may be causally related to drug resistance or susceptibility to develop treatment-related adverse toxicities . However, all these indications must be further confirmed because the polymorphic gene status was revealed almost only in patients’ normal (germline) cells at the DNA level without providing information on how these changes were maintained at the RNA level, and influenced RNA and protein expression in tumor cells. The aim of this study was to explore the genotype status of 28 SNPs in 14 genes related to processes involved in DNA repair, CDDP transport and detoxification, or involved in CDDP-related toxicity in a panel of 6 CDDP-resistant and 12 drug-sensitive human HGOS cell lines . In particular, we focused our study on both pharmacogenetic (germline) and pharmacogenomic (tumor-associated, somatic) markers, which had been indicated to influence treatment response and susceptibility to CDDP-related ototoxicity in HGOS patients, thus appearing as promising candidates for a translation to clinical practice. This selection was performed by taking into consideration the body of evidence reported so far, which has also been recently reviewed . This analysis was performed by using an innovative multimodal targeted next generation sequencing (mmNGS) approach that allowed for the contemporary study of the selected SNPs on both DNA- and RNA-derived libraries. Data obtained by mmNGS on DNA-derived libraries were validated by TaqMan genotyping. RNA expression level of the 14 genes in CDDP-resistant variants compared to their parental cell lines was also determined. Heatmap analysis was performed, including all CDDP-resistant and drug-sensitive cell lines. 2.1. Validation of Custom Multimodal NGS Panel Data obtained for the 28 SNPs on DNA-derived libraries by the custom mmNGS approach were validated by TaqMan genotyping in 24/28 SNPs. shows for each cell line the genotype status of all 28 SNPs, which were identified to be either heterozygous or homozygous by sequencing compared to the reference sequence . Variants with an allele frequency greater than 3% were considered reliable. Those SNPs that were homozygous wild-type for the reference allele were not reported in . By comparing the data obtained from sequencing and genotyping, we found that for 11/18 (61%) cell lines data obtained by both techniques matched in 100% of the SNPs, whereas in 4/18 (22%) cell lines the match ranged from 90 to 93%, and was below 90% in 3/18 (17%) cell lines. The fact that 39% of the cell lines did not show a complete match could be explained by the presence of different subpopulations within the same cell line. Interestingly, for five SNPs, ABCC2 rs17222723, ACYP2 rs1872328, TPMT rs12201199, rs1142345, and rs1800460 the homozygous wild-type genotype was identified in all 18 cell lines. 2.2. DNA SNP Evaluation in Relation to Level of CDDP Resistance 2.2.1. Comparison between U-2OS CDDP-Resistant Variants to Parental U-2OS Cell Line The comparison of polymorphisms identified in the group of U-2OS CDDP-resistant variants in comparison with their parental cells, identified two polymorphisms of the ERCC2 gene, (rs13181 and rs1799793), which exhibited a genotype change in relation to the acquisition of CDDP resistance . In the CDDP-sensitive, parental cell line and in the two variants with the lower level of CDDP resistance (U-2OS/CDDP300 and U-2OS/CDDP1µg), the genotype of ERCC2 rs13181 was heterozygous variant (GT), while in the variant with the highest resistance level (U-2OS/CDDP4µg) the genotype of the polymorphism shifted to homozygous wild-type (TT). shows the graphical representation of the mmNGS data obtained by the DNA variant calling identifier tool of the CLC Genomics Workbench (GWB) analysis. The data obtained for these SNPs by TaqMan genotyping and mmNGS were concordant for all cell lines except for ERCC2 rs13181 in U-2OS/CDDP1µg variant for which mmNGS reported GT but TaqMan genotyping a TT genotype. This apparent discordance may be due to the different sensitivity of the techniques and the presence of subpopulations with TT and GT genotypes. However, these data indicate that in these cells the transition toward a TT genotype is associated with development of CDDP resistance. For ERCC2 rs1799793, the sensitive cell line and the two U-2OS/CDDP300 and U-2OS/CDDP1µg resistant cell lines showed a heterozygous variant genotype CT, which became homozygous (CC) in the variant with the highest resistance level ( and ). As shown in and , both SNPs of ERCC2 were non-synonymous and caused amino acid changes. The SNP rs13181 caused the substitution of Lys by Gln and the rs1799793 the substitution of Asp with Asn. 2.2.2. Comparison between Saos-2 CDDP-Resistant Variants to Parental Saos-2 Cell Line The comparison of DNA variant calling data between Saos-2 CDDP-resistant variants and their parental Saos-2 CDDP-sensitive cell line identified genotype changes of ERCC2 rs13181 and ERCC1 rs11615 . These genotype changes were confirmed by both TaqMan genotyping and mmNGS. For ERCC2 rs13181, the genotype of the detected polymorphism was heterozygous variant GT in the sensitive and the two Saos-2 resistant variants with lower resistance levels, while in the Saos-2/CDDP6µg variant the genotype changed to homozygous variant GG . The same situation occurred for ERCC1 rs11615, which was heterozygous variant GA in the sensitive cell line and the two resistant variants with lower resistance levels, whereas homozygous variant GG in Saos-2/CDDP6µg . Different to the ERCC2 rs13181 variant, which caused an amino acid change from Lys to Gln, no amino acid changes were revealed by the CLC GWB analysis for the synonymous ERCC1 rs11615 variant . 2.3. RNA SNP Evaluation in Relation to Level of CDDP Resistance 2.3.1. Comparison between U-2OS Cell Line and U-2OS CDDP-Resistant Variants All genotype variations identified at the DNA level and described above were also identified on the RNA level, indicating that these changes had been selected and maintained during development of CDDP resistance. Differently, the GSTP1 rs1695 SNP changed in the RNA-derived libraries of U-2OS cell line and U-2OS/CDDP1µg variant compared to the DNA-derived libraries . The genotype of the GSTP1 rs1695 detected on DNA remained AG in the sensitive and in the three resistant cell lines. At the RNA level, the genotype of GSTP1 rs1695 was homozygous wild-type AA in U-2OS and heterozygous variant AG in U-2OS/CDDP300 and U-2OS/CDDP4µg variants, while in the U-2OS/CDDPP1µg variant, a multi nucleotide variant (MNV) GAT, was detected. Interestingly, the amino acid change Ile105Val caused by the GSTP1 rs1695 variant allele was identified by the CLC GWB in all three CDDP-resistant U-2OS variants . 2.3.2. Comparison between Saos-2 Cell Line and Saos-2 CDDP-Resistant Variants All genotype changes identified at the DNA level were also identified on RNA except for the GSTP1 rs1695 SNP. As with U-2OS cell lines, the rs1695 genotype was AG at DNA level whereas homozygous AA at RNA level in the Saos-2 parental cell line . No difference was found at DNA and RNA level for all CDDP-resistant variants . Accordingly, the amino acid change Ile105Val caused by the GSTP1 rs1695 variant allele was identified by the CLC GWB in all three CDDP-resistant Saos-2 variants with the AG genotype in the RNA-derived libraries . 2.4. RNA Expression Analysis Targeted RNAseq was performed for the 14 genes related to either CDDP drug response or toxicity reported after CDDP therapy. The fold-change of transcripts per million (TPM), which estimates the fold-change in RNA expression, for each CDDP-resistant variant compared to its drug-sensitive parental cell line is graphically shown in . In U2OS-derived CDDP-resistant variants, six genes, ABCC2, ABCC3, ACYP2, COMT, ERCC2 , and XRCC3 emerged to be increased more than 2-fold compared to the parental U-2OS cell line, whereas four genes, ATM, ATR, TP53 , and XPA were downregulated in CDDP-resistant variants. Considering all three CDDP-resistant variants together, the differential gene expression tool of the CLC GWB identified the downregulation of ATM, ATR , and TP53 as significant with a Bonferroni corrected p -value < 0.05. In Saos-2-derived CDDP-resistant variants, six genes were increased more than 2-fold: ABCB1, ABCC2 and XRCC3 in all three variants, whereas ACYP2, COMT and ERCC2 only in Saos-2/CDDP300, the variant with the lowest resistance level. CDDP-resistant variants also presented downregulation of ATM, ATR, GSTP1, TPMT, and XPA genes. Evaluating all three CDDP-resistant variants together, a significant difference after Bonferroni correction with a p -value < 0.05 was identified for upregulation of ABCB1 and downregulation of TPMT and XPA . Similarities between CDDP-resistant and CDDP-sensitive cell lines were assessed by using the heatmap tool of the CLC GWB, including all 14 genes . Two main clusters were revealed. One consisted of two clusters formed by all six CDDP-resistant variants clearly separated from their two parental cell lines. The 10 drug-sensitive cell lines formed the second main cluster, which was mostly separated from that of CDDP-resistant variants. As also shown in , the group of 14 genes resulted to be divided in 6 clusters, with genes belonging to the same family mostly grouped together. Data obtained for the 28 SNPs on DNA-derived libraries by the custom mmNGS approach were validated by TaqMan genotyping in 24/28 SNPs. shows for each cell line the genotype status of all 28 SNPs, which were identified to be either heterozygous or homozygous by sequencing compared to the reference sequence . Variants with an allele frequency greater than 3% were considered reliable. Those SNPs that were homozygous wild-type for the reference allele were not reported in . By comparing the data obtained from sequencing and genotyping, we found that for 11/18 (61%) cell lines data obtained by both techniques matched in 100% of the SNPs, whereas in 4/18 (22%) cell lines the match ranged from 90 to 93%, and was below 90% in 3/18 (17%) cell lines. The fact that 39% of the cell lines did not show a complete match could be explained by the presence of different subpopulations within the same cell line. Interestingly, for five SNPs, ABCC2 rs17222723, ACYP2 rs1872328, TPMT rs12201199, rs1142345, and rs1800460 the homozygous wild-type genotype was identified in all 18 cell lines. 2.2.1. Comparison between U-2OS CDDP-Resistant Variants to Parental U-2OS Cell Line The comparison of polymorphisms identified in the group of U-2OS CDDP-resistant variants in comparison with their parental cells, identified two polymorphisms of the ERCC2 gene, (rs13181 and rs1799793), which exhibited a genotype change in relation to the acquisition of CDDP resistance . In the CDDP-sensitive, parental cell line and in the two variants with the lower level of CDDP resistance (U-2OS/CDDP300 and U-2OS/CDDP1µg), the genotype of ERCC2 rs13181 was heterozygous variant (GT), while in the variant with the highest resistance level (U-2OS/CDDP4µg) the genotype of the polymorphism shifted to homozygous wild-type (TT). shows the graphical representation of the mmNGS data obtained by the DNA variant calling identifier tool of the CLC Genomics Workbench (GWB) analysis. The data obtained for these SNPs by TaqMan genotyping and mmNGS were concordant for all cell lines except for ERCC2 rs13181 in U-2OS/CDDP1µg variant for which mmNGS reported GT but TaqMan genotyping a TT genotype. This apparent discordance may be due to the different sensitivity of the techniques and the presence of subpopulations with TT and GT genotypes. However, these data indicate that in these cells the transition toward a TT genotype is associated with development of CDDP resistance. For ERCC2 rs1799793, the sensitive cell line and the two U-2OS/CDDP300 and U-2OS/CDDP1µg resistant cell lines showed a heterozygous variant genotype CT, which became homozygous (CC) in the variant with the highest resistance level ( and ). As shown in and , both SNPs of ERCC2 were non-synonymous and caused amino acid changes. The SNP rs13181 caused the substitution of Lys by Gln and the rs1799793 the substitution of Asp with Asn. 2.2.2. Comparison between Saos-2 CDDP-Resistant Variants to Parental Saos-2 Cell Line The comparison of DNA variant calling data between Saos-2 CDDP-resistant variants and their parental Saos-2 CDDP-sensitive cell line identified genotype changes of ERCC2 rs13181 and ERCC1 rs11615 . These genotype changes were confirmed by both TaqMan genotyping and mmNGS. For ERCC2 rs13181, the genotype of the detected polymorphism was heterozygous variant GT in the sensitive and the two Saos-2 resistant variants with lower resistance levels, while in the Saos-2/CDDP6µg variant the genotype changed to homozygous variant GG . The same situation occurred for ERCC1 rs11615, which was heterozygous variant GA in the sensitive cell line and the two resistant variants with lower resistance levels, whereas homozygous variant GG in Saos-2/CDDP6µg . Different to the ERCC2 rs13181 variant, which caused an amino acid change from Lys to Gln, no amino acid changes were revealed by the CLC GWB analysis for the synonymous ERCC1 rs11615 variant . The comparison of polymorphisms identified in the group of U-2OS CDDP-resistant variants in comparison with their parental cells, identified two polymorphisms of the ERCC2 gene, (rs13181 and rs1799793), which exhibited a genotype change in relation to the acquisition of CDDP resistance . In the CDDP-sensitive, parental cell line and in the two variants with the lower level of CDDP resistance (U-2OS/CDDP300 and U-2OS/CDDP1µg), the genotype of ERCC2 rs13181 was heterozygous variant (GT), while in the variant with the highest resistance level (U-2OS/CDDP4µg) the genotype of the polymorphism shifted to homozygous wild-type (TT). shows the graphical representation of the mmNGS data obtained by the DNA variant calling identifier tool of the CLC Genomics Workbench (GWB) analysis. The data obtained for these SNPs by TaqMan genotyping and mmNGS were concordant for all cell lines except for ERCC2 rs13181 in U-2OS/CDDP1µg variant for which mmNGS reported GT but TaqMan genotyping a TT genotype. This apparent discordance may be due to the different sensitivity of the techniques and the presence of subpopulations with TT and GT genotypes. However, these data indicate that in these cells the transition toward a TT genotype is associated with development of CDDP resistance. For ERCC2 rs1799793, the sensitive cell line and the two U-2OS/CDDP300 and U-2OS/CDDP1µg resistant cell lines showed a heterozygous variant genotype CT, which became homozygous (CC) in the variant with the highest resistance level ( and ). As shown in and , both SNPs of ERCC2 were non-synonymous and caused amino acid changes. The SNP rs13181 caused the substitution of Lys by Gln and the rs1799793 the substitution of Asp with Asn. The comparison of DNA variant calling data between Saos-2 CDDP-resistant variants and their parental Saos-2 CDDP-sensitive cell line identified genotype changes of ERCC2 rs13181 and ERCC1 rs11615 . These genotype changes were confirmed by both TaqMan genotyping and mmNGS. For ERCC2 rs13181, the genotype of the detected polymorphism was heterozygous variant GT in the sensitive and the two Saos-2 resistant variants with lower resistance levels, while in the Saos-2/CDDP6µg variant the genotype changed to homozygous variant GG . The same situation occurred for ERCC1 rs11615, which was heterozygous variant GA in the sensitive cell line and the two resistant variants with lower resistance levels, whereas homozygous variant GG in Saos-2/CDDP6µg . Different to the ERCC2 rs13181 variant, which caused an amino acid change from Lys to Gln, no amino acid changes were revealed by the CLC GWB analysis for the synonymous ERCC1 rs11615 variant . 2.3.1. Comparison between U-2OS Cell Line and U-2OS CDDP-Resistant Variants All genotype variations identified at the DNA level and described above were also identified on the RNA level, indicating that these changes had been selected and maintained during development of CDDP resistance. Differently, the GSTP1 rs1695 SNP changed in the RNA-derived libraries of U-2OS cell line and U-2OS/CDDP1µg variant compared to the DNA-derived libraries . The genotype of the GSTP1 rs1695 detected on DNA remained AG in the sensitive and in the three resistant cell lines. At the RNA level, the genotype of GSTP1 rs1695 was homozygous wild-type AA in U-2OS and heterozygous variant AG in U-2OS/CDDP300 and U-2OS/CDDP4µg variants, while in the U-2OS/CDDPP1µg variant, a multi nucleotide variant (MNV) GAT, was detected. Interestingly, the amino acid change Ile105Val caused by the GSTP1 rs1695 variant allele was identified by the CLC GWB in all three CDDP-resistant U-2OS variants . 2.3.2. Comparison between Saos-2 Cell Line and Saos-2 CDDP-Resistant Variants All genotype changes identified at the DNA level were also identified on RNA except for the GSTP1 rs1695 SNP. As with U-2OS cell lines, the rs1695 genotype was AG at DNA level whereas homozygous AA at RNA level in the Saos-2 parental cell line . No difference was found at DNA and RNA level for all CDDP-resistant variants . Accordingly, the amino acid change Ile105Val caused by the GSTP1 rs1695 variant allele was identified by the CLC GWB in all three CDDP-resistant Saos-2 variants with the AG genotype in the RNA-derived libraries . All genotype variations identified at the DNA level and described above were also identified on the RNA level, indicating that these changes had been selected and maintained during development of CDDP resistance. Differently, the GSTP1 rs1695 SNP changed in the RNA-derived libraries of U-2OS cell line and U-2OS/CDDP1µg variant compared to the DNA-derived libraries . The genotype of the GSTP1 rs1695 detected on DNA remained AG in the sensitive and in the three resistant cell lines. At the RNA level, the genotype of GSTP1 rs1695 was homozygous wild-type AA in U-2OS and heterozygous variant AG in U-2OS/CDDP300 and U-2OS/CDDP4µg variants, while in the U-2OS/CDDPP1µg variant, a multi nucleotide variant (MNV) GAT, was detected. Interestingly, the amino acid change Ile105Val caused by the GSTP1 rs1695 variant allele was identified by the CLC GWB in all three CDDP-resistant U-2OS variants . All genotype changes identified at the DNA level were also identified on RNA except for the GSTP1 rs1695 SNP. As with U-2OS cell lines, the rs1695 genotype was AG at DNA level whereas homozygous AA at RNA level in the Saos-2 parental cell line . No difference was found at DNA and RNA level for all CDDP-resistant variants . Accordingly, the amino acid change Ile105Val caused by the GSTP1 rs1695 variant allele was identified by the CLC GWB in all three CDDP-resistant Saos-2 variants with the AG genotype in the RNA-derived libraries . Targeted RNAseq was performed for the 14 genes related to either CDDP drug response or toxicity reported after CDDP therapy. The fold-change of transcripts per million (TPM), which estimates the fold-change in RNA expression, for each CDDP-resistant variant compared to its drug-sensitive parental cell line is graphically shown in . In U2OS-derived CDDP-resistant variants, six genes, ABCC2, ABCC3, ACYP2, COMT, ERCC2 , and XRCC3 emerged to be increased more than 2-fold compared to the parental U-2OS cell line, whereas four genes, ATM, ATR, TP53 , and XPA were downregulated in CDDP-resistant variants. Considering all three CDDP-resistant variants together, the differential gene expression tool of the CLC GWB identified the downregulation of ATM, ATR , and TP53 as significant with a Bonferroni corrected p -value < 0.05. In Saos-2-derived CDDP-resistant variants, six genes were increased more than 2-fold: ABCB1, ABCC2 and XRCC3 in all three variants, whereas ACYP2, COMT and ERCC2 only in Saos-2/CDDP300, the variant with the lowest resistance level. CDDP-resistant variants also presented downregulation of ATM, ATR, GSTP1, TPMT, and XPA genes. Evaluating all three CDDP-resistant variants together, a significant difference after Bonferroni correction with a p -value < 0.05 was identified for upregulation of ABCB1 and downregulation of TPMT and XPA . Similarities between CDDP-resistant and CDDP-sensitive cell lines were assessed by using the heatmap tool of the CLC GWB, including all 14 genes . Two main clusters were revealed. One consisted of two clusters formed by all six CDDP-resistant variants clearly separated from their two parental cell lines. The 10 drug-sensitive cell lines formed the second main cluster, which was mostly separated from that of CDDP-resistant variants. As also shown in , the group of 14 genes resulted to be divided in 6 clusters, with genes belonging to the same family mostly grouped together. In this study, a custom mmNGS approach has been used to study 28 SNPs of 14 genes, contemporarily on the DNA and RNA level, in 6 CDDP-resistant and 12 CDDP-sensitive human HGOS cell lines. To our knowledge, this innovative approach has not been used so far for pharmacogenomic studies. The successful validation of the DNA variant calling by TaqMan genotyping confirmed that this approach is an appropriate method to study even rare SNPs. Compared to genotyping by single TaqMan assays, the custom mmNGS approach is faster and also offers the possibility to identify additional SNPs mapping to the target region. Moreover, small targeted panels allow the pooling of higher sample numbers compared to whole-genome NGS. Another advantage of the mmNGS approach is the low amount of starting material that is required for library preparation, which facilitates the application of this method to tumor tissue samples. In addition, the simultaneous analysis of SNPs on the DNA and RNA level, as well as the possibility to estimate the level of RNA expression associated with the polymorphic gene status allow for a direct correlation between the genotype status with the biological function of each SNP. The frequencies of variant alleles per cell line ranged from 9 (found in IOR/OS15 and MG-63) to 22 (detected in IOR/SARG), confirming the heterogeneity and high genetic instability of HGOS. Particular genotype distributions were found for 9 SNPs. Five SNPs that had been reported in association with CDDP-related ototoxicity, ACYP2 rs1872328 , ABCC2 rs17222723 , TPMT rs12201199, rs1142345, and rs1800460 , were present only in the wild-type status in all cell lines. Two SNPs, ABCC2 rs717620 and GSTP1 rs1695, were found as homozygous wild-type or heterozygous but not as homozygous variant. These findings suggest that the variant allele of these seven SNPs could be of biological disadvantage in HGOS tumor cells. Differently, the two SNPs of TP53 , rs1042522 and rs1642785, were identified either in a homozygous wild-type or variant but not heterozygous status. The most relevant SNPs that emerged in this study to be associated with the development of CDDP resistance were GSTP1 rs1695, ERCC2 rs13181, ERCC2 rs1799793, and ERCC1 rs11615. The GSTP1 rs1695 was the only SNP for which the genotype changes found in the RNA-derived libraries differed from those revealed in DNA-derived libraries. Interestingly, in all five drug-sensitive cell lines with the heterozygous genotype in the DNA (U-2OS, Saos-2, IOR/10, IOR/14, IOR/18), the genotype status in the RNA was homozygous wild-type. The presence of the variant also in the RNA of all six CDDP-resistant cell lines, with the consequent amino acid change Ile105Val, strongly suggests that the AG genotype is associated with reduced CDDP response. These findings further support the previously demonstrated relevance of GSTP1 enhanced enzymatic activity in these CDDP-resistant HGOS cell lines . The pharmacogenomic findings emerged from the present study thus indicate that the increase in GSTP1 activity observed in CDDP-resistant variants is correlated with the transition to the AG genotype of the rs1695 polymorphism and the consequent Ile105Val amino acid change. This observation is concordant with the data reported in almost all germline studies. A significant association between AG+GG genotypes and poor histological response as well as decreased event-free and overall survival was observed in five studies whereas one study reported the GG genotype to be associated with good response . Interestingly, GSTP1 rs1695 was excluded from further analyses in the study by Goricar and co-workers because the genotype frequencies of rs1695 were not in Hardy Weinberg Equilibrium . Since their study was performed on paraffin embedded HGOS tumor tissue samples and not on DNA extracted from lymphocytes, as almost all pharmacogenetic analyses, their observation is concordant with our data obtained on HGOS cell lines. Genotype changes in relation to CDDP resistance were found for the two non-synonymous SNPs ERCC2 rs1799793 and rs13181 and the synonymous ERCC1 rs11615 at the DNA and RNA level. All three SNPs have been reported to be associated with survival and toxicity, but the data are quite discordant . However, the ERCC2 rs1799793 GG genotype was reported in association with poor event-free survival compared to the GA +AA genotypes and ERCC2 rs13181 AA with poor response to chemotherapy compared to AC+CC . Our data obtained in CDDP-resistance cell lines confirm the relevance of these two SNPs and suggest that they could serve as biomarkers. In one germline study performed on 130 patients with osteosarcoma treated with neoadjuvant cisplatin-based therapy in combination with doxorubicin, methotrexate, and ifosfamide the ERCC2 rs13181 and ERCC2 rs1799793 SNPs were associated with survival . The authors suggested that the amino acid change that occurred as a result of the mutation reduced the ability of the enzyme ERCC2 to repair DNA thus resulting in greater efficacy of cisplatin. In our study this finding seems to be confirmed by the fact that the most resistant cell line returned to the wild-type genotype, restoring the repair capacity of the enzyme with the consequent increased resistance to the chemotherapeutic agent. The functional consequences of ERCC2 rs1799793 and rs13181 on the protein structure and stability have recently been elucidated . Molecular dynamics simulation of the native ERCC2 protein and the variant protein with the substitution of Asp by Asn revealed that rs1799793 resulted in a destabilized, less active protein compared to the native. In addition, the ERCC2 rs13181 variant caused the loss of C-terminal alpha-helix and beta-sheet . Although these secondary structures were lost, the overall folding was not disrupted, suggesting that this polymorphic variation has a less relevant impact on protein function. For ERCC1 rs11615, which changed to the homozygous wild-type genotype status in the Saos-2 CDDP-resistant variant with the highest level of resistance, two germline studies reported similar evidence being better survival associated with the TT compared to the CC genotype . However, five studies reported the opposite evidence for overall survival . Differential gene expression analysis identified dysregulations in CDDP-resistant variants compared to their parental cell lines suggesting that development of CDDP resistance influences not only genes of the NER pathway, which is known to be mainly responsible for the removal of CDDP-associated DNA adducts, but also genes belonging to other DNA repair mechanisms, such as ATM and ATR . It has been shown that cancer cells that are deficient in one DNA repair pathway can activate other functional repair pathways, which underlines the importance to study not only one of them for treatment optimization . The biological consequence of the significant downregulation of ATM and ATR observed in Saos-2/CDDP-resistant variants is a relevant finding that needs to be further explored, since inhibitors against ATR have already been used in clinical trials in other cancers . On the other hand, also the upregulation of ACYP2 and COMT , although not significant, warrants attention because SNPs of these two genes had been described to be associated with ototoxicity after CDDP treatment . In conclusion, the mmNGS approach emerged to be an innovative, reliable tool to detect genetic polymorphisms at both DNA and RNA level, allowing for the identification of genetic changes causally related to CDDP resistance in HGOS cells. Once further validated in tumor samples series, these SNPs could be useful to identify patients with reduced sensitivity to CDDP-based therapy and/or increased susceptibility to CDDP-related adverse toxicities. 4.1. Cell Lines The study was performed on a panel of 12 drug-sensitive human HGOS cell lines: U-2OS, Saos-2, MG-63, and HOS (purchased from the American Type Culture Collection ATCC, Rockville, MD, USA) and IOR/OS9, IOR/OS10, IOR/OS14, IOR/OS15, IOR/OS18, IOR/OS20, IOR/MOS, IOR/SARG, which were established from tumor specimens at the Laboratory of Experimental Oncology of the Orthopaedic Rizzoli Institute . The panel of 6 CDDP-resistant variants derived either from U-2OS (U-2OS/CDDP300, U-2OS/CDDP1μg, U-2OS/CDDP4μg) or Saos-2 (Saos-2/CDDP300, Saos-2/CDDP1μg, Saos-2/CDDP6μg) CDDP-sensitive cell lines, as previously reported . Resistant variants were established by exposing parental cells to step-by-step increases of CDDP concentrations. The in vitro continuous drug exposure resulted in the establishment of variants resistant to 300 ng/mL CDDP (U-2OS/CDDP300 and Saos-2/CDDP300), 1 µg/mL (U-2OS/CDDP1µg and Saos-2/CDDP1µg), 4 µg/mL (U-2OS/CDDP4µg), or 6 µg/mL CDDP (Saos-2/CDDP6µg). Establishment of an adequate in vitro growth at each new CDDP concentration required approximately 10–12 weeks (corresponding to 8–10 in vitro passages), and variants were considered as definitely stabilized when reaching the 20th in vitro passage. CDDP sensitivity of each cell line was expressed as IC50 (drug concentration resulting in 50% inhibition of cell growth after 96 h of in vitro treatment). The fold-increase in CDDP resistance of each variant was determined by comparing its IC50 value with that of its corresponding parental cell line and, as previously described, ranged from 4.0- to 62.5-fold for U-2OS variants and from 7.4- to 112.1-fold for Saos-2 variants . All cell lines were cultured in Iscove’s modified Dulbecco’s medium (IMDM) added with 10% fetal bovine serum (Biowhittaker Europe, Cambrex-Verviers, Belgium) and maintained in a humified atmosphere with 5% CO 2 at 37 °C. Drug resistant variants were continuously cultured in the presence of the CDDP concentrations used for their selection. Cell pellets were prepared according to standard procedures when cells were confluent, snap-frozen, and stored at −80 °C. DNA fingerprint analyses were performed for all cell lines using 17 polymorphic short tandem repeat sequences confirming their identity. 4.2. Extraction of Nucleic Acids DNA and RNA were simultaneously isolated and purified from the same pellet obtained from each cell line by using the AllPrep DNA/RNA mini kit (Qiagen, Hilden, Germany) according to the manufacturers’ instructions. During this process, the DNA and RNA were isolated from the entire sample by passing the lysate first to the AllPrep DNA spin column to isolate high molecular weight total genomic DNA and through the AllPrep RNA spin column to isolate total RNA. A DNA and RNA quality check was performed for all samples by spectrophotometry (NP-80, Implen, Munich, Germany). All RNA samples were run on a 2100 Bioanalyzer system (Agilent, Santa Clara, CA, USA) using the RNA 6000 kit (Agilent, Santa Clara, CA, USA). 4.3. Custom Multi-Modal Targeted Next Generation Sequencing (mmNGS) Library preparation was performed according to the QIAseq Multimodal Panel handbook v06/2020 (Qiagen, Hilden, Germany) for small panels. The primers for the libraries derived from DNA were designed for 28 SNPs of 14 genes related to DNA repair, CDDP transport and detoxification, and TP53 . For the libraries prepared from RNA, primers were designed for the SNPs mapping to exons of the 14 genes, thus allowing RNA variant calling. The specific design of the RNA panel enabled also RNA expression analysis of these 14 genes (technical service Qiagen, Hilden, Germany). This approach uses integrated unique molecular indices (UMIs) which improves the specificity of variant detection. DNA- and RNA-derived libraries were prepared for all 18 cell lines. Prior to library preparation, the nucleic acid concentrations were determined fluorometrically by Qubit high-sensitivity assays on a Qubit reader version 4.0 (Thermo Fisher Scientific by Life Technologies Italia, Monza, Italy). For library preparation, the input amount was 40 ng of DNA and 100 ng of RNA. All libraries were run on a 2100 Bioanalyzer system (Agilent, Santa Clara, CA, USA) using the High Sensitivity DNA kit (Agilent, Santa Clara, CA, USA) to check the profile of the samples. The fragment lengths of all libraries ranged between 400 and 600 base pairs, as expected according to the protocol. In order to provide an accurate quantification of the amplifiable libraries, the QIAseq Library Quant Assay kit (Qiagen, Hilden, Germany) was performed on a real-time PCR system (7900HT Fast Real-time PCR system; (Thermo Fisher Scientific by Life Technologies Italia, Monza, Italy) for all of them. For sequencing, libraries were diluted to 1.2 pM, pooled together and analyzed by paired-end sequencing on a NextSeq 500 instrument (Illumina Inc., San Diego, CA, USA) using a mid-output reagent kit v2.5 (300 cycles) with a custom sequencing primer provided with the library preparation kit. 4.4. mmNGS Data Analysis by CLC Genomics Workbench All bioinformatic analyses were performed using the CLC GWB software (Qiagen Bioinformatics, Aarhus, Denmark) v22.04. FastQ files were downloaded from the BaseSpace cloud (Illumina Inc., San Diego, CA, USA) and imported in the CLC GWB (Qiagen Bioinformatics, Aarhus, Denmark). For the detection of DNA variants and gene expression, the FastQ files were analyzed using the Biomedical Genomics Analysis plugin running the Qiaseq Multimodal Analysis workflow. The DNA and RNA reads were aligned to the human genome hg38 reference sequence and filtered using a coverage of 100× and a variant allele frequency (VAF) higher than 3%. For RNA variant calling a custom workflow was provided by the Qiagen bioinformatics support. This workflow worked with UMIs, mapped the reads on the human genome hg38 and filtered with specific parameters for rare RNA variant calling. For differential gene expression analysis between the groups of drug-resistant variants and their respective parental cell line, the differential expression tools of the CLC GWB were used and changes with a Bonferroni corrected p -value < 0.05 were considered significant. For hierarchical clustering analysis the tool for creating heatmaps of the CLC GWB was used with Euclidean distance and complete linkage. 4.5. SNP Genotyping by Real-Time PCR In total, 24 of the 28 selected polymorphisms were validated by real-time genotyping PCR . TaqMan SNP genotyping assays (Thermo Fisher Scientific by Life Technologies Italia, Monza, Italy) or drug metabolizing enzymes (DMEs) assays, which had functionally been tested, were used to validate the performance of the mmNGS approach. The genotyping experiments were performed according to standard protocols using 10 ng DNA as input material using the VIIA 7 DX realtime PCR system (Thermo Fisher Scientific by Life Technologies Italia, Monza, Italy) and the results were analyzed with the TaqMan Genotyper software (Thermo Fisher Scientific by Life Technologies Italia, Monza, Italy), which generated allelic discrimination cluster plots to determine the genotype of each SNP. The study was performed on a panel of 12 drug-sensitive human HGOS cell lines: U-2OS, Saos-2, MG-63, and HOS (purchased from the American Type Culture Collection ATCC, Rockville, MD, USA) and IOR/OS9, IOR/OS10, IOR/OS14, IOR/OS15, IOR/OS18, IOR/OS20, IOR/MOS, IOR/SARG, which were established from tumor specimens at the Laboratory of Experimental Oncology of the Orthopaedic Rizzoli Institute . The panel of 6 CDDP-resistant variants derived either from U-2OS (U-2OS/CDDP300, U-2OS/CDDP1μg, U-2OS/CDDP4μg) or Saos-2 (Saos-2/CDDP300, Saos-2/CDDP1μg, Saos-2/CDDP6μg) CDDP-sensitive cell lines, as previously reported . Resistant variants were established by exposing parental cells to step-by-step increases of CDDP concentrations. The in vitro continuous drug exposure resulted in the establishment of variants resistant to 300 ng/mL CDDP (U-2OS/CDDP300 and Saos-2/CDDP300), 1 µg/mL (U-2OS/CDDP1µg and Saos-2/CDDP1µg), 4 µg/mL (U-2OS/CDDP4µg), or 6 µg/mL CDDP (Saos-2/CDDP6µg). Establishment of an adequate in vitro growth at each new CDDP concentration required approximately 10–12 weeks (corresponding to 8–10 in vitro passages), and variants were considered as definitely stabilized when reaching the 20th in vitro passage. CDDP sensitivity of each cell line was expressed as IC50 (drug concentration resulting in 50% inhibition of cell growth after 96 h of in vitro treatment). The fold-increase in CDDP resistance of each variant was determined by comparing its IC50 value with that of its corresponding parental cell line and, as previously described, ranged from 4.0- to 62.5-fold for U-2OS variants and from 7.4- to 112.1-fold for Saos-2 variants . All cell lines were cultured in Iscove’s modified Dulbecco’s medium (IMDM) added with 10% fetal bovine serum (Biowhittaker Europe, Cambrex-Verviers, Belgium) and maintained in a humified atmosphere with 5% CO 2 at 37 °C. Drug resistant variants were continuously cultured in the presence of the CDDP concentrations used for their selection. Cell pellets were prepared according to standard procedures when cells were confluent, snap-frozen, and stored at −80 °C. DNA fingerprint analyses were performed for all cell lines using 17 polymorphic short tandem repeat sequences confirming their identity. DNA and RNA were simultaneously isolated and purified from the same pellet obtained from each cell line by using the AllPrep DNA/RNA mini kit (Qiagen, Hilden, Germany) according to the manufacturers’ instructions. During this process, the DNA and RNA were isolated from the entire sample by passing the lysate first to the AllPrep DNA spin column to isolate high molecular weight total genomic DNA and through the AllPrep RNA spin column to isolate total RNA. A DNA and RNA quality check was performed for all samples by spectrophotometry (NP-80, Implen, Munich, Germany). All RNA samples were run on a 2100 Bioanalyzer system (Agilent, Santa Clara, CA, USA) using the RNA 6000 kit (Agilent, Santa Clara, CA, USA). Library preparation was performed according to the QIAseq Multimodal Panel handbook v06/2020 (Qiagen, Hilden, Germany) for small panels. The primers for the libraries derived from DNA were designed for 28 SNPs of 14 genes related to DNA repair, CDDP transport and detoxification, and TP53 . For the libraries prepared from RNA, primers were designed for the SNPs mapping to exons of the 14 genes, thus allowing RNA variant calling. The specific design of the RNA panel enabled also RNA expression analysis of these 14 genes (technical service Qiagen, Hilden, Germany). This approach uses integrated unique molecular indices (UMIs) which improves the specificity of variant detection. DNA- and RNA-derived libraries were prepared for all 18 cell lines. Prior to library preparation, the nucleic acid concentrations were determined fluorometrically by Qubit high-sensitivity assays on a Qubit reader version 4.0 (Thermo Fisher Scientific by Life Technologies Italia, Monza, Italy). For library preparation, the input amount was 40 ng of DNA and 100 ng of RNA. All libraries were run on a 2100 Bioanalyzer system (Agilent, Santa Clara, CA, USA) using the High Sensitivity DNA kit (Agilent, Santa Clara, CA, USA) to check the profile of the samples. The fragment lengths of all libraries ranged between 400 and 600 base pairs, as expected according to the protocol. In order to provide an accurate quantification of the amplifiable libraries, the QIAseq Library Quant Assay kit (Qiagen, Hilden, Germany) was performed on a real-time PCR system (7900HT Fast Real-time PCR system; (Thermo Fisher Scientific by Life Technologies Italia, Monza, Italy) for all of them. For sequencing, libraries were diluted to 1.2 pM, pooled together and analyzed by paired-end sequencing on a NextSeq 500 instrument (Illumina Inc., San Diego, CA, USA) using a mid-output reagent kit v2.5 (300 cycles) with a custom sequencing primer provided with the library preparation kit. All bioinformatic analyses were performed using the CLC GWB software (Qiagen Bioinformatics, Aarhus, Denmark) v22.04. FastQ files were downloaded from the BaseSpace cloud (Illumina Inc., San Diego, CA, USA) and imported in the CLC GWB (Qiagen Bioinformatics, Aarhus, Denmark). For the detection of DNA variants and gene expression, the FastQ files were analyzed using the Biomedical Genomics Analysis plugin running the Qiaseq Multimodal Analysis workflow. The DNA and RNA reads were aligned to the human genome hg38 reference sequence and filtered using a coverage of 100× and a variant allele frequency (VAF) higher than 3%. For RNA variant calling a custom workflow was provided by the Qiagen bioinformatics support. This workflow worked with UMIs, mapped the reads on the human genome hg38 and filtered with specific parameters for rare RNA variant calling. For differential gene expression analysis between the groups of drug-resistant variants and their respective parental cell line, the differential expression tools of the CLC GWB were used and changes with a Bonferroni corrected p -value < 0.05 were considered significant. For hierarchical clustering analysis the tool for creating heatmaps of the CLC GWB was used with Euclidean distance and complete linkage. In total, 24 of the 28 selected polymorphisms were validated by real-time genotyping PCR . TaqMan SNP genotyping assays (Thermo Fisher Scientific by Life Technologies Italia, Monza, Italy) or drug metabolizing enzymes (DMEs) assays, which had functionally been tested, were used to validate the performance of the mmNGS approach. The genotyping experiments were performed according to standard protocols using 10 ng DNA as input material using the VIIA 7 DX realtime PCR system (Thermo Fisher Scientific by Life Technologies Italia, Monza, Italy) and the results were analyzed with the TaqMan Genotyper software (Thermo Fisher Scientific by Life Technologies Italia, Monza, Italy), which generated allelic discrimination cluster plots to determine the genotype of each SNP. |
Transcending time: the forensic anthropological case study of three unidentified transgender women in Italy in the early 1990s | 62f24a40-ffc4-4dae-9a28-f0068fd1605f | 11003900 | Forensic Medicine[mh] | Biological sex is determined at conception and dictates dimorphic morphometric traits, while gender is a social construct tied to self-perception and identity. Gender dysphoria (GD) is the incongruence between the gender identity and the biological sex of an individual . People experiencing GD are broadly referred to as transgender, and they may grapple with severe psychological and emotional distress and uneasiness with themselves, due to this disconnection and the subsequent feeling of internal and social rejection such as transphobia . Transgender individuals may perform a transition from male to female (MtF) or from females to males (FtM), opting for surgical treatment like Gender Confirmation Surgery (GCS). In particular, MtF women can choose to modify masculine facial features via Facial Feminization Surgery (FFS) . At present, only biological sex can be inferred from bone analysis , leaving out the determination of gender identities which cannot be reliably inferred solely from associated evidence, such as clothing or personal items . Despite these limitations, the issue is paramount, especially in the light of the spreading of violence targeting trans and gender-diverse individuals with a reported 327 murders in 2022 alone. According to the Trans Murder Monitoring report from 2008 to 2022, Italy has the highest rate per capita of transgender murders. A 10-year retrospective study presented 20 cases of homicide of transgender individuals handled by the Institute of Legal Medicine of Milan, Italy . Most victims were sex workers from South America, with three being native Italians. Signs of overkill and postmortem symbolical manipulation were found, stressing the brutality perpetrated by the offenders toward these victims. In cases where soft tissues are preserved and have not been extensively damaged by postmortem decay, the possible presence of external genital organs and breast implants, or signs of mastectomy, coupled with circumstantial evidence, may suggest the individual’s transgender status. However, for highly decomposed or skeletonized remains, such evidence is often absent. This challenges forensic anthropologists to consider the possibility that the remains may belong to individuals who did not identify with their birth-assigned sex. When a discrepancy is noticed within a body or between anatomical and material evidence, it is essential to record and share this information with investigators. Such details can aid in refining potential identity matches and support the identification process. However, a recent survey indicates that some practitioners are reluctant to report such findings, deeming it beyond their scope . By presenting the postmortem examinations and anthropological analyses of the remains of three unidentified transgender women from the Institute of Legal Medicine of Milan (Italy), this report aims at: (i) presenting the challenging task of determining the transgender status from skeletal remains; (ii) comparing the outcome of postmortem and anthropological examinations; (iii) providing additional evidence that could guide the identification and interpretation of bone signs possibly related to GCS; (iv) increasing practitioners’ awareness of this issue.
Three skeletons of transgender individuals that died between 1992 and 1996 and underwent post-mortem examinations at the Institute of Legal Medicine of Milan were analyzed. The transgender status of the individuals was derived from the autopsy records. As the investigations did not lead to personal identification, the bodies were then buried as unidentified individuals in cemeteries across the city of Milan. According to the Italian cemetery policies, after ten years from burial, unclaimed human remains are to be disposed of in cumulative ossuaries. In order to avoid this, the Laboratory of Forensic Anthropology and Odontology ( Laboratorio di Antropologia e Odontologia Forense —LABANOF) of the University of Milan retrieves these remains for identification purposes . This activity aligns with Article 43 of the Presidential Decree of the Italian Republic (DPR) n.285 of September 10th, 1990, of the National Police Mortuary Regulation, and the accord between LABANOF and the University of Milan, and the special Commissioner of the Government for Missing Persons. Consequently, after complete skeletonization, the skeletal remains were exhumed between 2016 and 2019 and transferred to laboratory facilities in zinc boxes. The recovered remains presented different taphonomic patterns of preservation both in terms of quantity and quality of the remains. The completeness of the skeletons was assessed following the Bone Representation Index (BRI) which calculates the ratio between the number of bones recovered and the total number that should be anatomically present at the time of recovery . The remains were laid down, and the biological profile was created, including ancestry , sex [ – ], age , and stature estimations. Given that some skeletons were missing portions (e.g., pubic symphysis) due to postmortem damage, different methods were used based on the available skeletal material. Moreover, the bones were examined looking for evidence of gender-confirmation surgery (GCS) and facial feminization surgery (FFS), in accordance with the available literature [ – , , ].
For each case, the results of the post-mortem examination and anthropological study are reported below. Tables and summarize the postmortem examination findings and the biological profile of the three skeletons, respectively. Case 1 Case 1 was a 30/35-year-old black male (length 183 cm and weight 70 kg). The body was reported to be wearing a black bodysuit and blue pants. Earring holes were noted on the earlobes bilaterally. Bilateral breast prosthesis with associated scars were present. The external genitalia were male. The cause of death was overdose. Skeleton 1 was estimated to be a European, male individual, aged between 23 and 57 years. The zygomatic bones present as medially curved toward the infratemporal fossa. On the posterior third of the left zygomatic bone, a circular area (11 mm of diameter) with deposition of porous bone was found (Fig. ). The surrounding surface was compromised with exfoliation of the cortical bone due to postmortem damage, hampering a thorough evaluation. Case 2 Case 2 was a 25/30-year-old white male (length 182 cm and weight 74 kg). The body had no visible decomposition features. The body was naked as it came from a city hospital. Bilateral breast prosthesis with associated scars was observed. The external genitalia were male. Breast implants were found bilaterally at autopsy dissection. The cause of death was undetermined. Skeleton 2 was estimated to be a European, male individual, aged between 35 and 44 years. This skeleton did not present any macroscopic or microscopic alterations of the cortical bone affecting the cranium. No sign that could be associated with feminization surgery was found. Several small fragments of gelatinous silicone-like material, embedded in soil, were recovered when sieving the burial soil. Case 3 Case 3 was a 30/35-year-old male (length 170 cm and weight 67 kg). The body had no visible decomposition features. The ancestry was described as Austromelanesoid, according to the classification available at the time of the autopsy. The body was reported to be wearing a wig, with approximately 20-cm-long brown-reddish hair, several hairpins, and a leopard-type clip. Other garments included a scarf around the neck, a black corset, white pants, a black panty girdle, black leathery belts legging, black hold-up stockings, gray socks and leathery stiletto boots with metal heels. Make-up was found on the face, and earring holes were noted on the earlobes bilaterally. The victim was wearing a ring on the fourth finger of the right hand. Pink and red polish on the nails of hands and feet respectively was noted. On the face, two flat surgical scars were described in the right parotid and in the left mandibular regions. Bilateral breast prosthesis with associated scars was found. The external genitalia were male. The cause of death was overdose. Skeleton 3 was estimated to be a European, male individual, aged between 21 and 30 years. The extensive fragmentation of the cranium hampered a comprehensive analysis both for the biological profile and the detection of feminization signs. In its middle third, the right zygomatic bone presented a compact bone area, measuring 17 mm on a horizontal axis and 15 mm on a vertical axis, surrounded by porotic cortical bone. The bone here was remodeled, highly dense, possibly meaning that the timing of remodeling was not recent. On the left zygomatic bone, in its middle third, a porotic circular area extended for 13 mm on a horizontal axis and 11 mm on a vertical axis, with evidence of macro and microporosity. Figure displays the relevant skeletal findings. A breast prosthesis (Fig. ) was found after examining and sieving the burial soil.
Case 1 was a 30/35-year-old black male (length 183 cm and weight 70 kg). The body was reported to be wearing a black bodysuit and blue pants. Earring holes were noted on the earlobes bilaterally. Bilateral breast prosthesis with associated scars were present. The external genitalia were male. The cause of death was overdose. Skeleton 1 was estimated to be a European, male individual, aged between 23 and 57 years. The zygomatic bones present as medially curved toward the infratemporal fossa. On the posterior third of the left zygomatic bone, a circular area (11 mm of diameter) with deposition of porous bone was found (Fig. ). The surrounding surface was compromised with exfoliation of the cortical bone due to postmortem damage, hampering a thorough evaluation.
Case 2 was a 25/30-year-old white male (length 182 cm and weight 74 kg). The body had no visible decomposition features. The body was naked as it came from a city hospital. Bilateral breast prosthesis with associated scars was observed. The external genitalia were male. Breast implants were found bilaterally at autopsy dissection. The cause of death was undetermined. Skeleton 2 was estimated to be a European, male individual, aged between 35 and 44 years. This skeleton did not present any macroscopic or microscopic alterations of the cortical bone affecting the cranium. No sign that could be associated with feminization surgery was found. Several small fragments of gelatinous silicone-like material, embedded in soil, were recovered when sieving the burial soil.
Case 3 was a 30/35-year-old male (length 170 cm and weight 67 kg). The body had no visible decomposition features. The ancestry was described as Austromelanesoid, according to the classification available at the time of the autopsy. The body was reported to be wearing a wig, with approximately 20-cm-long brown-reddish hair, several hairpins, and a leopard-type clip. Other garments included a scarf around the neck, a black corset, white pants, a black panty girdle, black leathery belts legging, black hold-up stockings, gray socks and leathery stiletto boots with metal heels. Make-up was found on the face, and earring holes were noted on the earlobes bilaterally. The victim was wearing a ring on the fourth finger of the right hand. Pink and red polish on the nails of hands and feet respectively was noted. On the face, two flat surgical scars were described in the right parotid and in the left mandibular regions. Bilateral breast prosthesis with associated scars was found. The external genitalia were male. The cause of death was overdose. Skeleton 3 was estimated to be a European, male individual, aged between 21 and 30 years. The extensive fragmentation of the cranium hampered a comprehensive analysis both for the biological profile and the detection of feminization signs. In its middle third, the right zygomatic bone presented a compact bone area, measuring 17 mm on a horizontal axis and 15 mm on a vertical axis, surrounded by porotic cortical bone. The bone here was remodeled, highly dense, possibly meaning that the timing of remodeling was not recent. On the left zygomatic bone, in its middle third, a porotic circular area extended for 13 mm on a horizontal axis and 11 mm on a vertical axis, with evidence of macro and microporosity. Figure displays the relevant skeletal findings. A breast prosthesis (Fig. ) was found after examining and sieving the burial soil.
In forensic anthropology, the identification of remains that might belong to transgender individuals is not commonplace, and such a circumstance might not always be given adequate consideration . Forensic anthropologists have only recently begun to grapple with the binary nature of sex assessment, advocating for a more inclusive methodology for sex estimation that encompasses the variability found in transgender individuals [ , , , ]. This estimation can be challenging, as solely the skeletal features can be evaluated, and no reliable tool is available to assume a gender identity . As expected, in this study, the skeletal sex estimation based on morphological evaluation of the cranium and the pelvis classified the individuals as males. Thus, inferring transgender status based solely on skeletal evidence was not feasible. This has some serious repercussions when attempts are made to identify skeletal remains, especially in the absence of other evidence. From a practical standpoint, there is the relevant issue of how transgender individuals are classified within the registers of missing persons and of unidentified corpses . In the registry of missing persons, a transgender person should be listed under their self-identified gender, as this reflects their lived identity and is typically the gender under which their disappearance would have been reported. This respects of course the individual’s affirmed identity but may not necessarily be able to effectively assist in making a successful match with unidentified remains. In fact, there may be the case of a corpse found in an advanced state of decomposition or skeletonized, on which it is not possible to make an adequate assessment of the skin surface and soft tissue. Thus, there could be a transgender individual who is classified according to the biological sex and then entered in the wrong database, significantly reducing the chances of achieving proper identification. Notably, in the three cases studied, no clothing or other items indicative/suggestive of transgender identity were found when only skeletal remains were examined. The question then arises: how can anthropologists gather skeletal and associated evidence that would support the hypothesis of the victims being transgender, so that the transition does not lessen the possibility of being identified ? This case series did not identify specific skeletal indicators, but supportive findings were observed. For example, in two cases fragments of gelatinous material were recovered embedded in the burial soil of each skeleton, after accurate sieving procedures. Such a finding highlights the importance of thorough recovery and examination of residues in the soil which may provide valuable insights to the investigations . After careful assessments, they were interpreted as possibly coming from the decomposition of breast implants, which is consistent with these individuals’ autopsy report describing their presence. In one case, however, no such gelatinous material was found, although the deceased had breast implants. Clothing or other items offer limited reliability in such cases , as they can be dislodged during skeletonization and lost, especially, when the body is in an outdoor environment. On the contrary, the presence of secondary identifiers or circumstantial evidence can serve as additional indicators, which could raise the suspicion of a transgender individual, given the limitations of skeletal analysis alone. MtF individuals may undergo surgical procedures like facial feminization surgery (FFS) reconstruction, which heavily involves modifications of the maxillo-facial skeleton . Common sites of modification include the glabellar region and the mandible . However, no modification altering the original morphology of the glabella or mental regions was observed in the skeletons of this study. Other anatomical portions considered in FFS involve the nose, hairline, and cheekbones: however, these changes may not leave evidence on the skeleton , so they are not always easily recognizable. In this series, the only anomalies observed included small porotic and remodeled areas on the zygomatic bones in two cases. The literature on FFS mentions the zygomatic bones for injection of fillers of placement of implants for cheek augmentation . Enhancement of the malar projection may also include the creation of a defect in the zygomatic bone through osteotomy and placement of a bone graft . The postmortem report of case 3 detailed two bilateral scars in the right and left parotid regions, which arguably are the results of a surgical procedure, and bilateral anomalies of the cortical surface were observed on the skeletal remains. In case 1, signs were observed only on one side, which is not consistent with FFS procedures. Therefore, considering the dearth of detailed reports, in addition to the discordant evidence collected, we are far from confidently confirming or ruling out that these signs are related to FFS, based only on the skeletal evidence. The bone marks observed here are not specific enough to link them to FFS procedures. Gender confirming surgical procedures were pioneered in the USA in the 1980s and 1990s , and the three cases represent victims from the early to mid 1990s in Italy. Possibly, FFS procedures at that time were not as widespread as they are today. Arguably, these women did not undergo or could not access FFS. Although a study on a limited sample indicated that pelvic measurements may reveal a shift toward a male metric configuration in FtM individuals and while there is variable evidence regarding the interplay between hormone therapies and bone tissue, how gender transition affects the skeleton is still poorly understood . More research is needed to better understand the effects of hormone replacement therapy and gender confirmation surgeries on the skeleton. Investigating specific changes in bone density, structure, and growth resulting from these treatments could potentially reveal identifiable markers for transgender individuals. Nevertheless, these procedures do not seem to significantly affect the parameters used in standard sex estimation methods. A recent application of Fordisc for craniometric sex estimation on CT-scans of MtF transgender individuals concluded that post-surgery, metric methods would most likely still determine the individual biological sex, albeit with reduced probabilities and typicalities . The anthropological examination on the remains did not produce any definitive skeletal evidence that could be confidently linked to FFS. However, circumstantial evidence and thorough recovery procedures arguably support the hypothesis of the transgender status. When properly recovered, these elements, could possibly serve as red flags for forensic anthropologists who are aware of the challenging issue of identifying transgender individuals. Transgender women and men encompass a diverse population with different transition journeys and medical histories which add to the complexity. While acknowledging certain limitations, this study offered a significant contribution to this cause, highlighting the significance of a comprehensive and multifaceted approach to the intricacies of gender identity within the forensic identification process. In addition to adopting a multidimensional forensic approach in such scenarios, it is imperative to maintain an inclusive perspective that takes these situations into account, especially in light of the escalating violence targeting this part of the society.
|
Immunohistochemical Characterization of Spermatogenesis in the Ascidian | b9d2fbab-5530-40b6-bb8a-39d4a6c8a44d | 11592721 | Anatomy[mh] | Spermatogenesis is a highly organized process that generates haploid sperm cells through the proliferation, meiosis, and differentiation of germ cells, accompanied by a significant morphological transformation known as spermiogenesis. Diploid spermatogonia function as stem cells, continuously replenishing the spermatogonial pool. A subset of these cells differentiates into subsequent stages of spermatogenic cells, including spermatocytes, spermatids, and, ultimately, mature sperm. In addition to these germline cells, two types of somatic cells, Sertoli and Leydig cells, play crucial roles in regulating the development and differentiation of the germ cell lineage . In mammals, spermatogenesis takes place in the seminiferous tubules, where it progresses from the basal lamina toward the lumen. In anamniote vertebrates, such as fish and frogs, spermatogenesis occurs within cysts, each containing a specific stage of developing spermatogenic cell syncytia. Each cyst is enclosed by a layer of Sertoli cells . Ascidians, marine invertebrates, provide valuable insights into vertebrate evolution and present unique models for understanding gamete function . For instance, Ciona sperm exhibit clear activation of motility and chemotaxis toward the egg during fertilization . The molecular mechanisms underlying intracellular Ca 2+ dynamics and dynein regulation during fertilization have been elucidated in Ciona sperm . In ascidian sperm, the mitochondrion is positioned laterally on the nucleus, and during a distinctive event known as sperm reaction, the mitochondrion translocates along the flagellum, contributing to the exclusion of paternal mitochondria from embryonic development . Moreover, ascidians are hermaphroditic and exhibit gamete self-incompatibility mechanisms to prevent self-fertilization . Thus, Ciona sperm serve as a unique system for studying sperm behavior and sperm-egg interactions during fertilization. Previous studies have reported that gonad rudiment formation begins early in juvenile stages after metamorphosis and subsequently differentiates into the ovotestis . The later stages of spermatogenesis have been described through light and electron microscopy . However, despite significant advancements in understanding sperm physiology, the process of sperm formation in Ciona remains less well understood, and somatic cells supporting spermatogenesis, such as Sertoli and Leydig cells, have yet to be identified. We previously identified a set of genes highly and uniquely expressed in the testis, termed testis-expressed class D (TD) genes . In this paper, we characterize spermatogenic cells in Ciona robusta using immunohistochemical methods, employing antibodies against two TD gene products. 2.1. Dissection of Testis Tissue from Adult Animals Adults of Ciona robusta were cultivated either at the National Bioresource Project (NBRP) within the Field Science Education and Research Center, Maizuru Fisheries Experimental Station, Kyoto University (Maizuru City, Kyoto Prefecture), or at the Misaki Marine Biological Station, The University of Tokyo (Miura City, Kanagawa Prefecture). In some experiments, wild adults were collected at the Tohoku University Graduate School of Agricultural Science’s Complex Ecology Field Education and Research Center (Onagawa Town, Miyagi Prefecture) and transferred to an aquarium at the Shimoda Marine Research Center, University of Tsukuba. To prevent spawning, the specimens were maintained under controlled lighting conditions at seawater temperatures below 18 °C. Testes were dissected using dissection scissors and removed from the gut wall with ophthalmic forceps. The collected testes were preserved in filtered seawater on ice. 2.2. Isolation of Spermatogenic Cells Testes from 3 to 5 individuals were placed in a 1.5 mL tube containing three volumes of filtered seawater. The tissue was gently minced with ophthalmic forceps, and the mixture was tapped to suspend the contents. The sample was briefly centrifuged at 1260× g for 10 s to separate the tissue fragments. Spermatogenic cells were collected from the supernatant. All procedures were performed on ice. 2.3. Histological Observation of Testes The dissected testes were fixed overnight in Bouin’s solution, and then transferred to 70% ethanol. The tissues were dehydrated through an ethanol series, replaced with xylene, and embedded in paraffin. Sections of 8 μm thickness were prepared and stained with Hematoxylin and Eosin (HE). 2.4. Preparation of Frozen Sections Frozen sections were prepared from both unfixed and PFA-fixed testes. Dissected testes were washed in artificial seawater (ASW) and then placed in embedding medium White Tissue Coat (WTC, Yuai Kasei Co., Amagasaki, Japan) at 4 °C for 1 h before being frozen in MC802. For PFA fixation, testes were washed in ASW, then in 0.1 M MOPS and 0.5 M NaCl, and fixed in 4% PFA in phosphate-buffered saline (PBS) at 4 °C overnight. The fixed testes were washed three times for 20 min each in PBS, and then sequentially incubated in 10% sucrose in PBS at 4 °C for 30 min, 15% sucrose in PBS at 4 °C for 30 min, and 20% sucrose in PBS at 4 °C overnight. After an additional 30 min in 20% sucrose in PBS, the tissues were embedded in WTC at 4 °C for 1 h and frozen in MC802. Sections of 4 μm thickness were cut using a cryostat (REM-710 Retratome equipped with MC802, Yamato Kohki, Asaka, Japan), mounted on MAS-coated slides (Matsunami Glass Ind., Kishiwada, Japan), air-dried at room temperature for 1 h, and used for immunofluorescence staining. Samples not used immediately were stored at −80 °C. 2.5. Primary Antibodies For immunofluorescence analysis, we used five primary antibodies at the following dilutions: Anti-CiVH antibody (mouse monoclonal, 1:100) ; Anti-TD09 antibody (mouse polyclonal, 1:100) ; Anti-TD02 antibody (mouse polyclonal, 1:100) ; Anti-acetylated α-tubulin antibody (T6793, Sigma-Aldrich Co., Japan, mouse monoclonal, 1:1000); and Anti-MIS antibody H-300 (SC-28912, Biotechnology Inc., Santa Cruz, TX, USA, rabbit polyclonal, 1:1000). 2.6. Immunohistochemistry Spermatogenic cells adhered to MAS-coated slides were washed with a buffer containing 0.1 M MOPS and 0.5 M NaCl, and fixed with 4% paraformaldehyde in PBS at room temperature for 1 h, followed by fixation with 100% methanol at −20 °C for 10 min. The samples were quickly dried with cold air and rehydrated in PBS at room temperature for 5 min or more. Frozen sections, either freshly prepared or stored at −80 °C, were air-dried at room temperature for 1 h, fixed with 100% methanol at −20 °C for 10 min, quickly dried with cold air, and rehydrated in PBS for 5 min or more. The sections were washed three times in PBS to remove WTC. Blocking was performed with 10% goat serum in PBS, followed by incubation with primary antibodies at room temperature for 1 h or at 4 °C overnight. After washing with 0.05% Triton X-100 in PBS (T-PBS) or PBS three times for 5 min each, sections were incubated with secondary antibodies (goat Alexa-546-conjugated anti-mouse or anti-rabbit antibody, Thermo Fisher Scientific Inc., Tokyo, Japan). Nuclear staining was performed with 10 μM DAPI in PBS. Images were captured and analyzed using a fluorescent microscope with differential interference contrast objectives (BX51, Olympus Corp., Tokyo, Japan). The stages of spermatogenic cells were determined according to the criteria shown in . 2.7. Electron Microscopy Trimmed testis tissues were fixed in 0.45 M sucrose, 2.5% glutaraldehyde, and 0.1 M sodium cacodylate (pH 7.2) at 4 °C for 2 h, postfixed with 1% OsO 4 buffered in 0.1 M sodium cacodylate (pH 7.2) at 4 °C for 2 h, and dehydrated through a graded ethanol and propylene oxide series. Samples were embedded in Quetol 812 (Nisshin EM Co., Tokyo, Japan), sectioned, stained with uranyl acetate and lead citrate, and observed under a transmission electron microscope (JEM 1010EX; JEOL, Tokyo, Japan), as described previously . Adults of Ciona robusta were cultivated either at the National Bioresource Project (NBRP) within the Field Science Education and Research Center, Maizuru Fisheries Experimental Station, Kyoto University (Maizuru City, Kyoto Prefecture), or at the Misaki Marine Biological Station, The University of Tokyo (Miura City, Kanagawa Prefecture). In some experiments, wild adults were collected at the Tohoku University Graduate School of Agricultural Science’s Complex Ecology Field Education and Research Center (Onagawa Town, Miyagi Prefecture) and transferred to an aquarium at the Shimoda Marine Research Center, University of Tsukuba. To prevent spawning, the specimens were maintained under controlled lighting conditions at seawater temperatures below 18 °C. Testes were dissected using dissection scissors and removed from the gut wall with ophthalmic forceps. The collected testes were preserved in filtered seawater on ice. Testes from 3 to 5 individuals were placed in a 1.5 mL tube containing three volumes of filtered seawater. The tissue was gently minced with ophthalmic forceps, and the mixture was tapped to suspend the contents. The sample was briefly centrifuged at 1260× g for 10 s to separate the tissue fragments. Spermatogenic cells were collected from the supernatant. All procedures were performed on ice. The dissected testes were fixed overnight in Bouin’s solution, and then transferred to 70% ethanol. The tissues were dehydrated through an ethanol series, replaced with xylene, and embedded in paraffin. Sections of 8 μm thickness were prepared and stained with Hematoxylin and Eosin (HE). Frozen sections were prepared from both unfixed and PFA-fixed testes. Dissected testes were washed in artificial seawater (ASW) and then placed in embedding medium White Tissue Coat (WTC, Yuai Kasei Co., Amagasaki, Japan) at 4 °C for 1 h before being frozen in MC802. For PFA fixation, testes were washed in ASW, then in 0.1 M MOPS and 0.5 M NaCl, and fixed in 4% PFA in phosphate-buffered saline (PBS) at 4 °C overnight. The fixed testes were washed three times for 20 min each in PBS, and then sequentially incubated in 10% sucrose in PBS at 4 °C for 30 min, 15% sucrose in PBS at 4 °C for 30 min, and 20% sucrose in PBS at 4 °C overnight. After an additional 30 min in 20% sucrose in PBS, the tissues were embedded in WTC at 4 °C for 1 h and frozen in MC802. Sections of 4 μm thickness were cut using a cryostat (REM-710 Retratome equipped with MC802, Yamato Kohki, Asaka, Japan), mounted on MAS-coated slides (Matsunami Glass Ind., Kishiwada, Japan), air-dried at room temperature for 1 h, and used for immunofluorescence staining. Samples not used immediately were stored at −80 °C. For immunofluorescence analysis, we used five primary antibodies at the following dilutions: Anti-CiVH antibody (mouse monoclonal, 1:100) ; Anti-TD09 antibody (mouse polyclonal, 1:100) ; Anti-TD02 antibody (mouse polyclonal, 1:100) ; Anti-acetylated α-tubulin antibody (T6793, Sigma-Aldrich Co., Japan, mouse monoclonal, 1:1000); and Anti-MIS antibody H-300 (SC-28912, Biotechnology Inc., Santa Cruz, TX, USA, rabbit polyclonal, 1:1000). Spermatogenic cells adhered to MAS-coated slides were washed with a buffer containing 0.1 M MOPS and 0.5 M NaCl, and fixed with 4% paraformaldehyde in PBS at room temperature for 1 h, followed by fixation with 100% methanol at −20 °C for 10 min. The samples were quickly dried with cold air and rehydrated in PBS at room temperature for 5 min or more. Frozen sections, either freshly prepared or stored at −80 °C, were air-dried at room temperature for 1 h, fixed with 100% methanol at −20 °C for 10 min, quickly dried with cold air, and rehydrated in PBS for 5 min or more. The sections were washed three times in PBS to remove WTC. Blocking was performed with 10% goat serum in PBS, followed by incubation with primary antibodies at room temperature for 1 h or at 4 °C overnight. After washing with 0.05% Triton X-100 in PBS (T-PBS) or PBS three times for 5 min each, sections were incubated with secondary antibodies (goat Alexa-546-conjugated anti-mouse or anti-rabbit antibody, Thermo Fisher Scientific Inc., Tokyo, Japan). Nuclear staining was performed with 10 μM DAPI in PBS. Images were captured and analyzed using a fluorescent microscope with differential interference contrast objectives (BX51, Olympus Corp., Tokyo, Japan). The stages of spermatogenic cells were determined according to the criteria shown in . Trimmed testis tissues were fixed in 0.45 M sucrose, 2.5% glutaraldehyde, and 0.1 M sodium cacodylate (pH 7.2) at 4 °C for 2 h, postfixed with 1% OsO 4 buffered in 0.1 M sodium cacodylate (pH 7.2) at 4 °C for 2 h, and dehydrated through a graded ethanol and propylene oxide series. Samples were embedded in Quetol 812 (Nisshin EM Co., Tokyo, Japan), sectioned, stained with uranyl acetate and lead citrate, and observed under a transmission electron microscope (JEM 1010EX; JEOL, Tokyo, Japan), as described previously . 3.1. Observation and Classification of Spermatogenic Cells in Ciona Testis We initially examined HE-stained paraffin sections of the testes. The Ciona testis consists of numerous club-like compartments known as testicular follicles ( A). Each follicle was densely packed with spermatogenic cells at various developmental stages. As previously observed in Ciona and Botryllus , spermatogenic cells were distributed from larger cells near the basal lamina to smaller, later-stage cells in the center, indicating that the testis is non-cystic rather than cystic. However, the alignment of spermatogenic cells is less strictly organized compared to the mammalian seminiferous tubules. Late spermatids and mature sperm were observed in the region from the center to the efferent duct ( B). Based on cell size ( C,D) and the immunohistochemical characterizations described below, we classified the spermatogenic cells into round stages (RI to RIV) and elongated stages (EI to EIII) . 3.2. Classification of Spermatogenic Cells Using Immunohistochemical Staining 3.2.1. MIS H-300 Antibody Sertoli cells are essential for the growth and differentiation of spermatogenic cells in mammalian testes. However, previous studies on Boltenia villosa suggested the absence of Sertoli-like cells in ascidians . To investigate the presence of Sertoli-like cells and potential interactions between germ cells and somatic cells in Ciona , we used the MIS H-300 antibody, which targets Müllerian Inhibiting Substance (MIS), a TGFβ superfamily member secreted by Sertoli cells in vertebrates. Immunofluorescence staining revealed that free cells located between testicular follicles were positively stained with the MIS H-300 antibody ( A), while no spermatogenic cells within the follicles were stained. Further analysis of isolated cells showed that the signal was localized in the cytoplasm of these free cells, particularly near the nucleus ( B). 3.2.2. Anti-CiVH Antibody Vasa protein is a marker for primordial germ cells and spermatogonia. The anti-CiVH antibody, a monoclonal antibody that recognizes primordial germ cells and early-stage germ cells in Ciona juveniles , was used to stain adult Ciona testes. Immunofluorescence revealed a few large cells along the walls of the testicular follicles ( A). These isolated cells appeared round, with a diameter greater than 7.3 μm, and the immunofluorescence signal was present throughout the cytoplasm but not in the nucleus ( B). 3.2.3. Anti-TD09 Antibody The TD09 gene, highly expressed in Ciona testis, is homologous to the synap-tonemal complex protein SYCP3, which is expressed in meiotic cells . An an-tibody developed against TD09 recognized broad areas of the testis section, excluding the region containing mature sperm near the efferent duct ( A). In isolated spermatogenic cells, the nuclei of round-stage cells (RII to RIV) and short-flagellated elongated cells (EI) were clearly stained ( B,C). TD09-positive cells in the round stages ranged in diameter from 3.2 μm to 6.2 μm, while the flagella of elongated cells were 5.0 μm or less. 3.2.4. Anti-TD02 Antibody TD02 is another gene highly expressed in Ciona testis, showing homology to a histone H1-like protein involved in chromatin condensation during spermiogenesis . An antibody against TD02 stained the nucleus of late spermatids and mature sperm in the central area of the testicular follicle near the efferent duct ( A). In isolated cells, the antibody recognized the nucleus of late spermatids and mature sperm but did not stain early spermatids (E1), even those with long flagella ( B–D). The appearance of the TD02 protein coincided with the onset of nuclear elongation, which began in spermatids with flagella measuring ~5.8 μm. Spermatids with flagella longer than 28 μm were all stained by the anti-TD02 antibody. 3.2.5. Anti-Acetylated α-Tubulin Antibody Post-translational modifications of tubulins, such as acetylation, are critical for the formation and regulation of ciliary axonemes . The acetylated α-tubulin antibody is commonly used to detect cilia and flagella . The immunofluorescence analysis of testis sections showed strong staining in the central area of the testicular follicle, similar to anti-TD02 staining, but with a broader distribution, including non-flagellated spermatogenic cells ( A). In isolated cells, the antibody stained the cytoplasm of non-flagellated spermatids ( B). Approximately 75% of round-stage cells, with diameters ranging from 2.1 μm to 3.9 μm (RIII), were stained by the antibody. The flagella of spermatids and mature sperm were strongly stained ( C), consistent with previous findings in Ciona sperm . 3.3. Observation by Thin-Sectioned Electron Microscopy Thin-sectioned electron microscopy confirmed the non-cystic distribution of spermatogenic cells in a testicular follicle ( A). We also confirmed the presence of free cells outside the follicle, which was recognized by the MIS H300 antibody . These cells are distinctly different from spermatogenic cells in that they have a nucleus with many spotted heterochromatin regions ( A,B). Some of these cells appeared to be tightly associated with the lamina of testicular follicles ( C,D). We also found the presence of ciliated cells in the epithelia of testicular follicles. The cilia were long and protruded from flat epithelial cells. A long striated rootlet extended from the basal body of a flagellum ( E,F). The cilia possessed motile 9+2 structured axonemes with dynein arms ( G). We initially examined HE-stained paraffin sections of the testes. The Ciona testis consists of numerous club-like compartments known as testicular follicles ( A). Each follicle was densely packed with spermatogenic cells at various developmental stages. As previously observed in Ciona and Botryllus , spermatogenic cells were distributed from larger cells near the basal lamina to smaller, later-stage cells in the center, indicating that the testis is non-cystic rather than cystic. However, the alignment of spermatogenic cells is less strictly organized compared to the mammalian seminiferous tubules. Late spermatids and mature sperm were observed in the region from the center to the efferent duct ( B). Based on cell size ( C,D) and the immunohistochemical characterizations described below, we classified the spermatogenic cells into round stages (RI to RIV) and elongated stages (EI to EIII) . 3.2.1. MIS H-300 Antibody Sertoli cells are essential for the growth and differentiation of spermatogenic cells in mammalian testes. However, previous studies on Boltenia villosa suggested the absence of Sertoli-like cells in ascidians . To investigate the presence of Sertoli-like cells and potential interactions between germ cells and somatic cells in Ciona , we used the MIS H-300 antibody, which targets Müllerian Inhibiting Substance (MIS), a TGFβ superfamily member secreted by Sertoli cells in vertebrates. Immunofluorescence staining revealed that free cells located between testicular follicles were positively stained with the MIS H-300 antibody ( A), while no spermatogenic cells within the follicles were stained. Further analysis of isolated cells showed that the signal was localized in the cytoplasm of these free cells, particularly near the nucleus ( B). 3.2.2. Anti-CiVH Antibody Vasa protein is a marker for primordial germ cells and spermatogonia. The anti-CiVH antibody, a monoclonal antibody that recognizes primordial germ cells and early-stage germ cells in Ciona juveniles , was used to stain adult Ciona testes. Immunofluorescence revealed a few large cells along the walls of the testicular follicles ( A). These isolated cells appeared round, with a diameter greater than 7.3 μm, and the immunofluorescence signal was present throughout the cytoplasm but not in the nucleus ( B). 3.2.3. Anti-TD09 Antibody The TD09 gene, highly expressed in Ciona testis, is homologous to the synap-tonemal complex protein SYCP3, which is expressed in meiotic cells . An an-tibody developed against TD09 recognized broad areas of the testis section, excluding the region containing mature sperm near the efferent duct ( A). In isolated spermatogenic cells, the nuclei of round-stage cells (RII to RIV) and short-flagellated elongated cells (EI) were clearly stained ( B,C). TD09-positive cells in the round stages ranged in diameter from 3.2 μm to 6.2 μm, while the flagella of elongated cells were 5.0 μm or less. 3.2.4. Anti-TD02 Antibody TD02 is another gene highly expressed in Ciona testis, showing homology to a histone H1-like protein involved in chromatin condensation during spermiogenesis . An antibody against TD02 stained the nucleus of late spermatids and mature sperm in the central area of the testicular follicle near the efferent duct ( A). In isolated cells, the antibody recognized the nucleus of late spermatids and mature sperm but did not stain early spermatids (E1), even those with long flagella ( B–D). The appearance of the TD02 protein coincided with the onset of nuclear elongation, which began in spermatids with flagella measuring ~5.8 μm. Spermatids with flagella longer than 28 μm were all stained by the anti-TD02 antibody. 3.2.5. Anti-Acetylated α-Tubulin Antibody Post-translational modifications of tubulins, such as acetylation, are critical for the formation and regulation of ciliary axonemes . The acetylated α-tubulin antibody is commonly used to detect cilia and flagella . The immunofluorescence analysis of testis sections showed strong staining in the central area of the testicular follicle, similar to anti-TD02 staining, but with a broader distribution, including non-flagellated spermatogenic cells ( A). In isolated cells, the antibody stained the cytoplasm of non-flagellated spermatids ( B). Approximately 75% of round-stage cells, with diameters ranging from 2.1 μm to 3.9 μm (RIII), were stained by the antibody. The flagella of spermatids and mature sperm were strongly stained ( C), consistent with previous findings in Ciona sperm . Sertoli cells are essential for the growth and differentiation of spermatogenic cells in mammalian testes. However, previous studies on Boltenia villosa suggested the absence of Sertoli-like cells in ascidians . To investigate the presence of Sertoli-like cells and potential interactions between germ cells and somatic cells in Ciona , we used the MIS H-300 antibody, which targets Müllerian Inhibiting Substance (MIS), a TGFβ superfamily member secreted by Sertoli cells in vertebrates. Immunofluorescence staining revealed that free cells located between testicular follicles were positively stained with the MIS H-300 antibody ( A), while no spermatogenic cells within the follicles were stained. Further analysis of isolated cells showed that the signal was localized in the cytoplasm of these free cells, particularly near the nucleus ( B). Vasa protein is a marker for primordial germ cells and spermatogonia. The anti-CiVH antibody, a monoclonal antibody that recognizes primordial germ cells and early-stage germ cells in Ciona juveniles , was used to stain adult Ciona testes. Immunofluorescence revealed a few large cells along the walls of the testicular follicles ( A). These isolated cells appeared round, with a diameter greater than 7.3 μm, and the immunofluorescence signal was present throughout the cytoplasm but not in the nucleus ( B). The TD09 gene, highly expressed in Ciona testis, is homologous to the synap-tonemal complex protein SYCP3, which is expressed in meiotic cells . An an-tibody developed against TD09 recognized broad areas of the testis section, excluding the region containing mature sperm near the efferent duct ( A). In isolated spermatogenic cells, the nuclei of round-stage cells (RII to RIV) and short-flagellated elongated cells (EI) were clearly stained ( B,C). TD09-positive cells in the round stages ranged in diameter from 3.2 μm to 6.2 μm, while the flagella of elongated cells were 5.0 μm or less. TD02 is another gene highly expressed in Ciona testis, showing homology to a histone H1-like protein involved in chromatin condensation during spermiogenesis . An antibody against TD02 stained the nucleus of late spermatids and mature sperm in the central area of the testicular follicle near the efferent duct ( A). In isolated cells, the antibody recognized the nucleus of late spermatids and mature sperm but did not stain early spermatids (E1), even those with long flagella ( B–D). The appearance of the TD02 protein coincided with the onset of nuclear elongation, which began in spermatids with flagella measuring ~5.8 μm. Spermatids with flagella longer than 28 μm were all stained by the anti-TD02 antibody. Post-translational modifications of tubulins, such as acetylation, are critical for the formation and regulation of ciliary axonemes . The acetylated α-tubulin antibody is commonly used to detect cilia and flagella . The immunofluorescence analysis of testis sections showed strong staining in the central area of the testicular follicle, similar to anti-TD02 staining, but with a broader distribution, including non-flagellated spermatogenic cells ( A). In isolated cells, the antibody stained the cytoplasm of non-flagellated spermatids ( B). Approximately 75% of round-stage cells, with diameters ranging from 2.1 μm to 3.9 μm (RIII), were stained by the antibody. The flagella of spermatids and mature sperm were strongly stained ( C), consistent with previous findings in Ciona sperm . Thin-sectioned electron microscopy confirmed the non-cystic distribution of spermatogenic cells in a testicular follicle ( A). We also confirmed the presence of free cells outside the follicle, which was recognized by the MIS H300 antibody . These cells are distinctly different from spermatogenic cells in that they have a nucleus with many spotted heterochromatin regions ( A,B). Some of these cells appeared to be tightly associated with the lamina of testicular follicles ( C,D). We also found the presence of ciliated cells in the epithelia of testicular follicles. The cilia were long and protruded from flat epithelial cells. A long striated rootlet extended from the basal body of a flagellum ( E,F). The cilia possessed motile 9+2 structured axonemes with dynein arms ( G). Prior to this study, no characterization of spermatogenic cells in the ascidian Ciona robusta had been conducted at the molecular level. Here, we characterized these cells using antibodies against acetylated tubulin, Müllerian Inhibiting Substance (MIS), Vasa homolog, and the laboratory-made testis-specific proteins TD02 and TD09. Spermatogenic cells were classified into four round stages (RI to RIV) and three elongated stages (EI to EIII) . Based on previous morphological studies , cells in stages RI, RII/RIII, and RIV/EI/EII/EIII are identified as spermatogonia, spermatocytes, and spermatids, respectively. These antibodies clearly recognized structures specific to each stage, demonstrating that the method described here can be useful for further characterization of spermatogenic cells. We further characterized the distribution of each spermatogenic cell type in the non-cystic testis of Ciona . Unlike cystic testes, where cells are encapsulated, the spermatogenic cells in Ciona are distributed as a mass within testicular follicles, with no clear boundaries between cell stages. Some cell populations are dispersed into adjacent areas. In vertebrates, the transition from lobule-type to tubule-type testes is thought to have occurred during evolution . Lobule-type testes, seen in anamniotes such as fish and frogs, are characterized by cystic compartments that house specific stages of spermatogenic cells, each enclosed by Sertoli cells. In contrast, spermatogenesis in amniotes occurs within seminiferous tubules, where Sertoli cells are associated with spermatogenic cells throughout their development . The lobule-type testis can be further divided into restricted and unrestricted lobules . Three types of testes are observed in teleosts: tubule-type, restricted-lobule type, and unrestricted-lobule type. Lower teleosts, such as salmonids, exhibit tubular testis types, while higher teleosts have unrestricted lobular testes . The arrangement of spermatogenic cells in the testicular follicles of Ciona resembles the tubular type found in lower fishes and vertebrates. Spermatogenesis in ascidians may provide key insights into the evolutionary emergence of cystic formation in testes. We identified MIS-positive cells located outside the testicular follicles in Ciona . In lobule-type fish testes, the cystic formation of spermatogenic cells is thought to be supported by epithelial Sertoli-like cells . In Ciona , these MIS-positive cells are scattered between the testicular follicles, likely corresponding to cells described in early gonad formation . In sea urchins, another marine invertebrate, spermatogonia are located at the periphery of the gonad, and spermatogenesis proceeds from the epithelial basal wall toward the lumen in association with nutritive phagocytes . It is possible that a primitive form of Sertoli cells, similar to nutritive phagocytes, emerged to support spermatogenesis before becoming fully associated with spermatogenic cells within cysts. The Ciona testicular follicle may represent a transitional stage from unorganized cell arrangements to a Sertoli-cell-assisted tubular testis structure. Further studies on the spatial organization and functional specialization of Sertoli cells in Ciona could shed light on an important event in the evolutionary transition from invertebrates to vertebrates within the chordate lineage. We observed strong cytoplasmic staining with anti-acetylated tubulin antibody in round spermatid stages. Tubulin acetylation, which stabilizes microtubules, is typically localized in centrioles and ciliary axonemes. The cytoplasmic acetylation of tubulin was first reported in Chlamydomonas and has since been observed during specific cell cycle stages in most cells, including axonal microtubules in neurons and certain microtubule structures in cancer cells . In Drosophila spermatogenesis, which involves cytoplasmic ciliogenesis , acetylated tubulin is found in short elongating axonemes in the cytoplasm during the round spermatid stage , similar to our observations in Ciona testes. While the current study does not definitively show cytoplasmic ciliogenesis during Ciona spermatogenesis, it is possible that tubulin acetylation occurs prior to axoneme formation. Future studies should clarify whether cytoplasmic tubulin acetylation is associated with the pre-formation of doublet microtubules or is a prerequisite for their assembly from the basal body. In this study, we classified spermatogenic cells in the Ciona robusta testis into seven distinct stages using immunostaining with several antibodies. The MIS H-300 antibody detected free cells between the testicular follicles, which are likely Sertoli-like cells that potentially support the differentiation of spermatogenic cells. Acetylated tubulins were identified not only in the flagella of elongating spermatids and mature sperm but also in round spermatids prior to flagellar elongation. This suggests that tubulin acetylation may play a role in the cytoplasmic formation of doublet microtubules or axonemes before the onset of flagellation. These findings provide new insights into the mechanisms of spermatogenesis and offer clues about the evolutionary transition of testicular structures in animals. |
3e3696ba-33bf-4076-bdbc-96fe471dff0b | 11495376 | Microbiology[mh] | Carbon accounts for about half of the dry weight of microbial cells ; therefore, carbon source is the most needed nutrient and also an energy source, which is the key and prerequisite for the survival of microorganisms. The different abilities of microbes to utilize organic carbon sources determine their survival differences and ecological niches in the environment . In the natural environment, low molecular weight sugars, organic acids, amino acids, and fatty acids are generally considered to be the most common components in the organic carbon pool , which are widely used by most microbial species. Their metabolic pathways have been well studied and have become classical biochemical pathways. In addition to these traditional carbon sources, there are likely many other important components that play a role in driving the exchange between microorganisms and the environment. For example, the natural product trans -aconitic acid (TAA) , the geometric isomer of cis -aconitic acid (CAA) in the tricarboxylic acid (TCA) cycle, is one such under-appreciated carbon source. In terms of molecular property of TAA, due to its structural similarity to CAA, TAA can competitively inhibit aconitase , an enzyme that uses CAA as both a product and a substrate in the TCA cycle. However, in nature, TAA can be synthesized and secreted by plants as components of root exudates . For example, in barley ( Hordeum leporinum ), reed grass ( Phalaris tuberose ), and western larkspur ( Delphinium hesperium ), TAA concentrations reach astonishing levels of 3.5%, 4.2%, and 12.2% of dry weight, respectively ; in maize, TAA accounts for 95% of total aconitic acid and often reaches milligram of fresh weight . Bacteria also synthesize TAA. For example, in the nematode pathogen Bacillus thuringiensis , TAA can be synthesized as one of the main nematicidal toxins and accumulated extracellularly . From 1961 to the present, research on TAA has focused almost exclusively on revealing its remarkable natural occurrence and biological effects, such as killing of plant-parasitic nematodes , protection against brown planthopper feeding , inhibition of growth and transformation of Leishmania parasites , and anti-inflammatory effects on various mammalian diseases . However, as a naturally occurring high-yield molecule, TAA was known to be metabolized by bacteria as early as 1961 , but unfortunately, few studies have explored how TAA is metabolized, whether this ability is widespread in bacteria, and whether it is active in mediating interactions with the environment and microbiota. B. velezensis FZB42 has become a model bacterium for the study of microbe-environment interaction and is also a successful biofertilizer and antimicrobial strain used in agriculture . In this work, we found that FZB42 has a strong and efficient TAA assimilation ability and encodes a metabolic regulatory system that can sense and assimilate TAA. The system consists of a TAA assimilation-related ( tar ) operon and a regulatory gene tarR . The TarR protein receives the TAA signal and then activates the expression of the tar operon, which includes aconitate isomerase TarA (a core enzyme that converts TAA) and a TAA importer protein TarB. Further, the TAA assimilation system is found to provide significant growth and competitive advantages to host bacteria when colonized in soil and is active and widespread in the bacterial domain and natural ecosystems. Therefore, this study reveals the role of TAA molecules as important carbon sources in nature and the role of their catabolism in bacterial carbon acquisition and microbe-environment crosstalk, which will provide new perspectives for understanding the metabolic adaptations of microorganisms and designing tailored interactions between specific microbes and target environments.
Bacterial strains, plasmids, and culture conditions The bacterial strains and plasmids used in this study are listed in . For general cell propagation, Escherichia coli and B. velezensis strains were cultured in lysogeny broth (LB) medium at 37°C. For the TAA assimilation ability test, strains were collected in the logarithmic growth phase to prevent spore formation and false-negative results caused by dormancy, washed with ultrapure water to remove residual nutrients, and then inoculated into minimal ACO medium (7.5 g/L TAA, 2 g/L (NH 4 ) 2 SO 4 , 1 g/L K 2 HPO 4 , 0.5 g/L MgSO 4 , and 0.1 g/L FeCl 3 ·6H 2 O, pH 7.0) , or modified ACO medium (containing 1 g/L glucose and 2 g/L TAA). Antibiotics were added at appropriate final concentrations: 100 μg/ml ampicillin, 25 μg/ml erythromycin, 30 μg/ml (for E. coli ) and 5 μg/ml (for B. velezensis ) chloramphenicol, 50 μg/ml (for E. coli ) and 10 μg/ml (for B. velezensis ) kanamycin. Gene deletion and complementation in B. velezensis FZB42 To construct the tar gene deletion and complementation vectors, Cm and Kan resistance cassettes ( and ) were used to construct the corresponding pΔ tar and pC tar vectors based on the pMD19-T vector, respectively. In the gene complementation, the α-amylase gene locus ( amyE , RBAM_RS01650 ) of the FZB42 genome was used to integrate the intact tar gene. Transformation of B. velezensis was conducted as previously described . Generally, B. velezensis cells were cultured in 20 ml of SPI medium (0.19% (NH 4 ) 2 SO 4 , 1.36% K 2 HPO 4 ·3H 2 O, 0.58% KH 2 PO 4 , 0.10% Trisodium citrate dihydrate, 202 μl 5% MgSO 4 ·7H 2 O, 202 μl 50% w/v glucose, and 202 μl 1% w/v CAYE) at 37°C, 200 rpm, to an OD 600 value of 1.4–1.5, and 2.5 ml of the culture was then inoculated into 20 ml of SPII medium (19.60 ml SPI, 200 μl 50 mM CaCl 2 , and 200 μl 250 mM MgCl 2 ). After culturing at 37°C and 100 rpm for 1.5 h, cells were induced with 250 μl of 10 mM EGTA for 10 min to make the cells competent. The linearized vector of pΔ tar or pC tar digested with Bam HI was added and cultured at 37°C and 100 rpm for 45 min. Then 800 μl of LB medium was added and cultured at 37°C and 200 rpm for 2 h. The cells were then spread on resistant plates for transformant screening. RNA manipulation for TarR regulatory assays Total RNA of Bacillus strains cultured in modified LB liquid medium (5 g/L peptone, 5 g/L NaCl, 2.5 g/L yeast extract, with or without 7.5 g/L TAA, pH 7.0) for 20 h was extracted, digested, and reversely transcribed as previously described . For reverse transcription PCR (RT-PCR) and quantitative real-time PCR (qRT-PCR) assays, the internal reference gene 23S rRNA and the 2 -ΔΔCt method were used. To determine the transcription start site (TSS), a terminal deoxynucleotidyl transferase-based 5′-RACE experiment was conducted as previously described . Expression and purification of TarA, TarB, and TarR proteins For purification of TarB membrane protein, a transformed E. coli C43(DE3) strain was induced with isopropyl-β-D-thiogalactoside (IPTG) at a final concentration of 0.5 mM, harvested after 8 h of growth at 37°C, and resuspended in buffer A (25 mM Tris-HCl, 150 mM NaCl, and 10 mM imidazole, pH 7.6). After high-pressure homogenization, the supernatant was obtained by centrifugation at 34 570 g for 1 h at 4°C. The precipitate was collected by ultracentrifugation (Beckman Coulter Optimal XE-100, USA) at 200 000 rpm for 1 h, dissolved in buffer A (1% w/v n -dodecyl β-D-maltoside), stirred for 5 h, and centrifuged at 34 570 g to obtain the supernatant containing TarB, which was then purified by Ni-NTA affinity , quickly concentrated by ultrafiltration (10 kDa), and further purified by size exclusion chromatography using a Superdex 200 Increase 10/300 GL column. Except for TarB, all other proteins prepared in this work were induced in E. coli Rosetta(DE3) with 0.2 mM IPTG (final concentration) at 16°C overnight and purified by Ni-NTA affinity column. For Western blot assays, cell lysis and centrifugation methods were similar to those described above. The supernatant and pellet collected after ultracentrifugation at 200 000 rpm (Beckman Coulter Optimal XE-100, USA) for 1 h were samples of the cytoplasm fraction and membrane fraction, respectively. Enzymatic assays of AI activity Chemical standards of CAA and TAA (purity >98%) were purchased from Aladdin (Shanghai, China) and Tokyo Chemical Industry (Japan), respectively. During the preparation of the reaction system, the final concentration of each test enzyme was adjusted and normalized to 1 μM. When TarA activity was set to 100%, the relative AI activities of 16 TarA homologs were calculated based on the ratio of the peak area of product formed by TarA homologues to the peak area of product formed by TarA. Fluorescence microscopy The coding sequences (CDSs) of tarB and gfp digested from plasmid pAD43-25 at the Xba I and Hin dIII sites and connected by a 10-aa peptide linker coding sequence were spliced by overlap extension PCR (SOE-PCR), inserted into pAD43-25, and transformed into FZB42 to generate the recombinant FZB42-TarB-GFP strain. The strain FZB42-GFP containing the empty pAD43-25 plasmid was used as a control. Cell preparation and image processing were conducted using a Nikon structured illumination super-resolution microscope (N-SIM; Nikon Corporation, Japan) as previously described . Assays of B. velezensis FZB42 and Δ tarB in modified ACO liquid medium FZB42 and Δ tarB cultures were adjusted to a final OD 600 value of 0.01, inoculated into modified ACO liquid medium, and grown at 37°C for 48 h. Samples were taken every 2 h, and the OD 600 value of cell growth was determined using a microplate reader, and the residual levels of glucose and TAA in the bacterial supernatants were determined using a Glucose Content Assay kit (Biosharp, China) and as previously described , respectively. Meanwhile, samples were taken at 10 h, washed three times with ultrapure water, ground with liquid nitrogen, and then dissolved in 1 ml ultrapure water. After centrifugation at 17 220 g for 15 min, the supernatant was used for LC-Q-TOF-MS detection of intracellular TAA as previously described . Microscale thermophoresis (MST) The procedure was done according to the instructions of the Monolith His-Tag labeling kit (NanoTemper Technologies, Germany). Generally, TarB or TarR protein (800 nM) and dye (100 nM) were mixed evenly, incubated for 30 min, and then centrifuged to obtain the supernatant. One micromolar TAA ligand was set as the highest concentration, followed by the preparation of 16 gradient concentrations. Finally, 10 μl of labeled protein was added to each tube and loaded onto a model with a standard capillary. The scanning parameters are MST Power: 40% and LED Power: 80%. Fluorescence was measured using a Monolith NT.115 and data analyzed using MO. Affinity Analysis V2.3 * (NanoTemper Technologies, Germany). Electrophoretic mobility-shift assay (EMSA) The promoter DNA was fluorescently labeled and purified using a Nucleic Acid Purification kit (Axygen, USA). The working concentrations of DNA probes and TarR protein are detailed in the corresponding figures. To determine the role of the “TTATAA” sequence in the P tar -TarR interaction, “TTATAA” was mutated to “CCGCGG,” and the P tar DNA mutants were de novo synthesized (GenScript Biotech Corporation, China). The procedures for EMSA were conducted as previously described . DNA foot-printing assay In a 200-μl reaction system, 1 000 ng of 5′-FAM-labeled P tar DNA and 5 μM TarR (final concentration) was mixed (in the control experiment, bovine serum albumin was used instead of TarR) in 10 mM Tris-HCl (pH 7.8) with 10 mM MgCl 2 , 1 mM CaCl 2 , 0.4 mM dithiothreitol, 100 mM KCl, and 5% glycerol, and incubated at room temperature. After 30 min, 0.5 U of RNase-free DNase I (Roche, Basel, Switzerland) was added for digestion at 25°C for 3 min. The reaction termination, precipitation, and analysis procedures were done as previously described . Growth assays in soil Soil was collected from the vegetable field of Hubei Engineering University (30°94′N and 113°91′E), subjected to physical and chemical characterization , thoroughly crushed and passed through a 40-mesh sieve, and finally sterilized at 121°C for 1 h. Twenty grams of sterile soil was inoculated in a sterile Petri dish, and 10 7 cfu/g of bacterial suspension (FZB42 or Δ tarA vegetative cells, cultured at 37°C for 9 h, washed and resuspended in sterile ultrapure water, adjusted to the same cell turbidity). Then, 0 (control group) and 100 mg/kg (test group) of TAA aqueous solution (pH 7.0) were added to the soil. The soil was further cultured at 25°C and 40% relative humidity. One gram of soil was sampled from each treatment every 1–5 days and shaken in 9 ml of sterile ultrapure water at 25°C and 200 rpm for 20 min. The supernatant was diluted 10-fold in a gradient, spread on LB plates, incubated at 30°C overnight, and then counted. Competition assays in soil FZB42 or Δ tarA cultures were adjusted to a cell density of 10 7 cfu/ml using turbidimetry, then mixed in equal volumes and inoculated into 20 g of sterile soil pre-mixed with 100 mg/kg of TAA (test group) and another 20 g of sterile soil without TAA addition (control group). The procedures for culturing and sample preparation were described above. Δ tarA colonies were picked using Cm r selective plates (5 μg/ml) and counted as previously described . Phylogenetic analysis of TAA assimilation genes in bacterial taxa For genomic data acquisition, a total of 16 986 complete bacterial genomes were obtained from the RefSeq database (updated to March 2023) of NCBI (National Center for Biotechnology Information). To search for homologous proteins of TarA, TarB, and TarR in bacterial genomes, we used the hmmsearch tool of the HMMER (Version 3.3.2) software package to create an hmm seed file. The specific operations are as follows: Taking TarA, TarB, and TarR protein sequences as query sequences, blastp (Version 2.13.0+) was used to search for homologous proteins in the NCBI database, and the 100 sequences with the highest scores were selected as original sequences. These sequences were then aligned using MAFFT software (Version 7.505), and the hmmbuild tool was used to create an hmm file. Next, blastp was used to align the potential Tar proteins screened from the genomes with the corresponding TarA, TarB, and TarR protein sequences, and proteins with a similarity >30% were defined as Tar homologous proteins. To construct the phylogenetic tree of TarA and its homologous proteins, all TarA homologous proteins were first clustered at 50% sequence identity using the CD-HIT tool (Version 4.8.1). Then, the protein sequences were aligned using MAFFT software, and the maximum likelihood trees were constructed using Fasttree (Version 2.1.11). Finally, the phylogenetic trees were modified using the ggtree package (Version 3.8.2) in the R language (Version 4.3.1). Isolation and identification of TAA-assimilating environmental bacteria A total of 32 environmental samples were collected from Xiaogan city (32°92′N and 113°91′E) and Wuhan city (30°48′N and 114°37′E) in Hubei Province, China . One gram or 1 ml of environmental sample was inoculated into 20 ml of LB liquid medium. After vigorous growth at 30°C for 2 h, 1 ml of bacterial supernatant was spread on ACO plates and grown at 30°C for 24 h. Colonies were picked and sent to Tsingke Biotechnology Co., Ltd (China) for species identification. Identification of AI-encoding genes in TAA-assimilating bacteria Twelve out of sixty-six bacterial isolates representing different living environments, including plant-associated (beneficial and pathogenic), animal-associated (pathogenic), and free-living (neutral), were selected for whole-genome sequencing by Bioyi Biotechnology Co., Ltd (China). Each genome sequence of the 12 strains was aligned with the TarA protein sequence using the tblastn tool to identify AI homologous genes. Statistical analysis All our experiments were conducted in three biological replicates and three technical replicates for each treatment, and values are expressed as mean ± standard deviation. One-way ANOVA was conducted using Tukey’s honest significant difference test with an error probability of P < .01 and P < .0001 to determine statistically significant differences in comparison of AI protein activities and in tar gene expression levels in the TarR regulation studies. IBM SPSS (Statistical Package for the Social Sciences) software (Version 20.0) was used for these analyses.
The bacterial strains and plasmids used in this study are listed in . For general cell propagation, Escherichia coli and B. velezensis strains were cultured in lysogeny broth (LB) medium at 37°C. For the TAA assimilation ability test, strains were collected in the logarithmic growth phase to prevent spore formation and false-negative results caused by dormancy, washed with ultrapure water to remove residual nutrients, and then inoculated into minimal ACO medium (7.5 g/L TAA, 2 g/L (NH 4 ) 2 SO 4 , 1 g/L K 2 HPO 4 , 0.5 g/L MgSO 4 , and 0.1 g/L FeCl 3 ·6H 2 O, pH 7.0) , or modified ACO medium (containing 1 g/L glucose and 2 g/L TAA). Antibiotics were added at appropriate final concentrations: 100 μg/ml ampicillin, 25 μg/ml erythromycin, 30 μg/ml (for E. coli ) and 5 μg/ml (for B. velezensis ) chloramphenicol, 50 μg/ml (for E. coli ) and 10 μg/ml (for B. velezensis ) kanamycin.
B. velezensis FZB42 To construct the tar gene deletion and complementation vectors, Cm and Kan resistance cassettes ( and ) were used to construct the corresponding pΔ tar and pC tar vectors based on the pMD19-T vector, respectively. In the gene complementation, the α-amylase gene locus ( amyE , RBAM_RS01650 ) of the FZB42 genome was used to integrate the intact tar gene. Transformation of B. velezensis was conducted as previously described . Generally, B. velezensis cells were cultured in 20 ml of SPI medium (0.19% (NH 4 ) 2 SO 4 , 1.36% K 2 HPO 4 ·3H 2 O, 0.58% KH 2 PO 4 , 0.10% Trisodium citrate dihydrate, 202 μl 5% MgSO 4 ·7H 2 O, 202 μl 50% w/v glucose, and 202 μl 1% w/v CAYE) at 37°C, 200 rpm, to an OD 600 value of 1.4–1.5, and 2.5 ml of the culture was then inoculated into 20 ml of SPII medium (19.60 ml SPI, 200 μl 50 mM CaCl 2 , and 200 μl 250 mM MgCl 2 ). After culturing at 37°C and 100 rpm for 1.5 h, cells were induced with 250 μl of 10 mM EGTA for 10 min to make the cells competent. The linearized vector of pΔ tar or pC tar digested with Bam HI was added and cultured at 37°C and 100 rpm for 45 min. Then 800 μl of LB medium was added and cultured at 37°C and 200 rpm for 2 h. The cells were then spread on resistant plates for transformant screening.
Total RNA of Bacillus strains cultured in modified LB liquid medium (5 g/L peptone, 5 g/L NaCl, 2.5 g/L yeast extract, with or without 7.5 g/L TAA, pH 7.0) for 20 h was extracted, digested, and reversely transcribed as previously described . For reverse transcription PCR (RT-PCR) and quantitative real-time PCR (qRT-PCR) assays, the internal reference gene 23S rRNA and the 2 -ΔΔCt method were used. To determine the transcription start site (TSS), a terminal deoxynucleotidyl transferase-based 5′-RACE experiment was conducted as previously described .
For purification of TarB membrane protein, a transformed E. coli C43(DE3) strain was induced with isopropyl-β-D-thiogalactoside (IPTG) at a final concentration of 0.5 mM, harvested after 8 h of growth at 37°C, and resuspended in buffer A (25 mM Tris-HCl, 150 mM NaCl, and 10 mM imidazole, pH 7.6). After high-pressure homogenization, the supernatant was obtained by centrifugation at 34 570 g for 1 h at 4°C. The precipitate was collected by ultracentrifugation (Beckman Coulter Optimal XE-100, USA) at 200 000 rpm for 1 h, dissolved in buffer A (1% w/v n -dodecyl β-D-maltoside), stirred for 5 h, and centrifuged at 34 570 g to obtain the supernatant containing TarB, which was then purified by Ni-NTA affinity , quickly concentrated by ultrafiltration (10 kDa), and further purified by size exclusion chromatography using a Superdex 200 Increase 10/300 GL column. Except for TarB, all other proteins prepared in this work were induced in E. coli Rosetta(DE3) with 0.2 mM IPTG (final concentration) at 16°C overnight and purified by Ni-NTA affinity column. For Western blot assays, cell lysis and centrifugation methods were similar to those described above. The supernatant and pellet collected after ultracentrifugation at 200 000 rpm (Beckman Coulter Optimal XE-100, USA) for 1 h were samples of the cytoplasm fraction and membrane fraction, respectively.
Chemical standards of CAA and TAA (purity >98%) were purchased from Aladdin (Shanghai, China) and Tokyo Chemical Industry (Japan), respectively. During the preparation of the reaction system, the final concentration of each test enzyme was adjusted and normalized to 1 μM. When TarA activity was set to 100%, the relative AI activities of 16 TarA homologs were calculated based on the ratio of the peak area of product formed by TarA homologues to the peak area of product formed by TarA.
The coding sequences (CDSs) of tarB and gfp digested from plasmid pAD43-25 at the Xba I and Hin dIII sites and connected by a 10-aa peptide linker coding sequence were spliced by overlap extension PCR (SOE-PCR), inserted into pAD43-25, and transformed into FZB42 to generate the recombinant FZB42-TarB-GFP strain. The strain FZB42-GFP containing the empty pAD43-25 plasmid was used as a control. Cell preparation and image processing were conducted using a Nikon structured illumination super-resolution microscope (N-SIM; Nikon Corporation, Japan) as previously described .
B. velezensis FZB42 and Δ tarB in modified ACO liquid medium FZB42 and Δ tarB cultures were adjusted to a final OD 600 value of 0.01, inoculated into modified ACO liquid medium, and grown at 37°C for 48 h. Samples were taken every 2 h, and the OD 600 value of cell growth was determined using a microplate reader, and the residual levels of glucose and TAA in the bacterial supernatants were determined using a Glucose Content Assay kit (Biosharp, China) and as previously described , respectively. Meanwhile, samples were taken at 10 h, washed three times with ultrapure water, ground with liquid nitrogen, and then dissolved in 1 ml ultrapure water. After centrifugation at 17 220 g for 15 min, the supernatant was used for LC-Q-TOF-MS detection of intracellular TAA as previously described .
The procedure was done according to the instructions of the Monolith His-Tag labeling kit (NanoTemper Technologies, Germany). Generally, TarB or TarR protein (800 nM) and dye (100 nM) were mixed evenly, incubated for 30 min, and then centrifuged to obtain the supernatant. One micromolar TAA ligand was set as the highest concentration, followed by the preparation of 16 gradient concentrations. Finally, 10 μl of labeled protein was added to each tube and loaded onto a model with a standard capillary. The scanning parameters are MST Power: 40% and LED Power: 80%. Fluorescence was measured using a Monolith NT.115 and data analyzed using MO. Affinity Analysis V2.3 * (NanoTemper Technologies, Germany).
The promoter DNA was fluorescently labeled and purified using a Nucleic Acid Purification kit (Axygen, USA). The working concentrations of DNA probes and TarR protein are detailed in the corresponding figures. To determine the role of the “TTATAA” sequence in the P tar -TarR interaction, “TTATAA” was mutated to “CCGCGG,” and the P tar DNA mutants were de novo synthesized (GenScript Biotech Corporation, China). The procedures for EMSA were conducted as previously described .
In a 200-μl reaction system, 1 000 ng of 5′-FAM-labeled P tar DNA and 5 μM TarR (final concentration) was mixed (in the control experiment, bovine serum albumin was used instead of TarR) in 10 mM Tris-HCl (pH 7.8) with 10 mM MgCl 2 , 1 mM CaCl 2 , 0.4 mM dithiothreitol, 100 mM KCl, and 5% glycerol, and incubated at room temperature. After 30 min, 0.5 U of RNase-free DNase I (Roche, Basel, Switzerland) was added for digestion at 25°C for 3 min. The reaction termination, precipitation, and analysis procedures were done as previously described .
Soil was collected from the vegetable field of Hubei Engineering University (30°94′N and 113°91′E), subjected to physical and chemical characterization , thoroughly crushed and passed through a 40-mesh sieve, and finally sterilized at 121°C for 1 h. Twenty grams of sterile soil was inoculated in a sterile Petri dish, and 10 7 cfu/g of bacterial suspension (FZB42 or Δ tarA vegetative cells, cultured at 37°C for 9 h, washed and resuspended in sterile ultrapure water, adjusted to the same cell turbidity). Then, 0 (control group) and 100 mg/kg (test group) of TAA aqueous solution (pH 7.0) were added to the soil. The soil was further cultured at 25°C and 40% relative humidity. One gram of soil was sampled from each treatment every 1–5 days and shaken in 9 ml of sterile ultrapure water at 25°C and 200 rpm for 20 min. The supernatant was diluted 10-fold in a gradient, spread on LB plates, incubated at 30°C overnight, and then counted.
FZB42 or Δ tarA cultures were adjusted to a cell density of 10 7 cfu/ml using turbidimetry, then mixed in equal volumes and inoculated into 20 g of sterile soil pre-mixed with 100 mg/kg of TAA (test group) and another 20 g of sterile soil without TAA addition (control group). The procedures for culturing and sample preparation were described above. Δ tarA colonies were picked using Cm r selective plates (5 μg/ml) and counted as previously described .
For genomic data acquisition, a total of 16 986 complete bacterial genomes were obtained from the RefSeq database (updated to March 2023) of NCBI (National Center for Biotechnology Information). To search for homologous proteins of TarA, TarB, and TarR in bacterial genomes, we used the hmmsearch tool of the HMMER (Version 3.3.2) software package to create an hmm seed file. The specific operations are as follows: Taking TarA, TarB, and TarR protein sequences as query sequences, blastp (Version 2.13.0+) was used to search for homologous proteins in the NCBI database, and the 100 sequences with the highest scores were selected as original sequences. These sequences were then aligned using MAFFT software (Version 7.505), and the hmmbuild tool was used to create an hmm file. Next, blastp was used to align the potential Tar proteins screened from the genomes with the corresponding TarA, TarB, and TarR protein sequences, and proteins with a similarity >30% were defined as Tar homologous proteins. To construct the phylogenetic tree of TarA and its homologous proteins, all TarA homologous proteins were first clustered at 50% sequence identity using the CD-HIT tool (Version 4.8.1). Then, the protein sequences were aligned using MAFFT software, and the maximum likelihood trees were constructed using Fasttree (Version 2.1.11). Finally, the phylogenetic trees were modified using the ggtree package (Version 3.8.2) in the R language (Version 4.3.1).
A total of 32 environmental samples were collected from Xiaogan city (32°92′N and 113°91′E) and Wuhan city (30°48′N and 114°37′E) in Hubei Province, China . One gram or 1 ml of environmental sample was inoculated into 20 ml of LB liquid medium. After vigorous growth at 30°C for 2 h, 1 ml of bacterial supernatant was spread on ACO plates and grown at 30°C for 24 h. Colonies were picked and sent to Tsingke Biotechnology Co., Ltd (China) for species identification.
Twelve out of sixty-six bacterial isolates representing different living environments, including plant-associated (beneficial and pathogenic), animal-associated (pathogenic), and free-living (neutral), were selected for whole-genome sequencing by Bioyi Biotechnology Co., Ltd (China). Each genome sequence of the 12 strains was aligned with the TarA protein sequence using the tblastn tool to identify AI homologous genes.
All our experiments were conducted in three biological replicates and three technical replicates for each treatment, and values are expressed as mean ± standard deviation. One-way ANOVA was conducted using Tukey’s honest significant difference test with an error probability of P < .01 and P < .0001 to determine statistically significant differences in comparison of AI protein activities and in tar gene expression levels in the TarR regulation studies. IBM SPSS (Statistical Package for the Social Sciences) software (Version 20.0) was used for these analyses.
tarA and tarB genes constitute an operon and are responsible for TAA assimilation in B. velezensis FZB42 AI was proposed as a new enzyme as early as 1961 and was considered to be responsible for TAA utilization in bacteria. In 2017, we identified an AI-encoding gene, tbrA (TAA biosynthesis-related gene A); however, it mediates TAA biosynthesis in B. thuringiensis . Although tbrA mediates reversely, it is the only AI sequence available to date. Therefore, we used the TbrA protein sequence to do tblastn analysis against the FZB42 genome, and fortunately found the only homolog of this gene, RBAM_RS16125 (1 131 bp, 376 aa) . Its protein sequence shares 33% identity and 94% coverage with B. thuringiensis TbrA and belongs to the same PrpF superfamily as TbrA after conserved domain (CD) search . Based on these findings, we proposed that RBAM_RS16125 is an AI-encoding gene with the potential to isomerize TAA in B. velezensis and therefore named it TAA assimilation-related gene A ( tarA ). Meanwhile, we noticed that the gene RBAM_RS16130 (1 377 bp, 458 aa) located 43 bp downstream of tarA formed an operon with tarA and determined that the TSS of the operon tar was located 52 bp upstream of the start codon of the tarA CDS . CD search further reveals that RBAM_RS16130 encodes a major facilitator superfamily (MFS) transporter with 12 transmembrane-helices . Considering the domain and function annotations, as well as the similar gene organization to the TAA biosynthesis operon tbr in B. thuringiensis (containing the AI-encoding gene tbrA and the TAA exporter-encoding gene tbrB ), we proposed that RBAM_RS16130 was associated with the intracellular transport of TAA in TAA assimilation and named it tarB . To test whether the tar operon is responsible for TAA assimilation in FZB42, we tested the growth of single gene deletion mutants Δ tarA and Δ tarB separately and found that although they grew normal in liquid LB medium like FZB42 , both mutants could not grow on solid or liquid ACO media . However, complementation of tarA or tarB restored TAA assimilation ability . These results genetically confirmed that the operon composed of tarA and tarB is responsible for the assimilation of the TAA carbon source in B. velezensis FZB42. TarA is an AI in B. velezensis that converts TAA to CAA To confirm the AI activity of the TarA protein, the interconversion activity of the recombinant TarA-His 6 protein was tested in vitro using TAA and CAA as substrates. Distinct formations of CAA product with TAA substrate and TAA product with CAA substrate were detected by both HPLC and LC-Q-TOF-MS , demonstrating that TarA catalyzes the reversible reaction from TAA to CAA, but is much more efficient in forming TAA. These properties are consistent with the classic characteristics of AI enzymes. We further determined the effects of pH, ionic strength, and various metal ions on TarA activity, as well as the K m , v max , k cat , and k cat /K m kinetic constants of the TarA isomerization reaction to further elucidate its catalytic characteristics . The constants measured via Lineweaver–Burk plots are listed in , quantitatively confirming the equilibrium preference of TarA in TAA formation. These results demonstrated that B. velezensis TarA is an AI that can isomerize TAA to CAA. TarB is a membrane importer of TAA in B. velezensis TarB is predicted as a MFS transporter with 12 transmembrane helices via bioinformatics. To verify its function as a cell membrane importer of TAA, the membrane subcellular localization of the TarB protein was first determined. Through a microscope, we observed a strong green fluorescent signal on the cell membrane of the FZB42-TarB-GFP strain, indicating the localization of TarB-GFP fusion protein on the membrane. In addition, only a weak green signal was observed in the cytoplasm due to the dynamic balance of production and loss of TarB-GFP protein . In contrast, as a control protein, GFP was only uniformly distributed in the cytoplasm . Western blot confirmed the presence of TarB-GFP in cell membrane extract . A molecular binding assay of purified TarB protein and TAA substrate by MST further reveals significant ligand dose-dependent changes in fluorescence intensity ( K d = 5.66 ± 1.38 μM) , demonstrating that TarB, a B. velezensis membrane protein, binds to the TAA molecule with moderate strength. Then, the intracellular contents of FZB42 and Δ tarB strains cultured in modified ACO liquid medium were analyzed by LC-Q-TOF-MS. TAA was only detected in FZB42 cells but not in Δ tarB cells , suggesting that TAA is unable to enter the cells when tarB is deleted. From 0 to 6 h of the culture , FZB42 and Δ tarB appear to consume only glucose and not TAA ( and ). From 6 to 10 h, Δ tarB continues to consume glucose that is depleted at 10 h . However, starting at 6 h, FZB42 used TAA and glucose as carbon sources, which are depleted up at 14 h and 10 h , respectively. In contrast, Δ tarB shows a complete impairment in the use of TAA, as Δ tarB is unable to transport TAA into the cells throughout the culture period . Therefore, when the only available carbon source, glucose, was depleted, Δ tarB reaches its maximum OD 600 value at 10 h ; whereas FZB42 underwent a diauxic growth within 10–14 h and reaches a population twice larger than that of Δ tarB , further confirming the TAA-importing function of TarB. Taken together, these results demonstrate that TarB is a membrane importer of TAA molecules in B. velezensis . TAA signaling activated positive control of TAA assimilation by TarR in B. velezensis We noticed that one gene, RBAM_RS16120 (861 bp), annotated as encoding a LysR-type transcriptional regulator (LTTR), is located 173 bp upstream of the AI gene tarA and is transcribed in an opposite direction with tarA . Further secondary structure analysis of the RBAM_RS16120 gene product shows that the amino terminus (1–58 aa) contains a helix-turn-helix (HTH) DNA binding motif, and the carboxyl terminus (91–281 aa) contains a substrate binding region connected to the HTH motif via a 32-aa hinge region (59–90 aa) . This structure exhibits the typical domain organization of bacterial LTTRs . Considering the functional annotation and the sharing of the promoter region with tarA , the RBAM_RS16120 gene was considered to have a potential regulatory function on the TAA assimilation ability in B. velezensis and was therefore named tarR . To test the regulatory function of tarR on TAA assimilation, we compared the growth of the Δ tarR mutant with that of the starting strain FZB42. Although showing a normal growth in LB medium like FZB42, Δ tarR is unable to grow on ACO medium ( and ), implying positive regulation of TAA assimilation by tarR . Then, qRT-PCR analysis shows that in FZB42, when TAA was added, the transcription of tarA and tarB is significantly upregulated , confirming that the function of TAA assimilation in B. velezensis requires TAA molecules for activation. However, when tarR is deleted in FZB42, the expression of the tar operon cannot be activated despite the presence of TAA . Upon complementation with tarR in C tarR , transcription of tarA and tarB restores normal response to the TAA molecule . These results confirm the positive regulation of the inducible expression of tar by TarR and suggest a role for TarR in the reception of TAA signals. Moreover, tarR expression also responds positively to TAA induction . Next, to determine whether TarR functions as a direct regulator of the tar operon, EMSAs were conducted on the TarR protein with tar promoter region ( P tar ), and the 16S rRNA promoter ( P 16S ) region as a control. Specific interactions appear in the TarR- P tar system but not in the control . In the competition assay, when fixed amounts of TarR and labeled probe were incubated with increasing amounts of non-labeled cold probe, the shifted bands gradually disappeared , indicating that the cold probe competitively binds TarR, resulting in reduced amounts of labeled probe-TarR complex. These results show that TarR recognizes and binds P tar specifically. To further define the binding site(s) of TarR in P tar , two independent protected regions are identified by DNase I footprinting assay ( and ), showing a dual-site interaction between TarR and the tar promoter. Within the two regions, three 6-bp palindromic motifs “TTATAA” were identified. Mutation of one or three motifs caused the shifted bands to weaken or even disappear ( and ), demonstrating that “TTATAA” is a key sequence mediating P tar -TarR interaction and that all three palindrome sequences are essential. As mentioned before, the TarR protein belongs to the LysR transcriptional factor family, and its members usually regulate cellular functions by binding to certain small signaling molecules . Furthermore, considering the bioinformatical prediction that the carboxyl domain of the TarR protein is responsible for substrate binding , and the potential regulation of TarR by TAA shown in qRT-PCR analyses , we propose that TAA is a signaling molecule that directly binds to TarR to activate the tar operon. Indeed, both MST and thermal shifting assays reveal a strong interaction between the TAA molecule and TarR protein ( K d = 11.08 ± 4.35 μM), whereas as a negative control, citric acid (C 6 ) displays no binding ( and ). The amino and carboxyl domains of TarR, TarR N1–91 and TarR C59–286 , were further shown to be responsible for P tar -DNA binding and TAA-substrate binding , respectively. We then further conducted an EMSA assay to test the effect of TAA ligands on the binding efficiency of TarR to P tar and found that formation of TarR-DNA complexes is significantly enhanced as the levels of added TAA increase , demonstrating that TAA can stimulate TarR- P tar interaction. In summary, the above results illustrate that the TarR protein first binds to TAA signaling molecules and then binds to P tar to activate the expression of the tar operon. TAA assimilation system is widely present in bacteria To define the phylogenetic distribution of TAA-assimilation function in bacteria, we screened for the presence of tarA , tarB , and tarR homologous genes. Targets containing tar homolog were found to span across 16 bacterial phyla, 973 genera, and 4 570 species in total ( and ). Of the 16 phyla, 11 contain tarRAB -type species that simultaneously harboring tarA , tarB , and tarR homologs (1 330 in total) ( and ), indicating that these species have a complete assimilation system. Of the 11 phyla, Pseudomonadota , Actinomycetota , Bacillota , Campylobacterota , Bacteroidota , Thermodesulfobacteriota , Cyanobacteriota , and Spirochaetota are the top eight phyla with the largest number of target species among all the 16 phyla . Meanwhile, we noticed that among Pseudomonadota , Actinomycetota , Bacillota , Bacteroidota , and Cyanobacteriota phyla, 142 species are of tarAB -type containing only ( tarA + tarB ) homologs, namely, with complete isomerization and transport functional organization ( and ). For the remaining five of the 16 phyla, the target species are not only few in number but also belong to the tarA -type containing only unique tarA homologs ( and ). Despite lacking a transport element, these bacteria are still considered to have the potential to use TAA. First, they encode the key element of the AI enzyme in the TAA assimilation process. Second, the lack of tarB homologs may be due to the high diversity of MFS sequences between and even within subfamilies . There may be importers in alternative subfamilies with low similarity to TarB. To further reveal the coverage of TAA assimilation ability within a phylum, the analysis of the ratio of target species within a phylum to total sequenced species of the phylum is also shown . In the top 1 phylum Pseudomonadota , more than half of the species (3 411 target /6 600 total ) encode AI ( and ), and more than 31.1% (1 060/3 411) further encode tarB or even tarR alternatives , indicating that Pseudomonadota is a representative phylum that generally adopts TAA assimilation strategy; in Actinomycetota and Campylobacterota , the above ratios are 22.9% (813 target /3 553 total ) and 34.4% (280/813), 22.2% (39 target /176 total ) and 48.7% (19/39), respectively ( and ), which also indicate that TAA assimilation function is dominant in these phyla. Detailed information on the classification, homology type, and protein sequence identities of all bacterial targets are listed in . TAA-assimilating bacteria are also widely distributed in the natural environment We collected 32 samples to explore the distribution of TAA assimilation bacteria in the natural environments and found that many TAA-utilizing bacteria could be isolated from these samples, yielding a total of 77 strains, and 66 of them were identified . These bacteria belong to three phyla, namely Pseudomonadota , Bacillota , and Actinomycetota , involving 12 genera and 27 species, which successfully verified the widespread nature of TAA assimilation species in bacteria taxa. These isolates are particularly concentrated in the phylum Pseudomonadota (46/66, 69.7%), which ranks first in the phylogenetic results and is consistent with the early reports that identified many Pseudomonas spp. using TAA molecules . However, our isolation results differ from that report, which indicated that Gram-positive bacteria (involving only two phyla of Bacillota and Actinomycetota ) rarely use TAA. In our work, we found that 29.6% (8/27) of the isolated species are Gram-positive, which may reflect that uptake of TAA carbon sources by Gram-positive bacteria in nature is real. Subsequently, the genomes of 12 representatives of the 66 strains were analyzed by tblastn. These strains differ in species, living environments, and lifestyles, including plant- and animal- pathogenic , plant-beneficial , and free-living . Except for strain 56 ( Pantoea agglomerans sp.) and strain 60 ( Serratia marcescens sp.), the two species newly identified in this work that were able to use TAA but did not have TarA-homologs, the remaining 10 strains were found to have at least one homologous sequence . In total, we obtained 16 candidate sequences from 10 strains and found that 70% of the strains (7/10; strain 7, 16, 23, 32, 41, 68, and 69) encode active AI enzymes, and two of them, AI-7-1 ( ** P = .008) and AI-16-1 ( ** P = .004) in strain 7 and strain 16, exhibit significantly higher catalytic efficiency than TarA of FZB42 . In comparison, the forward or reverse isomerization activities of AI-4 (3.5%, 8.6%), AI-22 (8.2%, 24.7%), and AI-79 (0.7%, 0.4%) proteins are quite low , although strains 4 ( Aeromonas veronii sp.), strain 22 ( Acinetobacter pittii sp.), and strain 79 ( Providencia rettgeri sp.) encoding these genes grew well on the ACO plate , implying that weak enzymatic performance may be sufficient for survival in nature, or these AIs may require special biochemical conditions for catalysis, or these bacteria may encode unique and unknown AI sequence(s). Together, these results suggest that TAA assimilation was active and prevalent in bacterial carbon utilization and underscore the importance of TAA as a natural carbon nutrient. Growth advantage of TAA-assimilating B. velezensis strain in soil Based on experimental results, we deduced that the function of the TAA assimilation system is to confer a survival advantage to assimilators in natural environments where TAA is present. To test this hypothesis, we examined the growth rate and final populations of the TAA-assimilating strain FZB42 and the non-assimilating strain Δ tarA in sterile soil (naturally TAA-absent, ) without other competing microbiota. This approach allowed us to focus solely on the effects of TAA assimilation ability on competing bacteria. When examined individually in soils without TAA addition, the two strains show indistinguishable growth rates and similar final population levels over the course of the tests (30 days) . However, when TAA was added and present in soils, the TAA assimilation strain FZB42 reproducibly grew to 2–3 times the size of the non-assimilation strain , suggesting that TAA-assimilating strain has a considerable survival advantage in the presence of TAA. Competition between TAA-assimilating and non-assimilating B. velezensis strains in soil To further determine the competitive advantage conferred by the TAA assimilation system in FZB42 when competing with Δ tarA during survival in soil, cultures of both strains were mixed at a population ratio of 1:1 and inoculated into sterile soils with or without TAA. In the absence of TAA, neither strain becomes dominant , and the distribution of strains in the total bacterial population remains ~1:1 throughout the experiment . However, when TAA was added in the soils, strains utilizing TAA became the dominant members of the test group , as the ratio favors the FZB42 assimilator (from 0.94 to 3.86 during 0–19 days) . This confirms that bacteria equipped with a TAA-assimilation system have a clearly competitive advantage in environments where TAA is present.
and tarB genes constitute an operon and are responsible for TAA assimilation in B. velezensis FZB42 AI was proposed as a new enzyme as early as 1961 and was considered to be responsible for TAA utilization in bacteria. In 2017, we identified an AI-encoding gene, tbrA (TAA biosynthesis-related gene A); however, it mediates TAA biosynthesis in B. thuringiensis . Although tbrA mediates reversely, it is the only AI sequence available to date. Therefore, we used the TbrA protein sequence to do tblastn analysis against the FZB42 genome, and fortunately found the only homolog of this gene, RBAM_RS16125 (1 131 bp, 376 aa) . Its protein sequence shares 33% identity and 94% coverage with B. thuringiensis TbrA and belongs to the same PrpF superfamily as TbrA after conserved domain (CD) search . Based on these findings, we proposed that RBAM_RS16125 is an AI-encoding gene with the potential to isomerize TAA in B. velezensis and therefore named it TAA assimilation-related gene A ( tarA ). Meanwhile, we noticed that the gene RBAM_RS16130 (1 377 bp, 458 aa) located 43 bp downstream of tarA formed an operon with tarA and determined that the TSS of the operon tar was located 52 bp upstream of the start codon of the tarA CDS . CD search further reveals that RBAM_RS16130 encodes a major facilitator superfamily (MFS) transporter with 12 transmembrane-helices . Considering the domain and function annotations, as well as the similar gene organization to the TAA biosynthesis operon tbr in B. thuringiensis (containing the AI-encoding gene tbrA and the TAA exporter-encoding gene tbrB ), we proposed that RBAM_RS16130 was associated with the intracellular transport of TAA in TAA assimilation and named it tarB . To test whether the tar operon is responsible for TAA assimilation in FZB42, we tested the growth of single gene deletion mutants Δ tarA and Δ tarB separately and found that although they grew normal in liquid LB medium like FZB42 , both mutants could not grow on solid or liquid ACO media . However, complementation of tarA or tarB restored TAA assimilation ability . These results genetically confirmed that the operon composed of tarA and tarB is responsible for the assimilation of the TAA carbon source in B. velezensis FZB42.
B. velezensis that converts TAA to CAA To confirm the AI activity of the TarA protein, the interconversion activity of the recombinant TarA-His 6 protein was tested in vitro using TAA and CAA as substrates. Distinct formations of CAA product with TAA substrate and TAA product with CAA substrate were detected by both HPLC and LC-Q-TOF-MS , demonstrating that TarA catalyzes the reversible reaction from TAA to CAA, but is much more efficient in forming TAA. These properties are consistent with the classic characteristics of AI enzymes. We further determined the effects of pH, ionic strength, and various metal ions on TarA activity, as well as the K m , v max , k cat , and k cat /K m kinetic constants of the TarA isomerization reaction to further elucidate its catalytic characteristics . The constants measured via Lineweaver–Burk plots are listed in , quantitatively confirming the equilibrium preference of TarA in TAA formation. These results demonstrated that B. velezensis TarA is an AI that can isomerize TAA to CAA.
B. velezensis TarB is predicted as a MFS transporter with 12 transmembrane helices via bioinformatics. To verify its function as a cell membrane importer of TAA, the membrane subcellular localization of the TarB protein was first determined. Through a microscope, we observed a strong green fluorescent signal on the cell membrane of the FZB42-TarB-GFP strain, indicating the localization of TarB-GFP fusion protein on the membrane. In addition, only a weak green signal was observed in the cytoplasm due to the dynamic balance of production and loss of TarB-GFP protein . In contrast, as a control protein, GFP was only uniformly distributed in the cytoplasm . Western blot confirmed the presence of TarB-GFP in cell membrane extract . A molecular binding assay of purified TarB protein and TAA substrate by MST further reveals significant ligand dose-dependent changes in fluorescence intensity ( K d = 5.66 ± 1.38 μM) , demonstrating that TarB, a B. velezensis membrane protein, binds to the TAA molecule with moderate strength. Then, the intracellular contents of FZB42 and Δ tarB strains cultured in modified ACO liquid medium were analyzed by LC-Q-TOF-MS. TAA was only detected in FZB42 cells but not in Δ tarB cells , suggesting that TAA is unable to enter the cells when tarB is deleted. From 0 to 6 h of the culture , FZB42 and Δ tarB appear to consume only glucose and not TAA ( and ). From 6 to 10 h, Δ tarB continues to consume glucose that is depleted at 10 h . However, starting at 6 h, FZB42 used TAA and glucose as carbon sources, which are depleted up at 14 h and 10 h , respectively. In contrast, Δ tarB shows a complete impairment in the use of TAA, as Δ tarB is unable to transport TAA into the cells throughout the culture period . Therefore, when the only available carbon source, glucose, was depleted, Δ tarB reaches its maximum OD 600 value at 10 h ; whereas FZB42 underwent a diauxic growth within 10–14 h and reaches a population twice larger than that of Δ tarB , further confirming the TAA-importing function of TarB. Taken together, these results demonstrate that TarB is a membrane importer of TAA molecules in B. velezensis .
B. velezensis We noticed that one gene, RBAM_RS16120 (861 bp), annotated as encoding a LysR-type transcriptional regulator (LTTR), is located 173 bp upstream of the AI gene tarA and is transcribed in an opposite direction with tarA . Further secondary structure analysis of the RBAM_RS16120 gene product shows that the amino terminus (1–58 aa) contains a helix-turn-helix (HTH) DNA binding motif, and the carboxyl terminus (91–281 aa) contains a substrate binding region connected to the HTH motif via a 32-aa hinge region (59–90 aa) . This structure exhibits the typical domain organization of bacterial LTTRs . Considering the functional annotation and the sharing of the promoter region with tarA , the RBAM_RS16120 gene was considered to have a potential regulatory function on the TAA assimilation ability in B. velezensis and was therefore named tarR . To test the regulatory function of tarR on TAA assimilation, we compared the growth of the Δ tarR mutant with that of the starting strain FZB42. Although showing a normal growth in LB medium like FZB42, Δ tarR is unable to grow on ACO medium ( and ), implying positive regulation of TAA assimilation by tarR . Then, qRT-PCR analysis shows that in FZB42, when TAA was added, the transcription of tarA and tarB is significantly upregulated , confirming that the function of TAA assimilation in B. velezensis requires TAA molecules for activation. However, when tarR is deleted in FZB42, the expression of the tar operon cannot be activated despite the presence of TAA . Upon complementation with tarR in C tarR , transcription of tarA and tarB restores normal response to the TAA molecule . These results confirm the positive regulation of the inducible expression of tar by TarR and suggest a role for TarR in the reception of TAA signals. Moreover, tarR expression also responds positively to TAA induction . Next, to determine whether TarR functions as a direct regulator of the tar operon, EMSAs were conducted on the TarR protein with tar promoter region ( P tar ), and the 16S rRNA promoter ( P 16S ) region as a control. Specific interactions appear in the TarR- P tar system but not in the control . In the competition assay, when fixed amounts of TarR and labeled probe were incubated with increasing amounts of non-labeled cold probe, the shifted bands gradually disappeared , indicating that the cold probe competitively binds TarR, resulting in reduced amounts of labeled probe-TarR complex. These results show that TarR recognizes and binds P tar specifically. To further define the binding site(s) of TarR in P tar , two independent protected regions are identified by DNase I footprinting assay ( and ), showing a dual-site interaction between TarR and the tar promoter. Within the two regions, three 6-bp palindromic motifs “TTATAA” were identified. Mutation of one or three motifs caused the shifted bands to weaken or even disappear ( and ), demonstrating that “TTATAA” is a key sequence mediating P tar -TarR interaction and that all three palindrome sequences are essential. As mentioned before, the TarR protein belongs to the LysR transcriptional factor family, and its members usually regulate cellular functions by binding to certain small signaling molecules . Furthermore, considering the bioinformatical prediction that the carboxyl domain of the TarR protein is responsible for substrate binding , and the potential regulation of TarR by TAA shown in qRT-PCR analyses , we propose that TAA is a signaling molecule that directly binds to TarR to activate the tar operon. Indeed, both MST and thermal shifting assays reveal a strong interaction between the TAA molecule and TarR protein ( K d = 11.08 ± 4.35 μM), whereas as a negative control, citric acid (C 6 ) displays no binding ( and ). The amino and carboxyl domains of TarR, TarR N1–91 and TarR C59–286 , were further shown to be responsible for P tar -DNA binding and TAA-substrate binding , respectively. We then further conducted an EMSA assay to test the effect of TAA ligands on the binding efficiency of TarR to P tar and found that formation of TarR-DNA complexes is significantly enhanced as the levels of added TAA increase , demonstrating that TAA can stimulate TarR- P tar interaction. In summary, the above results illustrate that the TarR protein first binds to TAA signaling molecules and then binds to P tar to activate the expression of the tar operon.
To define the phylogenetic distribution of TAA-assimilation function in bacteria, we screened for the presence of tarA , tarB , and tarR homologous genes. Targets containing tar homolog were found to span across 16 bacterial phyla, 973 genera, and 4 570 species in total ( and ). Of the 16 phyla, 11 contain tarRAB -type species that simultaneously harboring tarA , tarB , and tarR homologs (1 330 in total) ( and ), indicating that these species have a complete assimilation system. Of the 11 phyla, Pseudomonadota , Actinomycetota , Bacillota , Campylobacterota , Bacteroidota , Thermodesulfobacteriota , Cyanobacteriota , and Spirochaetota are the top eight phyla with the largest number of target species among all the 16 phyla . Meanwhile, we noticed that among Pseudomonadota , Actinomycetota , Bacillota , Bacteroidota , and Cyanobacteriota phyla, 142 species are of tarAB -type containing only ( tarA + tarB ) homologs, namely, with complete isomerization and transport functional organization ( and ). For the remaining five of the 16 phyla, the target species are not only few in number but also belong to the tarA -type containing only unique tarA homologs ( and ). Despite lacking a transport element, these bacteria are still considered to have the potential to use TAA. First, they encode the key element of the AI enzyme in the TAA assimilation process. Second, the lack of tarB homologs may be due to the high diversity of MFS sequences between and even within subfamilies . There may be importers in alternative subfamilies with low similarity to TarB. To further reveal the coverage of TAA assimilation ability within a phylum, the analysis of the ratio of target species within a phylum to total sequenced species of the phylum is also shown . In the top 1 phylum Pseudomonadota , more than half of the species (3 411 target /6 600 total ) encode AI ( and ), and more than 31.1% (1 060/3 411) further encode tarB or even tarR alternatives , indicating that Pseudomonadota is a representative phylum that generally adopts TAA assimilation strategy; in Actinomycetota and Campylobacterota , the above ratios are 22.9% (813 target /3 553 total ) and 34.4% (280/813), 22.2% (39 target /176 total ) and 48.7% (19/39), respectively ( and ), which also indicate that TAA assimilation function is dominant in these phyla. Detailed information on the classification, homology type, and protein sequence identities of all bacterial targets are listed in .
We collected 32 samples to explore the distribution of TAA assimilation bacteria in the natural environments and found that many TAA-utilizing bacteria could be isolated from these samples, yielding a total of 77 strains, and 66 of them were identified . These bacteria belong to three phyla, namely Pseudomonadota , Bacillota , and Actinomycetota , involving 12 genera and 27 species, which successfully verified the widespread nature of TAA assimilation species in bacteria taxa. These isolates are particularly concentrated in the phylum Pseudomonadota (46/66, 69.7%), which ranks first in the phylogenetic results and is consistent with the early reports that identified many Pseudomonas spp. using TAA molecules . However, our isolation results differ from that report, which indicated that Gram-positive bacteria (involving only two phyla of Bacillota and Actinomycetota ) rarely use TAA. In our work, we found that 29.6% (8/27) of the isolated species are Gram-positive, which may reflect that uptake of TAA carbon sources by Gram-positive bacteria in nature is real. Subsequently, the genomes of 12 representatives of the 66 strains were analyzed by tblastn. These strains differ in species, living environments, and lifestyles, including plant- and animal- pathogenic , plant-beneficial , and free-living . Except for strain 56 ( Pantoea agglomerans sp.) and strain 60 ( Serratia marcescens sp.), the two species newly identified in this work that were able to use TAA but did not have TarA-homologs, the remaining 10 strains were found to have at least one homologous sequence . In total, we obtained 16 candidate sequences from 10 strains and found that 70% of the strains (7/10; strain 7, 16, 23, 32, 41, 68, and 69) encode active AI enzymes, and two of them, AI-7-1 ( ** P = .008) and AI-16-1 ( ** P = .004) in strain 7 and strain 16, exhibit significantly higher catalytic efficiency than TarA of FZB42 . In comparison, the forward or reverse isomerization activities of AI-4 (3.5%, 8.6%), AI-22 (8.2%, 24.7%), and AI-79 (0.7%, 0.4%) proteins are quite low , although strains 4 ( Aeromonas veronii sp.), strain 22 ( Acinetobacter pittii sp.), and strain 79 ( Providencia rettgeri sp.) encoding these genes grew well on the ACO plate , implying that weak enzymatic performance may be sufficient for survival in nature, or these AIs may require special biochemical conditions for catalysis, or these bacteria may encode unique and unknown AI sequence(s). Together, these results suggest that TAA assimilation was active and prevalent in bacterial carbon utilization and underscore the importance of TAA as a natural carbon nutrient.
B. velezensis strain in soil Based on experimental results, we deduced that the function of the TAA assimilation system is to confer a survival advantage to assimilators in natural environments where TAA is present. To test this hypothesis, we examined the growth rate and final populations of the TAA-assimilating strain FZB42 and the non-assimilating strain Δ tarA in sterile soil (naturally TAA-absent, ) without other competing microbiota. This approach allowed us to focus solely on the effects of TAA assimilation ability on competing bacteria. When examined individually in soils without TAA addition, the two strains show indistinguishable growth rates and similar final population levels over the course of the tests (30 days) . However, when TAA was added and present in soils, the TAA assimilation strain FZB42 reproducibly grew to 2–3 times the size of the non-assimilation strain , suggesting that TAA-assimilating strain has a considerable survival advantage in the presence of TAA.
B. velezensis strains in soil To further determine the competitive advantage conferred by the TAA assimilation system in FZB42 when competing with Δ tarA during survival in soil, cultures of both strains were mixed at a population ratio of 1:1 and inoculated into sterile soils with or without TAA. In the absence of TAA, neither strain becomes dominant , and the distribution of strains in the total bacterial population remains ~1:1 throughout the experiment . However, when TAA was added in the soils, strains utilizing TAA became the dominant members of the test group , as the ratio favors the FZB42 assimilator (from 0.94 to 3.86 during 0–19 days) . This confirms that bacteria equipped with a TAA-assimilation system have a clearly competitive advantage in environments where TAA is present.
Here, by studying the model strain B. velezensis FZB42 for microbe-environment interaction, we identified the bacterial genetic determinants of tarR and the tar operon responsible for activation of TAA assimilation and TAA sensing, import, and isomerization, confirming the growth and competitive survival advantages provided by this assimilation system , as well as its widespread distribution in the bacterial domain and in a variety of environmental bacteria. The tar operon is a new member of the bacterial inducible operon group. It shows a similar regulatory pattern to some classical operons (such as the lac for lactose catabolism , rha for rhamnose catabolism , and ttd for tartrate catabolism ), whose expression targets a substrate molecule for degradation and requires the molecule itself to first serve as an inducer for activation. Thus, TAA molecule, TarR protein, and the P tar promoter may provide new regulatory alternatives for synthetic biology applications. To our knowledge, all bacterial AIs identified to date exhibit the same enzymatic properties in vitro , namely, interconversion between CAA and TAA, and more uniquely, a preference for TAA formation . However, once inside the cell, these AIs “differentiated” in functions to specifically mediate TAA consumption (e.g. TarA) or biosynthesis (e.g. TbrA). Comparison of the in vitro enzymatic data for TarA and TbrA reveals similar and classic AI catalytic behavior . By expressing B. thuringiensis tbrA in B. velezensis and B. velezensis tarA in B. thuringiensis , we found that TarA and TbrA both belong to the PrpF superfamily and can functionally substitute for each other , suggesting that there are other factors besides AI enzymes that determine the assimilation or production of TAA. TAA transporters are one of the most likely determinants. In bacteria, TAA transporters may be highly differentiated in the transport directions of TAA, which determines the source and function of TAA molecules, which are nutrients imported from the environment or toxin synthesized for export. As in the nematode pathogen B. thuringiensis , once the AI enzyme in the cytoplasm isomerizes CAA to TAA, the exporter protein TbrB immediately pumps TAA out of the cell to alleviate its cumulative toxicity to the TCA cycle, thereby continuously driving the isomerization continuously toward and forming TAA nematicides . In B. velezensis , the delivery of TAA from the environment into the cell by the TarB importer may lead to a rapid increase of TAA level in the cytoplasm. Although AI favors CAA substrates, high concentrations of TAA substrates would repress TAA formation and shift AI equilibrium to TAA consumption. AI enzyme holds promise for industrial production of TAA via biosynthesis. Novel AI sequences distinct from TarA have the potential to be isolated from bacterial species clustering in Classes I, II, or III in the AI evolutionary analysis , or simply by ACO plate selection. Directed evolution of AI enzymes with desirable properties is also expected to accelerate the future applications of TAA in more challenging fields such as medicine and agriculture. The identification of the TarB importer following the discovery of the B. thuringiensis exporter TbrB further underscores the important role of orientation-specific transporters in mediating TAA-related physiology in bacteria. TarB and TbrB were found to have no sequence similarity and were classified into two independent subfamilies of 2.A.1 and 2.A.7, respectively, of the MFS (Transport Classification Database). These findings reflect the functional and sequence divergence of TAA transporters and indicate that highly independent groups of MFS or more complex subgroups that unidirectionally import or export TAA may have evolved in bacteria; in addition, bidirectional TAA transporters or non-specific transporters of TAA and its structural analogs may also exist. Therefore, the limitations of using the only available importer gene sequence, tarB , to predict TAA-assimilating bacteria containing importer homologs should be recognized. To solve these issues, the identification of more TAA transporters in bacteria is urgently needed. As early as 1961, it was discovered that TAA metabolism is inducible in bacteria . The identification of TarR has unraveled the regulatory mechanism, increased the regulatory scope of LTTRs, and deepened our understanding of microbial economic strategies for survival. Recent studies have shown that nutrients, in addition to being valuable sources of carbon and/or nitrogen, also act as signaling molecules in nutrient-assimilating bacteria to express genes that promote survival . Similarly, in the TAA-TarR signaling pathway of the important plant growth-promoting rhizobacterium (PGPR) B. velezensis FZB42, we further found 90 TarR potential regulon genes whose promoter regions contained at least two key recognition motifs “TTATAA” as the P tar , implying that TarR or TAA may regulate more bacterial functions required for survival in plant-associated environments. Characterization of the genes and functions is essential to understand the story of nutrition initiation. The widespread distribution of TAA assimilation systems in the bacterial domain reflects that TAA is a common and important component of the carbon source pool in nature. In addition to free-living or plant-associated environments, TAA molecule was also detected in animals . In this study, we found that many animal-endophytical bacteria were predicted to be TAA-assimilating bacteria , such as Acidaminococcus intestine and Fusobacterium mortiferum , Peptoniphilus ovalis , Desulfovibrio porci , Cloacibacillus porcorum , Sporomusa termitida and Intestinirhabdus alba , and Shewanella marinintestina , which have been reported to colonize the intestines of humans , monkeys , pigs , rabbits , insects , and marine fish , respectively. Can TAA nutrition influence the growth and colonization of assimilating bacteria in the animal host? Does TAA in the animal come from exogenous food intake or endogenous biosynthesis, and if the latter, what are the genes and pathways? These questions are unknown and valuable research topics. Metabolic adaptation, especially to the carbon sources, is a critical determinant of bacterial growth, colonization, and ecological function in different environments ; indeed, bacteria employ a variety of flexible strategies to achieve this. Bioinformatics evidence shows that genes encoding carbon metabolism categories have the strongest evolutionary conservation among all gene function categories in bacteria , supporting the idea that carbon metabolism is a fundamental and precondition in survival interactions. Experimental evidence from other plant-associated carbon source cases, such as opines , mimosine , and tartaric acid , also confirms that the greater and more diverse the carbon catabolism, the better the growth and colonization. Strengthening the link between carbon production and assimilation between the environments and microorganisms may help improve the performance of target microorganisms in the microbiota and shape the related structures. On this basis, beneficial microorganisms that are often unable to compete with the resident microbiota can be modified through synthetic biology to utilize specific nutrients, such as TAA , to greatly improve the application potential for probiotics.
Supplemental_material20241013_wrae198 TableS1_wrae198 TableS2_wrae198 TableS3_wrae198 TableS4_20241002_wrae198 TableS5_wrae198 TableS6_wrae198 TableS7_wrae198
|
|
Analysis of Transgender and Gender‐Diverse Topics Within Diversity, Equity, and Inclusion Curricular Content in Pediatric Anesthesiology Fellowship Programs in the United States and Canada—A Prospective Survey | 053dde5d-0395-4c59-9901-b23a1f51276f | 11806206 | Pediatrics[mh] | Introduction Transgender and gender‐diverse (TGD) individuals experience higher burdens of health disparities compared to their cisgender counterparts. Contributing factors include decreased access to care, denial of care, experiences with and fear of medical violence, and increasing legislative barriers. According to the 2021 Census of Population, 0.33% of the Canadian population age 15 and older are transgender or non‐binary, the majority of which are 15–34 years old . In the United States (USA), 0.6% of the population age 13 and older are transgender according to the Williams institute . Youth between the ages of 13–17 comprise about 18% of the USA TGD population and account for 1.43% of their age group . The Trans Pulse project, a community‐based research project in Ontario, Canada, found that 33.2% of respondents experienced an unmet health need within the past year compared to 10.7% of their cisgender counterparts . Similarly, the 2022 USA Transgender Survey revealed several barriers to healthcare including delay in seeking care due to fear of mistreatment or refusal of care due to transgender status, abusive language and misgendering from providers, and denial of insurance coverage for gender affirming and routine healthcare due to being transgender . In addition to barriers within the healthcare system, increasing legislative attacks on TGD individuals' rights to access healthcare, education, and legal recognition contribute to and exacerbate discrimination against the TGD community . On top of these already significant burdens, TGD patients often report having to educate health care professionals due to lack of expertise within the medical community. In a qualitative interview study of TGD individuals on their perioperative experiences, about half of the participants discussed the need for increased provider knowledge on the care of TGD patients . Both the Accreditation Council for Graduate Medical Education (ACGME) and the American Board of Anesthesiology (ABA) have included diversity, equity, and inclusion (DEI) topics as core aspects of graduate medical education curricula . In the 2024 Competencies guide, the ACGME lists “respect and responsiveness to diverse patient populations” as a professionalism competency . The ABA content outline includes DEI in health care, the workplace, and academia as topics that may be included on licensing exams . However, several studies have demonstrated a critical deficit of training in medical care for TGD patients at all levels of medical education and across multiple specialties . Common themes include a desire for increased education time, insufficient resident cultural sensitivity, and lack of faculty expertise. These trends have been noted within anesthesiology residencies and subspecialty fellowships as well . Anesthesiologists have a unique opportunity to interact with TGD patients in the already vulnerable perioperative setting. One study found that while anesthesiologists showed an overall high intent to provide culturally competent care to TGD patients, their lack of knowledge had a significantly negative influence . Given the relatively limited time that anesthesiologists have to establish a positive patient‐physician relationship, providing culturally sensitive care is of the utmost importance. Pediatric anesthesiologists will likely care for TGD patients more frequently, as evidenced by the overrepresentation of pediatric individuals in the TGD population. In fact, in a survey by Roque et al. , only 12% of pediatric anesthesia respondents had not taken care of a TGD person in the last five years. Improving pediatric anesthesiology trainee education is a vital step toward improving TGD patient care. We set out to describe the current state of TGD curricular content within DEI education across pediatric anesthesiology fellowship training programs in the United States and Canada. We additionally sought to characterize the existing DEI and TGD curricular content, the impact of known LGBTQ+ faculty champions and/or experts within the department, and the perceived adequacy of fellow preparedness. Methods A 25‐question descriptive, cross‐sectional, electronic survey was created, which asked participants about the inclusion of DEI educational content and content related to TGD patient populations in the pediatric anesthesiology fellowship curricula at their institutions. The questionnaire was constructed with input from the investigators' previous experiences caring for TGD populations and after reviewing other surveys in the literature . Informed consent was obtained on the opening page of the survey; participants consented by opting to proceed with the questionnaire. Survey questions were pretested to confirm appropriateness and understandability with a small group of content experts, including several anesthesia providers with prior work with the population and an adolescent medicine physician who co‐directs a gender clinic. Questions queried program demographics and curricular content related to DEI and TGD topics. The study was reviewed by the institutional review board at Seattle Children's Hospital (STUDY 00004062) and determined to be exempt. The final version of the survey was entered into Seattle Children's RedCap for distribution. The study population was a purposive sample of all pediatric anesthesiology fellowship program directors in the United States and Canada. The population was limited to the United States and Canada for convenience and to build upon prior research completed by MacCormick et al. in the field of obstetric anesthesiology . Program directors were invited to take the survey via email. In the case of USA participants, emails were distributed through the Pediatric Anesthesiology Program Directors' Association (PAPDA) listserve; Canadian participants were emailed directly by the study team. Reminder emails were sent approximately halfway through the open period (02/15/2023–04/30/2023). Participation was voluntary, anonymous, and no compensation was provided. Participants were not required to answer all questions, and skipped questions were removed from the denominator during analysis. Therefore, n values differ among questions. Descriptive statistical analysis was completed using Microsoft Excel. The Checklist for Reporting Results of Internet E‐Surveys (CHERRIES) was used to guide the development and reporting of this study . Results 3.1 Survey Response Data A total of 33 responses were collected from 69 total invited programs (response rate 33/69 = 48%). This represented 3 partial and 30 complete datasets (completion rate 30/33 = 91%). Of note, six responses were excluded because no questions were answered after agreeing to participate; technical difficulty or the intent to complete at a later time were assumed as potential explanations. 3.2 Program Setting In total, 33 of 69 pediatric anesthesiology fellowship programs were represented—29 USA programs and 4 Canadian programs (out of 62 and 7 total programs, respectively). Program size ranged from 1 to 16 fellows per year. Regional response rates were highest among the Western USA programs (9/9 = 100%) and lowest among the Midwestern USA programs (4/18 = 22.2%). Among all responses, Western USA and Canadian programs were overrepresented and Northeastern, Midwestern, and Southern USA were underrepresented (Table ). Nearly all programs that responded were affiliated with academic hospitals (29/30 = 97%) and the majority were housed within free‐standing children's hospitals (18/32 = 56%). Most programs were in hospitals that included TGD‐specific clinics (23/32 = 72%) and half (16/32 = 50%) reported clinical exposure to gender‐affirming surgeries and procedures for their fellows. The number of anesthesiologists within the pediatric divisions at the programs varied, with 11–20 anesthesiologists being the most frequent answer (12/29 = 41%) followed by > 50 (8/29 = 28%). 3.3 Program Faculty and Faculty Education Twenty‐eight percent of program directors (9/32) reported having faculty within their division who were active in presenting or publishing work related to TGD individuals and/or their care. Pediatric anesthesiology faculty known to be members of the LGBTQ+ community were present in 71% of programs (22/31). Among those programs with known LGBTQ+ faculty, 82% (18/22) had their known LGBTQ+ faculty represented within hospital or department leadership. Of the programs with known LGBTQ+ faculty in leadership positions, only 27.8% (5/18) include TGD content in their fellowship curriculum, but out of the 9 programs with TGD curriculum, 66.7% (6/9) have known LGBTQ+ faculty in leadership positions. The significance of this data is unclear. No sample size calculations or further data analysis were performed given the primary aim of the study was to describe the current state of TGD education in pediatric anesthesia fellowships. In the year prior to the study period, DEI training for faculty occurred in 76.7% (23/30) of programs and education related to TGD patient care occurred in only 40% (12/30) (Figure ). Among the 12 programs with TGD faculty education, participation was mandatory in 50% (6/12) and optional in 50% (6/12). TGD education for faculty was repeated annually in 25% (3/12) of programs, repeated at another interval in 33% (4/12) of programs and offered on a one‐time, non‐recurring basis in 42% (5/12) of programs. 3.4 Fellowship Curricula DEI curricula were present in 93.5% (29/31) of fellowship programs but only 29% (9/31) of programs included content specific to TGD populations (Figure ). In programs with this content, 1 h was dedicated to TGD content in 33% (3/9), 2 h in 33% (3/9), 3 h in 11% (1/9), and 4 h in 22% (2/9) of programs. TGD patient care curricula were mandatory in 78% (7/9) and optional in 22% (2/9) of programs. Similarly, 78% (7/9) of programs reported repeating these curricula with each fellow class. Of the subject areas asked about, there was notable heterogeneity of topics covered in the nine programs with TGD‐specific content (Figure ). In order of inclusion from most to least frequent, the topics covered were perioperative communication strategies, health disparities, anesthetic implications of hormones, anesthetic implications after gender‐affirming surgery and transphobia within healthcare. The most popular educational format used for teaching was traditional didactic sessions. Notably, no programs reported utilizing simulation (Figure ). 3.5 Perceived Barriers and Possible Strategies to Increasing TGD Content in Fellowship Curricula Only 17% (5/29) of program directors thought their curriculum adequately prepared their graduates to care for TGD patients. Sixty‐nine percent (20/29) of programs expressed a desire to see more educational content included in the future (Figure ). Of these programs, 35% (7/20) already teach TGD content, 55% (11/20) teach a DEI curriculum without TGD content, and 10% (2/20) do not teach any DEI or TGD curricula. Of the programs that were unsure if they wanted to increase their TGD curriculum ( n = 7), 14.3% (1/7) already teach TGD content while 85% (6/7) teach a DEI curriculum without TGD topics. Of the programs that selected they did not want more TGD curriculum ( n = 2), 50% (1/2) already teach TGD content and 50% (1/2) teach a DEI curriculum without TGD topics. Didactic sessions, problem‐based learning discussions (PBLDs), and online modules were the most sought‐after formats (Figure ). Lack of knowledgeable faculty educators and time were the most chosen perceived barriers to inclusion (Figure ). The strategies that fellowship program directors thought would be most successful in increasing TGD‐specific content were access to ready‐made curricular material focusing on TGD content, faculty willing and able to teach TGD content, and more evidence‐based research regarding anesthetic outcomes in TGD patients (Figure ). Survey Response Data A total of 33 responses were collected from 69 total invited programs (response rate 33/69 = 48%). This represented 3 partial and 30 complete datasets (completion rate 30/33 = 91%). Of note, six responses were excluded because no questions were answered after agreeing to participate; technical difficulty or the intent to complete at a later time were assumed as potential explanations. Program Setting In total, 33 of 69 pediatric anesthesiology fellowship programs were represented—29 USA programs and 4 Canadian programs (out of 62 and 7 total programs, respectively). Program size ranged from 1 to 16 fellows per year. Regional response rates were highest among the Western USA programs (9/9 = 100%) and lowest among the Midwestern USA programs (4/18 = 22.2%). Among all responses, Western USA and Canadian programs were overrepresented and Northeastern, Midwestern, and Southern USA were underrepresented (Table ). Nearly all programs that responded were affiliated with academic hospitals (29/30 = 97%) and the majority were housed within free‐standing children's hospitals (18/32 = 56%). Most programs were in hospitals that included TGD‐specific clinics (23/32 = 72%) and half (16/32 = 50%) reported clinical exposure to gender‐affirming surgeries and procedures for their fellows. The number of anesthesiologists within the pediatric divisions at the programs varied, with 11–20 anesthesiologists being the most frequent answer (12/29 = 41%) followed by > 50 (8/29 = 28%). Program Faculty and Faculty Education Twenty‐eight percent of program directors (9/32) reported having faculty within their division who were active in presenting or publishing work related to TGD individuals and/or their care. Pediatric anesthesiology faculty known to be members of the LGBTQ+ community were present in 71% of programs (22/31). Among those programs with known LGBTQ+ faculty, 82% (18/22) had their known LGBTQ+ faculty represented within hospital or department leadership. Of the programs with known LGBTQ+ faculty in leadership positions, only 27.8% (5/18) include TGD content in their fellowship curriculum, but out of the 9 programs with TGD curriculum, 66.7% (6/9) have known LGBTQ+ faculty in leadership positions. The significance of this data is unclear. No sample size calculations or further data analysis were performed given the primary aim of the study was to describe the current state of TGD education in pediatric anesthesia fellowships. In the year prior to the study period, DEI training for faculty occurred in 76.7% (23/30) of programs and education related to TGD patient care occurred in only 40% (12/30) (Figure ). Among the 12 programs with TGD faculty education, participation was mandatory in 50% (6/12) and optional in 50% (6/12). TGD education for faculty was repeated annually in 25% (3/12) of programs, repeated at another interval in 33% (4/12) of programs and offered on a one‐time, non‐recurring basis in 42% (5/12) of programs. Fellowship Curricula DEI curricula were present in 93.5% (29/31) of fellowship programs but only 29% (9/31) of programs included content specific to TGD populations (Figure ). In programs with this content, 1 h was dedicated to TGD content in 33% (3/9), 2 h in 33% (3/9), 3 h in 11% (1/9), and 4 h in 22% (2/9) of programs. TGD patient care curricula were mandatory in 78% (7/9) and optional in 22% (2/9) of programs. Similarly, 78% (7/9) of programs reported repeating these curricula with each fellow class. Of the subject areas asked about, there was notable heterogeneity of topics covered in the nine programs with TGD‐specific content (Figure ). In order of inclusion from most to least frequent, the topics covered were perioperative communication strategies, health disparities, anesthetic implications of hormones, anesthetic implications after gender‐affirming surgery and transphobia within healthcare. The most popular educational format used for teaching was traditional didactic sessions. Notably, no programs reported utilizing simulation (Figure ). Perceived Barriers and Possible Strategies to Increasing TGD Content in Fellowship Curricula Only 17% (5/29) of program directors thought their curriculum adequately prepared their graduates to care for TGD patients. Sixty‐nine percent (20/29) of programs expressed a desire to see more educational content included in the future (Figure ). Of these programs, 35% (7/20) already teach TGD content, 55% (11/20) teach a DEI curriculum without TGD content, and 10% (2/20) do not teach any DEI or TGD curricula. Of the programs that were unsure if they wanted to increase their TGD curriculum ( n = 7), 14.3% (1/7) already teach TGD content while 85% (6/7) teach a DEI curriculum without TGD topics. Of the programs that selected they did not want more TGD curriculum ( n = 2), 50% (1/2) already teach TGD content and 50% (1/2) teach a DEI curriculum without TGD topics. Didactic sessions, problem‐based learning discussions (PBLDs), and online modules were the most sought‐after formats (Figure ). Lack of knowledgeable faculty educators and time were the most chosen perceived barriers to inclusion (Figure ). The strategies that fellowship program directors thought would be most successful in increasing TGD‐specific content were access to ready‐made curricular material focusing on TGD content, faculty willing and able to teach TGD content, and more evidence‐based research regarding anesthetic outcomes in TGD patients (Figure ). Discussion In 2022, the Society for Pediatric Anesthesia released a statement calling on members to “increase their understanding of gender‐affirming care for TGD youth and ensure equitable access to gender‐affirming care” noting that “the inability to access such care can result in morbidity and mortality.” This is in alignment with the American Academy of Pediatrics' gender‐affirming care policy that was reaffirmed in 2023 . Our data suggests that there is still a significant amount of progress to be made in this area. While 76.7% of programs had DEI and cultural competency training for faculty, only 40% had curriculum specific to TGD patient care. The contrast was even more stark when it came to fellow education with 93.5% of programs including DEI curricula, less than 30% of which included TGD specific content. While it is impressive that nearly all of the responding programs have adopted DEI content into their curriculum, it is clear that topics relating to the TGD population are often neglected. Given this deficit of training, it is disheartening, though perhaps not surprising, that only 17% of program directors felt that their curriculum adequately prepared their graduates to care for TGD patients. Fortunately, our data suggest motivation for improvement with 69% of respondents desiring to see more TGD‐specific educational content included in the future. In order for progress to be made, it is important to identify and address impeding factors. One of the most chosen perceived barriers to the inclusion of TGD educational content was lack of knowledgeable and willing faculty educators. Less than 30% of respondents reported having faculty within their division who were active in presenting or publishing work related to TGD individuals and/or their care. In a survey of pediatric anesthesiologists by Roque et al., many respondents had basic knowledge of TGD youth, yet 26% (100/389) of respondents were undecided or agreed with the statement “Sex and gender are the same thing,”, 8% (31/388) did not feel it was appropriate to use a transgender child's affirmed name and pronouns, and only 51% (191/374) of respondents felt equipped with the knowledge to take care of TGD patients . These data corroborates a lack of knowledgeable faculty educators, which limits programs' abilities to provide education on TGD subjects . It is interesting to note that of the programs with TGD content, two thirds have known LGBTQ+ faculty in leadership positions. This suggests that promotion, sponsorship, and mentorship of known LGBTQ+ faculty may create an environment of psychological safety and empower faculty to educate their peers and learners on care for the TGD community. While anesthetic care of the TGD patient represents an excellent avenue for LGBTQ+ anesthesiologists to contribute to the care of their community, it should not be expected that LGBTQ+ anesthesiologists are solely responsible for these efforts. The impact of minority tax, the implicit or explicit expectation for minoritized faculty to contribute to diversity efforts, must be considered. LGBTQ+ attending anesthesiologists experience additional minority stress contributing to a higher rate of burnout and workplace discrimination . It is vital that contributions to diversity efforts are recognized and rewarded by academic departments and that faculty outside of the LGBTQ+ community aid in championing these efforts. In addition to increasing the number of engaged faculty, another strategy favored to increase TGD content included utilizing ready‐made curricular material. While the World Professional Association for Transgender Health (WPATH) Standards of Care, version 8, calls for education on the care of TGD patients in all specialties, there are no WPATH education modules specific to anesthesiology . Currently, two educational resources designed for pediatric anesthesiology fellowship education, one lecture and one problem‐based learning discussion, are available via the American Society of Anesthesiologists (ASA) Anesthesia Toolbox . Efforts should be made to create and distribute educational modules to guide faculty in content presentation and creation. Ready‐made content may also mitigate the current lack of content experts described above. Additionally, respondents noted a desire to increase evidence‐based research in anesthetic outcomes for TGD patients in order to increase TGD educational content. To do so, it is vital that hospitals track sexual orientation and gender identity (SOGI) data, to the extent that patients are comfortable disclosing such information, in their clinical practice. This would not only facilitate research to better understand anesthetic outcomes but would also provide a better understanding of the frequency at which TGD patients are seen in clinical practice, both in the context of gender‐affirming surgery and not. Due to the content and small sample size, our study has some limitations. Although the survey was pretested with content experts and constructed utilizing similar previously published surveys, the possibility remains that the survey instrument did not capture the intended information. Additionally, the study did not investigate the perspective of pediatric anesthesiology fellows, which would offer another important viewpoint. While there was a reasonable response rate of 48%, several factors may have impacted this rate, including using email as a contact method, some programs not having current fellows at the time of the survey, survey fatigue, and concerns of anonymity. The sample size is inherently restricted with only 69 total pediatric anesthesiology fellowship programs in the United States and Canada, which limited more advanced statistical analyses including a comparison of responses between United States and Canadian programs. There was a difference in response rate by region, resulting in the Midwestern and Southern USA programs being underrepresented. This is important to consider in the context of regional differences in governmental policies restricting access to gender‐affirming care and cultural acceptance of the TGD population. Although some of the TGD population may be migrating toward regions with more accepting policies, TGD patients will continue to be present in all practice settings . For trainees in regions with legislative restrictions that limit access to gender‐affirming care, it may be of increased importance to include more TGD education considering they will have fewer opportunities to provide this care during training. Anesthesiologists are the guardians of perioperative safety, which not only includes physiological safety, but also emotional and psychological safety. As the known population of TGD youth increases, it is vital that all pediatric anesthesiologists are equipped to treat this population with appropriate clinical knowledge and cultural sensitivity. To our knowledge, this study is the first to evaluate pediatric anesthesiology fellowship curricula for inclusion of TGD patient‐specific education, the results of which suggest that there is a significant amount of progress still to be made. In order to increase pediatric anesthesiology fellow education on the care of TGD patients, future endeavors must include robust patient SOGI data collection, access to ready‐made educational materials from content experts, recruitment and empowerment of LGBTQ+ anesthesiologists, consideration of TGD patient perspectives in the development of hospital policies, and more dedicated time toward DEI education, specifically in regard to the TGD population. This human study was reviewed by the Institutional Review Board at Seattle Children's Hospital (STUDY 00004062) and determined to be exempt. Informed consent was obtained for all participants as described in the manuscript. This article is original work, has not been previously published, and is not being considered for publication elsewhere. The authors declare no conflicts of interest. Data S1. Data S2. |
Manual small-incision cataract surgery under topical anesthesia | c96cb192-fbac-476b-8b95-395c3fe3f825 | 9907278 | Ophthalmology[mh] | It was a prospective analytical study of 33 patients who underwent MSICS surgery from March 2022 to June 2022 using topical proparacaine 0.5% eye drops. Informed written consent for MSICS under topical proparacaine anesthesia was taken after explaining the procedure. Inclusion criteria Cases of adult cataract patients of grade 1 to grade 3, who were cooperative during slit-lamp examination, and who could obey commands and with pupil dilatation >7 mm were included. Exclusion criteria Cataract patients with grade 4 brown cataract, who are hard of hearing and uncooperative during slit-lamp examination, patients with <7-mm pupil, pediatric patients, and patients with associated complications such as uveitis, lens-induced glaucoma, small pupil, existing ocular pathologies such as corneal opacity, pseudoexfoliation, complicated cataract, subluxated lens, glaucoma, traumatic cataract, and any other complications were excluded.
Cases of adult cataract patients of grade 1 to grade 3, who were cooperative during slit-lamp examination, and who could obey commands and with pupil dilatation >7 mm were included.
Cataract patients with grade 4 brown cataract, who are hard of hearing and uncooperative during slit-lamp examination, patients with <7-mm pupil, pediatric patients, and patients with associated complications such as uveitis, lens-induced glaucoma, small pupil, existing ocular pathologies such as corneal opacity, pseudoexfoliation, complicated cataract, subluxated lens, glaucoma, traumatic cataract, and any other complications were excluded.
Topical anesthesia proparacaine 0.5% eye drops were instilled every 5 min for 15 min prior to surgery and once or twice per op if the patient was sensitive to pain. Superior rectus bridal suture was avoided to prevent pain associated with traction on the muscle. A small nick was made on the sclera to hold with Hoskin’s forceps to stabilize the globe and also for adequate exposure to form the sclerocorneal tunnel. A superior scleral straight incision of 6.5 mm was done 2 mm behind the limbus, and the sclerocorneal tunnel was made by entering approximately 2 mm into clear cornea with deep side pockets. Care was taken to prevent premature entry thus avoiding iris prolapse during surgery. The side port was made at 3 o’clock, and staining of the anterior capsule was done using Trypan blue. Continuous curvilinear capsulorhexis was done using a bent 26-G needle mounted on a syringe. Gentle hydrodissection was done, and the nucleus was brought to the anterior chamber without touching or manipulating the iris by using a single dialer. Adequate viscoelastic (HPMC 2%) was used to maintain anterior chamber depth and prevent traction on the iris. Nucleus delivery was done using vectis and dialer by sandwich technique, and cortical wash was done using Simcoe’s cannula avoiding iris touch and manipulation. Rigid PMMA IOL was placed through the sclerocorneal tunnel and dialed into the bag. Viscoelastic was washed adequately, and the anterior chamber was formed using BSS. Subconjunctival antibiotic and steroid injection was avoided, and topical antibiotic steroid drops were used before applying the pad and bandage. If the patient complained of severe pain during surgery, or in the case of a very uncooperative patient, supplementation with 2% lignocaine injection periocular block was done. All the surgeries were performed by a single surgeon. The surgeon’s and patient’s experiences were noted immediately after surgery by using a valid scoring system .
Study population included 20 women (60.6%) and 13 men (39.4%). The mean age was 58.7 ± 8.3 years (range: 41–72 years). The majority of the patients had a rural background (54.6%). There were an equal number of people who were illiterate and studied up to secondary school (39.4%) . The average comfort score based on patient feedback after surgery was 3.45 ± 0.96, and the average patient cooperation score based on surgeon assessment was 3.42 ± 1.07. Comfort score was slightly higher in men compared to women. Comfort score was more in the urban population and patients who were educated more than secondary school. Cooperation score was also higher in the male population compared to females and higher in the urban population compared to rural and patients educated more than in secondary school [Tables and ]. In two patients with a comfort score and cooperation score of 1, intraoperative augmentation of anesthesia, peribulbar using 2% lignocaine 5 mL was done. One of these patients had intraoperative iris prolapse because of premature entry, which was the reason for non-cooperation during surgery and increased discomfort. Periocular block was used, and iris reposition and tunnel suturing were done in this case. The other patient was very anxious about the use of topical anesthesia and was uncooperative during surgery and thus was given periocular block, and the surgery was completed.
MSICS is one of the most common surgical procedures in ophthalmology. Most cataract surgeries are performed under local anesthesia. Advantages of topical anesthesia using drops or gel are that it provides adequate analgesia, early patient recovery, lack of injection-related complications, and lack of complications such as retrobulbar hemorrhage, injury to extraocular muscles, and other complications seen with retrobulbar, peribulbar, and intracameral anesthesia. A study conducted by Rewri et al . at Maharaja Agrasen Medical College, Haryana to determine satisfactory experience from the patient’s as well surgeon’s perspective in cataract surgery by phacoemulsification technique under topical anesthesia concluded that acceptance for topical anesthesia for cataract is high in both patient’s and surgeon’s perspectives. A prospective comparative study conducted by Gupta et al . comparing the patient’s experience of pain, surgical outcome, and surgeon’s experience in MSICS under topical anesthesia supplemented with intracameral lignocaine concluded that the performance of these surgeries under topical anesthesia has acceptable patient comfort and pain experienced in both techniques is comparable. Topical anesthesia does not compromise surgical results in MSICS. A study conducted by Agarwal at All India Institute of Medical Sciences, Delhi to determine satisfactory experience from the patient’s and surgeon’s perspectives in phacoemulsification cataract surgery under topical anesthesia concluded that the acceptance for topical anesthesia for cataract surgery is high. Patient satisfaction and cooperation depend on several sociodemographic and psychological factors. Gupta et al . demonstrated that both phacoemulsification and MSICS can be done under topical anesthesia supplemented with intracameral lignocaine with acceptable patient comfort, and the pain experienced in both techniques is comparable. Another study by Dr. Suresh K M concluded that the method of topical anesthesia in performing SICS is acceptable both for the patient and the operating surgeon. In this prospective analytical study conducted on 33 patients, the average comfort score based on patient feedback after surgery was 3.45 ± 0.96, and the average patient cooperation score based on surgeon assessment was 3.42 ± 1.07. Comfort score was slightly higher in men compared to women. Comfort score was more in the urban population and patients educated more than secondary school. Cooperation score was higher in the male population compared to females. Cooperation score was higher in urban population compared to rural and patients educated more than secondary school. Surgeon’s perspective was also noted. The surgeon was comfortable doing the surgery with topical anesthesia even though it needs some learning curve.
MSICS using only topical proparacaine 0.5% eye drops without the use of any intracameral anesthesia or periocular block can provide sufficient patient comfort and can avoid complications related to injection and peribulbar and retrobulbar anesthesia. The use of topical anesthetic eye drops cuts down the time taken and cost for the administration of local anesthesia and its complications. Hence, it can be used in large-scale cataract surgeries and provides economical utilization of resources and early postoperative recovery without compromising the surgical outcome. This technique, once mastered, can be routinely done and popularized as a daycare procedure for large-scale cataract surgeries, which not only brings down the complication rates but also cuts down the cost of surgery. Financial support and sponsorship Nil. Conflicts of interest There are no conflicts of interest.
Nil.
There are no conflicts of interest.
www.ijo.in
|
Reviewing deaths in British and US hospitals: a study of two scales for assessing preventability | 839b691e-5420-4e7f-bc3f-226ca1485400 | 5530333 | Preventive Medicine[mh] | Quality control and improvement strategies have often focused on death rates—including overall hospital death rates, as an indisputably severe consequence of poor care. Yet, a large body of work suggests that the signal (preventable deaths) is likely to be buried in the noise (inevitable deaths), even after risk adjustment. When overall deaths (measured in standardised mortality ratios) are used to assess healthcare quality, they are only functioning as a crude proxy for the preventable component. The lack of specificity and sensitivity raises the question—why not directly measure only the preventable death rates? A number of studies have now been conducted to directly measure preventable death rates in the USA, the UK, and in continental Europe. In fact, the English National Health Service (NHS) have recently announced a policy to examine hospital deaths routinely and classify them as preventable or not using the methods from these studies. Direct measurement most commonly involves the review of each death by an expert, typically by scrutinising charts (case notes) to determine if it was preventable. Most studies define deaths as preventable or not on the balance of probabilities, as in a case of tort which requires the probability of causation to exceed 50% in order for the exposure to be considered a contributory cause of death. The assessment is based on a measurement scale that asks the reviewers to estimate the probability that the death could have been prevented. There are two candidates for such a measurement scale: a Likert scale incorporating the ‘more likely than not’ standard among the anchors or a more fine-grained continuous scale of probabilities—0–100. Virtually, all published works on preventable deaths have used the Likert scale. We know of no papers that have measured the probability of preventable deaths on a continuous (0–100) scale, or which have investigated the correspondence of the Likert scale measurement to the underlying continuous probability construct supposedly captured by the Likert scale. The purpose of this study is to compare both the correspondence and the relative reliability of measurements obtained using the different scales. If, as expected, the two scales are highly correlated, it would provide some degree of construct validity, by demonstrating that these two measures are related, as one would expect based on theory. Nevertheless, one scale would be preferred if it is more reliable than the other. Finally, having characterised the reliability of the scales, we can illustrate the distribution of the preventability assessments in the absence of reviewer effects and measurement noise. This removes variability in the measurement due to systematic differences across cases in the overall severity of a reviewer and the noise resulting from the inconsistency of multiple reviewers looking at a specific case. Using two existing datasets (not created specifically for this study) where case notes had been reviewed by multiple reviewers, and where each reviewer estimated preventability on both Likert and continuous scales, our objectives are to present the following: The correspondence between the two scales and the more likely than not cut-off point. The reliability of each scale—the extent to which different reviews result in the same assessment. The impact of low reliability on individual and population measurement of preventable deaths. And, in the discussion, some implications and recommendations for the design of a programme that might use case note review to assess the preventability of deaths in a health system. UK study, 2009 Background Data were obtained from deaths reviewed as part of the Safer Patients Initiative (SPI), an evaluation of a large intervention to improve the quality of care in UK hospitals between 2003 and 2009. The study and its evaluation involved 22 hospitals in England and Wales, and details are published elsewhere. Case notes Case notes of 191 deceased patients admitted for respiratory complaints and over the age of 65 years were anonymised and scanned. These were then independently scrutinised by 22 reviewers in a separate exercise not previously reported in the above publications. The case notes were randomly assigned to at least two trained reviewers who used a pro forma to guide an implicit review (see below) of the notes. Each case note was reviewed by between three and seven reviewers in order to ascertain the reviewer effect on reliability. In total, 653 reviews were carried out. Selection of the reviewers Reviewers (n=22) were experienced, qualified, working clinicians (consultants in Medicine and Intensive Care Units) who were trained as described below. US Lab Indicators Study, 1997 Background This second dataset was a de-identified subset of data originally collected as part of a study done in the late 1990s (referred to as the ‘Lab Indicators Study’), the primary goal of which was to develop quality screens based on hospital-acquired metabolic derangements and drug toxicity, as assessed by laboratory testing records. All deaths occurring in the much larger Lab Indicators Study population, along with a sample of deaths from a control population, were included in this substudy on reviewer assessments of potentially preventable mortality. This dataset thus effectively oversampled for deaths in which potentially preventable adverse events occurred. The sample was drawn from seven hospitals in the Veterans Affairs system, which at that time had one of the most advanced electronic clinical database systems in the USA, thus considerably easing the implementation of the study protocol. Details of the study are published in prior publications, including analyses presenting estimates of preventable mortality, based on the Likert scale, but a comparative analysis of the continuous and Likert scale has not been published. Case notes Case notes of 179 deaths were reviewed, with all reviews done and data entered in the late 1990s. Of the initial sample, 66 cases (37%) were excluded by an initial review that identified and removed admissions for comfort care or palliative care and all advanced cancers, along with two additional cases in which the discharge record coding was incorrect and death had not actually occurred during the inpatient stay, leaving 111 case notes. The number of reviews per case note varied, ranging from 1 to 14, with a total of 383 reviews. Of the 111 deaths that were reviewed, 35 cases were reviewed as part of a substudy in the original project, with a target of 4–14 reviews per record. The remaining 88 records had a target of two reviews per record. Reviewers took records from randomly ordered lists of the much larger number of records targeted for review by the parent study and provided at each session to each reviewer. Because of the challenges of scheduling clinically active reviewers outside of their hospital responsibilities, the availability of charts being reviewed by other reviewers and the lack of stratification by death in assigning reviews, considerable variability in the number of reviews per record resulted in this subsample of deaths. Selection of the reviewers Thirteen practicing physicians, actively engaged in hospital medical practice, were recruited and trained in the use of the instrument (see below). Both studies Pro forma for data extraction Each case note was reviewed using a semistructured implicit (holistic) review (see online 1). The UK study used a pro forma adapted from the US study, which was itself based on the original instrument developed by RAND for the diagnosis-related group pre-post study in 1989. The UK pro forma included a number of questions not included in the US pro forma, covering the diagnosis and cause of death. The US pro forma did not include this information, but did elicit information about whether or not a ‘do not resuscitate’ order had been placed. These data items are not used in the analysis in this paper. The US pro forma asks for a percentage estimate for the likelihood that a death could have been prevented, whereas the UK pro forma asks the reviewer to indicate this quantity on a 0–100 scale. Importantly, the modified instrument for both studies requested reviewers to classify mortality events by the level of preventability on both a Likert and a continuous scale, with similar wording for both items. The instrument guided the doctor to systematically review and evaluate different parts of the medical case note as a minimum before they came to give their overall preventability assessment. 10.1136/bmjqs-2015-004849.supp1 Supplementary appendix In the UK study, the Likert categorisation was obtained using the question: ‘On the balance of probability (ie, >50% chance), was the death preventable? 1=Definitely Yes; 2=Probably Yes; 3=Uncertain; 4=Probably Not; 5=Definitely Not’. The percentage preventability on the continuous scale was elicited as the ‘best estimate of likelihood of preventability of death’. Similarly, but with somewhat different wording, in the US study, the Likert categorisation was obtained using the question: ‘Was patient death preventable by better quality of care? 1=Definitely Yes; 2=Probably Yes; 3=Uncertain; 4=Probably Not; 5=Definitely Not’. The percentage preventability was elicited as ‘What do you estimate the likelihood of prevention of death to be if care had been optimal?’ These specifications enabled us to do a test of ‘logical consistency’. It is clearly logically inconsistent to either: (a) choose Likert 5 (definitely not preventable) if the best preventability estimate is >50% or (b) choose Likert 1 (definitely preventable) if the best estimate is <50%. Training of the reviewers Training for the reviewers in the US study is described in prior publications, but involved reviewing sample charts and ensuring not the absence of disagreements in rating care, but rather that differences were primarily a matter of opinion, not a result of lack of understanding of the instrument or overlooking information available in different parts of the medical case note. Training of the reviewers in the UK study mimicked that of the US study, though using a different set of exemplar case notes. All the reviewers in the UK study attended two training sessions (a full-day and a half-day), with practice case-note reviews at home in between, while the US reviewers had a single-day training. The purpose of training in both studies was therefore to ensure that the reviewers were trained in the following: The use of the implicit review instrument. Referring consistently to certain pertinent sections of the case notes while conducting their reviews. Practicing reviewing notes for the purpose of this study and to discuss ambiguities with colleague reviewers and trainers in order to come to a common understanding of how difficult situations can be dealt with and what threshold others may place on quality and preventability. Statistical methods The Likert and the continuous scale measurements were compared graphically , and analysis of variance techniques (continuous scale by Likert category) were applied after excluding instances of logical inconsistency (ie, Likert 5 with preventability >50% or Likert 1 with preventability <50%). The calibration of the Likert scale with the continuous scale was described with a multilevel ordinal logistic regression analysis using a cubic polynomial function of the continuous scale with random coefficients at the reviewer level. The form of this polynomial was determined initially within a fixed effects ordinal logistic model. This polynomial function of the continuous scale was then used as an explanatory variable in a mixed effects version of the model with random slopes and intercepts. Separately for each scale, the variance was decomposed into three components, corresponding to reviewers, case notes and a residual error term. This was done using a linear mixed effects model with (crossed) random effects for reviewers and case notes for the continuous scale data, and a corresponding mixed effects ordinal logistic model for the Likert data. The estimation of the case-note variance from this model defines the distribution of the latent variable (or latent scale) we are trying to measure. This latent scale represents the average reviewer's estimate of the probability of preventing death for each case (if the entire population of reviewers from which our reviewer sample was drawn had reviewed each record). Predictive distributions for the case-note effects on the Likert scale were derived by setting the reviewer and residual variances to zero in the ordinal logistic model, representing the predicted value for an ‘average’ reviewer, measured without error. Predictive distributions on the continuous scale were estimated in a similar way, from ordinal logistic models fitted to a categorised (representing 10 percentage points each) version of the continuous scale. Models were fitted using the STATA V.14 and MLwiN V.2.28 packages. Background Data were obtained from deaths reviewed as part of the Safer Patients Initiative (SPI), an evaluation of a large intervention to improve the quality of care in UK hospitals between 2003 and 2009. The study and its evaluation involved 22 hospitals in England and Wales, and details are published elsewhere. Case notes Case notes of 191 deceased patients admitted for respiratory complaints and over the age of 65 years were anonymised and scanned. These were then independently scrutinised by 22 reviewers in a separate exercise not previously reported in the above publications. The case notes were randomly assigned to at least two trained reviewers who used a pro forma to guide an implicit review (see below) of the notes. Each case note was reviewed by between three and seven reviewers in order to ascertain the reviewer effect on reliability. In total, 653 reviews were carried out. Selection of the reviewers Reviewers (n=22) were experienced, qualified, working clinicians (consultants in Medicine and Intensive Care Units) who were trained as described below. Data were obtained from deaths reviewed as part of the Safer Patients Initiative (SPI), an evaluation of a large intervention to improve the quality of care in UK hospitals between 2003 and 2009. The study and its evaluation involved 22 hospitals in England and Wales, and details are published elsewhere. Case notes of 191 deceased patients admitted for respiratory complaints and over the age of 65 years were anonymised and scanned. These were then independently scrutinised by 22 reviewers in a separate exercise not previously reported in the above publications. The case notes were randomly assigned to at least two trained reviewers who used a pro forma to guide an implicit review (see below) of the notes. Each case note was reviewed by between three and seven reviewers in order to ascertain the reviewer effect on reliability. In total, 653 reviews were carried out. Reviewers (n=22) were experienced, qualified, working clinicians (consultants in Medicine and Intensive Care Units) who were trained as described below. Background This second dataset was a de-identified subset of data originally collected as part of a study done in the late 1990s (referred to as the ‘Lab Indicators Study’), the primary goal of which was to develop quality screens based on hospital-acquired metabolic derangements and drug toxicity, as assessed by laboratory testing records. All deaths occurring in the much larger Lab Indicators Study population, along with a sample of deaths from a control population, were included in this substudy on reviewer assessments of potentially preventable mortality. This dataset thus effectively oversampled for deaths in which potentially preventable adverse events occurred. The sample was drawn from seven hospitals in the Veterans Affairs system, which at that time had one of the most advanced electronic clinical database systems in the USA, thus considerably easing the implementation of the study protocol. Details of the study are published in prior publications, including analyses presenting estimates of preventable mortality, based on the Likert scale, but a comparative analysis of the continuous and Likert scale has not been published. Case notes Case notes of 179 deaths were reviewed, with all reviews done and data entered in the late 1990s. Of the initial sample, 66 cases (37%) were excluded by an initial review that identified and removed admissions for comfort care or palliative care and all advanced cancers, along with two additional cases in which the discharge record coding was incorrect and death had not actually occurred during the inpatient stay, leaving 111 case notes. The number of reviews per case note varied, ranging from 1 to 14, with a total of 383 reviews. Of the 111 deaths that were reviewed, 35 cases were reviewed as part of a substudy in the original project, with a target of 4–14 reviews per record. The remaining 88 records had a target of two reviews per record. Reviewers took records from randomly ordered lists of the much larger number of records targeted for review by the parent study and provided at each session to each reviewer. Because of the challenges of scheduling clinically active reviewers outside of their hospital responsibilities, the availability of charts being reviewed by other reviewers and the lack of stratification by death in assigning reviews, considerable variability in the number of reviews per record resulted in this subsample of deaths. Selection of the reviewers Thirteen practicing physicians, actively engaged in hospital medical practice, were recruited and trained in the use of the instrument (see below). This second dataset was a de-identified subset of data originally collected as part of a study done in the late 1990s (referred to as the ‘Lab Indicators Study’), the primary goal of which was to develop quality screens based on hospital-acquired metabolic derangements and drug toxicity, as assessed by laboratory testing records. All deaths occurring in the much larger Lab Indicators Study population, along with a sample of deaths from a control population, were included in this substudy on reviewer assessments of potentially preventable mortality. This dataset thus effectively oversampled for deaths in which potentially preventable adverse events occurred. The sample was drawn from seven hospitals in the Veterans Affairs system, which at that time had one of the most advanced electronic clinical database systems in the USA, thus considerably easing the implementation of the study protocol. Details of the study are published in prior publications, including analyses presenting estimates of preventable mortality, based on the Likert scale, but a comparative analysis of the continuous and Likert scale has not been published. Case notes of 179 deaths were reviewed, with all reviews done and data entered in the late 1990s. Of the initial sample, 66 cases (37%) were excluded by an initial review that identified and removed admissions for comfort care or palliative care and all advanced cancers, along with two additional cases in which the discharge record coding was incorrect and death had not actually occurred during the inpatient stay, leaving 111 case notes. The number of reviews per case note varied, ranging from 1 to 14, with a total of 383 reviews. Of the 111 deaths that were reviewed, 35 cases were reviewed as part of a substudy in the original project, with a target of 4–14 reviews per record. The remaining 88 records had a target of two reviews per record. Reviewers took records from randomly ordered lists of the much larger number of records targeted for review by the parent study and provided at each session to each reviewer. Because of the challenges of scheduling clinically active reviewers outside of their hospital responsibilities, the availability of charts being reviewed by other reviewers and the lack of stratification by death in assigning reviews, considerable variability in the number of reviews per record resulted in this subsample of deaths. Thirteen practicing physicians, actively engaged in hospital medical practice, were recruited and trained in the use of the instrument (see below). Pro forma for data extraction Each case note was reviewed using a semistructured implicit (holistic) review (see online 1). The UK study used a pro forma adapted from the US study, which was itself based on the original instrument developed by RAND for the diagnosis-related group pre-post study in 1989. The UK pro forma included a number of questions not included in the US pro forma, covering the diagnosis and cause of death. The US pro forma did not include this information, but did elicit information about whether or not a ‘do not resuscitate’ order had been placed. These data items are not used in the analysis in this paper. The US pro forma asks for a percentage estimate for the likelihood that a death could have been prevented, whereas the UK pro forma asks the reviewer to indicate this quantity on a 0–100 scale. Importantly, the modified instrument for both studies requested reviewers to classify mortality events by the level of preventability on both a Likert and a continuous scale, with similar wording for both items. The instrument guided the doctor to systematically review and evaluate different parts of the medical case note as a minimum before they came to give their overall preventability assessment. 10.1136/bmjqs-2015-004849.supp1 Supplementary appendix In the UK study, the Likert categorisation was obtained using the question: ‘On the balance of probability (ie, >50% chance), was the death preventable? 1=Definitely Yes; 2=Probably Yes; 3=Uncertain; 4=Probably Not; 5=Definitely Not’. The percentage preventability on the continuous scale was elicited as the ‘best estimate of likelihood of preventability of death’. Similarly, but with somewhat different wording, in the US study, the Likert categorisation was obtained using the question: ‘Was patient death preventable by better quality of care? 1=Definitely Yes; 2=Probably Yes; 3=Uncertain; 4=Probably Not; 5=Definitely Not’. The percentage preventability was elicited as ‘What do you estimate the likelihood of prevention of death to be if care had been optimal?’ These specifications enabled us to do a test of ‘logical consistency’. It is clearly logically inconsistent to either: (a) choose Likert 5 (definitely not preventable) if the best preventability estimate is >50% or (b) choose Likert 1 (definitely preventable) if the best estimate is <50%. Training of the reviewers Training for the reviewers in the US study is described in prior publications, but involved reviewing sample charts and ensuring not the absence of disagreements in rating care, but rather that differences were primarily a matter of opinion, not a result of lack of understanding of the instrument or overlooking information available in different parts of the medical case note. Training of the reviewers in the UK study mimicked that of the US study, though using a different set of exemplar case notes. All the reviewers in the UK study attended two training sessions (a full-day and a half-day), with practice case-note reviews at home in between, while the US reviewers had a single-day training. The purpose of training in both studies was therefore to ensure that the reviewers were trained in the following: The use of the implicit review instrument. Referring consistently to certain pertinent sections of the case notes while conducting their reviews. Practicing reviewing notes for the purpose of this study and to discuss ambiguities with colleague reviewers and trainers in order to come to a common understanding of how difficult situations can be dealt with and what threshold others may place on quality and preventability. Each case note was reviewed using a semistructured implicit (holistic) review (see online 1). The UK study used a pro forma adapted from the US study, which was itself based on the original instrument developed by RAND for the diagnosis-related group pre-post study in 1989. The UK pro forma included a number of questions not included in the US pro forma, covering the diagnosis and cause of death. The US pro forma did not include this information, but did elicit information about whether or not a ‘do not resuscitate’ order had been placed. These data items are not used in the analysis in this paper. The US pro forma asks for a percentage estimate for the likelihood that a death could have been prevented, whereas the UK pro forma asks the reviewer to indicate this quantity on a 0–100 scale. Importantly, the modified instrument for both studies requested reviewers to classify mortality events by the level of preventability on both a Likert and a continuous scale, with similar wording for both items. The instrument guided the doctor to systematically review and evaluate different parts of the medical case note as a minimum before they came to give their overall preventability assessment. 10.1136/bmjqs-2015-004849.supp1 Supplementary appendix In the UK study, the Likert categorisation was obtained using the question: ‘On the balance of probability (ie, >50% chance), was the death preventable? 1=Definitely Yes; 2=Probably Yes; 3=Uncertain; 4=Probably Not; 5=Definitely Not’. The percentage preventability on the continuous scale was elicited as the ‘best estimate of likelihood of preventability of death’. Similarly, but with somewhat different wording, in the US study, the Likert categorisation was obtained using the question: ‘Was patient death preventable by better quality of care? 1=Definitely Yes; 2=Probably Yes; 3=Uncertain; 4=Probably Not; 5=Definitely Not’. The percentage preventability was elicited as ‘What do you estimate the likelihood of prevention of death to be if care had been optimal?’ These specifications enabled us to do a test of ‘logical consistency’. It is clearly logically inconsistent to either: (a) choose Likert 5 (definitely not preventable) if the best preventability estimate is >50% or (b) choose Likert 1 (definitely preventable) if the best estimate is <50%. Training for the reviewers in the US study is described in prior publications, but involved reviewing sample charts and ensuring not the absence of disagreements in rating care, but rather that differences were primarily a matter of opinion, not a result of lack of understanding of the instrument or overlooking information available in different parts of the medical case note. Training of the reviewers in the UK study mimicked that of the US study, though using a different set of exemplar case notes. All the reviewers in the UK study attended two training sessions (a full-day and a half-day), with practice case-note reviews at home in between, while the US reviewers had a single-day training. The purpose of training in both studies was therefore to ensure that the reviewers were trained in the following: The use of the implicit review instrument. Referring consistently to certain pertinent sections of the case notes while conducting their reviews. Practicing reviewing notes for the purpose of this study and to discuss ambiguities with colleague reviewers and trainers in order to come to a common understanding of how difficult situations can be dealt with and what threshold others may place on quality and preventability. The Likert and the continuous scale measurements were compared graphically , and analysis of variance techniques (continuous scale by Likert category) were applied after excluding instances of logical inconsistency (ie, Likert 5 with preventability >50% or Likert 1 with preventability <50%). The calibration of the Likert scale with the continuous scale was described with a multilevel ordinal logistic regression analysis using a cubic polynomial function of the continuous scale with random coefficients at the reviewer level. The form of this polynomial was determined initially within a fixed effects ordinal logistic model. This polynomial function of the continuous scale was then used as an explanatory variable in a mixed effects version of the model with random slopes and intercepts. Separately for each scale, the variance was decomposed into three components, corresponding to reviewers, case notes and a residual error term. This was done using a linear mixed effects model with (crossed) random effects for reviewers and case notes for the continuous scale data, and a corresponding mixed effects ordinal logistic model for the Likert data. The estimation of the case-note variance from this model defines the distribution of the latent variable (or latent scale) we are trying to measure. This latent scale represents the average reviewer's estimate of the probability of preventing death for each case (if the entire population of reviewers from which our reviewer sample was drawn had reviewed each record). Predictive distributions for the case-note effects on the Likert scale were derived by setting the reviewer and residual variances to zero in the ordinal logistic model, representing the predicted value for an ‘average’ reviewer, measured without error. Predictive distributions on the continuous scale were estimated in a similar way, from ordinal logistic models fitted to a categorised (representing 10 percentage points each) version of the continuous scale. Models were fitted using the STATA V.14 and MLwiN V.2.28 packages. Data summary UK study Assessments were collected for 191 case notes from 22 reviewers. Of the 653 reviews, there were 644 that provided assessments of preventability on the Likert or continuous scale. Both scales were completed in 628 of the returns. Reviewers returned between 10 and 78 assessments each (mean 29.7). Of the 191 case notes, 19 had two reviews, 101 had three reviews, 48 had four reviews, 19 had five reviews, 3 had six reviews and 1 had seven reviews. The average number of reviews per case note was 3.4.The five categories of Likert responses (L1–L5) are summarised in . There were 637 continuous preventability assessments in total. The mean preventability was 17.5%, but the distribution is positively skewed with a median of 10% (quartiles, 3%, 28%), and with a high proportion (17.7%) of zeroes (113/637). US study This earlier study was smaller with 111 case notes reviewed by 13 reviewers. There were 383 reviews with assessments of preventability on the Likert or continuous scale. Reviewers carried out between 20 and 80 reviews each (mean 29.4). Of the 111 cases, 49 had one review, 33 had two, 14 had between three and eleven, and 14 had twelve or more. The average number of reviews per case note was 3.5. The five categories of Likert responses (L1–L5) are summarised in and had relatively fewer patients in the ‘definitely not’ preventable category than the UK study (24% vs 42%). The 383 continuous preventability assessments had a mean preventability of 12.4%, with a median of 5% (quartiles, 0%, 20%) and 34% zeroes (133/383). Logical consistency test: both studies There were no instances of reviewers choosing Likert 5 (definitely not preventable) and providing a best preventability estimate of >50%, but in the UK study, there were three cases of reviewers choosing Likert 1 (definitely preventable) and providing a best estimate of <50%. This originated from two reviewers, where Likert 1 was chosen alongside best estimates of 0%, 4% and 5% ( A). These were excluded from all analyses, as consideration of other reviewers' responses for these case notes suggested that the inconsistency derived from (hopefully momentary) confusion about the meaning of the Likert scale. Objective 1: correspondence between scales The observed association between the Likert and per cent assessments in the UK study, where both scales are available (n=628), is illustrated in figure A. Figure B shows the US associations for 383 reviews where, again, both scales were available. There is clearly a strong correspondence between the scales. If the three inconsistent points indicated in A are excluded, the Likert categories account for 74% of the variation in the continuous scale, rising to 76% after adjustment for reviewer effects in the UK study and 73% of the variation in the US study. Relatively large differences were found in the correspondence of the two scales between different UK reviewers using a multilevel ordinal logistic regression analysis which allows us to estimate how much the correspondence between the two measurements varies across reviewer. A shows the best linear predictors of the calibration lines for the reviewers in the UK study, together with 95% prediction limits for calibration over the population of reviewers. A similar figure for the US study is shown in B. The variance components for the reviewers were smaller in magnitude (as seen in B). Objective 2: reliability of each scale Cross-classified models (case note by reviewer) were fitted to the two scales, as described above. The proportion of the variance attributed to each component of variance is shown in , estimated separately for each study. This represents the reliability of a single observation for estimating the true level of that component. For example, under the column ‘Case note’, we see the reliability of estimating the preventability of the death detailed in a particular case note using a single review by a randomly selected reviewer. The residual error term represents an amalgam of two variance components: sampling variation (variation between repeat readings of a case note by the same reviewer) and the interaction between case notes and reviewers (the tendency for some case note/reviewer combinations to generate unexpected responses). Since no repeat readings were made with the same case note and reviewer, it is impossible to tease apart these components. The analyses on different scales—Likert and continuous—in both studies yield somewhat different results for the ‘reviewer’ components, but return similar estimates (∼27% for the UK study and ∼23% for the US study) for the reliability of a single review. Therefore, if a random reviewer was selected to review a random patient death from the UK sample, 27% of the variation in the resulting measurement would be due to signal (how preventable the death was) and 73% would be due to noise. Objective 3—impact of low reliability: extracting the case-note effect from the noise The practical usefulness of the analysis depends on being able to describe the distribution of preventability scores across the case note sample, having removed the substantial amount of noise due to differences between reviewers on average across all their reviews (the reviewer component of variance in ) and the inconsistency from review to review (the residual error in ). This gives the estimates of the distribution of case-preventability ratings, conditional on the case notes all being reviewed by the reviewer most typical (the modal effect) of the reviewer population, and with the residual error term removed. The characteristics of these distributions could be used to compare samples of case notes arising from different times and institutions, if the reviewers are drawn from the same population. and show the predictive distribution of case note preventability across the categories used in the Likert analysis and a categorised version of the percentage analysis . The charts in and represent the distribution of preventability (with the reviewer effects removed) among the population of case notes from which the sample is drawn. In both cases, the predictive case note distribution is less extreme (ie, suggests fewer cases with high preventability) than the distribution of the raw data. The wider distribution in the raw data reflects lack of consistency across reviewers and reviews. For example, in the UK study, the median percentage preventability from the predictive distribution in Ai is estimated as 11.0%, which is similar to that for the raw data (10.0%), but the quartiles (7.0, 17.9) cover a much reduced range compared with the raw data (3.0, 28.0). Relevant to the standard of causation that the death was more likely than not to be preventable, after removing reviewer variation and measurement error, there were almost no case notes estimated as having more than uncertain likelihood of having patient death preventable by better quality of care in either the UK or US study when measured on the Likert scale, and few if any where the median reviewer would conclude that the ‘likelihood of prevention of death…if care had been optimal’ would exceed 50% ( Aii, Bii). UK study Assessments were collected for 191 case notes from 22 reviewers. Of the 653 reviews, there were 644 that provided assessments of preventability on the Likert or continuous scale. Both scales were completed in 628 of the returns. Reviewers returned between 10 and 78 assessments each (mean 29.7). Of the 191 case notes, 19 had two reviews, 101 had three reviews, 48 had four reviews, 19 had five reviews, 3 had six reviews and 1 had seven reviews. The average number of reviews per case note was 3.4.The five categories of Likert responses (L1–L5) are summarised in . There were 637 continuous preventability assessments in total. The mean preventability was 17.5%, but the distribution is positively skewed with a median of 10% (quartiles, 3%, 28%), and with a high proportion (17.7%) of zeroes (113/637). US study This earlier study was smaller with 111 case notes reviewed by 13 reviewers. There were 383 reviews with assessments of preventability on the Likert or continuous scale. Reviewers carried out between 20 and 80 reviews each (mean 29.4). Of the 111 cases, 49 had one review, 33 had two, 14 had between three and eleven, and 14 had twelve or more. The average number of reviews per case note was 3.5. The five categories of Likert responses (L1–L5) are summarised in and had relatively fewer patients in the ‘definitely not’ preventable category than the UK study (24% vs 42%). The 383 continuous preventability assessments had a mean preventability of 12.4%, with a median of 5% (quartiles, 0%, 20%) and 34% zeroes (133/383). Logical consistency test: both studies There were no instances of reviewers choosing Likert 5 (definitely not preventable) and providing a best preventability estimate of >50%, but in the UK study, there were three cases of reviewers choosing Likert 1 (definitely preventable) and providing a best estimate of <50%. This originated from two reviewers, where Likert 1 was chosen alongside best estimates of 0%, 4% and 5% ( A). These were excluded from all analyses, as consideration of other reviewers' responses for these case notes suggested that the inconsistency derived from (hopefully momentary) confusion about the meaning of the Likert scale. Assessments were collected for 191 case notes from 22 reviewers. Of the 653 reviews, there were 644 that provided assessments of preventability on the Likert or continuous scale. Both scales were completed in 628 of the returns. Reviewers returned between 10 and 78 assessments each (mean 29.7). Of the 191 case notes, 19 had two reviews, 101 had three reviews, 48 had four reviews, 19 had five reviews, 3 had six reviews and 1 had seven reviews. The average number of reviews per case note was 3.4.The five categories of Likert responses (L1–L5) are summarised in . There were 637 continuous preventability assessments in total. The mean preventability was 17.5%, but the distribution is positively skewed with a median of 10% (quartiles, 3%, 28%), and with a high proportion (17.7%) of zeroes (113/637). This earlier study was smaller with 111 case notes reviewed by 13 reviewers. There were 383 reviews with assessments of preventability on the Likert or continuous scale. Reviewers carried out between 20 and 80 reviews each (mean 29.4). Of the 111 cases, 49 had one review, 33 had two, 14 had between three and eleven, and 14 had twelve or more. The average number of reviews per case note was 3.5. The five categories of Likert responses (L1–L5) are summarised in and had relatively fewer patients in the ‘definitely not’ preventable category than the UK study (24% vs 42%). The 383 continuous preventability assessments had a mean preventability of 12.4%, with a median of 5% (quartiles, 0%, 20%) and 34% zeroes (133/383). There were no instances of reviewers choosing Likert 5 (definitely not preventable) and providing a best preventability estimate of >50%, but in the UK study, there were three cases of reviewers choosing Likert 1 (definitely preventable) and providing a best estimate of <50%. This originated from two reviewers, where Likert 1 was chosen alongside best estimates of 0%, 4% and 5% ( A). These were excluded from all analyses, as consideration of other reviewers' responses for these case notes suggested that the inconsistency derived from (hopefully momentary) confusion about the meaning of the Likert scale. The observed association between the Likert and per cent assessments in the UK study, where both scales are available (n=628), is illustrated in figure A. Figure B shows the US associations for 383 reviews where, again, both scales were available. There is clearly a strong correspondence between the scales. If the three inconsistent points indicated in A are excluded, the Likert categories account for 74% of the variation in the continuous scale, rising to 76% after adjustment for reviewer effects in the UK study and 73% of the variation in the US study. Relatively large differences were found in the correspondence of the two scales between different UK reviewers using a multilevel ordinal logistic regression analysis which allows us to estimate how much the correspondence between the two measurements varies across reviewer. A shows the best linear predictors of the calibration lines for the reviewers in the UK study, together with 95% prediction limits for calibration over the population of reviewers. A similar figure for the US study is shown in B. The variance components for the reviewers were smaller in magnitude (as seen in B). Cross-classified models (case note by reviewer) were fitted to the two scales, as described above. The proportion of the variance attributed to each component of variance is shown in , estimated separately for each study. This represents the reliability of a single observation for estimating the true level of that component. For example, under the column ‘Case note’, we see the reliability of estimating the preventability of the death detailed in a particular case note using a single review by a randomly selected reviewer. The residual error term represents an amalgam of two variance components: sampling variation (variation between repeat readings of a case note by the same reviewer) and the interaction between case notes and reviewers (the tendency for some case note/reviewer combinations to generate unexpected responses). Since no repeat readings were made with the same case note and reviewer, it is impossible to tease apart these components. The analyses on different scales—Likert and continuous—in both studies yield somewhat different results for the ‘reviewer’ components, but return similar estimates (∼27% for the UK study and ∼23% for the US study) for the reliability of a single review. Therefore, if a random reviewer was selected to review a random patient death from the UK sample, 27% of the variation in the resulting measurement would be due to signal (how preventable the death was) and 73% would be due to noise. The practical usefulness of the analysis depends on being able to describe the distribution of preventability scores across the case note sample, having removed the substantial amount of noise due to differences between reviewers on average across all their reviews (the reviewer component of variance in ) and the inconsistency from review to review (the residual error in ). This gives the estimates of the distribution of case-preventability ratings, conditional on the case notes all being reviewed by the reviewer most typical (the modal effect) of the reviewer population, and with the residual error term removed. The characteristics of these distributions could be used to compare samples of case notes arising from different times and institutions, if the reviewers are drawn from the same population. and show the predictive distribution of case note preventability across the categories used in the Likert analysis and a categorised version of the percentage analysis . The charts in and represent the distribution of preventability (with the reviewer effects removed) among the population of case notes from which the sample is drawn. In both cases, the predictive case note distribution is less extreme (ie, suggests fewer cases with high preventability) than the distribution of the raw data. The wider distribution in the raw data reflects lack of consistency across reviewers and reviews. For example, in the UK study, the median percentage preventability from the predictive distribution in Ai is estimated as 11.0%, which is similar to that for the raw data (10.0%), but the quartiles (7.0, 17.9) cover a much reduced range compared with the raw data (3.0, 28.0). Relevant to the standard of causation that the death was more likely than not to be preventable, after removing reviewer variation and measurement error, there were almost no case notes estimated as having more than uncertain likelihood of having patient death preventable by better quality of care in either the UK or US study when measured on the Likert scale, and few if any where the median reviewer would conclude that the ‘likelihood of prevention of death…if care had been optimal’ would exceed 50% ( Aii, Bii). Main findings In this paper, we have discussed some of the measurement characteristics of methods to judge the preventability of hospital deaths. In reference to our first objective, despite the two samples being very different in time period, country and design of the sample, the Likert scale and the continuous scale appear to behave in a similar fashion . The reviewers appear to stop assigning the ‘uncertain preventability’ category and start assigning the ‘possibly’ and ‘probably preventable’ categories at just about exactly when they estimate the preventability on a continuous scale exceeds 50%. If the goal is to determine whether an average reviewer would feel that the death was more likely than not to be preventable, the observed correspondence provides support for grouping the response of ‘uncertain’ on a 5-point Likert scale along with the ‘possibly not’ and ‘probably not preventable’ responses. In terms of our second objective, we find that the reliability of the Likert and continuous scale measurements were similar to each other in both datasets (0.27 vs 0.27 for the UK study; 0.23 vs 0.22 for the US study), suggesting no particular preference for one scale or the other in terms of precision. These low estimates of reliability, at about 0.2–0.3, are also consistent with almost all prior studies using expert review to estimate preventable deaths, quality of care or preventability of adverse events. Relating to our third objective, we show that the low reliability has considerable impact on drawing conclusions about the burden of preventable deaths, both at the individual and population level when using the ‘more probable than not’ standard of causation. To make a judgement at the individual level about a specific case, a reliability of 0.25 for a single measurement implies a need to average 12 independent reviews to achieve a reliability of 0.8 for a decision about whether any given death was preventable. However, for an estimate at the hospital or system level, one can average across cases as well as reviewers, and reasonably precise estimates could be made with more practical numbers of reviews per case and total cases. In addition, as seen in and , analyses that do not remove the noise or reviewer differences will significantly overestimate the degree to which reviewers think deaths are preventable. Limitations and strengths The study was constrained by the original datasets. The Likert scale was always completed before the continuous scale on the case note review forms, whereas, ideally, the order would be randomised to mitigate practice effects. There were (small) differences in the precise descriptions given for the Likert categories in the UK and USA. The strength of our study relies in its generalisability, given the different datasets in terms of time, place and clinical conditions. We were able to tease apart reviewer effects and residual errors in describing preventability across the case note sample. We carefully eliminated from the statistical analysis the few cases of ‘incoherence’—the provision of logically inconsistent answers. Implications Relating to our fourth objective, what are the implications of our findings for the design of a programme that would attempt to measure the burden of preventable deaths in a health system? Our findings suggest that a Likert scale can reasonably represent expert opinion about causality, and there are no clear advantages in terms of precision to a continuous scale. However, the low reliability would suggest the need for a detailed consideration as to how to design the measurement procedure to find the optimal number of reviews per patient and independent reviewers needed to generate the estimates that the programme is supposed to produce at the required precision. Furthermore, to allow monitoring of the reliability of measurement and the estimates adjusted for that reliability, both reviewers and the case notes that they review must be more or less randomly distributed. This would preclude the use of reviewers only from the hospital that provides the cases, and require standardising the selection and training of reviewers across the health system. However, it is crucial to point out that the interpretation of these numbers elicited from physician reviewers remains open to question. We would argue strongly against the interpretation that the measurements represent an objective probability that can be used to estimate a casualty count. It is critical to point out that there is really no evidence suggesting that this scale, elicited from physician experts by either of the two measurements, is anything more than ordinal with respect to the true probability of preventing death. Counterfactual reasoning is notoriously difficult, and there is evidence that physicians are not very good at estimating absolute prognostic probabilities and systematically increase their estimates of poor care when there is a bad outcome. If we can only assume that the measurement is ordinal with respect to the true probability of death, then it makes no sense to dichotomise it and consider the cases on one side of the cut-off as ‘truly’ preventable and those on the other as not. In fact, it is extraordinary that this measurement procedure, which has been used in numerous studies, has been assumed to represent what would have to be a ratio scale of measurement, with a true zero and equal intervals for one to be able to estimate the actual burden of preventable deaths, in the absence of any evidence supporting that inference. There may be a better solution. The measurement properties demonstrate that physicians, with a low, but not insignificant, level of reproducibility, can distinguish between patient case notes on a scale that is elicited by asking them to estimate the probability that a death could have been averted by optimal care. Why not give up on trying, after the fact, to estimate the probability of preventing death? It is, after all, a hubristic endeavour. Rather, ask the reviewers to estimate simply how good the care was in the cases of people who have died. This would not allow health systems to count up the number of deaths attributable to poor care, as they all seem to want to do. However, with a sufficient investment in a robust measurement system, it could allow systems to track relative performance across both hospitals and time, ensuring that attention could be focused on laggards and improvement of the system overall could be tracked. In that sense, review of deaths is really just a review of case notes enriched, one may suppose, to contain a higher proportion of serious errors. Review of deaths also enables doctors to be involved in quality assurance and to detect specific ‘bear traps’ to which they and others can be alerted. In this paper, we have discussed some of the measurement characteristics of methods to judge the preventability of hospital deaths. In reference to our first objective, despite the two samples being very different in time period, country and design of the sample, the Likert scale and the continuous scale appear to behave in a similar fashion . The reviewers appear to stop assigning the ‘uncertain preventability’ category and start assigning the ‘possibly’ and ‘probably preventable’ categories at just about exactly when they estimate the preventability on a continuous scale exceeds 50%. If the goal is to determine whether an average reviewer would feel that the death was more likely than not to be preventable, the observed correspondence provides support for grouping the response of ‘uncertain’ on a 5-point Likert scale along with the ‘possibly not’ and ‘probably not preventable’ responses. In terms of our second objective, we find that the reliability of the Likert and continuous scale measurements were similar to each other in both datasets (0.27 vs 0.27 for the UK study; 0.23 vs 0.22 for the US study), suggesting no particular preference for one scale or the other in terms of precision. These low estimates of reliability, at about 0.2–0.3, are also consistent with almost all prior studies using expert review to estimate preventable deaths, quality of care or preventability of adverse events. Relating to our third objective, we show that the low reliability has considerable impact on drawing conclusions about the burden of preventable deaths, both at the individual and population level when using the ‘more probable than not’ standard of causation. To make a judgement at the individual level about a specific case, a reliability of 0.25 for a single measurement implies a need to average 12 independent reviews to achieve a reliability of 0.8 for a decision about whether any given death was preventable. However, for an estimate at the hospital or system level, one can average across cases as well as reviewers, and reasonably precise estimates could be made with more practical numbers of reviews per case and total cases. In addition, as seen in and , analyses that do not remove the noise or reviewer differences will significantly overestimate the degree to which reviewers think deaths are preventable. The study was constrained by the original datasets. The Likert scale was always completed before the continuous scale on the case note review forms, whereas, ideally, the order would be randomised to mitigate practice effects. There were (small) differences in the precise descriptions given for the Likert categories in the UK and USA. The strength of our study relies in its generalisability, given the different datasets in terms of time, place and clinical conditions. We were able to tease apart reviewer effects and residual errors in describing preventability across the case note sample. We carefully eliminated from the statistical analysis the few cases of ‘incoherence’—the provision of logically inconsistent answers. Relating to our fourth objective, what are the implications of our findings for the design of a programme that would attempt to measure the burden of preventable deaths in a health system? Our findings suggest that a Likert scale can reasonably represent expert opinion about causality, and there are no clear advantages in terms of precision to a continuous scale. However, the low reliability would suggest the need for a detailed consideration as to how to design the measurement procedure to find the optimal number of reviews per patient and independent reviewers needed to generate the estimates that the programme is supposed to produce at the required precision. Furthermore, to allow monitoring of the reliability of measurement and the estimates adjusted for that reliability, both reviewers and the case notes that they review must be more or less randomly distributed. This would preclude the use of reviewers only from the hospital that provides the cases, and require standardising the selection and training of reviewers across the health system. However, it is crucial to point out that the interpretation of these numbers elicited from physician reviewers remains open to question. We would argue strongly against the interpretation that the measurements represent an objective probability that can be used to estimate a casualty count. It is critical to point out that there is really no evidence suggesting that this scale, elicited from physician experts by either of the two measurements, is anything more than ordinal with respect to the true probability of preventing death. Counterfactual reasoning is notoriously difficult, and there is evidence that physicians are not very good at estimating absolute prognostic probabilities and systematically increase their estimates of poor care when there is a bad outcome. If we can only assume that the measurement is ordinal with respect to the true probability of death, then it makes no sense to dichotomise it and consider the cases on one side of the cut-off as ‘truly’ preventable and those on the other as not. In fact, it is extraordinary that this measurement procedure, which has been used in numerous studies, has been assumed to represent what would have to be a ratio scale of measurement, with a true zero and equal intervals for one to be able to estimate the actual burden of preventable deaths, in the absence of any evidence supporting that inference. There may be a better solution. The measurement properties demonstrate that physicians, with a low, but not insignificant, level of reproducibility, can distinguish between patient case notes on a scale that is elicited by asking them to estimate the probability that a death could have been averted by optimal care. Why not give up on trying, after the fact, to estimate the probability of preventing death? It is, after all, a hubristic endeavour. Rather, ask the reviewers to estimate simply how good the care was in the cases of people who have died. This would not allow health systems to count up the number of deaths attributable to poor care, as they all seem to want to do. However, with a sufficient investment in a robust measurement system, it could allow systems to track relative performance across both hospitals and time, ensuring that attention could be focused on laggards and improvement of the system overall could be tracked. In that sense, review of deaths is really just a review of case notes enriched, one may suppose, to contain a higher proportion of serious errors. Review of deaths also enables doctors to be involved in quality assurance and to detect specific ‘bear traps’ to which they and others can be alerted. |
Effects of Message Framing and Time Discounting on Health Communication for Optimum Cardiovascular Disease and Stroke Prevention (EMT-OCSP): a protocol for a pragmatic, multicentre, observer-blinded, 12-month randomised controlled study | b2d63545-bdb8-4188-9098-587afcc6636d | 7993219 | Health Communication[mh] | Primary prevention of cardiovascular disease (CVD) and stroke often fails due to poor patient adherence to evidence-based recommendations and interventions, and poor motivation to engage in healthy behaviour, despite evidence that lifestyle changes and pharmacological therapy effectively modify risk factors. Poor adherence is multifactorial, with large contributions from physicians’ poor communication about CVD risks and individuals’ inaccurate perceptions. Many proposed means of improving adherence are insufficiently effective. The current approach to CVD risk communication usually centres on total CVD risk scores; however, many people incorrectly interpret statistical information, and poor comprehension likely limits their ability to make healthier choices. Appropriate message design may promote individuals’ perception and motivation, adherence to health plans and successful CVD risk control, but this depends on many factors (eg, age; risk, socioeconomic and educational levels; personality and psychological characteristics). CVD risk communication could be improved by intuitively visualising life expectancy changes or disease-free survival losses/gains. Considering people’s tendency to overweigh losses relative to gains, loss-framed health messages may be more likely to motivate healthy behaviour. However, health behaviour findings on framing impacts are inconsistent, and effects may depend on the behaviour in question. In addition, people tend to discount future (long-term/lifetime) effects, responding more to immediate (short-term) costs and benefits (time discounting or time preference), especially as they age, although this tendency varies among individuals. Time preference/discounting is the most common cognitive bias affecting intentions to engage in actual behaviour, and it may fluctuate asymmetrically for gains versus losses. Health behaviour researchers have sought to identify the contexts in which gain/loss-framed health messages are most likely to motivate healthy behaviour. Moreover, several small trials have investigated the combined effects of message framing and time discounting in the promotion of smoking and drinking cessation. Their findings have not produced consistent conclusions, due in part to limitations (ie, small samples, immediate measures of attitudes toward health behaviours or intentions to engage in behaviours as primary outcomes of interest). Few studies have examined actual behavioural changes, which enable better characterisation of the circumstances under which cognitive deviation is likely to make practical differences in health promotion, and the identification of contexts in which health message shaping is likely to maximally affect healthy behaviour. The use of smartphone applications (apps) for disease prevention is growing. It enhances CVD risk communication and intervention delivery, and increases access to effective CVD prevention. Although preliminary trials have documented benefits of app use among people with CVD, high-quality data are limited and long-term outcome data are not available. Rigorous evidence for the clinical validity of mobile healthcare is lacking, raising questions such as which app components are likely to facilitate behavioural changes and enable individuals to adhere to medical advice. Recently, a competing risk-adjusted lifetime-perspective model for CVD (LIFE-CVD) was developed and validated to aid survival gain calculation for primary CVD prevention and individuals’ understanding of risks and interventions. The model has three types of input data (demographic characteristics, existing and anticipated risk factors) and provides generic formulas for CVD risk, life expectancy and survival gain calculation according to individuals’ ethnic backgrounds, facilitating the intuitive framing of health communication. Gain-framed and loss-framed communications can be created by setting different reference points (actual and ideal CVD-free life expectancy, respectively), and short-term and long-term interventional benefits can be represented graphically. No study to date has investigated the effects of message framing and time discounting on CVD prevention adherence with a focus on long-term persistence of healthy behavioural changes. Large-scale randomised controlled trials (RCTs) assessing the effects of cognitive deviation in CVD risk communication on individuals’ risk perceptions, health motivation and sustained healthy behaviour changes, as well as on major clinical outcomes, should be prioritised. Considering that physician–patient risk communication time is commonly limited, clarity regarding the most effective messaging strategies for general and specific populations is valuable. The combined application of the LIFE-CVD model, a cognitive bias-based messaging strategy and a health communication app may optimise primary CVD prevention in target populations.
In the Effects of Message Framing and Time Discounting on Health Communication for Optimum Cardiovascular Disease and Stroke Prevention (EMT-OCSP) Study, we aim to determine whether risk/intervention communication strategies (gain-framed vs loss-framed, long-term vs short-term) and potential interaction thereof have different effects on the optimisation of primary prevention adherence among subjects with at least one CVD risk factor using a smartphone app based on the LIFE-CVD model. We have developed and pre-tested the ‘Health Keeper’ (HK) app (under the China National Key Research and Development Project), based on the LIFE-CVD model , and will employ it in this study. Strategic message design, taking advantage of cognitive bias, will be used to maximise participants’ uptake and engagement in primary prevention, and the outcomes of interventions will be evaluated according to participants’ sex, age, risk level, socioeconomic status and psychological characteristics. Evidence of variation in preventive effectiveness (CVD risk reduction) among study groups, regardless of the effect size, will indicate the importance of messaging strategy leveraging.
Study design and setting The EMT-OCSP trial is designed as a 2×2 factorial, observer-blinded multicentre RCT with four parallel groups. It will be integrated with the National Basic Public Health Service Program (NBPHSP) in China, initiated in 2009. NBPHSP activities include the full spectrum of population-based essential health services for all urban and rural inhabitants. For the EMT-OCSP Study, we will use data from 32 rural and 24 urban primary healthcare centres. The study has been registered at the trial registration. The HK app was developed at West China Hospital, Sichuan, China, with stroke neurologists and software developers following a user-centred, evidence-based approach. Based on the LIFE-CVD model, it enables the calculation of 10-year and lifetime CVD risks, and risk-adjusted gains/losses in life expectancy. The app has physician and patient portals, and is compatible with Apple and Android systems. Participants register and are given user IDs based on their cell phone numbers; they must accept the terms and conditions of app use and confirm that they have read the brief explanatory introduction. Participants then input profile data (birthdate, sex, ethnicity, height and weight (the body mass index is then calculated), total and high-density lipoprotein (HDL) cholesterol levels (the non-HDL cholesterol level is then calculated) and systolic blood pressure (BP); pull-down options and definitions are provided when applicable), which the participants and their physicians check for accuracy before submission. Referential data rules, valid values, range checks and consistency checks for the input data will be pre-set according to the LIFE-CVD model and stored in the software. After submission, these data (birthdate, sex, ethnicity and height) are locked and cannot be modified at will to avoid cross-use and contamination. Any necessary data modification will require physician approval. Users then click ‘next’ to view their estimated baseline CVD risk and CVD-free life expectancy in their randomly assigned format; these results will be saved. The app’s main screen shows estimated CVD-free life expectancy (represented as a thermometer) with a brief explanation . Participants will be required to click the ‘I want to be better’ button to set personal optimal values (eg, for BP and cholesterol; the app limits values to reasonable ranges) and interventional goals (to achieve motivation, these must be better than baseline, eg, lower systolic BP and cholesterol level). Possible gains derived from anticipated lifestyle and pharmacological interventions, which may decrease with age and increase with risk-factor burden, are displayed with brief text descriptions, permitting intuitive comparison of baseline and optimised CVD-free years ( ; models A and C, larger later outcomes; models B and D, smaller sooner outcomes). The optimised CVD-free life expectancy is saved until the next calculation. The user can click a reset button to return to baseline, with a pop-up ‘see what doctor says’ link. In this section, individuals receive brief individually tailored pictorial recommendations (eg, to quit smoking, control weight or attain a target BP) according to current primary prevention guidelines. This information is sent simultaneously to corresponding physicians’ mobile terminals. Individuals are required to use the app for this game-like estimation/desired goal setting, and accept the motivations for optimised goals, one to three times per month (thereafter, the app locks the estimation function to prevent cross-use). Unless the task has been completed, each participant will receive three reminders in the last week of each month (on Monday, Tuesday and Wednesday; two alarms with text messages sent automatically at 19:00, and a physician phone call). The software will record failure to complete this task as a truancy on the user’s activity calendar. The physician portal provides access to data for an average of 100 patients per physician. The community physicians will uniformly apply pooled cohort equations (PCEs) for risk assessment to guide decision making for primary prevention ; to avoid misunderstanding and with recognition of potential inaccuracy, PCE results will not be presented in the HK app. Participants Participants will be aged 45–80 years, personally own and use a smartphone (Apple or Android platform) with Internet access, and have at least one of the following CVD risk factors: history of CVD at age <60 years in a first-degree relative, smoking, diabetes, hypertension and low-density lipoprotein (LDL) cholesterol ≥4.5 mmol/L. Participants with histories of CVD, heart failure or chronic kidney disease (estimated glomerular filtration rate <30 mL/min/1.73 m 2 ); those with terminal malignancy at baseline and those with severe psychological or mental disorders will be excluded. Violation of the study protocol, including inadequate informed consent; enrollment of subjects not meeting the inclusion/exclusion criteria; improper breaking of the blinding; multiple visits missed or outside permissible windows; inadequate record-keeping; intentional deviation from the protocol; repeated non-compliance by the subject; and falsification, and participation in another clinical study during follow-up will prompt exclusion. Enrolment The research nurses will contact potentially eligible participants identified from the NBPHSP database, explaining the trial and ascertaining interest. Interested patients will be evaluated at their primary healthcare centres; those eligible for EMT-OCSP participation will be invited to participate during NBPHSP-related visits. The research nurses will send invitations for baseline visits with available appointment times by message, and register all forms documenting NBPHSP baseline visits. Reasons for individuals’ refusal to participate in the study will be documented. Randomisation Using a computer-generated randomisation schedule with permuted blocks of random sizes, study participants will be assigned to groups A–D , representing different CVD-free life expectancy formats, and receive corresponding smartphone app upgrade packages. Random numbers will be generated prior to the study by a computerised random number generator. Block sizes will not be disclosed until primary endpoint analysis. Randomisation will be requested by the staff member responsible for recruitment and clinical interviewing from the Clinical Trial Coordinating Centre. The intervention does not permit participant or provider allocation blinding, but providers will be instructed to not disclose allocation status at follow-up assessments. Staff responsible for data collection and analysis will be blinded to group allocation. At randomisation, participants will be assigned unique identification codes. Procedure The study flow, schedule and activities are presented in and , respectively. The same 12-month outcome schedule will be used for all randomised participants, regardless of intervention completion/discontinuation, except for the adherence assessment. The study will be closed on 21 January 2022. Data collection will be finalised in January 2022, or on completion of all 1-year visits. Study-related primary healthcare centre visits will be conducted using motivational interviewing methodology, with the aim of promoting health and CVD prevention. A pictorial depiction of the LIFE-CVD model will be used to facilitate individuals’ understanding of how their lifestyles and drug therapies are linked to CVD risk. Participants will be blinded to study hypotheses and theories underlying which interventions are considered to be active. At baseline visits, in addition to health assessment, participants will install and test the HK app, register and link their app accounts for this study. Following standardised, simplified procedures, community physicians will teach participants how to use the app calculator and communicate about calculation results using the ‘teach-back’ method, providing additional information when needed. Family members will be permitted to attend baseline visits. Participants will use the HK app to set goals, receive performance feedback and self-monitor primary CVD prevention. Physicians at primary healthcare centres will routinely manage participants according to primary CVD prevention guidelines. They will aim to provide accurate, accessible and understandable CVD-related information and confirm participants’ understanding of calculation results and conceptualisation of CVD-free life expectancy as dynamic, modifiable by lifestyle changes and pharmacological treatment. Participants will also be given written guidance on CVD-free life expectancy calculation. The intervention will be regarded as low intensity if fewer than 6 monthly CVD-free life expectancy calculations per year, or fewer than three calculations in the first 6 months, are completed. Using the app, research nurses will periodically inform healthcare centre staff and participants about CVD prevention, the current trial status and plans for the next phase, and acknowledge their support. At 9 months, participants will receive a message with information about the study proceedings and a reminder about the 1-year follow-up visit, irrespective of participation status at the 6-month follow-up, provided that they have not been excluded or withdrawn consent. At recruitment and 6 months, physicians will indicate the importance of following study guidelines (including once-monthly calculation) and contacting physicians with potentially study-related problems, and explain the app’s purpose, use and updating, what to do in the event of smartphone/app loss/damage and monthly task counts and study visits. Participants will be asked whether they are having problems with app use. At follow-up sessions, reasons for task non-completion and simple strategies for enhancing adherence (eg, linking calculation to normal routines) will be discussed briefly. Follow-up assessments and interviews will be conducted at 6 and 12 months after randomisation, within a 28-day window extending from 7 days before to 21 days after the due date. Research nurses will prompt the healthcare centres to invite participants and conduct the 6-month and 12-month examinations (ie, risk factor measurement and questionnaire administration) at the local primary healthcare centres. When appropriate, participants will receive recommendations for additional follow-up visits and/or referrals to community physicians for pharmacological treatment according to CVD prevention guidelines. Cross-group contamination will be assessed at 1 year (have you talked to other participants about your LIFE-CVD model, and if yes was your attitude changed? Are you aware of the model of a participant in another group, or vice versa?). Research nurses will send assessment results to the research centre. All participants will be informed about these results. Endpoints The primary endpoints are changes in the estimated 10-year CVD risk, estimated lifetime CVD risk and estimated CVD-free life expectancy from baseline to the 1-year follow-up, based on the levels of total, HDL and LDL cholesterol; systolic BP; body mass index; history of diabetes mellitus and/or early (age <60 years) parental myocardial infarction; antithrombotic therapy and high BP treatment; diabetes; smoking habit and age. Secondary endpoints are changes in CVD risk factors (BP and serum cholesterol, LDL, non-HDL, triglycerides and fasting glucose levels); lifestyle factors (physical activity, tobacco use, alcohol use and dietaryhabits); pharmacological treatments for hypertension, dyslipidaemia and diabetes; and anti-thrombotic drug prescriptions after 1 year. Examinations and measurements From the baseline questionnaire, data on age, sex, civil status, education, family histories of premature CVD and diabetes, and systolic and diastolic BP will be recorded. Socioeconomic status will be classified according to the per-capita disposable income of urban households (by quintile) and highest attained educational level. The following data will be collected: self-reported lifestyle data (tobacco use; alcohol use (Alcohol Use Disorders Identification Test) ; physical activity and time spent sitting; dietary habit (daily fruit, root, legume and vegetable consumption); health; diabetes and BP and lipid-lowering pharmacological treatments), physical measurements (height (in cm), weight (in kg), systolic and diastolic BP (in mm Hg)) and blood examination findings (serum total, HDL and LDL cholesterol, triglycerides and fasting glucose levels (in mmol/L), determined using standard biochemical methods). Height and weight will be measured with subjects barefoot in light clothing using calibrated scales and stadiometers. Systolic and diastolic BP will be measured two times with a calibrated digital gauge (2 mm precision) after 5 min rest with subjects seated; mean values will be recorded. Blood samples will be collected after overnight fasts and sent to the clinical chemistry departments of the nearest local hospitals. Data on prescriptions for antithrombotic medications and those used to treat hypertension, dyslipidaemia and diabetes will be retrieved from the digital medical records of community hospitals. Purchases of pharmacological products from community hospitals, registered in participants’ medical insurance records, will also be recorded. Psychological variables will be used as moderators/mediators of intervention effects on primary and secondary outcomes, and as outcomes at 1 year. At baseline and 1 year, participants’ health literacy, coping strategies (using the Brief Coping Orientation to Problems Experienced Instrument), general self-efficacy, anxiety and depression (using the Hospital Anxiety and Depression Scale), and optimism/pessimism (using the Life Orientation Test) will be measured. At baseline and follow-up visits, participants will be asked to self-rate their health (five alternatives), CVD risk and health-specific self-efficacy (ability to reduce CVD risk through preventive actions) using a visual analogue scale (0–10). Training and certification Before participant recruitment, community physicians and key personnel at participating sites will be trained and certified. Personnel at each centre will be trained centrally, and training session content will be standardised and presented by members of the study’s executive team. The beta version of the HK app, operating instructions, forms, training manuals and other materials will be provided. The training session will cover trial overview; participant recruitment, eligibility and exclusion criteria; participant visit procedures; app-based data entry; data collection; physician–patient health communication requirements (standardised trial introduction, app use instruction, education and counselling for adherence, uniform and reproducible elicitation and explanation of information from the app and data discrepancy management) and general information about obtaining high-quality research data, intervention providers’ responsibilities, workflow and strategies to minimise communication bias and cross-group contamination. The providers should meet well-defined performance criteria, assessed by observation during role play with standardised mock participants. Pilot study A 1-month pilot study was performed in 2019 to test the original HK app with 200 participants from a district far from the study region, with different socioeconomic statuses and at least one CVD risk factor. The study organisation worked well, and its protocols and procedures were developed further and showed good feasibility. Two per cent cross-group contamination was detected. Life expectancy and CVD risk information was calibrated according to participants’ experiences and suggestions, obtained through questionnaires and interviews. Community physicians’ experiences, derived in part from discussions of the results with participants, were also considered. Based on these findings, and as we expect <20% contamination in the EMT-OCSP Study, we chose to use an individually randomised design rather than a clustering method. Sample size calculation The sample size needed to detect clinically significant between-group differences in baseline–1-year changes in primary outcome variables, with sufficient power and considering 20% drop-out, was determined to be 15 000 . We used a power threshold of 0.9 and significance level of 0.05. Calculation was based on SD of CVD-free life expectancy, CVD risk and conventional risk factors derived from the China National Stroke Screening Survey and the best available data so far. Statistical analyses Data management and statistical analyses will be performed using SPSS software (V.20.0; SPSS). Analyses will include data from all study participants. Data from individuals who do not consent to participation (sex, age, socioeconomic status, clinical CVD risk factors and lifestyle factors) will be included in selection bias analysis. The baseline demographic and clinical characteristics of the four parallel groups will be compared. Continuous variables (represented as means and SDs) will be compared using analysis of variance (ANOVA) and categorical variables (represented by frequencies and percentages) will be compared using χ 2 tests. Wilcoxon test will be used if necessary. Primary and secondary outcomes will be evaluated according to intention-to-treat methodology. The 2×2 factorial design ANOVA will be used to assess effects of message framing and time discounting and interactions between them due to both independent variables having two levels. Differences in changes of primary outcomes from baseline to 1-year follow-up will be analysed with regression methods to identify predictors of the changes in the endpoints. In subgroup analyses, we will estimate the intervention’s effects on primary outcomes for different age groups, sexes, CVD risk levels, socioeconomic status and interventional intensities. We will also perform Bonferroni correction for multiple comparisons in our analysis and sensitivity analyses to assess the robustness of trial results obtained with different methods of handling missing data. Patient and public involvement No patient involved.
The EMT-OCSP trial is designed as a 2×2 factorial, observer-blinded multicentre RCT with four parallel groups. It will be integrated with the National Basic Public Health Service Program (NBPHSP) in China, initiated in 2009. NBPHSP activities include the full spectrum of population-based essential health services for all urban and rural inhabitants. For the EMT-OCSP Study, we will use data from 32 rural and 24 urban primary healthcare centres. The study has been registered at the trial registration. The HK app was developed at West China Hospital, Sichuan, China, with stroke neurologists and software developers following a user-centred, evidence-based approach. Based on the LIFE-CVD model, it enables the calculation of 10-year and lifetime CVD risks, and risk-adjusted gains/losses in life expectancy. The app has physician and patient portals, and is compatible with Apple and Android systems. Participants register and are given user IDs based on their cell phone numbers; they must accept the terms and conditions of app use and confirm that they have read the brief explanatory introduction. Participants then input profile data (birthdate, sex, ethnicity, height and weight (the body mass index is then calculated), total and high-density lipoprotein (HDL) cholesterol levels (the non-HDL cholesterol level is then calculated) and systolic blood pressure (BP); pull-down options and definitions are provided when applicable), which the participants and their physicians check for accuracy before submission. Referential data rules, valid values, range checks and consistency checks for the input data will be pre-set according to the LIFE-CVD model and stored in the software. After submission, these data (birthdate, sex, ethnicity and height) are locked and cannot be modified at will to avoid cross-use and contamination. Any necessary data modification will require physician approval. Users then click ‘next’ to view their estimated baseline CVD risk and CVD-free life expectancy in their randomly assigned format; these results will be saved. The app’s main screen shows estimated CVD-free life expectancy (represented as a thermometer) with a brief explanation . Participants will be required to click the ‘I want to be better’ button to set personal optimal values (eg, for BP and cholesterol; the app limits values to reasonable ranges) and interventional goals (to achieve motivation, these must be better than baseline, eg, lower systolic BP and cholesterol level). Possible gains derived from anticipated lifestyle and pharmacological interventions, which may decrease with age and increase with risk-factor burden, are displayed with brief text descriptions, permitting intuitive comparison of baseline and optimised CVD-free years ( ; models A and C, larger later outcomes; models B and D, smaller sooner outcomes). The optimised CVD-free life expectancy is saved until the next calculation. The user can click a reset button to return to baseline, with a pop-up ‘see what doctor says’ link. In this section, individuals receive brief individually tailored pictorial recommendations (eg, to quit smoking, control weight or attain a target BP) according to current primary prevention guidelines. This information is sent simultaneously to corresponding physicians’ mobile terminals. Individuals are required to use the app for this game-like estimation/desired goal setting, and accept the motivations for optimised goals, one to three times per month (thereafter, the app locks the estimation function to prevent cross-use). Unless the task has been completed, each participant will receive three reminders in the last week of each month (on Monday, Tuesday and Wednesday; two alarms with text messages sent automatically at 19:00, and a physician phone call). The software will record failure to complete this task as a truancy on the user’s activity calendar. The physician portal provides access to data for an average of 100 patients per physician. The community physicians will uniformly apply pooled cohort equations (PCEs) for risk assessment to guide decision making for primary prevention ; to avoid misunderstanding and with recognition of potential inaccuracy, PCE results will not be presented in the HK app.
Participants will be aged 45–80 years, personally own and use a smartphone (Apple or Android platform) with Internet access, and have at least one of the following CVD risk factors: history of CVD at age <60 years in a first-degree relative, smoking, diabetes, hypertension and low-density lipoprotein (LDL) cholesterol ≥4.5 mmol/L. Participants with histories of CVD, heart failure or chronic kidney disease (estimated glomerular filtration rate <30 mL/min/1.73 m 2 ); those with terminal malignancy at baseline and those with severe psychological or mental disorders will be excluded. Violation of the study protocol, including inadequate informed consent; enrollment of subjects not meeting the inclusion/exclusion criteria; improper breaking of the blinding; multiple visits missed or outside permissible windows; inadequate record-keeping; intentional deviation from the protocol; repeated non-compliance by the subject; and falsification, and participation in another clinical study during follow-up will prompt exclusion.
The research nurses will contact potentially eligible participants identified from the NBPHSP database, explaining the trial and ascertaining interest. Interested patients will be evaluated at their primary healthcare centres; those eligible for EMT-OCSP participation will be invited to participate during NBPHSP-related visits. The research nurses will send invitations for baseline visits with available appointment times by message, and register all forms documenting NBPHSP baseline visits. Reasons for individuals’ refusal to participate in the study will be documented.
Using a computer-generated randomisation schedule with permuted blocks of random sizes, study participants will be assigned to groups A–D , representing different CVD-free life expectancy formats, and receive corresponding smartphone app upgrade packages. Random numbers will be generated prior to the study by a computerised random number generator. Block sizes will not be disclosed until primary endpoint analysis. Randomisation will be requested by the staff member responsible for recruitment and clinical interviewing from the Clinical Trial Coordinating Centre. The intervention does not permit participant or provider allocation blinding, but providers will be instructed to not disclose allocation status at follow-up assessments. Staff responsible for data collection and analysis will be blinded to group allocation. At randomisation, participants will be assigned unique identification codes.
The study flow, schedule and activities are presented in and , respectively. The same 12-month outcome schedule will be used for all randomised participants, regardless of intervention completion/discontinuation, except for the adherence assessment. The study will be closed on 21 January 2022. Data collection will be finalised in January 2022, or on completion of all 1-year visits. Study-related primary healthcare centre visits will be conducted using motivational interviewing methodology, with the aim of promoting health and CVD prevention. A pictorial depiction of the LIFE-CVD model will be used to facilitate individuals’ understanding of how their lifestyles and drug therapies are linked to CVD risk. Participants will be blinded to study hypotheses and theories underlying which interventions are considered to be active. At baseline visits, in addition to health assessment, participants will install and test the HK app, register and link their app accounts for this study. Following standardised, simplified procedures, community physicians will teach participants how to use the app calculator and communicate about calculation results using the ‘teach-back’ method, providing additional information when needed. Family members will be permitted to attend baseline visits. Participants will use the HK app to set goals, receive performance feedback and self-monitor primary CVD prevention. Physicians at primary healthcare centres will routinely manage participants according to primary CVD prevention guidelines. They will aim to provide accurate, accessible and understandable CVD-related information and confirm participants’ understanding of calculation results and conceptualisation of CVD-free life expectancy as dynamic, modifiable by lifestyle changes and pharmacological treatment. Participants will also be given written guidance on CVD-free life expectancy calculation. The intervention will be regarded as low intensity if fewer than 6 monthly CVD-free life expectancy calculations per year, or fewer than three calculations in the first 6 months, are completed. Using the app, research nurses will periodically inform healthcare centre staff and participants about CVD prevention, the current trial status and plans for the next phase, and acknowledge their support. At 9 months, participants will receive a message with information about the study proceedings and a reminder about the 1-year follow-up visit, irrespective of participation status at the 6-month follow-up, provided that they have not been excluded or withdrawn consent. At recruitment and 6 months, physicians will indicate the importance of following study guidelines (including once-monthly calculation) and contacting physicians with potentially study-related problems, and explain the app’s purpose, use and updating, what to do in the event of smartphone/app loss/damage and monthly task counts and study visits. Participants will be asked whether they are having problems with app use. At follow-up sessions, reasons for task non-completion and simple strategies for enhancing adherence (eg, linking calculation to normal routines) will be discussed briefly. Follow-up assessments and interviews will be conducted at 6 and 12 months after randomisation, within a 28-day window extending from 7 days before to 21 days after the due date. Research nurses will prompt the healthcare centres to invite participants and conduct the 6-month and 12-month examinations (ie, risk factor measurement and questionnaire administration) at the local primary healthcare centres. When appropriate, participants will receive recommendations for additional follow-up visits and/or referrals to community physicians for pharmacological treatment according to CVD prevention guidelines. Cross-group contamination will be assessed at 1 year (have you talked to other participants about your LIFE-CVD model, and if yes was your attitude changed? Are you aware of the model of a participant in another group, or vice versa?). Research nurses will send assessment results to the research centre. All participants will be informed about these results.
The primary endpoints are changes in the estimated 10-year CVD risk, estimated lifetime CVD risk and estimated CVD-free life expectancy from baseline to the 1-year follow-up, based on the levels of total, HDL and LDL cholesterol; systolic BP; body mass index; history of diabetes mellitus and/or early (age <60 years) parental myocardial infarction; antithrombotic therapy and high BP treatment; diabetes; smoking habit and age. Secondary endpoints are changes in CVD risk factors (BP and serum cholesterol, LDL, non-HDL, triglycerides and fasting glucose levels); lifestyle factors (physical activity, tobacco use, alcohol use and dietaryhabits); pharmacological treatments for hypertension, dyslipidaemia and diabetes; and anti-thrombotic drug prescriptions after 1 year.
From the baseline questionnaire, data on age, sex, civil status, education, family histories of premature CVD and diabetes, and systolic and diastolic BP will be recorded. Socioeconomic status will be classified according to the per-capita disposable income of urban households (by quintile) and highest attained educational level. The following data will be collected: self-reported lifestyle data (tobacco use; alcohol use (Alcohol Use Disorders Identification Test) ; physical activity and time spent sitting; dietary habit (daily fruit, root, legume and vegetable consumption); health; diabetes and BP and lipid-lowering pharmacological treatments), physical measurements (height (in cm), weight (in kg), systolic and diastolic BP (in mm Hg)) and blood examination findings (serum total, HDL and LDL cholesterol, triglycerides and fasting glucose levels (in mmol/L), determined using standard biochemical methods). Height and weight will be measured with subjects barefoot in light clothing using calibrated scales and stadiometers. Systolic and diastolic BP will be measured two times with a calibrated digital gauge (2 mm precision) after 5 min rest with subjects seated; mean values will be recorded. Blood samples will be collected after overnight fasts and sent to the clinical chemistry departments of the nearest local hospitals. Data on prescriptions for antithrombotic medications and those used to treat hypertension, dyslipidaemia and diabetes will be retrieved from the digital medical records of community hospitals. Purchases of pharmacological products from community hospitals, registered in participants’ medical insurance records, will also be recorded. Psychological variables will be used as moderators/mediators of intervention effects on primary and secondary outcomes, and as outcomes at 1 year. At baseline and 1 year, participants’ health literacy, coping strategies (using the Brief Coping Orientation to Problems Experienced Instrument), general self-efficacy, anxiety and depression (using the Hospital Anxiety and Depression Scale), and optimism/pessimism (using the Life Orientation Test) will be measured. At baseline and follow-up visits, participants will be asked to self-rate their health (five alternatives), CVD risk and health-specific self-efficacy (ability to reduce CVD risk through preventive actions) using a visual analogue scale (0–10).
Before participant recruitment, community physicians and key personnel at participating sites will be trained and certified. Personnel at each centre will be trained centrally, and training session content will be standardised and presented by members of the study’s executive team. The beta version of the HK app, operating instructions, forms, training manuals and other materials will be provided. The training session will cover trial overview; participant recruitment, eligibility and exclusion criteria; participant visit procedures; app-based data entry; data collection; physician–patient health communication requirements (standardised trial introduction, app use instruction, education and counselling for adherence, uniform and reproducible elicitation and explanation of information from the app and data discrepancy management) and general information about obtaining high-quality research data, intervention providers’ responsibilities, workflow and strategies to minimise communication bias and cross-group contamination. The providers should meet well-defined performance criteria, assessed by observation during role play with standardised mock participants.
A 1-month pilot study was performed in 2019 to test the original HK app with 200 participants from a district far from the study region, with different socioeconomic statuses and at least one CVD risk factor. The study organisation worked well, and its protocols and procedures were developed further and showed good feasibility. Two per cent cross-group contamination was detected. Life expectancy and CVD risk information was calibrated according to participants’ experiences and suggestions, obtained through questionnaires and interviews. Community physicians’ experiences, derived in part from discussions of the results with participants, were also considered. Based on these findings, and as we expect <20% contamination in the EMT-OCSP Study, we chose to use an individually randomised design rather than a clustering method.
The sample size needed to detect clinically significant between-group differences in baseline–1-year changes in primary outcome variables, with sufficient power and considering 20% drop-out, was determined to be 15 000 . We used a power threshold of 0.9 and significance level of 0.05. Calculation was based on SD of CVD-free life expectancy, CVD risk and conventional risk factors derived from the China National Stroke Screening Survey and the best available data so far.
Data management and statistical analyses will be performed using SPSS software (V.20.0; SPSS). Analyses will include data from all study participants. Data from individuals who do not consent to participation (sex, age, socioeconomic status, clinical CVD risk factors and lifestyle factors) will be included in selection bias analysis. The baseline demographic and clinical characteristics of the four parallel groups will be compared. Continuous variables (represented as means and SDs) will be compared using analysis of variance (ANOVA) and categorical variables (represented by frequencies and percentages) will be compared using χ 2 tests. Wilcoxon test will be used if necessary. Primary and secondary outcomes will be evaluated according to intention-to-treat methodology. The 2×2 factorial design ANOVA will be used to assess effects of message framing and time discounting and interactions between them due to both independent variables having two levels. Differences in changes of primary outcomes from baseline to 1-year follow-up will be analysed with regression methods to identify predictors of the changes in the endpoints. In subgroup analyses, we will estimate the intervention’s effects on primary outcomes for different age groups, sexes, CVD risk levels, socioeconomic status and interventional intensities. We will also perform Bonferroni correction for multiple comparisons in our analysis and sensitivity analyses to assess the robustness of trial results obtained with different methods of handling missing data.
No patient involved.
Funding and approval The EMT-OCSP Study is supported by the China National Key Research and Development Project (2018YFC1311406). Funding covers only meetings and organisational costs. This funding source had no role in the study design and will have no role in study execution, data analysis or interpretation or the decision to publish results. The study application was submitted to the institutional review board of West China Hospital in 2019. Approval to conduct study activities and informed consent forms will be obtained from the regional ethics committee overseeing the participating healthcare centres, and attached to the application to the regional ethics board. The study will be conducted according to ethical principles based on the Helsinki Declaration, good clinical practice (GCP), national regulatory mandatory instructions and this protocol. Informed consent and anonymity Participants (or representatives) will be required to provide informed consent after receipt of written and oral study information and before any study procedure is performed. The informed consent form will describe study monitoring procedures and the handling of access to national registry data and participants’ medical records. Subjects will be informed verbally and in writing that participation is voluntary and can be withdrawn at any time, with no consequence for ordinary healthcare. District nurses will provide written confirmation that study information has been provided, and will collect informed consent forms and send them to the research nurses. Participant anonymity will be maintained through transcript masking and restriction of raw data access to the study team. Fear-arousing communication As with all screening targeting healthy populations, the presentation of CVD risk profiles is of concern. The app’s intuitive presentation of CVD-free life expectancy may generate more anxiety than would being informed of risk marker increases. The risk messages were confirmed in the pilot study to be easily understandable and accurately interpreted. Any negative reaction (eg, anxiety) that they generate will be addressed with individualised supportive counselling from research nurses. Participants may communicate questions to research nurses at any time. We expect that effective risk communication will increase CVD prevention guideline compliance. We do not consider this to be ethically problematic, as message shaping has not been demonstrated to improve prognoses among CVD-free subjects, and as all communication will follow current guidelines. The study will not compromise participants’ receipt of healthcare. On invitation to NBPHSP participation, individuals are informed that health screening will be combined with advice and description of measures to improve long-term health. The EMT-OCSP Study will adhere fully to all intentions of the NBPHSP. The expected improvements in risk stratification and communication outweigh any potential negative effect of the study. Individualisation and contamination control As phone numbers in China can be registered only under users’ real names and identification card numbers, HK app user IDs will be unique login credentials corresponding to individual mobile devices. The activities permitted for individual users will be regulated by these credentials, preventing unauthorised access to or loss of participant data. If a participant’s mobile phone is lost or broken, he/she will need to apply for a new account through the community physician to the research centre. Although the intervention will be housed on individuals’ smartphones, the EMT-OCSP Study is likely susceptible to contamination. App messages cannot be fully blinded, and friends randomised to different study groups are likely to discuss app results. To minimise contamination, we will educate physicians in contamination avoidance; instruct participants to not transfer, share or exchange intervention content; prevent cross-use through app design; separate different groups’ follow-up schedules; measure contamination at the individual level to permit analytical control and increase the sample size to compensate. If necessary, we will use contamination-adjusted intention-to-treat analysis, which employs instrumental variables analysis. Protocol modification Any significant protocol modification (eg, of eligibility criteria, sample size, outcomes assessed or analyses) will require formal approval by the EMT-OCSP research group and the institutional review board prior to implementation. Minor corrections/clarifications that do not affect study conduct will be approved by the research group and documented in memoranda. Data storage and access All data will be entered into the password-protected EMT-OCSP research database at the Clinical Research Centre of West China Hospital, Sichuan University, managed by a trained database manager. Different security levels (password access) may be granted to groups and individuals. To enable external data linkage, participants’ identification numbers will be retained in the database. The data manager will log study data in access files on computers used solely for this purpose, and import these files, consent forms and case report forms into the database. All records with participant ID numbers linked to other identifying information will be stored in a separate, locked file in an area with limited access. Identifying information will be redacted from data dispersed to team members. The core coordinating centres will have access only to their own data. The principal investigator will have direct access to datasets from her own site, and to data from other sites by request. The key code will be retained for a maximum of 3 years. Participant files will be stored for 3 years after study completion. All personal data will be processed in accordance with the Cybersecurity Law of People’s Republic of China, and will be available annually and free of charge to study participants, who can request error correction. Participants’ study information will not be released outside of the study without their written permission. Prevention of missing data We will seek to follow participants for the entire study period, regardless of intervention modification/discontinuation, to enable follow-up data collection and prevent missing data. The pilot study helped to identify and address potential problems to minimise missing data. The study’s executive team will set a priori targets for the acceptable level of missing data, and on-site data collection will be monitored and reported monthly. Reasons for non-adherence (eg, intervention discontinuation due to harms vs inefficacy) and non-retention (ie, consent withdrawal or loss to follow-up) will be recorded. Data monitoring and quality assurance In addition to app-based manual and automatic data checks, data quality and completeness will be ensured by monthly visual validation at the data coordinating centre (DCC) and regular on-site monitoring (at least once per participating site). Errors will be recorded in data query reports sent to data managers at core coordinating centres. The managers will check the original sources for inconsistency, check other sources to determine the correction required and correct the errors. DCC monitors will review source documents as needed to assess the completeness and accuracy of app data (patient initials, birthdate, sex, written informed consent provision, eligibility criteria fulfilment, randomisation date and intervention assignment). Through visits and electronic monitoring, the DCC will audit overall data quality and completeness and send monthly reports by email with information on missing data and missed visits. Personnel at the core coordinating centre and participating sites will review these reports for accuracy and report any discrepancies to the DCC. A qualified external monitor from the Clinical Trial Unit of West China Hospital, Sichuan University, will ensure that the data are reliable, accurate and complete; participants’ safety and rights are protected; and the study is conducted in accordance with current protocols and study agreements, as well as GCP and all applicable government requirements. The principal investigator has approved the monitor’s direct access to relevant documentation and will ensure co-investigators’ and departments’ cooperation with monitoring. He will be notified of corrective/preventive measures for any violation or negligence. Research data requests Researchers affiliated with the EMT-OCSP Study will have access to an internal webpage listing available aggregated datasets and variables; they will not have access to individual data or the original database. External researchers collaborating with at least one steering group member may submit standardised applications for data via this webpage. On steering group approval, the data manager will export anonymised aggregated data to authorised researchers. Publication policy The scientific integrity of the trial requires that data from all participating sites be analysed study-wide. Individual centres are not expected to report solely on data collected on site. A publications committee will approve all papers and abstracts based on study data before submission. The study results will be released to participating physicians and patients, and to the medical community.
The EMT-OCSP Study is supported by the China National Key Research and Development Project (2018YFC1311406). Funding covers only meetings and organisational costs. This funding source had no role in the study design and will have no role in study execution, data analysis or interpretation or the decision to publish results. The study application was submitted to the institutional review board of West China Hospital in 2019. Approval to conduct study activities and informed consent forms will be obtained from the regional ethics committee overseeing the participating healthcare centres, and attached to the application to the regional ethics board. The study will be conducted according to ethical principles based on the Helsinki Declaration, good clinical practice (GCP), national regulatory mandatory instructions and this protocol.
Participants (or representatives) will be required to provide informed consent after receipt of written and oral study information and before any study procedure is performed. The informed consent form will describe study monitoring procedures and the handling of access to national registry data and participants’ medical records. Subjects will be informed verbally and in writing that participation is voluntary and can be withdrawn at any time, with no consequence for ordinary healthcare. District nurses will provide written confirmation that study information has been provided, and will collect informed consent forms and send them to the research nurses. Participant anonymity will be maintained through transcript masking and restriction of raw data access to the study team.
As with all screening targeting healthy populations, the presentation of CVD risk profiles is of concern. The app’s intuitive presentation of CVD-free life expectancy may generate more anxiety than would being informed of risk marker increases. The risk messages were confirmed in the pilot study to be easily understandable and accurately interpreted. Any negative reaction (eg, anxiety) that they generate will be addressed with individualised supportive counselling from research nurses. Participants may communicate questions to research nurses at any time. We expect that effective risk communication will increase CVD prevention guideline compliance. We do not consider this to be ethically problematic, as message shaping has not been demonstrated to improve prognoses among CVD-free subjects, and as all communication will follow current guidelines. The study will not compromise participants’ receipt of healthcare. On invitation to NBPHSP participation, individuals are informed that health screening will be combined with advice and description of measures to improve long-term health. The EMT-OCSP Study will adhere fully to all intentions of the NBPHSP. The expected improvements in risk stratification and communication outweigh any potential negative effect of the study.
As phone numbers in China can be registered only under users’ real names and identification card numbers, HK app user IDs will be unique login credentials corresponding to individual mobile devices. The activities permitted for individual users will be regulated by these credentials, preventing unauthorised access to or loss of participant data. If a participant’s mobile phone is lost or broken, he/she will need to apply for a new account through the community physician to the research centre. Although the intervention will be housed on individuals’ smartphones, the EMT-OCSP Study is likely susceptible to contamination. App messages cannot be fully blinded, and friends randomised to different study groups are likely to discuss app results. To minimise contamination, we will educate physicians in contamination avoidance; instruct participants to not transfer, share or exchange intervention content; prevent cross-use through app design; separate different groups’ follow-up schedules; measure contamination at the individual level to permit analytical control and increase the sample size to compensate. If necessary, we will use contamination-adjusted intention-to-treat analysis, which employs instrumental variables analysis.
Any significant protocol modification (eg, of eligibility criteria, sample size, outcomes assessed or analyses) will require formal approval by the EMT-OCSP research group and the institutional review board prior to implementation. Minor corrections/clarifications that do not affect study conduct will be approved by the research group and documented in memoranda.
All data will be entered into the password-protected EMT-OCSP research database at the Clinical Research Centre of West China Hospital, Sichuan University, managed by a trained database manager. Different security levels (password access) may be granted to groups and individuals. To enable external data linkage, participants’ identification numbers will be retained in the database. The data manager will log study data in access files on computers used solely for this purpose, and import these files, consent forms and case report forms into the database. All records with participant ID numbers linked to other identifying information will be stored in a separate, locked file in an area with limited access. Identifying information will be redacted from data dispersed to team members. The core coordinating centres will have access only to their own data. The principal investigator will have direct access to datasets from her own site, and to data from other sites by request. The key code will be retained for a maximum of 3 years. Participant files will be stored for 3 years after study completion. All personal data will be processed in accordance with the Cybersecurity Law of People’s Republic of China, and will be available annually and free of charge to study participants, who can request error correction. Participants’ study information will not be released outside of the study without their written permission.
We will seek to follow participants for the entire study period, regardless of intervention modification/discontinuation, to enable follow-up data collection and prevent missing data. The pilot study helped to identify and address potential problems to minimise missing data. The study’s executive team will set a priori targets for the acceptable level of missing data, and on-site data collection will be monitored and reported monthly. Reasons for non-adherence (eg, intervention discontinuation due to harms vs inefficacy) and non-retention (ie, consent withdrawal or loss to follow-up) will be recorded.
In addition to app-based manual and automatic data checks, data quality and completeness will be ensured by monthly visual validation at the data coordinating centre (DCC) and regular on-site monitoring (at least once per participating site). Errors will be recorded in data query reports sent to data managers at core coordinating centres. The managers will check the original sources for inconsistency, check other sources to determine the correction required and correct the errors. DCC monitors will review source documents as needed to assess the completeness and accuracy of app data (patient initials, birthdate, sex, written informed consent provision, eligibility criteria fulfilment, randomisation date and intervention assignment). Through visits and electronic monitoring, the DCC will audit overall data quality and completeness and send monthly reports by email with information on missing data and missed visits. Personnel at the core coordinating centre and participating sites will review these reports for accuracy and report any discrepancies to the DCC. A qualified external monitor from the Clinical Trial Unit of West China Hospital, Sichuan University, will ensure that the data are reliable, accurate and complete; participants’ safety and rights are protected; and the study is conducted in accordance with current protocols and study agreements, as well as GCP and all applicable government requirements. The principal investigator has approved the monitor’s direct access to relevant documentation and will ensure co-investigators’ and departments’ cooperation with monitoring. He will be notified of corrective/preventive measures for any violation or negligence.
Researchers affiliated with the EMT-OCSP Study will have access to an internal webpage listing available aggregated datasets and variables; they will not have access to individual data or the original database. External researchers collaborating with at least one steering group member may submit standardised applications for data via this webpage. On steering group approval, the data manager will export anonymised aggregated data to authorised researchers.
The scientific integrity of the trial requires that data from all participating sites be analysed study-wide. Individual centres are not expected to report solely on data collected on site. A publications committee will approve all papers and abstracts based on study data before submission. The study results will be released to participating physicians and patients, and to the medical community.
Reviewer comments Author's manuscript
|
Evaluating the role of large language models in inflammatory bowel disease patient information | 1aef9b15-945a-434a-82b7-617f7d23940c | 11326091 | Patient Education as Topic[mh] | We are writing to express out thoughts on the recently published article by Gravina et al . Gravina et al assessed the capability of large language models (LLMs) like ChatGPT to provide plausible medical information to patients with inflammatory bowel disease (IBD). Despite identifying several limitations, the authors concluded that there is significant potential in using LLMs for this purpose. One of the key insights from the article is the potential for ChatGPT to offer immediate and accessible information to patients. The authors correctly note that this could be particularly beneficial in providing preliminary guidance and answering common queries that patients may have about their condition. This aligns with the increasing trend of patients seeking health information online before consulting their healthcare providers. However, the study also underscores significant limitations, such as the potential for outdated or inaccurate information. Given that medical knowledge is continuously evolving, it is crucial for artificial intelligence (AI) tools like ChatGPT to have mechanisms for regular updates to ensure the information provided is current and evidence-based. This is especially important for chronic conditions like IBD, where treatment guidelines and best practices frequently change. A pertinent question arises: Can LLMs truly perform inference? Current AI-based agents utilizing LLMs operate by either generating answers directly or referring to external tools if the LLM itself cannot provide an answer. These agents determine the necessary information, redefine the questions, call appropriate tools to extract information, analyze the extracted data, and iterate this process as needed to reach a final answer. This pattern, known as reasoning + action, closely mimics human problem-solving by iteratively refining questions and seeking relevant tools rather than merely retrieving similar past solutions. The effectiveness of such an approach often hinges on prompt engineering. Enhanced prompt engineering can significantly improve the accuracy of LLM-generated answers by aligning queries more closely with the model’s trained data and inference capabilities. Therefore, evaluating LLMs based on selected questions often reflects their proficiency in leveraging search tools to produce desired answers. Advanced prompt engineering techniques can potentially yield more accurate responses, indicating that simple question-and-answer testing might not fully capture an LLM’s capabilities. Moreover, the retrieval-augmented generation (RAG) technique enhances traditional LLMs by enabling real-time retrieval of external data not included in the training dataset, thus generating answers that integrate the latest information. This approach helps prevent hallucination and allows the model to utilize a broader knowledge base. However, standardized performance evaluation of these advanced techniques remains challenging due to the limited benchmarks available, making it difficult to assess using only a few representative questions. Another important point raised by the authors is the issue of contextual understanding and empathy, which AI currently lacks. The physician patient relationship is built on trust and understanding, and while AI can provide factual information, it cannot replace the nuanced, empathetic communication that healthcare providers offer. This aspect is particularly vital for managing chronic diseases that significantly impact patients’ quality of life. The authors’ recommendation for further refinement and alignment of AI outputs with reliable medical databases is essential. Such improvements could enhance the accuracy and reliability of AI-generated medical information, making it a more robust tool for both patients and healthcare providers. Despite these challenges, there is no doubt that LLMs, equipped with sophisticated learning datasets and RAG capabilities, hold promise for clinical application. However, evaluating their potential solely based on simple question-answer accuracy is inadequate. It is essential to consider the advanced techniques and iterative processes that significantly enhance the precision and reliability of LLM-generated medical information. In conclusion, the article by Gravina et al provides valuable insights into the current capabilities and limitations of AI in gastroenterology. While promising, further refinement and a more nuanced evaluation approach are crucial for realizing the full potential of AI in healthcare. Continued research and development, combined with rigorous validation against established medical standards, will be essential.
|
A cross-sectional study reporting concussion exposure, assessment and management in Western Australian general practice | bf76e7d6-da4e-479b-a556-663af2507b68 | 7927406 | Family Medicine[mh] | Sports-related concussion (SRC) has been the focus of increased publicity in recent years as a result of attention in the lay press, however, SRC accounts for approximately only 20% of concussion in the general population . Other major causes of concussion include falls, motor vehicle and bicycle crashes, and assault, [ – ] with falls the leading cause in older populations . General Practitioners (GPs) are often called upon to assess patients who have sustained a concussion and historical notes or a recent discharge summary from the Emergency Department may be unavailable at this assessment. The GP may, or may not, know the patient, and it may, or may not, be clear to the GP whether the patient’s presenting symptoms originate from the described concussive injury. This complicates a GP’s understanding of how a patient is affected by a concussion and limits their ability to optimise early clinical assessment and management with a view to improving long term outcomes. The diagnosis of concussion is clinical and is made following assessment by a medical practitioner. There are currently no blood tests or imaging techniques available to GPs to confirm a diagnosis of concussion . Symptoms such as dizziness, fogginess, headache, nausea, sleep and mood disturbance, memory impairment and neck pain are indicators of concussion after either a direct blow to the head, or after sustaining a transmitted force to the head from an impact to the body . Collateral history may also assist with diagnosis, with brief loss of consciousness, unsteadiness, posturing, a dazed appearance, disorientation and agitation all suggestive of concussion . Once a GP has excluded focal neurology, a diagnosis of concussion is considered. After making such a diagnosis a GP must then address how the patient can be best managed and make decisions around the need for involvement of other health care professionals in the patient’s management. Typically, concussed adults recover in 10–14 days . Normal recovery from concussion in children can take up to 28 days [ – ]. Approximately 85% of patients will recover within these timeframes . The role of the GP is pivotal to the patient’s safe return to activities of daily living (ADLs), and includes decisions around screen time, reading, exercise, school and learning, work, driving, sport, and the potential need for other intervention. Research providing support for these decisions can be difficult to interpret, contradictory, complex and difficult to implement, especially outside return to sport protocols . Furthermore, there is a potential risk of slowing recovery time, or exacerbating patient condition, by failing to optimise management or by returning to ADLs too soon . This creates a clinical environment prone to variation in care between clinicians and between institutions or jurisdictions, particularly outside SRC management. Information relating to how concussion is being managed is scarce, and it is unknown whether the care is optimal and whether the impost on the health care system can be reduced simply by improved medical education. There is a paucity of Australian literature exploring GPs’ exposure to patients presenting with concussion. This study aimed to identify characteristics of current Western Australian (WA) GP exposure to patients with concussion, factors associated with their knowledge, confidence in diagnosing and managing patients with concussion, referral practices and familiarity with guidelines. Such information is necessary to identify the challenges faced by GPs relating to concussion diagnosis and management. Identification of such challenges is crucial for targeting areas where GPs require increased support and improved GP education, with the ultimate goal of improving patient care.
Study design and sample recruitment This cross-sectional survey of WA GPs received ethics approval from Curtin University Ethics Committee (Approval Number HREC2019-0602). GPs were recruited using a convenience sample over a two-month period. To be included in this study, GPs were required to be working in a clinical role in a general practice setting within WA. Data collection An electronic survey was developed to determine the current approach to concussion diagnosis and management in general practice in WA (Supplementary Text ). The study investigators assessed the readability and content of the questionnaire to assure its face and content validity. Key areas were identified from current literature . Demographics of the respondents, as well as their reported exposure to, and identification of, concussion diagnoses were included, as were questions relating to GP knowledge of symptoms and signs commonly identified as potential flags for a concussion diagnosis encompassing all “sub-types” of concussion . The survey was disseminated via the Royal Australian College of General Practitioners (RACGP) WA newsletter and the WA ShareGP group. The survey (Supplementary Text ) was intended to briefly assess: scope of practice. This included age and professional level of the doctor, additional post graduate qualifications and special interest affiliations, hours worked, practice location (metropolitan/regional) and work outside general practice. exposure to patients with potential/confirmed concussion and frequency of diagnosis. knowledge of potential symptoms of concussion: possible symptoms and distractors were offered (Supplementary Text a). knowledge of potential clinical signs of concussion: possible signs and distractors were offered (Supplementary Text b). confidence in diagnosis, management, and identification of prolonged recovery follow up and referral patterns familiarity with any concussion guideline(s) or assessment tool(s) use of clinical coding in General Practice Data analyses Demographic characteristics, usual practice and knowledge of the GPs were compared by practice location: mean and standard deviation were reported for continuous variables whilst frequency and percentage were reported for categorical variables. Independent t-tests were performed to compare group means for normally distributed continuous variables. Associations between categorical variables and practice location, as well as knowledge, confidence and practices were performed using Chi-square tests, or Fisher’s Exact test when there was low number of expected frequencies. The significance level was set at 0.05. Data were analysed using Stata 14/IC (Stata Corp, Texas).
This cross-sectional survey of WA GPs received ethics approval from Curtin University Ethics Committee (Approval Number HREC2019-0602). GPs were recruited using a convenience sample over a two-month period. To be included in this study, GPs were required to be working in a clinical role in a general practice setting within WA.
An electronic survey was developed to determine the current approach to concussion diagnosis and management in general practice in WA (Supplementary Text ). The study investigators assessed the readability and content of the questionnaire to assure its face and content validity. Key areas were identified from current literature . Demographics of the respondents, as well as their reported exposure to, and identification of, concussion diagnoses were included, as were questions relating to GP knowledge of symptoms and signs commonly identified as potential flags for a concussion diagnosis encompassing all “sub-types” of concussion . The survey was disseminated via the Royal Australian College of General Practitioners (RACGP) WA newsletter and the WA ShareGP group. The survey (Supplementary Text ) was intended to briefly assess: scope of practice. This included age and professional level of the doctor, additional post graduate qualifications and special interest affiliations, hours worked, practice location (metropolitan/regional) and work outside general practice. exposure to patients with potential/confirmed concussion and frequency of diagnosis. knowledge of potential symptoms of concussion: possible symptoms and distractors were offered (Supplementary Text a). knowledge of potential clinical signs of concussion: possible signs and distractors were offered (Supplementary Text b). confidence in diagnosis, management, and identification of prolonged recovery follow up and referral patterns familiarity with any concussion guideline(s) or assessment tool(s) use of clinical coding in General Practice
Demographic characteristics, usual practice and knowledge of the GPs were compared by practice location: mean and standard deviation were reported for continuous variables whilst frequency and percentage were reported for categorical variables. Independent t-tests were performed to compare group means for normally distributed continuous variables. Associations between categorical variables and practice location, as well as knowledge, confidence and practices were performed using Chi-square tests, or Fisher’s Exact test when there was low number of expected frequencies. The significance level was set at 0.05. Data were analysed using Stata 14/IC (Stata Corp, Texas).
Characteristics, usual practice and knowledge of the GP respondents The survey was disseminated to 3880 WA GPs. Sixty-six GPs responded. The response rate was 1.7%. The respondent furthest north was from Karratha and furthest south from Albany. Of the respondents one was a GP registrar and 12 held dual Fellowships of the Australian College of Rural and Remote Medicine (ACRRM) and the RACGP. Eight respondents (12%) had additional post-graduate level Sports Medicine qualifications (Certificate/Diploma in Sports Medicine). Twenty-one respondents (32%) reported also working outside their general practice surgery setting. None of the respondents were members of the Sports and Exercise Medicine or Musculoskeletal Medicine RACGP Specific Interest groups. The demographics of GPs who practised in regional and metropolitan areas were comparable ( p > 0.05) (Table ). Diagnoses made, referral practices, knowledge of symptoms and signs of concussion, confidence in diagnosis and management, identification of prolonged recovery, and guideline and resource awareness were also comparable between GPs who practised in regional and metropolitan areas (Table ). Approximately 45% of the respondents reported requesting imaging to diagnose a concussion. Despite 81% of respondents reporting feeling confident making a concussion diagnosis, only 63% felt confident to manage this diagnosis (Table ). A lack of clarity around what constitutes a delayed recovery was also identified (Table ). When patients were deemed to be “slow to recover”, close observation and referral on to a specialist were identified as management strategies for 53% of responders. Thirty-five responders identified their referral pathways; the most commonly identified was the emergency department (38%), followed by sports medicine clinic (25%), physiotherapist (18%), neurologist (13%), neurosurgeon (5%) and concussion clinic (3%). Association between annual number of concussion diagnoses and frequency of concussion as a secondary diagnosis Among the GPs who made less than five concussion diagnoses per year, most (78%) never o r rarely made a diagnosis of concussion as a secondary diagnosis (Table ). Similarly, most GPs who made more than five concussion diagnoses per year (77%) never o r rarely diagnosed concussion as a secondary diagnosis. There was no association between the number of concussion diagnoses made per year and the frequency of concussion as a secondary diagnosis (Table ). Symptom and Sign Identification All respondents correctly identified headache, fogginess, dizziness, difficulty concentrating, irritability and drowsiness or sleep disturbance as potential symptoms of concussion. Sixty-three respondents (96%) identified sensitivity to light or sound as a possible symptom, 62 (94%) identified nausea and vomiting as potential symptoms and 41 (63%) identified neck pain as potentially important when making a concussion diagnosis. Twenty-three (35%) respondents both correctly identified all offered symptoms of concussion and correctly excluded the distractors (Table ). Characteristics of GPs were similar between those who identified all symptoms of concussion and distractors correctly and those who did not (Table ). However, familiarity with guidelines was associated with concussion knowledge; eighty-four percent of GPs who were unaware of any guidelines were less likely to answer all symptoms and distractors correctly whilst 87% of the GPs who correctly identified all symptoms of concussion and distractors were aware of guidelines (Table ). Sixty-four respondents correctly identified balance disturbance, vestibular-ocular impairment and objective memory impairment as signs of concussion. Sixty-three respondents (98%) suspected concussion when presented with a facial or scalp injury. Forty-six respondents (72%) identified exercise intolerance and 45 (70%) identified orthostatic hypotension as potential signs of concussion. Forty-three respondents (67%) would be suspicious of concussion in a patient with neck tenderness. Twenty-three respondents (35%) identified all positive signs of concussion but no respondents also correctly identified all distractors (Table ). As such, test of association could not be performed. Associations between GP characteristics and diagnostic confidence as well as referral practices GPs’ hours worked, age group, professional level, location of practice, having work outside sessional load and having a protocol of coding were comparable between those who were, and were not, confident in diagnosing concussion (Table ). However, the number of concussion diagnoses a GP made per year was associated with their confidence in diagnosing concussion (Table ). Ninety-two percent of the GPs who were not confident in diagnosing concussion made less than five concussion diagnoses per year (Table ). Having heard of concussion guidelines was also significantly associated with confidence in diagnosing concussion, with 78% of the GPs who were confident in their diagnoses aware of guidelines whilst 58% of the GPs who were not confident in their diagnoses having never heard of concussion guidelines (Table ). The number of concussion diagnoses a GP made per year was also associated with their confidence in managing concussion (Table ). Seventy-nine percent of the GPs who were not confident in managing concussion made less than five concussion diagnoses per year. Age group of the GPs was associated with their confidence in managing concussion with 75% of the GPs who were not confident aged 45 years old or younger, whilst 55% of the GPs who were confident aged over 45 years old (Table ). Contradictory to confidence in diagnosing concussion, confidence in managing concussion was not significantly associated with GPs exposure to guidelines (Table ). Characteristics of GPs were similar between the group who would refer for diagnostic imaging and the group who would not (Table ). Similarly, characteristics of GPs were comparable between the group who would refer to specialist and the group who would not (Table ). Associations between knowledge and confidence as well as referral practices and familiarity with guidelines The proportion of GPs who did answer, or did not answer, all symptoms and distractors correctly was equally distributed between the group who were and were not confident in diagnosing as well as managing concussion (Table ), and between referrers and non-referrers to a specialist (Table ) or diagnostic imaging (Table ). Answering all symptoms and distractors of concussion correctly was not associated with the GPs’ confidence in diagnosing or in managing concussion (Table ). Answering all symptoms and distractors correctly was not associated with GPs’ referral to specialist, referral for diagnostic imaging (Table ). Eighty-two percent of the GPs who did not answer all signs and distractors of concussion correctly were confident in diagnosing concussion (Table ). About 53% of the group who did, and 45% of the group who did not, answer all signs and distractors of concussion correctly would refer to a specialist, and diagnostic imaging, respectively (Table ). However, due to lack of GPs answering all the signs of concussion correctly, the tests of association could not be performed.
The survey was disseminated to 3880 WA GPs. Sixty-six GPs responded. The response rate was 1.7%. The respondent furthest north was from Karratha and furthest south from Albany. Of the respondents one was a GP registrar and 12 held dual Fellowships of the Australian College of Rural and Remote Medicine (ACRRM) and the RACGP. Eight respondents (12%) had additional post-graduate level Sports Medicine qualifications (Certificate/Diploma in Sports Medicine). Twenty-one respondents (32%) reported also working outside their general practice surgery setting. None of the respondents were members of the Sports and Exercise Medicine or Musculoskeletal Medicine RACGP Specific Interest groups. The demographics of GPs who practised in regional and metropolitan areas were comparable ( p > 0.05) (Table ). Diagnoses made, referral practices, knowledge of symptoms and signs of concussion, confidence in diagnosis and management, identification of prolonged recovery, and guideline and resource awareness were also comparable between GPs who practised in regional and metropolitan areas (Table ). Approximately 45% of the respondents reported requesting imaging to diagnose a concussion. Despite 81% of respondents reporting feeling confident making a concussion diagnosis, only 63% felt confident to manage this diagnosis (Table ). A lack of clarity around what constitutes a delayed recovery was also identified (Table ). When patients were deemed to be “slow to recover”, close observation and referral on to a specialist were identified as management strategies for 53% of responders. Thirty-five responders identified their referral pathways; the most commonly identified was the emergency department (38%), followed by sports medicine clinic (25%), physiotherapist (18%), neurologist (13%), neurosurgeon (5%) and concussion clinic (3%).
Among the GPs who made less than five concussion diagnoses per year, most (78%) never o r rarely made a diagnosis of concussion as a secondary diagnosis (Table ). Similarly, most GPs who made more than five concussion diagnoses per year (77%) never o r rarely diagnosed concussion as a secondary diagnosis. There was no association between the number of concussion diagnoses made per year and the frequency of concussion as a secondary diagnosis (Table ).
All respondents correctly identified headache, fogginess, dizziness, difficulty concentrating, irritability and drowsiness or sleep disturbance as potential symptoms of concussion. Sixty-three respondents (96%) identified sensitivity to light or sound as a possible symptom, 62 (94%) identified nausea and vomiting as potential symptoms and 41 (63%) identified neck pain as potentially important when making a concussion diagnosis. Twenty-three (35%) respondents both correctly identified all offered symptoms of concussion and correctly excluded the distractors (Table ). Characteristics of GPs were similar between those who identified all symptoms of concussion and distractors correctly and those who did not (Table ). However, familiarity with guidelines was associated with concussion knowledge; eighty-four percent of GPs who were unaware of any guidelines were less likely to answer all symptoms and distractors correctly whilst 87% of the GPs who correctly identified all symptoms of concussion and distractors were aware of guidelines (Table ). Sixty-four respondents correctly identified balance disturbance, vestibular-ocular impairment and objective memory impairment as signs of concussion. Sixty-three respondents (98%) suspected concussion when presented with a facial or scalp injury. Forty-six respondents (72%) identified exercise intolerance and 45 (70%) identified orthostatic hypotension as potential signs of concussion. Forty-three respondents (67%) would be suspicious of concussion in a patient with neck tenderness. Twenty-three respondents (35%) identified all positive signs of concussion but no respondents also correctly identified all distractors (Table ). As such, test of association could not be performed.
GPs’ hours worked, age group, professional level, location of practice, having work outside sessional load and having a protocol of coding were comparable between those who were, and were not, confident in diagnosing concussion (Table ). However, the number of concussion diagnoses a GP made per year was associated with their confidence in diagnosing concussion (Table ). Ninety-two percent of the GPs who were not confident in diagnosing concussion made less than five concussion diagnoses per year (Table ). Having heard of concussion guidelines was also significantly associated with confidence in diagnosing concussion, with 78% of the GPs who were confident in their diagnoses aware of guidelines whilst 58% of the GPs who were not confident in their diagnoses having never heard of concussion guidelines (Table ). The number of concussion diagnoses a GP made per year was also associated with their confidence in managing concussion (Table ). Seventy-nine percent of the GPs who were not confident in managing concussion made less than five concussion diagnoses per year. Age group of the GPs was associated with their confidence in managing concussion with 75% of the GPs who were not confident aged 45 years old or younger, whilst 55% of the GPs who were confident aged over 45 years old (Table ). Contradictory to confidence in diagnosing concussion, confidence in managing concussion was not significantly associated with GPs exposure to guidelines (Table ). Characteristics of GPs were similar between the group who would refer for diagnostic imaging and the group who would not (Table ). Similarly, characteristics of GPs were comparable between the group who would refer to specialist and the group who would not (Table ).
The proportion of GPs who did answer, or did not answer, all symptoms and distractors correctly was equally distributed between the group who were and were not confident in diagnosing as well as managing concussion (Table ), and between referrers and non-referrers to a specialist (Table ) or diagnostic imaging (Table ). Answering all symptoms and distractors of concussion correctly was not associated with the GPs’ confidence in diagnosing or in managing concussion (Table ). Answering all symptoms and distractors correctly was not associated with GPs’ referral to specialist, referral for diagnostic imaging (Table ). Eighty-two percent of the GPs who did not answer all signs and distractors of concussion correctly were confident in diagnosing concussion (Table ). About 53% of the group who did, and 45% of the group who did not, answer all signs and distractors of concussion correctly would refer to a specialist, and diagnostic imaging, respectively (Table ). However, due to lack of GPs answering all the signs of concussion correctly, the tests of association could not be performed.
In this study we examined the knowledge surrounding, and approach to, concussion diagnosis and management of GPs in WA. Overall, the findings suggest that knowledge and management practice amongst GPs is varied, albeit comparable between GPs who work in the metropolitan and regional areas. This supports previous literature where gaps in clinicians’ knowledge have been identified [ – ]. GPs are tasked with the provision of primary care to patients with both acute concussion and those with prolonged or persistent concussive symptoms. Failure to identify, diagnose or appropriately manage such presentations risks poorer outcomes. Likewise, the premature return of patients to ADLs where they may be at risk of further head injury poses a risk of second impact syndrome. Only one GP reported being exposed to more than ten concussions per year whilst 61% of GPs reported exposure to less than five episodes per year. This may be because patients are failing to present for medical assessment following a potentially concussive injury or it may be due to failure of patients to present specifically to general practice. Alternatively, it is possible that a diagnosis of concussion may be overlooked when a patient does present and is assessed. Previous research has suggested that whilst unusual for more than 20 patients to be seen in a year by a family medicine specialist, approximately half see more than ten cases per year . GPs play an important role in the acute phase of symptom management. Previous studies have highlighted the need for ongoing concussion education and awareness to best enable clinicians in this role and understanding the key features in assessment and management of concussion is essential for primary care providers . In this study most respondents identified a delayed recovery as symptoms or signs persisting after five days in both adults and children. It has been demonstrated in experimental models that concussion injury triggers a neurometabolic cascade of events resulting in abnormal potassium, calcium, glutamate, glucose, and lactate levels and altered cerebral blood flow which takes seven to ten days to resolve . Additional microglial and inflammatory responses can continue for considerably longer than this initial metabolic cascade . If GPs are expecting resolution of symptoms in a concussed patient within 5 days, it may be that patients are being allowed to return to activities where they risk sustaining a further concussive force too early and this may have clinical consequences. It has been suggested that phase of recovery should be considered in regards to treatment approaches: Acute (0–4 weeks), Post-Acute (4–12 weeks) and Persistent (> 3 months). . It was rare for a secondary diagnosis of concussion to be made by a respondent. It may be that this is due to the lack of exposure to multi-trauma patients in general practice, or may be because patients present later when concussion has been initially overlooked and other trauma-related primary diagnoses have been made elsewhere. Alternatively, GP exposure to concussion may be predominantly in patients who are improving post-event and requesting further management advice and input regarding return to usual ADLS, that is the concussion was diagnosed elsewhere previously. Thirty-five percent of respondents identified all symptoms of concussion and distractors correctly. It was unexpected that only 63% of respondents identified neck pain as a potential symptom of a concussive injury and only 67% identified neck tenderness as a suggestive sign. Given that concussion can occur as a result of a transmitted force from the body, neck symptoms and signs should be identified as an integral part of any concussion assessment. Education of GPs to assess for concussion in patients presenting with neck pain or tenderness after a potentially concussive injury may increase concussion diagnosis. Whilst GPs correctly identified signs of balance disturbance, objective memory impairment, vestibular-ocular impairment and facial/scalp injury as a sign that may indicate concussive injury, symptoms of autonomic dysfunction were less commonly identified as potential signs. Increasing evidence relating to dysautonomia has emerged in recent years and this has led to development of early subthreshold exercise programs which may improve patient recovery . Increasing GP knowledge in this area may reduce the numbers of patients suffering from prolonged symptoms and as such should be included in future GP concussion education programs. Concussion has been described as having multiple symptoms and signs which can be divided into different ‘sub-types’ or clusters . Diagnostic confidence and confidence in management was seen to increase with increasing exposure to patients with concussion in our study. The variation in concussion presentation means that no one strategy is appropriate for all and may explain why older GPs were more confident in management of concussion. Further research is required to confirm that the confidence in management brought about with age is reflected in patient outcomes. Concussion is a clinical diagnosis with no identifiable findings on standard CT or MRI protocols. Previous literature has suggested that, in clinical practice, cranial CT scanning is likely to be overused in the evaluation of mTBI . It is suspected that doctors using imaging are doing so to exclude other brain injuries and subsequently making a diagnosis of concussion upon receipt of normal results. It may be that imaging is requested due to lack of confidence to clinically diagnose focal neurology, or imaging may be driven by patient request, due to the threat of medico-legal repercussions of a missed alternative neurological diagnosis, or due to an alternative factor such as establishing a baseline in a person who may go on to sustain further concussions. It has been shown in youth concussion that utilisation of conventional neuroimaging results in identification of signs of TBI in 3.1% of cranial CT scans and 1.5% of MRI brain scans . If the imaging in not being performed to exclude alternative diagnosis, this variation in GP care implies that there may be an unnecessary burden on the health economy relating to imaging that may not be justified. Further research clarifying why a GP chooses, or chooses not, to image a patient is required and what modality of imaging is chosen and why. The timing of imaging, if it is requested, also requires further investigation. Previous literature has identified that in paediatric concussion the majority of CT scans have been shown to be obtained during the acute concussion period, whereas MRI scans were ordered later in a patient’s recovery . Failure to clarify these questions is a limitation of this study; prior to national distribution of this survey modification is required to allow this information to be gathered. An awareness of concussion guidelines was associated with confidence in diagnosis but was not associated with confidence in management. Knowledge surrounding best practice relating to concussion management is rapidly evolving and best practice guidelines for SRC are drawn from the four yearly International Conference on Concussion in Sport’s Consensus Statement on Concussion in Sport. Seventy percent of respondents were familiar with at least one concussion guideline (Table ), but only 28 respondents were familiar with the SCAT5 and one was still using a SCAT3. When concussions are sustained in a non-sporting environment inferences are drawn. There is a lack of clear guidance for general practitioners relating to management of concussion. Given that guideline awareness was linked to diagnostic confidence, our results suggest that there needs to be further dissemination of currently available guidelines . However, lack of link to confidence in management suggests that current guidelines may not be useful in this respect to GPs, and this is reflected in lack of information regarding when to allow patients, for example, to return to drive or work. Ideally, nationally consistent, and regularly updated concussion diagnosis and management guidelines relevant to concussion from all causes, for all medical practitioners, including GPs, with links to appropriate resources may be one way of addressing the current inconsistencies in awareness. Education relating to management strategies appears to be an area of need. In a Medline search for “concussion” for 2019 alone there were 867 English language articles identified and it is possible that lack of confidence in this area is related to concussion research rapidly evolving in recent years. Overseas literature has also highlighted gaps in concussion knowledge among family physicians, and deficiencies relating to concussion guideline knowledge, as well as implementation of recommendations, in family physicians treating sports-related concussion . Objective knowledge scores have been demonstrated not to predict self-reported concussion knowledge. . Other than one missing data point, 21 GPs (33%) reported using clinical coding when recording a diagnosis of concussion in patient notes. This presents a problem when collecting and collating epidemiological data. GPs should be encouraged to code all consultations. This may require modification of current electronic medical record programs and ideally should allow for consistency of codes through different providers in both public and private health settings enabling further data regarding the incidence and prevalence to be collected. Previous literature has demonstrated that despite presentations to the emergency department being higher than those to outpatient departments with minor head injury, outpatient presentations were still significant . Accurate GP coding of information is the first step, with subsequent routine data sharing which would facilitate further analysis for incidence estimates and research. There is a risk of selection bias in any study which relies on participants to volunteer to respond to a questionnaire. Despite respondents coming from varying geographical locations across WA, our response rate was low and the findings need to be interpreted with caution. Whilst low response rate is not unusual in research involving GPs, strategies identified by Parkinson et al (such as providing incentives and rewards, provision of paper format questionnaires with reply paid envelopes and contacting practice managers to encourage their general practitioners to complete the survey) will be incorporated into further national surveys with the aim of increasing response rate. Cross-sectional surveys may result in an over-representation of one group and this limitation is acknowledged, although none of the respondents were members of the sports and exercise or musculoskeletal medicine RACGP specific interest networks which may have been expected if this were the case. A further limitation is that the results reflected the knowledge and experience of each respondent on the day they completed the survey and the respondent may respond differently today. It is perceived that the immediate priority for future attention is the development of further guidance for GPs in relation to diagnostic and management decisions in patients presenting with concussion from all causes, not just sport. Collection of data from GP Registrars will provide further information relating to current registrars’ knowledge in this area, and identification of concussion knowledge in newly qualified doctors will provide information to how concussion teaching and knowledge varies amongst medical schools. Data from rural and remote general practices, and Aboriginal Medical Services, are required in future studies to determine the differences between these practices and services, if any. In addition, clarification needs to be sought as to why imaging is requested by some GPs but not others and qualitative semi-structured interviews of GPs may be of value.
Knowledge surrounding concussion guidelines, diagnosis and management varied across GPs in WA. Increasing the visibility of already available concussion guidelines to assist those GPs who are not confident making a diagnosis of concussion may result in increasing recognition of concussions presenting to general practice. The discrepancy between confidence in making a diagnosis of concussion and confidence in managing concussion indicates that GPs may need additional educational materials to assist in managing concussions. These materials may assist GPs working in metropolitan and regional areas with management of patients presenting early with concussion, and those with ongoing symptoms.
Additional file 1. Additional file 2: Supplementary Text 2 . a. Symptoms offered to respondents to identify their knowledge of symptoms of concussion. b. Signs offered to respondents to identify their knowledge of signs of concussion.
|
Medical Error: Using Storytelling and Reflection to Impact Resident Error Response Factors | 57c52ee1-905d-4dc8-bf77-27b86de878de | 11466310 | Family Medicine[mh] | By the end of this activity, learners will be able to: 1. Define medical error and integrate this topic into their understanding of the profession of medicine. 2. Identify ways that physicians cope and thrive after medical error. 3. Describe safety culture and identify ways that colleagues can help with error management and recovery. 4. Describe local policies and practices related to medical error. I am a healer, yet sometimes I do more harm than good. D. Hilfiker , “Facing Our Mistakes” The Kohn et al. landmark study in 2000 reported preventable medical errors in hospitals resulted in approximately 98,000 deaths across 33.6 million hospital admissions. More recent studies estimate that 440,000 people die in the United States each year due to preventable medical error. Negative outcomes related to error are evident: rising health care costs due to adverse events and the subsequent need for repeat tests, readmissions, increased length of stay, rising insurance premiums, and the American public's lack of trust in health care professionals and institutions. Other indirect costs that result from poor safety outcomes in the US include loss of income to patients, and sometimes-preventable intermediate or long-range disabilities or chronic health care challenges. Even though medical error is the third leading cause of death in the US, costing between $73.5 and $98 billion in quality adjusted life years, , and error experiences are common among residents, residents receive little formal training in error management and recovery. Maladaptive coping strategies appear frequently among learners , and practicing physicians, with impact on physician quality of life and patient care. Researchers have identified common physician trajectories after error, including the frequently reported phenomenon of second victimhood—fear of litigation, shame, self-blame, and guilt arising from acute awareness of human fallibility and its impact on patients. These consequences appear cumulative and build across one's tenure. Further, shame and embarrassment create barriers to disclosure, reducing opportunity for analysis and process improvement. Second victimhood recovery follows six stages: chaos and accident response, intrusive reflections, restoring personal integrity, enduring the inquisition, obtaining emotional first aid, and moving on (which can involve surviving in, thriving in, or dropping out of medicine). In contrast to maladaptive responses, speaking up about medical error is an important act that impacts patient safety, quality of patient care, and long-term error reduction as transparency improves. The act of disclosure can be healing for the physician as it provides the opportunity to connect with the patient in a meaningful way where care and compassion is expressed, and patient-centered care is the goal. Researchers have identified several factors associated with desired outcomes: error reporting, – disclosure, coping, , constructive change, and growth after error. More specifically, coping may be aided by professional counseling, discussing the error with trusted peers, and engaging in quality projects linked to the error. Coping is essential to safeguard the emotional well-being of physicians and to prevent burnout. Error management support and guidance from more experienced physicians impacts resident emotions and behaviors. Presumably, supporting and promoting resident exposure to medical error disclosure through engagement with more experienced physicians helps them to emulate constructive responses and to identify maladaptive ones. Previous curricular evaluations have demonstrated the value of simulated clinical scenarios to practice communication-based skills in a safe setting using a team approach. – Our curriculum contributes to the existing literature by exploring the dimension of storytelling. Storytelling creates the framework to engage residents in the discussion and practice of disclosure, and to integrate a challenging concept—personal fallibility—into their professional identity. To summarize, development of an approach to medical error is critical for personal and professional resilience and meaningful participation in quality improvement. , Curricula addressing error management and recovery are desired by learners , and can improve resident knowledge, skills, and abilities in this area. , , Mentor storytelling can be particularly effective as well as integrating error management and recovery into everyday activities. We organized key factors related to effective error management and physician growth after error using a logic model for developing health interventions called the predisposing, reinforcing, and enabling constructs in educational diagnosis and evaluation - policy, regulatory, and organizational constructs in educational and environmental development (PRECEDE-PROCEED) model. Predisposing factors included resident knowledge (awareness of mentor error, local policies and procedures, effective error disclosure steps, and related professional values) and beliefs (error recovery self-efficacy and attitude towards error). We also targeted enabling factors, specifically resident skills (error and cause identification, error disclosure, emotion management, and accessing support), as well as a reinforcing factor (talking among colleagues). The facilitator's guide includes additional details on the medical error response factors. The primary goal of our curriculum was to help residents appreciate the pervasiveness of medical error and practice productive error responses. Preexisting error curricula in our family medicine residency program included: (1) the Institute for Healthcare Improvement's patient safety module completed during resident orientation , ; (2) a longitudinal wellness program, Tending the Flame, with a session on medical error and a personal wellness plan (inclusive of coping strategies) development assignment ; and (3) a didactic session specifically on disclosure training (given once every 3 years). In addition, local rotation sites had policies related to medical error and disclosure although residents were not particularly aware of these before our curricular intervention. We used the AAMC Quality Improvement and Patient Safety Competencies and ACGME family medicine milestones to inform the development of the educational objectives and pedagogical techniques (i.e., facilitating reflection and role-modeling for affective or attitudinal targets like self-efficacy). Specific objectives are included in our facilitator's guide , which also describes the implementation for the three 60-minute resident sessions. Slides for each session are included in – , and the faculty survey and pre-/postmodule surveys are also included. Logistics We recruited family medicine residency faculty ( N = 7) in a medium-sized, urban, midwestern program with an e-mail and survey sent 1 week before session one. During weekly didactics we recruited PGY 1, 2, and 3 family medicine residents ( N = 30). We scheduled the three sessions to occur over a 5 months period between August 2022 and January 2023, in order to allow participants to have time to process the session content in between sessions. Session One We used mentor storytelling to demonstrate physician error, emotions/coping strategies, and the ways in which the physician's environment can impact error response (e.g., workload and error culture). Prior to the session, we identified a faculty volunteer to share a personal story of an error and the aftermath with our program recruitment and faculty survey . Once the faculty member was selected, we gave the following guidance to potential storytellers: (1) we believe the most impactful stories will come from program mentors and leadership; (2) the session is confidential and for learning; (3) the purpose of the story sharing is to reflect on ways that physicians react to medical error, amplifying strategies that result in good care for the patient and physician(s) involved; (4) stories do not have to be long, 5–10 minutes works well. The session began with the faculty story and then the faculty facilitator guided residents in reflection using several prompts . Next, the session facilitator presented a small amount of lecture material, facilitated small- and large-group discussions, and prompted timed self-reflection through writing. Session Two During the next session , we stimulated interest by discussing key professional values related to error and then had a large-group discussion of learner fears and barriers related to error. We introduced the concept of safety culture , and residents had the opportunity to reflect on strengths and weaknesses of the current culture of their practice. The faculty presenter engaged learners throughout the session using small- and large-group discussion, polling, and timed self-reflection through writing. Session Three During the third session , the faculty presenter led a review of local policies and procedures related to medical error, and residents practiced self-awareness, error disclosure, root cause reflection, and coping. The faculty member facilitated the review of sample error cases and also instructed students to write a letter to a colleague or future self in the wake of perceived error. Evaluation We emailed residents the premodule survey immediately prior to the first session. Residents were given time to complete the premodule survey immediately prior to the start of the session, but survey links remained active through the start of session three to allow residents who missed session one or two the opportunity to complete the survey. The postmodule survey was sent via email to residents at the end of session three, and residents were given time to complete this survey just after session three concluded. Anonymous pre- and postmodule survey responses were analyzed and compared. To measure knowledge and confidence we collated participant affirmative responses ( agree / strongly agree ) on a 5-point Likert scale (1 = strongly disagree , 5 = strongly agree ). We measured satisfaction by collating a percentage of residents selecting each rating of the emotional difficulty (0 = not emotionally difficult at all , 10 = extremely emotionally difficult ) and helpfulness (0 = not helpful at all, 10 = extremely helpful ) of the curriculum. Due to small sample size, analyses by PGY were not undertaken. The evaluation procedures were approved by the Wright State University Boonshoft School of Medicine Institutional Review Board (FWA# 00002427, approved August 26, 2022). We recruited family medicine residency faculty ( N = 7) in a medium-sized, urban, midwestern program with an e-mail and survey sent 1 week before session one. During weekly didactics we recruited PGY 1, 2, and 3 family medicine residents ( N = 30). We scheduled the three sessions to occur over a 5 months period between August 2022 and January 2023, in order to allow participants to have time to process the session content in between sessions. We used mentor storytelling to demonstrate physician error, emotions/coping strategies, and the ways in which the physician's environment can impact error response (e.g., workload and error culture). Prior to the session, we identified a faculty volunteer to share a personal story of an error and the aftermath with our program recruitment and faculty survey . Once the faculty member was selected, we gave the following guidance to potential storytellers: (1) we believe the most impactful stories will come from program mentors and leadership; (2) the session is confidential and for learning; (3) the purpose of the story sharing is to reflect on ways that physicians react to medical error, amplifying strategies that result in good care for the patient and physician(s) involved; (4) stories do not have to be long, 5–10 minutes works well. The session began with the faculty story and then the faculty facilitator guided residents in reflection using several prompts . Next, the session facilitator presented a small amount of lecture material, facilitated small- and large-group discussions, and prompted timed self-reflection through writing. During the next session , we stimulated interest by discussing key professional values related to error and then had a large-group discussion of learner fears and barriers related to error. We introduced the concept of safety culture , and residents had the opportunity to reflect on strengths and weaknesses of the current culture of their practice. The faculty presenter engaged learners throughout the session using small- and large-group discussion, polling, and timed self-reflection through writing. During the third session , the faculty presenter led a review of local policies and procedures related to medical error, and residents practiced self-awareness, error disclosure, root cause reflection, and coping. The faculty member facilitated the review of sample error cases and also instructed students to write a letter to a colleague or future self in the wake of perceived error. We emailed residents the premodule survey immediately prior to the first session. Residents were given time to complete the premodule survey immediately prior to the start of the session, but survey links remained active through the start of session three to allow residents who missed session one or two the opportunity to complete the survey. The postmodule survey was sent via email to residents at the end of session three, and residents were given time to complete this survey just after session three concluded. Anonymous pre- and postmodule survey responses were analyzed and compared. To measure knowledge and confidence we collated participant affirmative responses ( agree / strongly agree ) on a 5-point Likert scale (1 = strongly disagree , 5 = strongly agree ). We measured satisfaction by collating a percentage of residents selecting each rating of the emotional difficulty (0 = not emotionally difficult at all , 10 = extremely emotionally difficult ) and helpfulness (0 = not helpful at all, 10 = extremely helpful ) of the curriculum. Due to small sample size, analyses by PGY were not undertaken. The evaluation procedures were approved by the Wright State University Boonshoft School of Medicine Institutional Review Board (FWA# 00002427, approved August 26, 2022). Of the cohort of 30 PGY 1, 2, and 3 participating family medicine residents, 22 (73%) completed the premodule survey, and 15 (50%) completed the postmodule survey. Seven pre- and postmodule surveys were able to be matched using anonymous codes. Of those who completed the postmodule survey, nine completed all three sessions, five completed two sessions, and one completed one session. Additionally, seven out of seven faculty completed the faculty survey. State of Preexisting Environment Premodule responses demonstrated the pervasiveness of error experience among residents. Most residents reported having experienced error (55%, n = 12) and having had a mentor or peer share an error story with them (73%, n = 16). Among those with an error experience, 10 (77%) reported the error was disclosed to the patient. Many residents with error experience reported that their team or organization learned from the error (75%, n = 9), acknowledged the error (92%, n = 11), and debriefed as a team (83%, n = 10). Knowledge and Confidence Our curriculum was associated with an increase in residents who reported several target factors: disclosure confidence/self-efficacy, knowledge of local procedures, accessing support as an error response, faculty and peer story sharing, and acknowledgement that mentors have made errors. Specifically, disclosure self-efficacy ( I can be honest about the errors I make as a doctor. ) increased after the curriculum from 86% ( n = 19) to 93% ( n = 14; ). As expected, faculty reported higher levels of confidence compared to residents with error disclosure and personal relationship recovery . Knowledge of related local procedures ( I know what to do at my institution when faced with a medical error. ) increased from 46% ( n = 10) to 93% ( n = 14; ). The curriculum was also associated with an increase in reaching out to others as an error response from 36% ( n = 8) to 87% ( n = 13). Debriefing with the team remained common, but the rate of residents reporting feel bad about myself as an error response increased from 41% ( n = 9) to 60% ( n = 9). After the curriculum, rates of reported faculty and peer story sharing increased, and resident reported awareness of mentor error increased from 68% ( n = 15) to 87% ( n = 13). Incidentally, all faculty respondents reported I have made errors in my care for patients . Responses also demonstrated an increase in resident reported self-awareness ( I acknowledge when I am at increased risk for making errors ) from 77% ( n = 17) to 93% ( n = 14). Satisfaction Overall residents reported the training was helpful (ranking > 5), and six residents (40%) reported an emotionally difficult rating of 5 or greater for the curriculum . Prior to the module, residents were most interested in further training through personal stories of mentor error (73%, n = 16), and after the curriculum, residents reported most interest in additional training in legal and malpractice risk (73%, n = 11). Premodule responses demonstrated the pervasiveness of error experience among residents. Most residents reported having experienced error (55%, n = 12) and having had a mentor or peer share an error story with them (73%, n = 16). Among those with an error experience, 10 (77%) reported the error was disclosed to the patient. Many residents with error experience reported that their team or organization learned from the error (75%, n = 9), acknowledged the error (92%, n = 11), and debriefed as a team (83%, n = 10). Our curriculum was associated with an increase in residents who reported several target factors: disclosure confidence/self-efficacy, knowledge of local procedures, accessing support as an error response, faculty and peer story sharing, and acknowledgement that mentors have made errors. Specifically, disclosure self-efficacy ( I can be honest about the errors I make as a doctor. ) increased after the curriculum from 86% ( n = 19) to 93% ( n = 14; ). As expected, faculty reported higher levels of confidence compared to residents with error disclosure and personal relationship recovery . Knowledge of related local procedures ( I know what to do at my institution when faced with a medical error. ) increased from 46% ( n = 10) to 93% ( n = 14; ). The curriculum was also associated with an increase in reaching out to others as an error response from 36% ( n = 8) to 87% ( n = 13). Debriefing with the team remained common, but the rate of residents reporting feel bad about myself as an error response increased from 41% ( n = 9) to 60% ( n = 9). After the curriculum, rates of reported faculty and peer story sharing increased, and resident reported awareness of mentor error increased from 68% ( n = 15) to 87% ( n = 13). Incidentally, all faculty respondents reported I have made errors in my care for patients . Responses also demonstrated an increase in resident reported self-awareness ( I acknowledge when I am at increased risk for making errors ) from 77% ( n = 17) to 93% ( n = 14). Overall residents reported the training was helpful (ranking > 5), and six residents (40%) reported an emotionally difficult rating of 5 or greater for the curriculum . Prior to the module, residents were most interested in further training through personal stories of mentor error (73%, n = 16), and after the curriculum, residents reported most interest in additional training in legal and malpractice risk (73%, n = 11). To address the lack of a graduate medical education curriculum related to medical error response, we developed three sessions for family medicine residents. We found the use of a model for factor organization and assessment of the curriculum as a necessary initial step. From this evaluation we confirmed that most residents already had prior experience with team or personal error, our curricular intervention positively impacted specific targets, and residents found the curriculum helpful. However, sample size and the format and timing of our postmodule survey limited our ability to fully assess our curriculum. We plan to refine the curriculum based on initial results by adding several new evaluation techniques and offering the module more broadly to other graduate medical education specialty programs in our community. Because of the complexity of physician error response, we found the process of organizing predisposing, enabling, and reinforcing factors from the literature helpful for identifying potential curricular targets. Similarly, our exploration of the preexisting curriculum allowed us to select targets not already being fully addressed. Our residency, like many others, has a formal and informal curriculum (i.e., unspoken norms and lessons that residents learn) for medical error, and identifying this curriculum helped us partner with current faculty champions to develop a synergistic curriculum. Our curriculum evaluation suggests that most residents had prior experience with team or personal error, confirming the importance of the topic, and suggests that a brief curriculum can be effective at impacting important error response factors. Residents reported significant interest in error training, and many found our curriculum helpful. The interest in further training shifting from personal stories of mentor error (premodule) to legal and malpractice concerns (postmodule) could indicate a deficiency in the curriculum or a natural change in focus. From an experiential standpoint, the storytelling by faculty prompted residents to use empathic language to explore what they would have felt and done in the storyteller's shoes and voice the respect they had for the physician doing the right thing. After initial rumination regarding the error itself, residents shared surprising insights into error management and recovery, including a reframing of the error and a thankfulness for physician colleagues that can be called upon to help deal with complications of error. Based on our experience in this curriculum, residents can be remarkably open, especially when openness is modeled by mentors. Surprisingly, the rate of residents who report feeling bad about oneself after error increased from 41% ( n = 9) to 60% ( n = 9) after the curriculum. Perhaps the act of addressing this subject as a group was emotionally charging. The fact that 40% of residents reported an emotionally difficult rating of 5 or greater highlights the importance of including resident support resources during the sessions themselves. We are unsure if this finding will persist in larger sample sizes. Time was a barrier to accomplishing delivery of all components of the curriculum, and we hope to refine or remove less important components. For example, all residents (pre- and postmodule) and faculty reported that good doctors should be honest about errors they make, suggesting this belief may be a less important target for intervention. The time spent on discussion of values and underlying medical ethics could be used to give more time for other components. The literature has shown a disconnect between belief in the appropriateness of disclosure, intent to disclose, and actual disclosure rates. Therefore, resident and patient-oriented outcomes are an important next step in curriculum assessment (e.g., resident milestones data, error reporting rates at primary rotation sites, and institutional patient safety survey information). Limitations of our evaluation stemmed from an ambitious desire to develop a comprehensive curriculum, limited sample size and time, lack of direct skills and knowledge assessments, and missing team and patient-oriented outcomes. As such, our curriculum and evaluation do not allow for specific module content to be associated with changes in resident responses. Similarly, the complexity of error response and the scope and time for our curriculum evaluation limited our ability to associate our curriculum with specific changes in rates of error acknowledgement, disclosure, coping, and growth, or patient-oriented outcomes like satisfaction with error disclosure or care relationship after error. Our small sample size limits generalizability, and we hope to offer a refined curriculum to our entire graduate medical education community, which will help with sample size limitations. Our timeframe allowed only for short-term reassessment of survey responses, which could differ from long-term impacts. Future studies could use resident milestone data to assess measurable outcomes, and a 6-month postsurvey could be undertaken to see if changes in knowledge, skills, and beliefs persist. Organizational data like error reporting rates and patient safety survey data could be tracked by year to evaluate potential impacts organization-wide. The project also uncovered a need for faculty development in medical error response, and sessions may be adapted for this purpose. Our brief curriculum was associated with an increase in related resident reported knowledge, confidence, and story sharing. The medical community will benefit from further refining the model for error management and growth behaviors among residents. It will also be important to move from short-term, self-reported resident knowledge and beliefs to long-term resident and patient-oriented outcomes like error reporting rate, disclosure skills assessment by faculty and patient, and overall error rates. We intend to improve our postmodule assessment by adding direct skills assessment for error disclosure and incorporating resident and patient-oriented outcomes like reporting rates and patient safety survey data. Increasing our sample size by offering the curriculum to our local graduate medical education community including numerous specialty programs will allow us to better compare pre- and postmodule results. Ultimately, development of interdisciplinary and interprofessional error response training will best prepare learners to manage their future errors and their personal recovery. Facilitators Guide.docx Error Session 1.pptx Error Session 1 Handout.pdf Error Session 2.pptx Error Session 3.pptx Error Session 3 Handout - Error Cases.docx Faculty Survey.docx Premodule Resident Survey.docx Postmodule Resident Survey.docx All appendices are peer reviewed as integral parts of the Original Publication. |
Return of Clinically Actionable Pharmacogenetic Results From Molecular Tumor Board | 7efb2aba-ea63-4cbd-971a-71f7d29faa87 | 11924161 | Pharmacology[mh] | As part of the routine care for the patients seen in the MTB for our Health Precision Genomics Program, tissue and blood for sequencing are obtained at initial patient visits and sent to a commercial laboratory for sequencing. Binary alignment map (BAM) files are returned to the Precision Health Cloud database (LifeOmic, Inc., Indianapolis, IN). Patient consents for sequencing and treatment indicate that the results may provide information about other health problems; this includes PGx, such as drug metabolism. The PGx analysis was run once per week for the patients being seen at that week's MTB. The workflow begins by identifying new patients each week who have paired germline and somatic whole exome sequencing. The PGx variants from those patients' germline BAM files are then extracted using Aldy , , run in a Jupyter Notebook in the LifeOmic Precision Health Cloud environment. Aldy returns the star allele haplotypes for CYP2B6, CYP2C8, CYP2C9, CYP2C19, CYP2D6, CYP3A4, CYP3A5, CYP4F2, DPYD, G6PD, NUDT15, SLCO1B1, TPMT , and UGT1A1 . PGx experts evaluate the results for clinically actionable variants. Any potentially actionable results are reported at the MTB discussion. Since the bioinformatics analysis of the PGx results is not part of a CLIA (Clinical Laboratory Improvement Amendments) approved process, any variants that may impact clinical care are recommended to be confirmed by additional testing in a CLIA‐certified laboratory. The volume of cases analyzed by MTB was collected from our Precision Genomics Program from January 1, 2023, to December 31, 2023. Salaries for pharmacists in the United States were obtained from the Occupational Employment and Wage Statistics, 2023, by the US Bureau of Labor Statistics. The times for the PGx processing were estimated by an external observer who timed the process and also had follow‐up discussions with the PGx team.
An overview of the main steps to extract the PGx information from the sequencing data is illustrated in Figure . The entire hands‐on processing time was approximately 3 hours and 15 minutes. ( ) Patients with sequencing data were identified from the standard clinical workflow from our Precision Genomics Program clinic. Names and medical record numbers were used to identify patient BAM sequencing files in the Precision Health Cloud, and those BAM files were downloaded into the Jupyter Notebook computing environment. Initial time to get the patient information and locate the sequencing data was approximately 1 hour; this step included identifying patient BAM files and initializing the Jupyter environment to download BAMs. Downloading BAM files took approximately 2 hours, depending on the number of patients; however, this did not require human labor. After the download was complete, the running time for Aldy to extract the PGx diplotypes was approximately 45 minutes. Although the number of patients varied by week, that variability had little impact on the hands‐on time due to the automation. Assessment of the actionability of the PGx variants for all the patients took approximately 30 minutes. Participation in the weekly MTB, where the results were presented, took approximately 1 hour. Therefore, the overall time required for all preparation steps to extract PGx information for the weekly MTBs and report the clinically actionable results was approximately 3 hours and 15 minutes. On average in 2023, there were approximately 20.6 patients per week analyzed by the tumor board. To estimate the costs related to these practices above, we assumed that all tasks were performed by a clinical pharmacist. The median annual base salary of a pharmacist in the United States in 2023 was $134,665. We added an additional 40% to account for benefits. In addition, the incremental computing cost (over the required infrastructure for receiving, storing, and accessing the BAM files) for extracting PGx diplotypes within the LifeOmic Precision Health Cloud environment was approximately $20 per week. It was estimated based on the computing cost of the cloud computing services for using Jupyter Notebooks in the LifeOmic Precision Health Cloud environment. Collectively, the final estimated cost for extracting PGx information and reporting the clinically actionable results to the tumor board in the United States was $15.27 on average per patient.
In this report, we describe our process for extracting PGx information from germline whole exome sequencing that is being done as part of our Precision Genomics Program. Since we don't control the entire informatics pipeline at the commercial sequencing laboratory, it should be emphasized that this does not replace CLIA‐approved PGx reporting. Rather our program explores the value of testing patients with complicated medical profiles, who could benefit from more expensive PGx testing with its potential to reduce drug complications. With hands‐on time of only 3–4 hours per week (~20 patients per week), our estimates of cost to screen for those patients likely to have actionable variants were approximately $15 per patient. The pharmacist's time required could vary depending on the pharmacist's expertise in pharmacogenetics. These calculations are based on adding the PGx analysis superimposed on an existing informatics platform that stores and provides access to the BAM files and has to capability to run Aldy within the environment; it also requires that the germline sequencing BAM files from the sequencing lab are already received into the platform. Since we have shown that extracting these results from clinical sequencing is accurate, , this process provides a strategy to focus the more expensive CLIA‐based genotyping on those that are likely to carry clinically actionable variants for their current medications. This practice is likely to enhance the cost‐effectiveness of PGx testing, which is an established barrier to PGx adoption. , , According to our cost estimation, conducting confirmative PGx testing for patients having clinically actionable variants would be more cost‐effective than testing every patient ( Sections and ). Our $15 per patient calculations do not include the costs of the confirmatory testing of those actionable results, which would increase the cost primarily for those who have actionable variants and would benefit from the testing; however, the variable confirmatory testing is included in the supplementary materials. According to previous research, patients' willingness to pay for PGx testing is typically not more than $100. This is lower than most clinical laboratories charge for the testing; however, the amount they are willing to pay may be different if they knew they were highly likely to carry variants that would impact their care. Further research would be needed to understand the patients' willingness to pay if they were given this additional information. A previous investigation reported using research‐grade genotyping array data from a biobank to trigger alerts for DPYD testing when patients were to receive a fluoropyrimidine therapy. Their oncologists avoided a normal dose of a fluoropyrimidine therapy in a patient who carried a DPYD variant that would likely have caused severe toxicity. Our genotype extraction procedure builds upon this practice to allow return of extracted PGx results for 14 PGx genes with clinically actionable recommendations in clinical practice guidelines or drug labels. In addition, incidental cancer risk genetic variants are occasionally identified in the sequencing of MTBs. , Similar to our process, those are also confirmed by genetic testing in clinical laboratories. Thus, with the appropriate patient consent, this efficient and cost‐effective method to identify patients likely to benefit from confirmatory clinical genotyping may mitigate hesitation to implement genetic testing in at least some patients. It is important to recognize that this should not replace clinical genetic testing; PGx genotypes extracted by Aldy, in the absence of a CLIA‐approved workflow, should be considered preliminary, since false negatives would put patients at risk of extreme toxicities. So, this should be implemented in clinics where routine PGx testing is not yet implemented. This research has several limitations. First, to run this analysis, a substantial amount of infrastructure must already be in place to receive, store, and have the computing environment to run the Aldy analysis. It also requires the agreements with the laboratory to send the BAM files. The cost estimate does not include the cost of setting up that infrastructure, the agreements, or transferring the sequencing files. Second, many clinics sequence DNA only from the tumors but not from matched normal tissues or cells. Since previous results from tumor sequencing have shown that pharmacogenes occasionally undergo genetic changes in the tumors, using sequencing data from tumors would need to be interpreted with caution. Third, if the weekly patient volume increases significantly, there would be a proportional increase in the hands‐on time required for the PGx evaluation, the computation time, and the MTB discussion. Last, we did not account for the cost of educating the pharmacists and pharmacist salaries are likely to vary depending on geographic location, hospital size and type, and other factors not considered in this report. The salaries used included all types of pharmacist positions in the United States; however, the impact of the variable salaries across disciplines on the local cost is likely to have little impact since the final cost for extracting PGx information is low. Since we provided the details of each of the inputs used in the calculation, others could use their local salary and benefit values to calculate their local costs. Future studies are needed to optimize the inclusion of this PGx data extraction into the CLIA‐approved laboratory process and enhance the accuracy of the PGx data extracted from the tumor DNA sequencing.
This work was supported by funding from the Indiana University Precision Health Initiative Grand Challenge, Indianapolis, Indiana, USA; the Vera Bradley Foundation for Breast Cancer Research, Indianapolis, Indiana, USA; K23GM147805, National Institutes of Health, Bethesda, Maryland, USA; the Zerbe Chair in Pharmacoeconomics, Indianapolis, Indiana, USA; the NIH—National Institute of General Medical Sciences—T32GM144891, Bethesda, Maryland, USA; the NIH—Interdisciplinary Training in Cancer Prevention and Control—T32CA117865, Bethesda, Maryland, USA; and Indiana University Simon Comprehensive Cancer Center, Indianapolis, Indiana, USA.
Steven M. Bray is employed at LifeOmic, Inc. All other authors declared no competing interests for this work.
H.K. and T.C.S. wrote the manuscript. H.K., T.B.S., J.T.C., W.O., S.M.B., E.M.T., M.T.T., C.A.F., B.P.S., T.S., and T.C.S. designed and performed the research, and analyzed the data.
Data S1.
|
A Cross-Sectional Exploratory Study of Rat Sarcoid (Ras) Activation in Women with and Without Polycystic Ovary Syndrome | db1d4bce-212f-4afd-a4ef-ffd703f0cedd | 11898917 | Biochemistry[mh] | Menstrual irregularity, anovulatory infertility and hirsutism are classic features of polycystic ovary syndrome (PCOS) which is a multifactorial endocrine disorder affecting approximately 5–20% of women of reproductive age worldwide . It is recognized that there is an increase in the prevalence of metabolic features including type 2 diabetes (T2D), hypertension, and cardiovascular disease in PCOS (1). However, the inherent mechanism is still unclear, insulin resistance (IR) and obesity-related inflammation associated with PCOS have been implicated . Rat sarcoma (Ras) proteins, including Kirsten rat sarcoma viral oncogene homolog (KRAS), Harvey Rat sarcoma virus (HRAS), and neuroblastoma Ras viral oncogene homolog (NRAS), are a family of small GTPases that are key regulators of a wide array of cellular processes that involve growth, differentiation, and apoptosis through their actions in signal transduction pathways . PCOS is often associated with IR, and Ras proteins are involved in the insulin signaling pathway, particularly through the Ras-MAPK (Mitogen-Activated Protein Kinase) pathway , which regulates cellular responses such as growth and differentiation . Ras proteins play a role in follicular development and steroidogenesis by mediating signals from gonadotropins such as luteinizing hormone (LH) . Ras-MAPK signaling is involved in the pathways that regulate androgen production in the ovaries and thus may contribute to hyperandrogenism, one of the diagnostic features of PCOS . PCOS is associated with a pro-inflammatory state and Ras proteins are involved in regulating inflammatory responses, with activation of Ras contributing to the chronic low-grade inflammation seen in PCOS patients; KRAS is a potential pharmacological target to treat PCOS . Circulating Ras proteins, whether as fragments, exosome-associated proteins or Ras-related signaling components, are found in cancer and metabolic disorders ; however, circulatory Ras protein levels have not previously been reported in PCOS. Circulating growth factors that act through the Ras signaling system do, however, differ in PCOS, examples of which include Epidermal growth factor (EGF), which binds to the EGF Receptor (EGFR, also known as ErbB1) . Fibroblast growth factors (FGFs), which bind to FGF Receptors (FGFRs), affect ovarian function in PCOS . Platelet-derived growth factor (PDGF), binding to the PDGF Receptor (PDGFR), may modulate steroid production involved in fertility in PCOS. Vascular endothelial growth factor (VEGF), binding to the VEGF Receptor (VEGFR), plays a critical role in angiogenesis and ovarian folliculogenesis, with the suggestion that it is associated with increased PCOS risk . Insulin-like growth Factors (IGFs) activate Ras and play an important role in the pathogenesis of PCOS related to insulin resistance and inflammatory responses . Other growth factors acting through Ras include the Hepatocyte growth factor (HGF), binding to the HGF receptor (HGFR), and Nerve growth factor (NGF), binding to the nerve growth factor receptor (NGFR). The growth factors and their activation of Ras are shown in . This study explores Ras signaling in PCOS, with a particular focus on the potential importance of circulatory Ras proteins and growth factors that activate Ras intracellular signaling pathways in women with and without PCOS from a UK Biobank.
2.1. Study Design This study was designed as a cross-sectional analysis, incorporating a total of 234 Caucasian women aged between 18 and 36 years. Among them, 147 were diagnosed with polycystic ovary syndrome (PCOS), while 97 served as healthy controls; all subjects had been recruited to a PCOS biobank (ISRCTN70196169: 2012–2017) based in the Department of Endocrinology, Hull Royal Infirmary, UK, with approval from the Newcastle and North Tyneside Ethics Committee . Each participant provided written informed consent. PCOS diagnosis was established using the Rotterdam criteria , requiring the presence of at least two of the following: oligo/anovulation, hyperandrogenism (defined by a Ferriman-Gallwey score > 8, a free androgen index exceeding 4, or total testosterone levels above 1.5 nmol/L, based on local laboratory references), or polycystic ovarian morphology confirmed via transvaginal ultrasound. Secondary conditions such as nonclassical 21-hydroxylase deficiency were excluded through appropriate screening, as previously outlined . The demographic details of both PCOS and control groups are provided in . Control participants exhibited regular menstrual cycles, unremarkable physical examinations, and no evidence of polycystic ovaries upon ultrasound assessment. They were also medication-free. Fasting blood samples were collected at the time of informed consent, centrifuged at 3500× g for 15 min, and stored at −80 °C in aliquots. Various biomarkers were analyzed, including sex hormone-binding globulin (SHBG), insulin (measured using a DPC Immulite 200 analyzer, Euro/DPC, Llanberis UK), lipid measurements, and plasma glucose, which were utilized to compute the homeostasis model assessment of insulin resistance (HOMA-IR) (Synchron LX20 analyzer, Beckman-Coulter, High Wycombe, UK). The free androgen index (FAI) was derived by dividing total testosterone by SHBG and multiplying by 100. Serum testosterone levels were determined using isotope-dilution liquid chromatography tandem mass spectrometry (LC-MS/MS) Anti-Müllerian hormone was measured using a Beckman Coulter Access automated immunoassay. Plasma protein levels were assessed through the Slow Off-rate Modified Aptamer (SOMA)-scan platform , which facilitated the quantification of multiple proteins, including GTPase Kirsten rat sarcoma virus (KRAS), Ras GTPase-activating protein 1 (RASA1), Heparin-binding epidermal growth factor-like growth factor (HB-EGF), Epidermal growth factor receptor variant III (EGFRvIII), Epidermal growth factor (EGF), Epidermal growth factor receptor (EGFR), Fibroblast growth factor 8 isoform A (FGF-8A), Fibroblast growth factor 8 isoform B (FGF-8B), Fibroblast growth factor receptors 1–4 (FGFR1, FGFR2, FGFR3, FGFR4), Fibroblast growth factor 2 (FGF2), Fibroblast growth factors 4–7 (FGF4, FGF5, FGF6, FGF7), Fibroblast growth factors 9–10 (FGF9, FGF10), Fibroblast growth factor 12 (FGF12), Fibroblast growth factors 16–20 (FGF16, FGF17, FGF18, FGF19, FGF20), Fibroblast growth factor 23 (FGF23), Platelet-derived growth factor C (PDGFC), Platelet-derived growth factor subunit A (PDGFA), Platelet-derived growth factor subunit B (PDGFB), Platelet-derived growth factor receptor beta (PDGFRB), Platelet-derived growth factor receptor alpha (PDGFRA), Vascular endothelial growth factor A (VEGFA), Vascular endothelial growth factor A isoform 121 (VEGFA-121), Vascular endothelial growth factor C (VEGF-C), Vascular endothelial growth factor D (VEGF-D), Vascular endothelial growth factor receptor 2 (VEGF-sR2), Vascular endothelial growth factor receptor 3 (VEGF-sR3), Insulin (INS), Insulin-like growth factor I (IGF1), Macrophage colony-stimulating factor 1 (CSF1), Macrophage colony-stimulating factor 1 receptor (CSF1R), Granulocyte-macrophage colony-stimulating factor 2 (CSF2), Granulocyte colony-stimulating factor (CSF3), Granulocyte colony-stimulating factor receptor (CSF3R), Insulin-like growth factor 1 receptor (IGF1R), Cation-independent mannose-6-phosphate receptor (IGF2R), Hepatocyte growth factor (HGF), and beta-nerve growth factor (NGF). Calibration was based on standards as previously described . Protein quantification was performed using an aptamer-based approach known as the SOMAmer protein array . Plasma samples collected in EDTA tubes underwent a structured protocol: (1) SOMAmers bound to analytes using a photocleavable linker; (2) The complexes were immobilized on a streptavidin-coated surface; (3) Ultraviolet (UV) exposure released analyte-SOMAmer complexes into solution; (4) These complexes were immobilized once more through biotin-streptavidin interactions; (5) Eluted SOMAmers served as surrogates for analyte quantification; (6) Hybridization to complementary oligonucleotides enabled final quantification. Normalization and standardization of raw intensities, hybridization signals, medians, and calibration data followed established methodologies . 2.2. Statistics Continuous variables were presented as means ± standard deviations (SD). To compare circulating levels of KRAS, Ras GTPase-activating protein-1 (RASA1), and 45 growth factor-related proteins between the PCOS and control groups, independent two-sample t -tests were employed. A significance threshold of p < 0.05 was applied. The quantile normalized SOMAscan proteomic data was log-transformed for further statistical assessments. Linear models for microarray (limma) analysis in conjunction with t -tests were used to compare and identify the dysregulated proteins in PCOS versus controls. Any protein changes with a fold change of 1 and raw p -value < 0.05 were considered significant. Spearman’s rank correlations were computed to examine associations between growth factor-related proteins and body mass index (BMI). Statistical analyses were conducted using RStudio (version 2023.03.0), with all tests performed as two-tailed analyses, considering p -values below 0.05 as statistically significant.
This study was designed as a cross-sectional analysis, incorporating a total of 234 Caucasian women aged between 18 and 36 years. Among them, 147 were diagnosed with polycystic ovary syndrome (PCOS), while 97 served as healthy controls; all subjects had been recruited to a PCOS biobank (ISRCTN70196169: 2012–2017) based in the Department of Endocrinology, Hull Royal Infirmary, UK, with approval from the Newcastle and North Tyneside Ethics Committee . Each participant provided written informed consent. PCOS diagnosis was established using the Rotterdam criteria , requiring the presence of at least two of the following: oligo/anovulation, hyperandrogenism (defined by a Ferriman-Gallwey score > 8, a free androgen index exceeding 4, or total testosterone levels above 1.5 nmol/L, based on local laboratory references), or polycystic ovarian morphology confirmed via transvaginal ultrasound. Secondary conditions such as nonclassical 21-hydroxylase deficiency were excluded through appropriate screening, as previously outlined . The demographic details of both PCOS and control groups are provided in . Control participants exhibited regular menstrual cycles, unremarkable physical examinations, and no evidence of polycystic ovaries upon ultrasound assessment. They were also medication-free. Fasting blood samples were collected at the time of informed consent, centrifuged at 3500× g for 15 min, and stored at −80 °C in aliquots. Various biomarkers were analyzed, including sex hormone-binding globulin (SHBG), insulin (measured using a DPC Immulite 200 analyzer, Euro/DPC, Llanberis UK), lipid measurements, and plasma glucose, which were utilized to compute the homeostasis model assessment of insulin resistance (HOMA-IR) (Synchron LX20 analyzer, Beckman-Coulter, High Wycombe, UK). The free androgen index (FAI) was derived by dividing total testosterone by SHBG and multiplying by 100. Serum testosterone levels were determined using isotope-dilution liquid chromatography tandem mass spectrometry (LC-MS/MS) Anti-Müllerian hormone was measured using a Beckman Coulter Access automated immunoassay. Plasma protein levels were assessed through the Slow Off-rate Modified Aptamer (SOMA)-scan platform , which facilitated the quantification of multiple proteins, including GTPase Kirsten rat sarcoma virus (KRAS), Ras GTPase-activating protein 1 (RASA1), Heparin-binding epidermal growth factor-like growth factor (HB-EGF), Epidermal growth factor receptor variant III (EGFRvIII), Epidermal growth factor (EGF), Epidermal growth factor receptor (EGFR), Fibroblast growth factor 8 isoform A (FGF-8A), Fibroblast growth factor 8 isoform B (FGF-8B), Fibroblast growth factor receptors 1–4 (FGFR1, FGFR2, FGFR3, FGFR4), Fibroblast growth factor 2 (FGF2), Fibroblast growth factors 4–7 (FGF4, FGF5, FGF6, FGF7), Fibroblast growth factors 9–10 (FGF9, FGF10), Fibroblast growth factor 12 (FGF12), Fibroblast growth factors 16–20 (FGF16, FGF17, FGF18, FGF19, FGF20), Fibroblast growth factor 23 (FGF23), Platelet-derived growth factor C (PDGFC), Platelet-derived growth factor subunit A (PDGFA), Platelet-derived growth factor subunit B (PDGFB), Platelet-derived growth factor receptor beta (PDGFRB), Platelet-derived growth factor receptor alpha (PDGFRA), Vascular endothelial growth factor A (VEGFA), Vascular endothelial growth factor A isoform 121 (VEGFA-121), Vascular endothelial growth factor C (VEGF-C), Vascular endothelial growth factor D (VEGF-D), Vascular endothelial growth factor receptor 2 (VEGF-sR2), Vascular endothelial growth factor receptor 3 (VEGF-sR3), Insulin (INS), Insulin-like growth factor I (IGF1), Macrophage colony-stimulating factor 1 (CSF1), Macrophage colony-stimulating factor 1 receptor (CSF1R), Granulocyte-macrophage colony-stimulating factor 2 (CSF2), Granulocyte colony-stimulating factor (CSF3), Granulocyte colony-stimulating factor receptor (CSF3R), Insulin-like growth factor 1 receptor (IGF1R), Cation-independent mannose-6-phosphate receptor (IGF2R), Hepatocyte growth factor (HGF), and beta-nerve growth factor (NGF). Calibration was based on standards as previously described . Protein quantification was performed using an aptamer-based approach known as the SOMAmer protein array . Plasma samples collected in EDTA tubes underwent a structured protocol: (1) SOMAmers bound to analytes using a photocleavable linker; (2) The complexes were immobilized on a streptavidin-coated surface; (3) Ultraviolet (UV) exposure released analyte-SOMAmer complexes into solution; (4) These complexes were immobilized once more through biotin-streptavidin interactions; (5) Eluted SOMAmers served as surrogates for analyte quantification; (6) Hybridization to complementary oligonucleotides enabled final quantification. Normalization and standardization of raw intensities, hybridization signals, medians, and calibration data followed established methodologies .
Continuous variables were presented as means ± standard deviations (SD). To compare circulating levels of KRAS, Ras GTPase-activating protein-1 (RASA1), and 45 growth factor-related proteins between the PCOS and control groups, independent two-sample t -tests were employed. A significance threshold of p < 0.05 was applied. The quantile normalized SOMAscan proteomic data was log-transformed for further statistical assessments. Linear models for microarray (limma) analysis in conjunction with t -tests were used to compare and identify the dysregulated proteins in PCOS versus controls. Any protein changes with a fold change of 1 and raw p -value < 0.05 were considered significant. Spearman’s rank correlations were computed to examine associations between growth factor-related proteins and body mass index (BMI). Statistical analyses were conducted using RStudio (version 2023.03.0), with all tests performed as two-tailed analyses, considering p -values below 0.05 as statistically significant.
Baseline data for the 147 PCOS subjects and 97 controls are shown in . The two cohorts were age-matched, but subjects with PCOS had a greater BMI, increased IR, hyperandrogenemia, increased C-reactive protein (CRP, an inflammatory marker), and a raised anti-Mullerian hormone. PCOS subjects had higher systolic and diastolic blood pressures, a higher fasting blood glucose, and a lower high-density lipoprotein . The results of the Somascan analysis of those proteins that differed between PCOS and control women are shown in , and all proteins are shown in . Circulating KRas and RASA1 did not differ between PCOS and control women ( p > 0.05). EGF1, EGFR, and EGFRvIII were decreased in PCOS ( p = 0.04, p = 0.04, and p < 0.001, respectively). FGF8, FGF9, and FGF17 were increased in PCOS ( p = 0.02, p = 0.03 and p = 0.04, respectively), and FGFR1 was decreased in PCOS ( p < 0.001). VEGF-D ( p < 0.001), IGF1 ( p < 0.001), IGF-1sR ( p = 0.02), and PDGFRA ( p < 0.001) were decreased in PCOS compared to controls, as shown in and . Fibroblast growth factor receptor 1 (FGFR1; bFGF-R); Insulin-like growth factor I (IGF-1); Vascular endothelial growth factor D (VEGF-D); Platelet-derived growth factor receptor alpha (PDGFRA); Epidermal growth factor receptor variant III (EGFRvIII); Fibroblast growth factor 8 isoform A (FGF-8A); Insulin-like growth factor 1 receptor (IGF-I sR); Fibroblast growth factor 9 (FGF9); Epidermal growth factor receptor (EGFR); Epidermal growth factor (EGF); Fibroblast growth factor 17 (FGF-17); Fibroblast growth factor 19 (FGF-19). With stratification of the cohort based on BMI (≤29.9 kg/m 2 ), EGFR was decreased in PCOS ( p < 0.03), FGF8 was increased in PCOS ( p < 0.04), and FGFR1 was decreased in PCOS ( p < 0.001). VEGF-D ( p < 0.01), IGF1 ( p < 0.01), and IGF-1sR ( p < 0.03) were decreased in PCOS compared to controls, as shown in and . Fibroblast growth factor receptor 1 (FGFR1; bFGF-R); Insulin-like growth factor I (IGF-1); Vascular endothelial growth factor D (VEGF-D); Insulin-like growth factor 1 receptor (IGF-I sR); Epidermal growth factor receptor (EGFR); Fibroblast growth factor 8 isoform B ( FGF-8B ). With stratification of the cohort based on BMI (≤29.9 kg/m 2 ) and HOMA-IR, FGFR1 was decreased in PCOS ( p < 0.001), and VEGF-D ( p < 0.04), IGF1 ( p < 0.04), and IGF-1sR ( p < 0.02) were decreased in PCOS compared to controls, as shown in and . Fibroblast growth factor receptor 1 (FGFR1; bFGF-R); Insulin-like growth factor 1 receptor (IGF-I sR); Insulin-like growth factor I (IGF-1); Vascular endothelial growth factor D (VEGF-D). Correlation analysis of the unstratified groups showed that BMI was associated positively with the altered VEGF-sR3, FGF5, VEGF-sR2, and VEGF-C (r = 0.23, p = 0.0109; r = 0.19, p = 0.03; r = 0.2, p = 0.02; r = 0.19, p = 0.03, respectively) only in the PCOS subjects as shown in , but there was no correlation with BMI and the growth factor-related proteins that differed between PCOS and controls. Neither was there any correlation with inflammation, as adjudged by CRP, or of IR, as adjudged by the Homeostatic Model Assessment for IR (HOMA-IR), with the growth factor-related proteins that differed between PCOS and controls.
Circulatory Ras protein levels have not previously been reported in women with PCOS, making this study the first to explore their potential role in the pathophysiology of the condition. Prior investigations into Ras proteins in PCOS have focused on their expression in specific tissues rather than in the circulation. For instance, Xu et al. analyzed the gene expression of HRAS, NRAS, and KRAS in subcutaneous adipose tissue and found no significant differences in the expression of these Ras family genes between 22 women with PCOS and 13 controls . However, HRAS, NRAS, and KRAS expression showed positive correlations with testosterone levels, highlighting a potential link to androgenic activity. Interestingly, these genes also demonstrated inverse correlations with metabolic traits such as BMI, fasting glucose and IR, linking Ras proteins to the pathogenesis of PCOS. This study extends the investigation to circulating Ras proteins, providing a broader perspective on their systemic role in PCOS. Notably, circulating Ras proteins KRas and RASA1 did not differ significantly between women with PCOS and controls, indicating that systemic levels of these key Ras proteins may not be altered in the condition. However, despite the absence of significant differences in circulating Ras protein levels, findings from a drug-hub gene interaction network suggest that Ras-related genes, such as KRAS, may still have therapeutic potential . The network identified KRAS, along with other hub genes like Phosphatase and tensin homolog (PTEN), and MAPK1, as potential targets for treatment, with metformin specifically interacting with both PTEN and KRAS. These insights suggest that targeting these pathways, despite their unaltered expression in the circulation, could open new therapeutic avenues for PCOS management, highlighting the complexity of the condition and the potential for exploring Ras-mediated signaling pathways in treatment development. Of the 45 growth factor-related proteins analyzed for their role in activating Ras intracellular signaling pathways, only 11 demonstrated significant differences between the two groups. Interestingly, among these, three proteins were increased and eight were decreased in PCOS. These results collectively suggest that Ras-activated pathways are not broadly upregulated in PCOS and, therefore, Ras activation may not play a central role in mediating the pathogenesis or phenotype of the condition. Among the 11 growth factor-related proteins—EGF1, EGFR, EGFRvIII, FGF8, FGF9, FGF17, FGFR1, VEGF-D, IGF1, IGF-1sR, and PDGFRA—that showed significant differences between women with PCOS and controls, these changes were found to be independent of BMI, inflammation (indicated by the lack of correlation with CRP), or IR, suggesting that these differences are intrinsic to PCOS itself. However, when the cohort was stratified by BMI ≤ 29.9 kg/m 2 , only EGFR FGF8, FGFR1 VEGF-D, IGF1, and IGF-1sR remained significant with a loss of significance for EGF1, EGFRvIII, FGF8, FGFR1, and VEGF-D. This showed that these latter proteins are BMI dependent, and highlights that statistical adjustment for BMI may give different results compared to when cohorts are matched for BMI, as the statistical adjustment may result in an over or under estimate given that BMI and PCOS are so closely linked. When the cohorts were stratified for BMI ≤ 29.9 kg/m 2 and normal HOMA-IR (≤1.9), FGFR1, VEGF-D, IGF1, and IGF-1sR differed, suggesting that these factors are more likely to be inherently different in PCOS rather than an epiphenomenon of an associated PCOS feature such as BMI or IR. Conversely, EGFR and FGF8 were no longer significant indicating that these are dependent on IR, in accord with the literature for EGFR , while FGF8 modulation by insulin resistance is a novel finding. Our findings show reduced plasma EGF1 and EGFR levels in women with PCOS and, here, plasma EGF was no longer different when BMI was stratified and plasma EGFR was no longer different when BMI and IR were stratified, suggesting that EGF1 is BMI dependent; however, there are no comparable studies looking at serum EGF1 and EGFR in PCOS in the literature, and the effect of BMI on EGF1 and EGFR levels from other studies are discrepant. EGF1 and EGFR were negatively regulated with BMI and IR in adipose tissue , whilst, in patients with bipolar disorder, BMI did not affect EGF1 levels ; however, in breast cancer patients, there is a positive correlation between BMI and EGFR2 . A further study in participants having normal BMI reported higher EGFR levels in the cumulus granulosa cells of PCOS subjects versus controls, but the serum EGFR levels were not assessed in that study . Studies report that elevated EGF and EGFR expression is found in the granulosa cells and follicular fluid of the ovary in PCOS where EGF may inhibit granulosa cell estrogen synthesis, which is translated into arrest of follicle growth . EGFR signaling has been proposed as a therapeutic target, as its inhibition improved ovulatory function, estrous cyclicity, and hormone balance . Further studies are needed to determine the effect of BMI on circulating EGF1 and EGFR levels. Our study found increased levels of FGF8, FGF9, and FGF17 in PCOS patients, marking a novel observation as these growth factors have not been extensively explored in this context; however, as noted above, FGF8 appears to be related to IR. Research on FGF8’s ovarian functions has demonstrated its critical involvement in folliculogenesis, ovulation, and granulosa cell proliferation, suggesting that altered levels of FGF8 in PCOS could impact ovarian function and reproductive outcomes . FGF8’s role has also been studied in modulating granulosa cell function and its interaction with bone morphogenetic protein (BMP) signaling, which regulates ovarian steroidogenesis . FGF8 suppresses follicle-stimulating hormone (FSH)-induced estradiol production while enhancing BMP-Smad signaling, suggesting it contributes to disrupted oocyte–granulosa cell communication in PCOS . The increased FGF8 levels observed in our study may play a role in altered follicular dynamics and impaired estradiol production in PCOS, potentially representing a target for therapeutic intervention. However, a previous study investigating inflammatory proteins in non-obese, non-insulin-resistant PCOS women found no significant difference in FGF8 levels between PCOS and controls, suggesting that the role of FGF8 might vary depending on population characteristics, particularly IR . Conversely, FGF9 regulates granulosa cell function by inhibiting steroid hormone production and specific gene expressions, such as FSH receptor (FSHR) and cytochrome P450 family 11 subfamily A member 1 (CYP11A1), while promoting granulosa cell proliferation . It may also slow follicle development through pathways like IGF-I and cAMP, with hormonal factors such as IGF-I and estradiol influencing FGF9 production, highlighting its role in ovarian function and follicular differentiation . In our study, the increased expression of FGF9 in PCOS participants was lost when stratified for BMI, indicating that it is reflective of obesity rather than inherent to PCOS. FGF9 has been shown to enhance progesterone production in granulosa cells and regulate steroidogenic proteins such as steroidogenic acute regulatory protein (StAR) and cholesterol side-chain cleavage enzyme (P450scc) . Thus, the elevated FGF9 levels in obese PCOS patients could be an attempt to counteract the dysregulation of steroidogenesis and follicular maturation. However, this compensatory effect may be insufficient to overcome the broader metabolic and reproductive challenges characteristic of the condition. However, no studies have specifically investigated FGF9 levels in PCOS compared to control groups, making it challenging to directly compare our findings with existing literature. FGF17 is detected mainly in oocytes, but also in granulosa cells and there is no literature directly addressing its role in PCOS or metabolism; however, in this study, it can be seen that it appears to be related to BMI as, when BMI was accounted for, it did not differ between women with and without PCOS. In our study, we observed decreased FGFR1 expression in PCOS, which aligns with findings from another study showing reduced FGFR1 levels in granulosa-lutein cells of women with PCOS . This reduction in FGFR1 may contribute to impaired angiogenesis and follicular development, suggesting a link between FGFR1 dysregulation and ovarian dysfunction in PCOS. Further insights into the role of FGFR1 signaling in reproductive health come from studies indicating that the loss of FGFR1 disrupts cell cycle progression and downregulates key genes such as EZH2, which regulates cell proliferation and differentiation . The decreased expression of FGFR1 in PCOS participants may, therefore, reflect a disruption in cellular pathways that are essential for normal ovarian function and follicle maturation. Interestingly, a recent study has shown that impaired FGFR1 signaling is associated with IR and metabolic dysfunction, further supporting the relevance of our findings ; however, after stratification for both BMI and HOMA-IR, it still differed between the women with and without PCOS suggesting that it is independent of both BMI and IR. The decreased expression of FGFR1 in our PCOS participants could reflect a similar disruption in signaling, potentially contributing to the metabolic and reproductive challenges characteristic of the condition. Additionally, we found decreased levels of VEGF-D expression in PCOS patients, a factor primarily described as a lymphangiogenic molecule. Previous studies have reported significantly elevated concentrations of serum VEGF (also known as VEGF-A) in PCOS patients in comparison to controls . This increase in pro-angiogenic factors like VEGF can contribute to hyperplasia, hypervascularity, and the gradual development of endothelial dysfunction, key features of PCOS . Interestingly, however, our study did not find significant differences in VEGF expression between PCOS patients and controls. Instead, we identified a significant decrease in VEGF-D expression, a less well-researched member of the VEGF family in the context of PCOS. The observed decrease in VEGF-D expression in our study, in contrast to previous findings of elevated VEGF-A levels, may reflect differences in the roles of these VEGF family members in PCOS pathophysiology. Similarly, another study investigating adolescent PCOS patients found no significant differences in VEGF levels between cases and controls, suggesting that VEGF expression may vary across subpopulations and disease stages . Future research should investigate the specific regulatory mechanisms governing VEGF-D expression in PCOS and how these pathways interact with broader angiogenic and lymphangiogenic processes. We observed decreased levels of IGF1 expression in PCOS patients. This appears to contrast with previous studies that have reported increased serum levels of IGF1 in PCOS patients in comparison to controls ; however, in a meta-analysis of 20 studies, 5 of these showed higher IGF1 levels in the controls and when stratified for BMI less than or equal to 29 kg/m 2 , IGF1 levels were significantly higher in the control populations, suggesting that the serum IGF1 levels between groups depends on the proportion of subjects with a BMI less than or greater than 29 kg/m 2 . These results are therefore in accord with the literature when stratifying for BMI in our analysis. In addition, PCOS patients who are obese exhibit reduced growth hormone (GH) secretion, possibly resulting in diminished downstream effects on the regulation of IGF1 . Our findings suggest that IGF1 levels in PCOS are not universally elevated and that the relationship between PCOS and IGF1 dynamics is more complex than previously understood. PDGFRA was found to decrease in women with PCOS, but after stratifying for BMI, it no longer differed between groups, indicating that it was a feature of obesity rather than PCOS per se. PDGFR signaling is thought to be a common mechanism in the control of multiple steroidogenic lineages involved in fertility and has been suggested as a potential therapeutic target for PCOS . While our findings on growth factor-related proteins provide new insights into the complex hormonal interactions in PCOS, they also underscore the inherent challenges in isolating the effects of PCOS from other confounding factors. Limitations of this study include its focus on a predominantly Caucasian population, which may limit the generalizability of the findings to other ethnic groups. Additionally, while our study sought to examine the role of circulating growth factor-related proteins in PCOS, stratifying for BMI and IR allowed us to determine which factors may be inherent within PCOS. Both BMI and IR are highly correlated with PCOS and its metabolic and hormonal features. As such, regression models adjusting for these factors may show differing results than when covariate adjustment is made up front by accounting for BMI and IR, for instance. Further studies with larger sample sizes and refined statistical approaches, such as stratifying by BMI, insulin sensitivity and systemic inflammation could help address these limitations and provide more definitive insights into the pathways involved in PCOS pathophysiology.
Several growth factors that activate Ras differ between women with and without PCOS and, when stratified for BMI and HOMA-IR, only FGFR1, VEGF-D, IGF1, and IGF-1sR, differed and those appear to be inherent features of the pathophysiology of PCOS.
|
Impact of needle-point bipolar ionization system in the reduction of bioaerosols in collective transport | 7cec0be7-2c3c-4048-ac5c-e5fa1016209b | 9500091 | Microbiology[mh] | Introduction The airborne transmission model refers to the spread of a disease through aerosols up to 100 μm, which can be transmitted through the air over time and distance ( ). At the beginning of the COVID-19 pandemic, this route of infection was not considered relevant, and more attention was paid to the transmission through droplets and infected surfaces ( ). The global acceptance of the spread of SARS-CoV-2 by aerosols allowed an improvement in the preventive approach, including new techniques for epidemiological management, such as the measurement of exhaled carbon dioxide (CO 2 ) as an indicator of the risk of contagion ( ; ) and air purification technologies. The term ‘bioaerosol’ refers to those particles of biological origin or active particles that can affect living organisms, causing them some allergy, toxicity, or infection ( , ). Bioaerosols can include viruses, bacteria, spores, pollen, and, in general, any other microorganisms with an aerodynamic diameter between 0.5 and 100 μm ( ). In the context of airborne pathogen transmission, following mechanistic hypotheses about disease transmission, the spread of bioaerosols is associated with respiratory events ( ), mainly coughs and sneezes ( ). However, the generation of bioaerosols has also been demonstrated during breathing and speech. Disseminating aerosols size ranges from a few nanometers to hundreds of micrometers ( ). Regarding the COVID-19 pandemic, bioaerosols released by a person infected with SARS-CoV-2 may contain some virus or traces of it, and the viral load contained in the particles is a crucial factor in determining the relative contribution of airborne transmission ( ; ). The modal distributions of aerosols acquire great relevance in infection transmission. Particle size determines the aerodynamic characteristics and deposition dynamics and the variability in the viral colonization model depending on the depth of the reached respiratory tract ( ; ; ). A higher transmission rate of SARS-CoV-2 has been observed in means of transport compared to other public and shared spaces. Thus, analyzed the relationship between the transmission of COVID-19 and the type of work, pointing to an incidence of 18 % of cases in the transport sector, only behind the health sector (22 %). determined an association between the number of infected patients and the domestic transportation route of Wuhan by train, private car, and flights, finding a significant relationship between the infection rate and transportation by train, but not in flights and private vehicles. This trend was also studied using other infectious disease models ( , , ). The risk of contagion in public transport is determined by the seat's proximity to the infected person, the number of passengers, the number of interactions with other passengers, the duration of the trip, and the capacity performance of air renewal ( ). Generic measures have been adopted and have proven effective to reduce the virus spreading: surfaces sanitation, social responsibility measures (such as the use of a mask, social distancing, and hand hygiene), and ventilation measures ( ; ). Regarding basic interior sanitation, solutions based on chlorine and isopropyl alcohol effectively disinfect surfaces ( ). Recommendations from the Centers for Disease Control and Prevention (CDC) and The Transportation Research Board (TRB) include at least two daily cleanings and suggest having particular account high-touch zones ( ). The survival of SARS-CoV-2 varies depending on the material on which it is deposited. SARS-CoV-2 remains active for up to 72 h on plastic and stainless steel, up to 24 h on cardboard, and <4 h on copper ( ), therefore it would be necessary to consider heterogeneity within the different areas of the same public transport unit. HEPA filtration is very effective eliminating of microorganisms present in the ambient air in the means of transport since it allows efficient elimination of microorganisms reaching efficiencies close to 99.997 % ( ; ). However, the performance of this technology will depend on the number of air changes and the feasibility of incorporating these filters to the installed air conditioning system. The percentage of recirculated air is another aspect to consider in the air quality in transport ( ). Additionally, natural air renovation, such as opening windows, can considerably reduce the concentration of CO 2 inside a vehicle as well as the number of airborne containing microorganisms ( ). Concerning the COVID-19 pandemic, the preventive strategy to reduce the risk of contagion by aerosols has been managed along two different lines: on the one hand, based on the measurement of CO 2 as an indicator of air renovation. On the other hand, by purifying or cleaning the air by different means including ultraviolet C (UV-C) radiation technology, dry fumigation with hydrogen peroxide (H 2 O 2 ) or technology based on non-thermal plasma, among others. Technology based on the UV spectrum between 200 and 280 nm (UV-C) has been widely used in disinfection processes. The intracellular components of microorganisms (DNA, RNA, and proteins) present a high and variable sensitivity to absorb UV-C photons, producing critical damage to their genome and inhibiting their correct replication ( ). The efficiency of UV disinfection varies depending on the distance from the surface and the application time ( ). This technology has proven helpful against the SARS-CoV-2 virus ( ; ; ), reducing the viral load between 0 and 6 log orders of magnitude in culture media ( ). Although the performance of the technology is limited against bioaerosols loaded with SARS-CoV-2, where the required UV 254 dose is high ( ), some studies highlight its effectiveness for air purification ( ) and especially if combined with HEPA filtration ( ). Non-thermal plasma-based technology has positioned itself as one of the leading air purification strategies in the context of the pandemic caused by the SARS-CoV-2 virus. Plasma inactivation of microorganisms has been attributed to cell wall rupture and damage of the genetic material ( ). Its mechanism of action is multiple. The presence of charged particles, ions, reactive oxygen species (ROS) and oxygen-containing radicals, UV-C, vacuum ultraviolet (VUV), and localized heating events stand out, acting exclusively or in combination ( ; ; ; ). The antimicrobial effect of this technology has been widely exploited in the health sector to sterilize surgical instruments ( ). Moreover, it has been tested in controlled environments reporting excellent efficiency in the inactivation of specific bacterial species ( ) and enveloped/non-enveloped viruses ( ; ; ). However, a limited number of studies support its efficiency in other settings. Non-thermal plasma-based technology has been tested for other applications regarding the current pandemic, obtaining variable and lower results than those described by the manufacturers ( ). One of the major concerns and limitations of this technology is the generation of by-products in harmful concentrations ( ), where some authors suggest increases in actual conditions ( ). However, this aspect is not discussed in this article. This work provides information on possible systems to reduce the risk of infectious diseases in public transport. First, the efficacy of needle-point bipolar ionization against eliminating environmental bioaerosols in the Zaragoza Tram (Zaragoza, Spain) is evaluated. Given biosafety and ethical constraints, we decided to perform the assays using environmental bacteria (not artificially dusted). Afterward, the efficiency of the isolated filtration media is tested against non-biological particles. These tests carried out in the laboratory allow the efficiency of the filter to be measured against submicron matter (hardly characterizable using the previous tests). Finally, the combined action of both strategies is studied. In parallel, the antimicrobial performance of ionization for surface disinfection is evaluated. This work evaluates the air purification systems for their implementation in local public transport.
Materials and methods 2.1 Needle-tip bipolar ionization system As shown in , two PA604 ionization units of the 600 Series (Tayra SA, Spain) were installed in the suction vertices of the delivery fan of the two air conditioning units arranged in the Zaragoza Tram Unit (Model Urbos 3 CAF, ES). The two HVAC (Heating, Ventilation, and Air Conditioning) units installed in the Tram drive a total flow of 2800—3300 m 3 /h, with a fresh air ratio of 1:3 fresh/return air. The selected ionizers were of the needle tip brush type (PA604, Tayra SA Spain). They produce an equal amount of positive and negative ions. They have a maximum treated flow of 4100 m 3 /h and a <5 kV DC voltage between brushes. In the case of the tests where ionization was evaluated, an Air Ion Counter COM-3200PRO II (Com System INC, Tokyo JA) was used to ensure the correct generation of ions. 2.2 Conditions and preparation of the tram units The Urbos 3 tram model (CAF, Spain) has a total length of 33 m, a width of 2.65 m, and a height of 3.2 m. It has a capacity of 200 seats, of which 146 are standing seats (3.5 people per m 2 ) and 54 are seats. It operates in Zaragoza, the fifth biggest city in Spain, with a population close to 700 k inhabitants, located midway between Madrid and Barcelona. The tram units included in the study were in operation for a full day, and the usual protocol was not carried out night cleaning. During the morning of the following day, the tram unit provided partial service lasting approximately 4 h. During this time, the hydroalcoholic gel dispensers were removed to avoid disinfection of the tram surfaces included in the study, and the windows were closed during the journeys. Travelers wore masks. Upon arrival at the depot, all the tram doors were opened for 15 min to renew the interior air and replace it with fresh ambient air. Doors opening were performed to maximize the homogeneity and reproducibility of the assays. The conditions used for the preparation of the Tram Units are detailed in . The bipolar ionization units were turned on and stabilized for at least 15 min prior to the start of the test. The filter used during the tests was the usual one recommended by the manufacturer (Coarse 45 % according to UNE-EN ISO 16890, Merak Long Life Filter, Madrid SP). They were kept in the same usual conditions and were within the period of useful life determined by the manufacturer. The air conditioning system was kept off for the reference sampling, and the Tram was ventilated between tests, recirculating the depot air for at least 15 min. 2.3 Environmental sampling conditions and cultivation A total of 6000 l of air was sampled at a flow rate of 300 l/min in an initial 5 ml of phosphate buffered saline (PBS) solution (Sigma-Aldrich, Darmstadt DE) using a Coriolis μ (Bertin Technologies, Montigny-le-Bretonneux FR). Sampling was carried out at three different points of the Tram to homogenize the sampling and collect microorganisms at points characterized by different ionic concentrations. The manufacturers initially modeled the ion concentration throughout the Tram Unit, guaranteeing a sufficient concentration for the inactivation of microorganisms in any of their locations. In each test, three samplings were carried out with a frequency of 30 min to evaluate the system's efficiency for 90 min, corresponding to the travel time in each tram line. In addition, an initial sampling was carried out to calculate the relative efficiency. Once the sampling was finished, the volume corresponding to 450 l of air sample aspiration was seeded on a Plate Count Agar (PCA) plate (Scharlab, Spain), for 72 h at 30 °C. The final sampled solution was variable depending on the climatological conditions of the sampling. Thus, the counts were normalized to be comparable, taking the environmental sampling prior to the intervention as a reference. Five replicates of each sample were made. Colony-forming units (CFU) were manually counted after the incubation period (72 h at 30 °C). The CFU counted in each replica were averaged and normalized. Efficiency in CFU inactivation was calculated by comparing the percentage of surviving CFUs with the reference sample. 2.4 Identification and visualization of the bacterial species found Among the CFUs, the 15 most representative specimens collected during the samplings were analyzed. The identification of the microorganisms has been carried out using MALDI-TOF mass spectrometry technology (MALDI Biotyper, Bruker Massachusetts USA), comparing the results to databases (Bruker, Massachusetts USA). The morphology of the bacterial strains fibers was observed using a Scanning Electron Microscopy (SEM) JEOL 6360-LV (Deben UK Ltd., Edmunds, United Kindom). For sample preparation, a sample of the corresponding CFU was placed on a slide, fixed with carbon tape to a SEM microscope holder and sputtered with Au/Pd to promote electron conduction. The average of CFUs diameters and standard deviations (SD) were obtained from manual measurements with the free software Image-J (v1.52; ) for n = 50. 2.5 Sampling conditions of surfaces and cultivation Rodac-PCA plates (Plate Count Agar-Sharlab, Spain) were used for surface sampling. At least 30 surface samples per test were taken at different points of the Tram, at different levels, and considering different types of surfaces. Seats, backrests, grab bars, intermediate supports, walls, and windows were included. Despite being carried out initially, the soil samples were excluded due to their high variability associated with residual contamination and dirt. The final surface sample was compared with a nearby (5 cm proximity) reference sample, and the ionization efficiency was calculated following the same method as for air sampling. Samples collected by contact in the Rodac-PCA plates were cultivated for 72 h at 30 °C. 2.6 Determination of filtration efficiency against submicron particles As shown in -a, aerosols were produced using a Topas-ATM226 generator with a saline solution of sodium chloride (3 % NaCl in ddH 2 0). The obtained microdroplets pass through a tubular silica gel air dryer to evaporate the water and produce solid particles. The particle size distribution ( -b) inside the cabin was measured using an SMPS TSI 3936 composed of an electrostatic classifier (DMA TSI 3081) and a condensation particle counter (CPC TSI 3782). The particles were dragged at a 0.6 l/min flow rate. The filter was placed between bronze discs sealed with Teflon tape, with 30 × 20 mm Teflon washers on each side. The exposed filter area was variable (2.05, 4.1, and 8.1 mm) to adjust the desired flow rate. The measurements lasted 2 min and were made in duplicate. Due to the concentration of particles variation, measurements were made passing through a free tube between measurements to calculate relative efficiency according to Eq. . Where C up stands for concentration upstream and C down stands for concentration downstream. The retention efficiency is expressed in global efficiency as ‘number of particles’. (1) n = 100 × C up − C down C up Pressure drop testing was carried out using alcohol columns based on Bernoulli's principle. The free ends of the tubes have been inserted into the two quick couplings located on both sides of the filter sample holder. Measurements were made with a volumetric flow rate of 0.6 l/min. 2.7 Determination of air purification efficiency System performance has been characterized using different approaches. Firstly, the calculation of the relative efficiency in the air and surface samples was carried out according to Eq. . The determination of the final CFU ( CFU 2 ) was given as the average between the counts of the five replicates of each sampling. The initial CFU data ( CFU 1 ) was taken from the previous test to assess performance every 30 min. Those cultures plates replicates with CFUs with values higher than three times those of the rest of the plates of the same sample were eliminated as they were considered non-representative extreme values. Secondly, clean air delivery rate (CADR) and first order loss rates using a simple linear regression against ln-transformed mean concentration values was determined as described by . (2) n = 100 × CFU 1 − CFU 2 CFU 1
Needle-tip bipolar ionization system As shown in , two PA604 ionization units of the 600 Series (Tayra SA, Spain) were installed in the suction vertices of the delivery fan of the two air conditioning units arranged in the Zaragoza Tram Unit (Model Urbos 3 CAF, ES). The two HVAC (Heating, Ventilation, and Air Conditioning) units installed in the Tram drive a total flow of 2800—3300 m 3 /h, with a fresh air ratio of 1:3 fresh/return air. The selected ionizers were of the needle tip brush type (PA604, Tayra SA Spain). They produce an equal amount of positive and negative ions. They have a maximum treated flow of 4100 m 3 /h and a <5 kV DC voltage between brushes. In the case of the tests where ionization was evaluated, an Air Ion Counter COM-3200PRO II (Com System INC, Tokyo JA) was used to ensure the correct generation of ions.
Conditions and preparation of the tram units The Urbos 3 tram model (CAF, Spain) has a total length of 33 m, a width of 2.65 m, and a height of 3.2 m. It has a capacity of 200 seats, of which 146 are standing seats (3.5 people per m 2 ) and 54 are seats. It operates in Zaragoza, the fifth biggest city in Spain, with a population close to 700 k inhabitants, located midway between Madrid and Barcelona. The tram units included in the study were in operation for a full day, and the usual protocol was not carried out night cleaning. During the morning of the following day, the tram unit provided partial service lasting approximately 4 h. During this time, the hydroalcoholic gel dispensers were removed to avoid disinfection of the tram surfaces included in the study, and the windows were closed during the journeys. Travelers wore masks. Upon arrival at the depot, all the tram doors were opened for 15 min to renew the interior air and replace it with fresh ambient air. Doors opening were performed to maximize the homogeneity and reproducibility of the assays. The conditions used for the preparation of the Tram Units are detailed in . The bipolar ionization units were turned on and stabilized for at least 15 min prior to the start of the test. The filter used during the tests was the usual one recommended by the manufacturer (Coarse 45 % according to UNE-EN ISO 16890, Merak Long Life Filter, Madrid SP). They were kept in the same usual conditions and were within the period of useful life determined by the manufacturer. The air conditioning system was kept off for the reference sampling, and the Tram was ventilated between tests, recirculating the depot air for at least 15 min.
Environmental sampling conditions and cultivation A total of 6000 l of air was sampled at a flow rate of 300 l/min in an initial 5 ml of phosphate buffered saline (PBS) solution (Sigma-Aldrich, Darmstadt DE) using a Coriolis μ (Bertin Technologies, Montigny-le-Bretonneux FR). Sampling was carried out at three different points of the Tram to homogenize the sampling and collect microorganisms at points characterized by different ionic concentrations. The manufacturers initially modeled the ion concentration throughout the Tram Unit, guaranteeing a sufficient concentration for the inactivation of microorganisms in any of their locations. In each test, three samplings were carried out with a frequency of 30 min to evaluate the system's efficiency for 90 min, corresponding to the travel time in each tram line. In addition, an initial sampling was carried out to calculate the relative efficiency. Once the sampling was finished, the volume corresponding to 450 l of air sample aspiration was seeded on a Plate Count Agar (PCA) plate (Scharlab, Spain), for 72 h at 30 °C. The final sampled solution was variable depending on the climatological conditions of the sampling. Thus, the counts were normalized to be comparable, taking the environmental sampling prior to the intervention as a reference. Five replicates of each sample were made. Colony-forming units (CFU) were manually counted after the incubation period (72 h at 30 °C). The CFU counted in each replica were averaged and normalized. Efficiency in CFU inactivation was calculated by comparing the percentage of surviving CFUs with the reference sample.
Identification and visualization of the bacterial species found Among the CFUs, the 15 most representative specimens collected during the samplings were analyzed. The identification of the microorganisms has been carried out using MALDI-TOF mass spectrometry technology (MALDI Biotyper, Bruker Massachusetts USA), comparing the results to databases (Bruker, Massachusetts USA). The morphology of the bacterial strains fibers was observed using a Scanning Electron Microscopy (SEM) JEOL 6360-LV (Deben UK Ltd., Edmunds, United Kindom). For sample preparation, a sample of the corresponding CFU was placed on a slide, fixed with carbon tape to a SEM microscope holder and sputtered with Au/Pd to promote electron conduction. The average of CFUs diameters and standard deviations (SD) were obtained from manual measurements with the free software Image-J (v1.52; ) for n = 50.
Sampling conditions of surfaces and cultivation Rodac-PCA plates (Plate Count Agar-Sharlab, Spain) were used for surface sampling. At least 30 surface samples per test were taken at different points of the Tram, at different levels, and considering different types of surfaces. Seats, backrests, grab bars, intermediate supports, walls, and windows were included. Despite being carried out initially, the soil samples were excluded due to their high variability associated with residual contamination and dirt. The final surface sample was compared with a nearby (5 cm proximity) reference sample, and the ionization efficiency was calculated following the same method as for air sampling. Samples collected by contact in the Rodac-PCA plates were cultivated for 72 h at 30 °C.
Determination of filtration efficiency against submicron particles As shown in -a, aerosols were produced using a Topas-ATM226 generator with a saline solution of sodium chloride (3 % NaCl in ddH 2 0). The obtained microdroplets pass through a tubular silica gel air dryer to evaporate the water and produce solid particles. The particle size distribution ( -b) inside the cabin was measured using an SMPS TSI 3936 composed of an electrostatic classifier (DMA TSI 3081) and a condensation particle counter (CPC TSI 3782). The particles were dragged at a 0.6 l/min flow rate. The filter was placed between bronze discs sealed with Teflon tape, with 30 × 20 mm Teflon washers on each side. The exposed filter area was variable (2.05, 4.1, and 8.1 mm) to adjust the desired flow rate. The measurements lasted 2 min and were made in duplicate. Due to the concentration of particles variation, measurements were made passing through a free tube between measurements to calculate relative efficiency according to Eq. . Where C up stands for concentration upstream and C down stands for concentration downstream. The retention efficiency is expressed in global efficiency as ‘number of particles’. (1) n = 100 × C up − C down C up Pressure drop testing was carried out using alcohol columns based on Bernoulli's principle. The free ends of the tubes have been inserted into the two quick couplings located on both sides of the filter sample holder. Measurements were made with a volumetric flow rate of 0.6 l/min.
Determination of air purification efficiency System performance has been characterized using different approaches. Firstly, the calculation of the relative efficiency in the air and surface samples was carried out according to Eq. . The determination of the final CFU ( CFU 2 ) was given as the average between the counts of the five replicates of each sampling. The initial CFU data ( CFU 1 ) was taken from the previous test to assess performance every 30 min. Those cultures plates replicates with CFUs with values higher than three times those of the rest of the plates of the same sample were eliminated as they were considered non-representative extreme values. Secondly, clean air delivery rate (CADR) and first order loss rates using a simple linear regression against ln-transformed mean concentration values was determined as described by . (2) n = 100 × CFU 1 − CFU 2 CFU 1
Results and discussion 3.1 Effect of filtration and needle-tip brush bipolar ionization on air 3.1.1 CFU stability in the environmental air Airborne microorganisms are significantly affected by weather and environmental conditions ( ; ; ). To guarantee a stable concentration of environmental bacteria during the tests, air samples were taken every 30 min without filtration and ionization. As depicted in , the results of this test suggest that the environmental microbiota is stable over time during the study hours on the same day. An average of 40.4 ± 1.5 CFU/m 3 of air was obtained during the first test and 53.8 ± 3.0 CFU/m 3 during the second test. CFU counts were averaged using all five replicates of each sample. In the plates at 30 and 90 min from the first test, 40.0 CFU/m 3 were counted in each one with respect to the initial 42.5 CFU/m 3 . In the plates of the second test, 52.5 and 53.1 CFU/m 3 were found at 30 and 90 min, compared to the initial 57.5 CFU/m 3 . To determine the stability of the CFUs, the Relative Standard Deviation (RSD) of each set of plates was calculated. The obtained RSD of ~0.04 in the first test and ~0.06 in the second test suggests that the bacterial environmental contamination is stable in air during the 90 min of sampling, so that the rest of the tests presented below are performed under robust and reproducible conditions throughout each test. Loss-rate regression of CFU has been fitted to a polynomial to characterize stability ( c). The resulting CADR of 0.299 m 3 /min and the loss rate constant of 0.0012 per minute during the stability control condition reinforces the previous conclusion about the stability of environmental CFUs. 3.1.2 Bacterial species identification As enumerated in , the 15 most frequent colonies were isolated and identified. The bacterial identification is of interest since the efficiency of the BIU is evaluated predominantly for this subset of strains, despite the existence of others in a smaller proportion. In addition, its visualization allowed to evaluate its morphology and size, which is relevant for filtration studies. As depicted in , most bacterial strain (12/15) have sizes >1 μm in its longest dimension. The measured dimension represents the minimum size in which each strain of CFUs can be found in the air. 3.1.3 Bipolar ionization reduces aerial CFUs The effect of ionization was considered exclusively to evaluate the single efficiency of the bipolar ionization unit. The objective of this study was to isolate the efficiency of the ion system and quantify the improvement that it supposes by itself in the absence of other perturbations, such as the filtration of the air conditioning system. An increasing efficiency was observed from the initial sampling to the final samplings. Ion concentration varied (19.9 · 10 9 —31.5 · 10 9 ions/m 3 ) in all the samplings carried out due to the heterogeneity in the air distribution at the different points of the Tram. On average, during the first 30 min of the experiment, CFUs were reduced by 45.9 % (54.1 % persisted). As shown in , an efficiency of 61.8 % (38.2 % persisted) was estimated after 60 min. The efficiency increased to 69.2 % (30.8 % persisted) after 90 min. evaluated the efficiency of needle-point bipolar ionization systems (~10—20 kV) in airplanes, obtaining a reduction of CFU ( Staphylococcus epidermidis ) variable between 20.7 % and 60.0 % after 1 h of ionization, which is consistent with the results obtained in the present study. The resulting loss rate constant is 0.018 per minute under ionization conditions, and the estimated CADR was 5.153 m 3 /min. Regarding the control values, it is observed that the clean air delivery rate increases, which implies a greater amount of air free of bioaerosols. Specifically, this value increased in >17 times. 3.1.4 Filtration reduces aerial CFUs The electrostatic potential influence of the surface on the deposition of bacteria from the air has been demonstrated ( ). Various studies suggest that bipolar ionization reduces particulate matter in the air due to increased particle deposition and/or filtration efficiency associated with the increase in the aerodynamic diameter of aerosols due to the agglomeration of fine particles ( ; ; ). This phenomenon is mainly associated with electrostatic effects ( ; ) and is dependent on particle size and composition, relative humidity, ionization time, and surface material ( ). Filters can interfere in estimating the efficiency of the ionization system. Then, the effect of the filtration implemented in the air conditioning system of the Tram has been characterized separately. The tests were carried out with the filter usually installed on the Tram (Coarse 45 % filter medium). The Air Changes per Hour (ACH) of the Zaragoza Tram remained as they are in the normal operation of the unit at 25 ACH. This indicates that the air passes through the filter 12.5 times/h on average. As seen in , loss rate constant was 0.033 per minute, and the estimated CADR was 9.261 m 3 /min. These parameters imply that filtration is more effective than ionization. CFU amount upon filtration suggests a reduced efficiency considering the number of air changes inside the tram unit. On average, CFUs were reduced by 73.4 % (26.6 % persisted), 84.0 % (16.0 % persisted), and 92.0 % (8.0 % persisted) during the first 30, 60, and 90 min, respectively. In this sense, the CFUs are large enough to favor the filtration mechanisms since they are, for the most part, >1 μm. Faced with submicron particles, smaller in size than the bacteria tested, the efficiency of the filter is limited according to the conditions studied ( ). The filter complies with the UNE-EN 16890 standard and refers to a 75 % filtration for particles greater than ~10 μm. This filter is efficient for the retention of dust or pollen. However, it is inefficient against fine particles and viruses. The tests carried out are limited. Due to the conditions of the equipment used for the determination of filtration efficiency against submicron particles, the filter may have been ‘clogged’ by the concentration of NaCl particles, assuming a notable increase in pressure drop and an ornament in the efficiency result. The Coarse 45 % filter was evaluated in a more compact format than natural: 2—3 mm thick instead of 20—30 mm. The results suggest that the speed at which the air passes through the filter medium retains the larger particles. However, no notable difference is observed in the flows studied. The curves could not be very representative of this filter's actual efficiency (in working conditions in the air conditioning system of the Tram). In experimental sampling tests, SARS-CoV-2 viral RNA has been mainly found in particle sizes from 0.25 μm ( ; ). Therefore, determining the efficiency of the filter against submicronic sizes and above is relevant to quantify its performance. As shown in , the Coarse 45 % filter presented an approximate retention efficiency of ~30 % for 0.5 μm particles at a flow rate of ~4079 m 3 /h. For flow rates lower than this, the efficiency of the filter decreases to values <7 %. These tests suggest that larger particles have more inertia at high speeds (high flow rates) and, therefore, a higher retention rate in the filter medium ( , ). However, the clogging of NaCl particles observed in the head loss tests may have overestimated these results. In the tests where the highest speed is simulated, clogging is observed, which translates into an increase up to 380 Pa of pressure drop ( ). 3.1.5 Effect of combined ionization and filtration efficiency against aerial CFUs Once the effects of ionization and filtration were evaluated separately, the combination's performance was intended to study. However, no clear advantage was obtained from the combination of mechanisms. The first 30 min reduced the CFU concentration by 74.2 % (25.8 % persisted), while at 60 min, the efficiency increased to 82.8 % (17.2 % persisted), and at 90 min, 94.1 % was eliminated (5.9 % persisted) ( ). While these data are promising, the filtration alone reduced the CFU concentration by 73.4 % and 92.0 %, respectively, after 30 and 90 min. The maintenance of filtration efficiency can be explained because the highest rate of particle agglomeration due to electrostatic phenomena occurs in the finest particles (<0.15 μm). Then, there is a higher concentration of particles >0.3 μm ( ), which does not favor the performance of filtration ( , ). The loss rate constant was 0.047 per minute, and the estimated CADR was 13.208 m 3 /min. The value of CADR of ionization and filtration together was higher than filtration and ionization separately by 3.947 and 8.055 m 3 /min, respectively. 3.2 Effect of the bipolar ionization in the elimination of CFUs on surfaces 3.2.1 Evaluation of the CFUs stability on surfaces To guarantee the uniformity of the tests, four surface samples have been carried out, with a total of at least 60 Rodac PCA plates per test: 30 at t = 0 (CFU i ) and 30 at t = 2 h (CFU f ). Ideally, the CFU i should be equal to the CFU f (ie the percentage of CFUi = CFUf would ideally be 100 %). However, we have considered it relevant to analyze whether there are more or fewer CFU at the end of the experiment to determine if it is a matter of stability or simply variability across the surface (i.e., CFU i < CFU f or CFU i > CFU f ). If the CFUs had deteriorated over time, a trend greater than CFU i > CFU f . This has not been observed, so it is assumed that the CFUs are stable during the time of the experiment (2 h) but there is a non-negligible need between the neighboring test surfaces. To quantify this dispersion, the RSD of each set of plates has been calculated. RSD close to zero suggests high uniformity in the samples. The variability of the results ( ) required an increase in the sample to obtain representative conclusions. 3.2.2 Bipolar ionization efficiency on surface samples On the one hand, “effectiveness” refers to the percentage of PCA plates where final CFU number was lower than the initial (CFU i > CFU f ). That is, in this column the percentage of CFU plates that showed a potential action of ionization has been studied. On the other hand, “efficiency” represents the percentage of “removed” bacteria in these plates potentially due to ionization effect, which is calculated according to Eq. . It is relevant to differentiate between the efficiency for plates with CFU > 50, CFU > 20, and CFU > 10 to avoid distorting the results obtained. However, the global efficiency does not discriminate between values obtained ( ). According to our findings, in this experimental set-up, the ionization did not present a major role on the inactivation of microorganisms on surfaces. The average ionization effectiveness was 46 %, while the average CFU i > CFU f in the stability study was 36 %. We cannot exclude that the difference between initial and final CFUs could be due to effects other than ionization.
Effect of filtration and needle-tip brush bipolar ionization on air 3.1.1 CFU stability in the environmental air Airborne microorganisms are significantly affected by weather and environmental conditions ( ; ; ). To guarantee a stable concentration of environmental bacteria during the tests, air samples were taken every 30 min without filtration and ionization. As depicted in , the results of this test suggest that the environmental microbiota is stable over time during the study hours on the same day. An average of 40.4 ± 1.5 CFU/m 3 of air was obtained during the first test and 53.8 ± 3.0 CFU/m 3 during the second test. CFU counts were averaged using all five replicates of each sample. In the plates at 30 and 90 min from the first test, 40.0 CFU/m 3 were counted in each one with respect to the initial 42.5 CFU/m 3 . In the plates of the second test, 52.5 and 53.1 CFU/m 3 were found at 30 and 90 min, compared to the initial 57.5 CFU/m 3 . To determine the stability of the CFUs, the Relative Standard Deviation (RSD) of each set of plates was calculated. The obtained RSD of ~0.04 in the first test and ~0.06 in the second test suggests that the bacterial environmental contamination is stable in air during the 90 min of sampling, so that the rest of the tests presented below are performed under robust and reproducible conditions throughout each test. Loss-rate regression of CFU has been fitted to a polynomial to characterize stability ( c). The resulting CADR of 0.299 m 3 /min and the loss rate constant of 0.0012 per minute during the stability control condition reinforces the previous conclusion about the stability of environmental CFUs. 3.1.2 Bacterial species identification As enumerated in , the 15 most frequent colonies were isolated and identified. The bacterial identification is of interest since the efficiency of the BIU is evaluated predominantly for this subset of strains, despite the existence of others in a smaller proportion. In addition, its visualization allowed to evaluate its morphology and size, which is relevant for filtration studies. As depicted in , most bacterial strain (12/15) have sizes >1 μm in its longest dimension. The measured dimension represents the minimum size in which each strain of CFUs can be found in the air. 3.1.3 Bipolar ionization reduces aerial CFUs The effect of ionization was considered exclusively to evaluate the single efficiency of the bipolar ionization unit. The objective of this study was to isolate the efficiency of the ion system and quantify the improvement that it supposes by itself in the absence of other perturbations, such as the filtration of the air conditioning system. An increasing efficiency was observed from the initial sampling to the final samplings. Ion concentration varied (19.9 · 10 9 —31.5 · 10 9 ions/m 3 ) in all the samplings carried out due to the heterogeneity in the air distribution at the different points of the Tram. On average, during the first 30 min of the experiment, CFUs were reduced by 45.9 % (54.1 % persisted). As shown in , an efficiency of 61.8 % (38.2 % persisted) was estimated after 60 min. The efficiency increased to 69.2 % (30.8 % persisted) after 90 min. evaluated the efficiency of needle-point bipolar ionization systems (~10—20 kV) in airplanes, obtaining a reduction of CFU ( Staphylococcus epidermidis ) variable between 20.7 % and 60.0 % after 1 h of ionization, which is consistent with the results obtained in the present study. The resulting loss rate constant is 0.018 per minute under ionization conditions, and the estimated CADR was 5.153 m 3 /min. Regarding the control values, it is observed that the clean air delivery rate increases, which implies a greater amount of air free of bioaerosols. Specifically, this value increased in >17 times. 3.1.4 Filtration reduces aerial CFUs The electrostatic potential influence of the surface on the deposition of bacteria from the air has been demonstrated ( ). Various studies suggest that bipolar ionization reduces particulate matter in the air due to increased particle deposition and/or filtration efficiency associated with the increase in the aerodynamic diameter of aerosols due to the agglomeration of fine particles ( ; ; ). This phenomenon is mainly associated with electrostatic effects ( ; ) and is dependent on particle size and composition, relative humidity, ionization time, and surface material ( ). Filters can interfere in estimating the efficiency of the ionization system. Then, the effect of the filtration implemented in the air conditioning system of the Tram has been characterized separately. The tests were carried out with the filter usually installed on the Tram (Coarse 45 % filter medium). The Air Changes per Hour (ACH) of the Zaragoza Tram remained as they are in the normal operation of the unit at 25 ACH. This indicates that the air passes through the filter 12.5 times/h on average. As seen in , loss rate constant was 0.033 per minute, and the estimated CADR was 9.261 m 3 /min. These parameters imply that filtration is more effective than ionization. CFU amount upon filtration suggests a reduced efficiency considering the number of air changes inside the tram unit. On average, CFUs were reduced by 73.4 % (26.6 % persisted), 84.0 % (16.0 % persisted), and 92.0 % (8.0 % persisted) during the first 30, 60, and 90 min, respectively. In this sense, the CFUs are large enough to favor the filtration mechanisms since they are, for the most part, >1 μm. Faced with submicron particles, smaller in size than the bacteria tested, the efficiency of the filter is limited according to the conditions studied ( ). The filter complies with the UNE-EN 16890 standard and refers to a 75 % filtration for particles greater than ~10 μm. This filter is efficient for the retention of dust or pollen. However, it is inefficient against fine particles and viruses. The tests carried out are limited. Due to the conditions of the equipment used for the determination of filtration efficiency against submicron particles, the filter may have been ‘clogged’ by the concentration of NaCl particles, assuming a notable increase in pressure drop and an ornament in the efficiency result. The Coarse 45 % filter was evaluated in a more compact format than natural: 2—3 mm thick instead of 20—30 mm. The results suggest that the speed at which the air passes through the filter medium retains the larger particles. However, no notable difference is observed in the flows studied. The curves could not be very representative of this filter's actual efficiency (in working conditions in the air conditioning system of the Tram). In experimental sampling tests, SARS-CoV-2 viral RNA has been mainly found in particle sizes from 0.25 μm ( ; ). Therefore, determining the efficiency of the filter against submicronic sizes and above is relevant to quantify its performance. As shown in , the Coarse 45 % filter presented an approximate retention efficiency of ~30 % for 0.5 μm particles at a flow rate of ~4079 m 3 /h. For flow rates lower than this, the efficiency of the filter decreases to values <7 %. These tests suggest that larger particles have more inertia at high speeds (high flow rates) and, therefore, a higher retention rate in the filter medium ( , ). However, the clogging of NaCl particles observed in the head loss tests may have overestimated these results. In the tests where the highest speed is simulated, clogging is observed, which translates into an increase up to 380 Pa of pressure drop ( ). 3.1.5 Effect of combined ionization and filtration efficiency against aerial CFUs Once the effects of ionization and filtration were evaluated separately, the combination's performance was intended to study. However, no clear advantage was obtained from the combination of mechanisms. The first 30 min reduced the CFU concentration by 74.2 % (25.8 % persisted), while at 60 min, the efficiency increased to 82.8 % (17.2 % persisted), and at 90 min, 94.1 % was eliminated (5.9 % persisted) ( ). While these data are promising, the filtration alone reduced the CFU concentration by 73.4 % and 92.0 %, respectively, after 30 and 90 min. The maintenance of filtration efficiency can be explained because the highest rate of particle agglomeration due to electrostatic phenomena occurs in the finest particles (<0.15 μm). Then, there is a higher concentration of particles >0.3 μm ( ), which does not favor the performance of filtration ( , ). The loss rate constant was 0.047 per minute, and the estimated CADR was 13.208 m 3 /min. The value of CADR of ionization and filtration together was higher than filtration and ionization separately by 3.947 and 8.055 m 3 /min, respectively.
CFU stability in the environmental air Airborne microorganisms are significantly affected by weather and environmental conditions ( ; ; ). To guarantee a stable concentration of environmental bacteria during the tests, air samples were taken every 30 min without filtration and ionization. As depicted in , the results of this test suggest that the environmental microbiota is stable over time during the study hours on the same day. An average of 40.4 ± 1.5 CFU/m 3 of air was obtained during the first test and 53.8 ± 3.0 CFU/m 3 during the second test. CFU counts were averaged using all five replicates of each sample. In the plates at 30 and 90 min from the first test, 40.0 CFU/m 3 were counted in each one with respect to the initial 42.5 CFU/m 3 . In the plates of the second test, 52.5 and 53.1 CFU/m 3 were found at 30 and 90 min, compared to the initial 57.5 CFU/m 3 . To determine the stability of the CFUs, the Relative Standard Deviation (RSD) of each set of plates was calculated. The obtained RSD of ~0.04 in the first test and ~0.06 in the second test suggests that the bacterial environmental contamination is stable in air during the 90 min of sampling, so that the rest of the tests presented below are performed under robust and reproducible conditions throughout each test. Loss-rate regression of CFU has been fitted to a polynomial to characterize stability ( c). The resulting CADR of 0.299 m 3 /min and the loss rate constant of 0.0012 per minute during the stability control condition reinforces the previous conclusion about the stability of environmental CFUs.
Bacterial species identification As enumerated in , the 15 most frequent colonies were isolated and identified. The bacterial identification is of interest since the efficiency of the BIU is evaluated predominantly for this subset of strains, despite the existence of others in a smaller proportion. In addition, its visualization allowed to evaluate its morphology and size, which is relevant for filtration studies. As depicted in , most bacterial strain (12/15) have sizes >1 μm in its longest dimension. The measured dimension represents the minimum size in which each strain of CFUs can be found in the air.
Bipolar ionization reduces aerial CFUs The effect of ionization was considered exclusively to evaluate the single efficiency of the bipolar ionization unit. The objective of this study was to isolate the efficiency of the ion system and quantify the improvement that it supposes by itself in the absence of other perturbations, such as the filtration of the air conditioning system. An increasing efficiency was observed from the initial sampling to the final samplings. Ion concentration varied (19.9 · 10 9 —31.5 · 10 9 ions/m 3 ) in all the samplings carried out due to the heterogeneity in the air distribution at the different points of the Tram. On average, during the first 30 min of the experiment, CFUs were reduced by 45.9 % (54.1 % persisted). As shown in , an efficiency of 61.8 % (38.2 % persisted) was estimated after 60 min. The efficiency increased to 69.2 % (30.8 % persisted) after 90 min. evaluated the efficiency of needle-point bipolar ionization systems (~10—20 kV) in airplanes, obtaining a reduction of CFU ( Staphylococcus epidermidis ) variable between 20.7 % and 60.0 % after 1 h of ionization, which is consistent with the results obtained in the present study. The resulting loss rate constant is 0.018 per minute under ionization conditions, and the estimated CADR was 5.153 m 3 /min. Regarding the control values, it is observed that the clean air delivery rate increases, which implies a greater amount of air free of bioaerosols. Specifically, this value increased in >17 times.
Filtration reduces aerial CFUs The electrostatic potential influence of the surface on the deposition of bacteria from the air has been demonstrated ( ). Various studies suggest that bipolar ionization reduces particulate matter in the air due to increased particle deposition and/or filtration efficiency associated with the increase in the aerodynamic diameter of aerosols due to the agglomeration of fine particles ( ; ; ). This phenomenon is mainly associated with electrostatic effects ( ; ) and is dependent on particle size and composition, relative humidity, ionization time, and surface material ( ). Filters can interfere in estimating the efficiency of the ionization system. Then, the effect of the filtration implemented in the air conditioning system of the Tram has been characterized separately. The tests were carried out with the filter usually installed on the Tram (Coarse 45 % filter medium). The Air Changes per Hour (ACH) of the Zaragoza Tram remained as they are in the normal operation of the unit at 25 ACH. This indicates that the air passes through the filter 12.5 times/h on average. As seen in , loss rate constant was 0.033 per minute, and the estimated CADR was 9.261 m 3 /min. These parameters imply that filtration is more effective than ionization. CFU amount upon filtration suggests a reduced efficiency considering the number of air changes inside the tram unit. On average, CFUs were reduced by 73.4 % (26.6 % persisted), 84.0 % (16.0 % persisted), and 92.0 % (8.0 % persisted) during the first 30, 60, and 90 min, respectively. In this sense, the CFUs are large enough to favor the filtration mechanisms since they are, for the most part, >1 μm. Faced with submicron particles, smaller in size than the bacteria tested, the efficiency of the filter is limited according to the conditions studied ( ). The filter complies with the UNE-EN 16890 standard and refers to a 75 % filtration for particles greater than ~10 μm. This filter is efficient for the retention of dust or pollen. However, it is inefficient against fine particles and viruses. The tests carried out are limited. Due to the conditions of the equipment used for the determination of filtration efficiency against submicron particles, the filter may have been ‘clogged’ by the concentration of NaCl particles, assuming a notable increase in pressure drop and an ornament in the efficiency result. The Coarse 45 % filter was evaluated in a more compact format than natural: 2—3 mm thick instead of 20—30 mm. The results suggest that the speed at which the air passes through the filter medium retains the larger particles. However, no notable difference is observed in the flows studied. The curves could not be very representative of this filter's actual efficiency (in working conditions in the air conditioning system of the Tram). In experimental sampling tests, SARS-CoV-2 viral RNA has been mainly found in particle sizes from 0.25 μm ( ; ). Therefore, determining the efficiency of the filter against submicronic sizes and above is relevant to quantify its performance. As shown in , the Coarse 45 % filter presented an approximate retention efficiency of ~30 % for 0.5 μm particles at a flow rate of ~4079 m 3 /h. For flow rates lower than this, the efficiency of the filter decreases to values <7 %. These tests suggest that larger particles have more inertia at high speeds (high flow rates) and, therefore, a higher retention rate in the filter medium ( , ). However, the clogging of NaCl particles observed in the head loss tests may have overestimated these results. In the tests where the highest speed is simulated, clogging is observed, which translates into an increase up to 380 Pa of pressure drop ( ).
Effect of combined ionization and filtration efficiency against aerial CFUs Once the effects of ionization and filtration were evaluated separately, the combination's performance was intended to study. However, no clear advantage was obtained from the combination of mechanisms. The first 30 min reduced the CFU concentration by 74.2 % (25.8 % persisted), while at 60 min, the efficiency increased to 82.8 % (17.2 % persisted), and at 90 min, 94.1 % was eliminated (5.9 % persisted) ( ). While these data are promising, the filtration alone reduced the CFU concentration by 73.4 % and 92.0 %, respectively, after 30 and 90 min. The maintenance of filtration efficiency can be explained because the highest rate of particle agglomeration due to electrostatic phenomena occurs in the finest particles (<0.15 μm). Then, there is a higher concentration of particles >0.3 μm ( ), which does not favor the performance of filtration ( , ). The loss rate constant was 0.047 per minute, and the estimated CADR was 13.208 m 3 /min. The value of CADR of ionization and filtration together was higher than filtration and ionization separately by 3.947 and 8.055 m 3 /min, respectively.
Effect of the bipolar ionization in the elimination of CFUs on surfaces 3.2.1 Evaluation of the CFUs stability on surfaces To guarantee the uniformity of the tests, four surface samples have been carried out, with a total of at least 60 Rodac PCA plates per test: 30 at t = 0 (CFU i ) and 30 at t = 2 h (CFU f ). Ideally, the CFU i should be equal to the CFU f (ie the percentage of CFUi = CFUf would ideally be 100 %). However, we have considered it relevant to analyze whether there are more or fewer CFU at the end of the experiment to determine if it is a matter of stability or simply variability across the surface (i.e., CFU i < CFU f or CFU i > CFU f ). If the CFUs had deteriorated over time, a trend greater than CFU i > CFU f . This has not been observed, so it is assumed that the CFUs are stable during the time of the experiment (2 h) but there is a non-negligible need between the neighboring test surfaces. To quantify this dispersion, the RSD of each set of plates has been calculated. RSD close to zero suggests high uniformity in the samples. The variability of the results ( ) required an increase in the sample to obtain representative conclusions. 3.2.2 Bipolar ionization efficiency on surface samples On the one hand, “effectiveness” refers to the percentage of PCA plates where final CFU number was lower than the initial (CFU i > CFU f ). That is, in this column the percentage of CFU plates that showed a potential action of ionization has been studied. On the other hand, “efficiency” represents the percentage of “removed” bacteria in these plates potentially due to ionization effect, which is calculated according to Eq. . It is relevant to differentiate between the efficiency for plates with CFU > 50, CFU > 20, and CFU > 10 to avoid distorting the results obtained. However, the global efficiency does not discriminate between values obtained ( ). According to our findings, in this experimental set-up, the ionization did not present a major role on the inactivation of microorganisms on surfaces. The average ionization effectiveness was 46 %, while the average CFU i > CFU f in the stability study was 36 %. We cannot exclude that the difference between initial and final CFUs could be due to effects other than ionization.
Evaluation of the CFUs stability on surfaces To guarantee the uniformity of the tests, four surface samples have been carried out, with a total of at least 60 Rodac PCA plates per test: 30 at t = 0 (CFU i ) and 30 at t = 2 h (CFU f ). Ideally, the CFU i should be equal to the CFU f (ie the percentage of CFUi = CFUf would ideally be 100 %). However, we have considered it relevant to analyze whether there are more or fewer CFU at the end of the experiment to determine if it is a matter of stability or simply variability across the surface (i.e., CFU i < CFU f or CFU i > CFU f ). If the CFUs had deteriorated over time, a trend greater than CFU i > CFU f . This has not been observed, so it is assumed that the CFUs are stable during the time of the experiment (2 h) but there is a non-negligible need between the neighboring test surfaces. To quantify this dispersion, the RSD of each set of plates has been calculated. RSD close to zero suggests high uniformity in the samples. The variability of the results ( ) required an increase in the sample to obtain representative conclusions.
Bipolar ionization efficiency on surface samples On the one hand, “effectiveness” refers to the percentage of PCA plates where final CFU number was lower than the initial (CFU i > CFU f ). That is, in this column the percentage of CFU plates that showed a potential action of ionization has been studied. On the other hand, “efficiency” represents the percentage of “removed” bacteria in these plates potentially due to ionization effect, which is calculated according to Eq. . It is relevant to differentiate between the efficiency for plates with CFU > 50, CFU > 20, and CFU > 10 to avoid distorting the results obtained. However, the global efficiency does not discriminate between values obtained ( ). According to our findings, in this experimental set-up, the ionization did not present a major role on the inactivation of microorganisms on surfaces. The average ionization effectiveness was 46 %, while the average CFU i > CFU f in the stability study was 36 %. We cannot exclude that the difference between initial and final CFUs could be due to effects other than ionization.
Conclusions The increasing understanding of the mechanisms behind COVID-19 infection transmission should drive a shift in the way we address the prevention of the transmission of this disease and other respiratory infections. A higher transmission rate of SARS-CoV-2 has been observed in transport compared to other public and shared spaces. It is required to implement preventive measures for different transportation settings and countries. In previous studies we found that the Zaragoza Tram did not represent a high risk of contagion (for more information see ). Nevertheless, after the pandemic, preventive strategies are relaxed and the potential risk of contagion increases, so it is necessary to find long-term solutions for the improve the quality of the air we breathe. Although masks are very efficient in expelling fewer potentially pathogenic microorganisms into the air, it is a temporary measure that probably will not be used indefinitely. Studies on environmental bioaerosols in air samples suggest that BIU systems have a beneficial effect on eliminating CFUs. The main limitation for the implantation of ionization systems is a possible generation of by-products due to the electrostatic interaction between elements, although this aspect is not discussed here, and it is out of the scope of this work. Under the conditions studied, which are favorable (including closed tram unit and long exposure times), efficiencies close to 69.2 % were obtained after 90 min of ionization. The implemented filtration system offers better results in the same period against microorganisms present in the air (92.0 %). The combined action of both systems slightly improved the filtration performance (94.1 %), not representing a major improvement. According to the literature, the predominance of agglomerates is close to 300 nm. However, the Tram filters tested in this study did not perform well in laboratory analyses against those particle sizes (<10 %). We cannot exclude a synergy of bipolar ionization with other filter types that could retain these agglomerates. The filtration tests assayed in the studied Tram units are not very efficient for submicron particles. The Coarse 45 %-type filters are designed to eliminate dust, pollen, and larger particles. However, considering the number of air changes per hour (~25 ACH, ~1450 m 3 /h) approximately 6 % of the CFUs are retained each time the air passes through the filter leading to a ~75 % efficiency observed during the first half-hour (note that no CFU sources existed during the tests). In addition, the main bacterial families found almost always presented sizes >1 um, so the filtration of submicron matter would be more representative in the case of viruses. A limited filtration of fine particles is observed in the tests carried out with submicron particles, where the retention of 0.1 μm particles was <6 %. Faced with these sizes, HEPA filters are an effective strategy. However, it must be considered that the quality of the filter is as important as its performance (measured in m 3 /h). This is why the implementation of HEPA filters in the HVAC system of the Tram is unthinkable due to its pressure drop. Installation of stand-alone HEPA purification systems to reduce airborne CFUs and viruses could be considered, although this would require separate studies. Through the technical feature recently agreed upon by the American Society of Heating, Refrigerating and Air-Conditioning Engineers (ASHRAE), it was possible to quantify the impact of the technologies (ionization and filtration) on the interior air of the Tram. It was performed using the loss rate value and the CADR parameter. Combined action of ionization and filtration (CADR = 13.208 m 3 /min) present a substantially advantage over untreated air (CADR = 0.299 m 3 /min). Even so, the filtration was also effective on its own (CADR = 9.261 m 3 /min). Exclusive ionization did not have such a notable effect (CADR = 5.153 m 3 /h), although also reduced the aerial CFU concentration. All the assays described here were performed under favorable conditions for air purification since there was no external dissemination of bioaerosols and the air was constantly recirculated. Of note, our findings highlight the importance of filtration and air cleaning in public transport. Importantly, the development and validation of new effective air purification systems may be essential for minimizing the spread of infectious diseases in the future and this field merits further research.
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
|
Teleneuropsychology in the time of COVID-19: The experience of The Australian Epilepsy Project | f0f4a9de-63ac-4d40-91c7-f4400557e15f | 7561524 | Physiology[mh] | Introduction The emergence of COVID-19 has resulted in significant changes in the day-to-day functioning of society the world over. Social distancing policies have had profound effects on people’s day-to-day movements and interpersonal interactions with enormous implications for workplaces, organisations, institutions and social activities. In the health sector this has meant that, wherever possible, clinical consultations have moved from face-to-face to a virtual environment (e.g. telephone, videoconferencing). It also appears that fear of infection is leading to avoidance of the hospital system by individuals who would otherwise seek medical care for conditions such as stroke and cardiac arrest . Thus, one might ask how can the health sector, including clinical research, best continue to operate in this environment? In this communication we describe how a large-scale clinical research project in epilepsy has adapted to the emergence of COVID-19, in recognition of the fact that face-to-face interactions in clinical research will have to be reduced for the foreseeable future. We focus on the use of teleneuropsychology (TeleNP) – the application of audiovisual technologies to enable remote clinical encounters with patients to conduct neuropsychological (NP) assessments – to acquire research-based neuropsychological datasets. While our emphasis is on research-based data collection via teleNP, we also touch upon issues relevant to the clinical application of TeleNP in epilepsy.
The Australian Epilepsy Project The Australian Epilepsy Project (AEP) is a large-scale clinical research project shortlisted for funding by the Medical Research Future Fund of the Australian Government. The vision of the AEP is to develop predictive epilepsy-specific decision support tools for use by clinicians. Machine learning / artificial intelligence (AI) methods will be applied to prospectively acquired neuropsychological, genetics and advanced imaging data obtained from 8000+ adults living with epilepsy, to predict their epilepsy-related two-year outcomes . Sharing of de-identified datasets will further maximise breakthrough opportunities in research. We have commented elsewhere on the role of machine learning/AI in the analysis of such datasets, and in the health sector more broadly, and do not consider this issue further here. The AEP commenced a pilot study in February of 2020 to evaluate recruitment feasibility and participant tolerability of the protocols for collection of neuropsychological and imaging data. The first case of COVID-19 was reported in Australia in late January 2020, and a State of Emergency was declared in Victoria in mid-March, around six weeks into AEP recruitment. Despite the introduction of Government-mandated COVID-related restrictions, the AEP pilot study has been able to continue by switching to the use of TeleNP for all participants.
TeleNP can enable research in the era of COVID-19 Prior to the arrival of COVID-19 in Australia, the AEP Pilot Study relied almost exclusively on face-to-face interactions for its data collection and analysis activities. The institutional and governmental response to COVID-19 in Australia demanded a re-evaluation and adjustment to each of these activities to ensure the safety of all persons involved, while preserving the scientific integrity and health care objectives. All our activities accord with guidance from Federal and State regulatory authorities, Institutional clinical governance, and local human research ethics committee approval. provides a condensed description of our current operating protocol, outlining the key conceptual features. Box 1 Data collection activities of the AEP Pilot Study during COVID-19. Referral • Recruitment occurs through neurologists during routine clinical practice. For the majority of patients this is now via telehealth consultation. Recruitment • Information provision and the establishing of informed consent are conducted via email and telephone. Participant consent is confirmed verbally and documented electronically and in writing by the researcher. Data collection • Cognition: Neuropsychological assessment is performed entirely via TeleNP. This includes a combination of neuropsychologist-supervised oral and computer-assisted testing, in conjunction with purely computer-administered web-based testing. Further detail is provided in the main text. • MRI: MRI is performed in-person, at research-dedicated scanners. All participants are screened for COVID-19 symptoms or risk factors before they attend the premises. On-site, physical distancing strategies, appropriate personal protective equipment use, and cleaning procedures are all applied according to up-dated research facility protocols. Images are transmitted electronically to the hospital radiology departments for standard reporting and clinical use. This approach eliminates the need for participants to physically attend hospital premises for clinical scans and reduces the burden on hospital radiology at a time of increased strain on the hospital system. • While genetics and epilepsy follow-up (e.g. seizure diaries, medications, psychological and quality of life questionnaires, adverse events, health economic data) are not collected in the AEP Pilot Study, in the full project genetic samples will be obtained by blood-draw at a local community pathology provider (typically at the same time as routine clinical blood tests), and epilepsy follow-up will occur via smart device app/web survey and telephone call. Reporting • Data analysis and transfer is performed by research staff accessing secure server platforms remotely via encrypted network connections, enabling this work to be safely performed from home. Key team decision making activities are supported via teleconferencing. Alt-text: Box 1
Selecting epilepsy-relevant tools for TeleNP The most substantive protocol changes have involved the transition to collecting all neuropsychological data via TeleNP. Traditional, face-to-face neuropsychological testing carries elevated COVID-19 risk, both for participants and the neuropsychologist, and is clearly unacceptable from both a community safety and occupational health viewpoint. The examiner and examinee may spend several hours in close proximity, passing materials back and forth (e.g. stimulus materials, response forms), usually in a small enclosed room for privacy. Further, the examinee must also travel to the physical premises for the assessment, which can necessitate additional interpersonal interactions (e.g. public transport, waiting areas). The neuropsychological measures used in our TeleNP protocol are listed in . These measures were selected for their evidence base in epilepsy (as acquired through traditional, face-to-face assessments), and their compatibility with TeleNP administration. The experience gained from the AEP Pilot Study will be used to further empirically refine instrument selection. In our pre−COVID face-to-face protocol we had been administering EpiTrack and the Reaction Time task from the CANTAB. The trail making test (TMT), inhibition task and maze task within EpiTrack cannot be administered via telehealth, and the Reaction Time task is not available via CANTAB Connect (indeed the variability in the hardware possessed by participants would almost certainly preclude accurate measurement of reaction times in any home delivered, web based platform). We include the oral version of the TMT in our telehealth protocol, as a measure comparable to the written TMT . We have also trialled various versions of the Stroop task (Victoria Stroop , Dodrill Stroop ), to use as a measure of inhibition, but have ultimately abandoned it due to insensitivity (Victoria version) and difficulty presenting the stimuli appropriately via videoconference (Dodrill version).
Practical considerations for TeleNP The potential benefits of TeleNP have long been recognised, including convenience, user satisfaction, potential cost-reductions and improved access (geographic; availability of interpreter services ). Nonetheless, the neuropsychological community has not uniformly embraced the adoption of TeleNP necessitated by the emergence of COVID-19. One of the most obvious concerns relates to whether TeleNP departs sufficiently from standardised face-to-face administration to invalidate test results and interpretation. There is accumulating evidence that telehealth delivered neuropsychological assessments can yield reliable and valid evaluations . Since the emergence of COVID-19, a number of journal articles and statements from professional bodies have provided guidelines and experience-based recommendations regarding the use of TeleNP (via position papers , webinars and online resources; see, for instance, the Australian Psychological Society [ https://www.psychology.org.au/Event/21454 ], the International Neuropsychological Society [ https://www.the-ins.org/webinars/ ], the Inter Organizational Practice Committee [ https://iopc.online] ). While the purpose of the present paper is not to provide a comprehensive review of TeleNP, we do provide a brief discussion on some issues relevant to our implementation. The best TeleNP evidence concerns the use of tasks that are predominantly verbal in nature . This encompasses the majority of measures we have selected for use , and includes measures such as paragraph and word list learning tasks , verbal span/working memory tasks (such as digit span) , verbal fluency tasks , and measures of crystallized intelligence (e.g. measures of word reading and vocabulary ). There is also evidence for tasks that rely upon verbal responses to visually presented stimuli, such as visual confrontation naming and visuoperceptual reasoning tasks (e.g. WAIS Matrix Reasoning ). While supported by good evidence, it is worth noting that purely verbal tasks do not guarantee immunity to issues when administered via telehealth. Transient interruptions of the connection can interfere, especially with ‘one shot’ (e.g. digit span) or timed tasks (e.g. oral versions of SDMT and TMT). Our experience to date has been that poor connections are often apparent from the outset, and in many instances can be remedied simply by re-establishing the call or asking other users on the network to minimise their own network usage (e.g. streaming). We have had one participant whose computer microphone proved to be faulty but were able to proceed by using a concurrent telephone call accompanying the computer’s video feed without appreciable lag (a solution, incidentally, that we have employed in clinical practice also). Another participant was unable to establish a videoconference link from home, despite repeated attempts, and ultimately completed their teleNP assessment onsite. More difficult to administer are tasks that require physical interaction with stimuli provided by the examiner (e.g. paper forms, three dimensional blocks). While there is some evidence for administration of such tasks via TeleNP (e.g. Grooved Pegboard ; written version of the Symbol Digit Modalities Test , Complex Figure Copy and Recall Tests , Clock Drawing Test ) we have ultimately elected not to include them for a variety of reasons (impracticalities of providing materials to participants: Grooved Pegboard; poor sensitivity and reliability: Rey Complex Figure ; suitable oral version available: SDMT). In other instances, we opted to retain the conceptual element of a traditional pen-and-paper task, but change the mode of delivery and response (e.g. using the oral versions of the Trail Making Test and Symbol Digit Modalities Test ), though we note that this likely alters what the task is actually measuring (e.g. pen-and-paper versus oral Trail Making Test ). TeleNP is not without its challenges. Familiarity with the required technology – on the part of both the examinee and clinician – can influence the degree of engagement with, and the flow and ease of, the interaction. Indeed, we have found it essential to factor in ∼15 min of initial set up time at the beginning of appointments to ensure participants are able to log onto the videoconference call and that their technology is functioning appropriately (assisting them via phone as necessary). A single neuropsychologist administered the TeleNP assessments for our protocol, with this individual completing multiple supervised practice administrations prior to commencing participant data collection (and reviewing the aforementioned TeleNP webinars provided by the Australian Psychological Association and the International Neuropsychological Society once these were available). These practice sessions were essential to ensuring familiarity with the testing technology and practicalities of administration via telehealth. We have also developed a set of Standard Operating Procedures for telehealth to facilitate the training of new staff as the project expands. The suitability of TeleNP for specific patient groups is an important issue, such as paediatric populations, people with intellectual disability or severe cognitive compromise, and linguistically and culturally diverse groups. Indeed, many of these concerns are also relevant to traditional face-to-face consultations. This complexity has not yet been fully addressed by the field and remains a critical challenge to the broad application of clinical TeleNP. However, the acquisition of uniform test data for machine-learning analysis is a narrower problem, where these issues are partly avoided through assessment of a necessarily more targeted cohort in which TeleNP administration is appropriate. Access to technology is another issue of concern, since not all individuals possess the hardware required to support videoconference-based TeleNP. The use of technology at a local facility (e.g. GP clinic or research site) can increase availability and address issues of social equity, while simultaneously ensuring the quality of technology and connectivity . Indeed, the majority of evidence for telehealth administration comes from studies where examinees are tested via technology at a local research facility . We have made this approach available to participants, in order to improve participant access to the study (we offer free parking for participants who are able to travel to the facility by car and offer taxi vouchers for those who require transport). To date, roughly 20 % of participants have opted to complete their TeleNP testing on-site at the research facility (at the time of their MRI scan); the remainder have completed the TeleNP assessment using their own technology at home. Operational changes we implemented to enable TeleNP are outlined in . TeleNP appointments are conducted via Zoom (using a HIPAA compliant Education account; zoom.us), with the ‘password’ and ‘waiting room’ features enabled. TeleNP sessions are delivered by a qualified clinical neuropsychologist. For tasks using oral or screenshare-based stimulus delivery, responses are recorded by the examiner on original test record forms. The web-delivered, computer-administered CANTAB tests are recorded and scored by the software itself. Throughout the TeleNP session (including CANTAB testing), the examinee remains in audiovisual contact with the examiner, enabling monitoring of behaviours and the occurrence of potential distractions. To date, our TeleNP participants have responded positively, reporting the experience to be smooth and efficient, and appreciating the opportunity to carry out the assessment without leaving their home. Those who have previously undergone face-to-face assessment have noted the telehealth experience to be similar. Box 2 TeleNP within the AEP. Prior to their TeleNP appointment participants are: • Screened to check that they have adequate technology to support TeleNP, defined at minimum as: o Computer or iPad with web cam, microphone and internet connection (smartphones and non-iPad tablets are not suitable, given software requirements of the web-delivered computer-administered testing used in the AEP, see below). o Quiet, distraction free room in which to complete the TeleNP assessment • Emailed a link to a set of electronically-hosted surveys tapping elements of psychosocial functioning germane to epilepsy (e.g. Neurological Disorders Depression in Epilepsy ; Patient Health Questionnaire GAD-7 ; Epilepsy Anxiety Survey Instrument ; QoLiE-31 ; Liverpool Adverse Events Profile ; ABNAS ). The surveys are hosted on a REDCap database server at our institution. • Emailed a telehealth information sheet, along with text describing an ‘agreement to telehealth’ whereby participant and researcher agree that they will not “record, reproduce, publish or make copies of the materials used during the neuropsychology telehealth session” . Participants are advised that their TeleNP session cannot proceed until they confirm acceptance of this agreement. At the beginning of the TeleNP session the neuropsychologist: • Verifies the participant’s identity • Checks the participant’s telehealth technology setup (e.g. microphone and webcam setup and testing; quality of connection; disabling of other apps and notifications; suitability of environment) • Explains what will happen in the event of lost connection (attempt to reconnect; if unable to, will call mobile phone; if no contact within 10 min, session considered aborted and will be rescheduled) • Confirms the participant’s current location and obtains additional contact information in event of emergency (e.g. seizure). We explain that in the event of a seizure, if we cannot reach one of the contacts provided, or if we feel a more urgent response is appropriate, that we will call an ambulance. This information is summarised in a teleNP information sheet provided to all participants in advance of their session. We are yet to have a participant experience a seizure during testing. • Re-iterates terms of agreement to participate in telehealth (e.g. participant will not record or reproduce any materials) Neuropsychological testing is administered via the following methods : 1 via ScreenShare linked to a high resolution document camera: Test of Premorbid Function ; Matrix Reasoning from WASI-II ; Boston Naming Test ; oral Symbol Digit Modalities Test . 2 via oral stimulus delivery: letter and category verbal fluency, reverse digit span, oral Trail Making Test , Rey Auditory Verbal Learning Test , Vocabulary from WASI-II . 3 via web-delivered computer-administered testing: using the CANTAB web platform. While testing via this platform can be completed by the participant in a standalone manner, we have the examiner remain on the videocall throughout, to handle any unanticipated problems that might arise and also to monitor behaviour during the testing. For each task, the neuropsychologist records observations of anything that might invalidate a test (e.g. temporary connection loss; distraction). All data is recorded on response forms coding using a random six digit participant identification code, and then transcribed onto a secure central database (REDCap). Alt-text: Box 2
TeleNP reveals a typical pattern of impairments in epilepsy summarises the cognitive data we have acquired at the time of writing. Given the relatively small sample to date ( n = 29), data have not been subgrouped according to AEP referral type (first unprovoked seizure: n = 17, new diagnosis epilepsy: n = 6, refractory epilepsy: n = 6), or into method of neuropsychological test administration (teleNP-home: n = 18, teleNP-onsite: n = 6; face-to-face: n = 5). Data are expressed as z-scores, calculated relative to normative data. The distributions shown in are as expected for an epilepsy cohort, with reductions in the domains of processing speed, working memory, executive function, language, and anterograde memory . summarises the data for each cognitive task. Performance is considered for the sample as a whole (overall mean z-score, SD, n and result of one sample t -test/Wilcoxon signed rank test [relative to a mean/median z-score = 0, one-sided test], and percentage of cases performing >1.5 SDs below expectation), and also separately for each method of administration: teleNP-at home, teleNP-onsite, face-to-face ( n for each method of administration, result of a one-way ANOVA/Kruskall-Wallis test). The ANOVA/Kruskall-Wallis tests indicate that no cognitive test shows a significant effect of administration method (p > .05, albeit with small group sizes). Collapsed across administration method, one sample t-tests/Wilcoxon signed rank tests confirm a pattern of impairments typical of those seen in epilepsy, with significant (p < 0.05) reductions in processing speed (oral SDMT), executive function (oral TMT B, COWAT), language (BNT, Animal fluency) and anterograde memory (RAVLT measures). Trends (p < 0.08) were also apparent for sustained rapid information processing (CANTAB RVP) and working memory (DSB, CANTAB SWM). Screening of mood and anxiety also confirmed a relatively high proportion of individuals at risk for these disorders: 10 of 29 participants (34 %) were at high risk of mood disorder based on the NDDIE [total score > 15, see reference ]; and between 8 and 11 (28 % and 38 %) were at high risk of anxiety disorder based on the brEASI [total score > 7, see reference ] and GAD7 [total score > 7, see reference ], respectively.
The future of TeleNP in epilepsy research and clinical trials The social distancing requirements stemming from COVID-19 are likely to be with us for a long time. This highlights the importance of expanding the testing options available to neuropsychology, by developing assessment tools explicitly designed for use via telehealth. Such developments would be of great benefit even once the need for social distancing has passed, improving access to neuropsychological services. Existing, evidence-based screening tools already used in epilepsy (e.g. EpiTrack ) could be adapted and validated for delivery via online platforms. In the process, such tools could be extended, ensuring coverage of epilepsy relevant neuropsychological domains and exploiting the response sampling available via a computerised medium. The guiding principle should be to target those domains most affected or important to people with epilepsy , using measures sensitive to the lifetime variability of the condition, from disease onset through introduction of anti-seizure medications (ASMs) to chronic refractoriness and surgery. These domains include: • Anterograde memory: the most common cognitive complaint in epilepsy ; the majority of focal epilepsies affect the temporal lobe . • Executive functions/fluid intelligence: sensitive to ASM effects ; the frontal lobes are frequently involved in focal and genetic epilepsies. • Crystallized intelligence: considered less susceptible to ASM effects ; can be affected by age of epilepsy onset ; provides a measure of cognitive reserve . • Mood: frequently disturbed in people with epilepsy . • Adverse treatment side effects Such screening assessments cannot replace comprehensive assessments and may miss subtle problems for some individuals. However, unlike other existing tools that have been developed for dementia screening, the tools would be validated for epilepsy, be age and education adjusted, address functions most often affected during the course of epilepsy and its treatment, and be useable for the remote assessment of patients unable to attend face-to-face assessment or who would otherwise be lost to follow-up. Ultimately this kind of approach will facilitate large-scale collection of data that would not otherwise be practical using traditional methods.
TeleNP in the clinical management of epilepsy To this point we have emphasised the role of TeleNP in research. What of the role of TeleNP in clinical care? While COVID-19 strains the health systems of many countries around the world, individuals nonetheless continue to experience seizures and associated cognitive and psychological comorbidities. In response to COVID-19 and the rush towards telehealth, the ILAE Neuropsychology Task Force underscored that comprehensive telehealth neuropsychological assessments for epilepsy surgery candidates have not yet been carried out or validated . The Task Force recommends that any surgical candidates proceed to surgery only after a comprehensive, face-to-face, neuropsychological work up, concluding that, “whilst compromise and new ways of working are necessary for urgent neurosurgical procedures, epilepsy surgery should not be conducted as an emergency procedure.” While this position aspires to an ideal, the effects of COVID-19 will in all likelihood be with us for a long time yet, and epilepsy surgery cannot simply cease to occur for the foreseeable future, especially not if the only barrier is failure to carry out a neuropsychological evaluation. In our view, a reasonable position would accommodate TeleNP using epilepsy-relevant, evidence-based instruments, augmented by shorter face-to-face assessments where required to address specific clinical issues. The impetus to move towards telehealth stimulated by COVID-19 should be viewed as an opportunity to expand the reach and breadth of neuropsychology . The clinical neuropsychologist’s role in an epilepsy surgery program extends beyond psychometric documentation. The delivery of counselling, psychoeducation, advocacy, and psychotherapy via telehealth has a solid evidence base , including in epilepsy specific contexts , and is encouraged in recent set of epilepsy-specific consensus recommendations . Speaking from the local perspective, within the Department of Clinical Neuropsychology at Austin Health we already employ telehealth (telephone, videoconferencing using the Coviu platform) routinely in the pre- and post-surgical counselling of patients . Anecdotally, a number of our patients have commented that they feel more comfortable in their home environment, and find it easier to be open about their experiences when communicating through the intermediary of technology. The average age of our surgical cohort is in the mid-30′s representing a generation for whom technology is ‘second-nature’. Approximately 20 % of our patients live outside of metropolitan centres. Further expanding the role of TeleNP would be of great benefit to such a patient demographic, improving timely access to care. For example, it would enable important post-operative evaluations and follow-up without the burden of travel and the attendant costs and psychosocial disruption. These anecdotal benefits are substantiated by our recent experience in Germany using phone or videoconference telemedicine in the counselling of people with epilepsy during the COVID pandemic . Overall 82 % of the 239 adult epilepsy patients participating in the audit were satisfied with their telemedicine experience, with high rates of satisfaction especially for time, comprehensibility, and opportunity to get answers to current questions. The participants considered immediate convenience and shortfall of travel expenses as advantages of telemedicine. Approximately three quarters of participants reported that they would appreciate the opportunity for future telemedical counselling, but at the same consider telemedicine as an add-on service rather than a permanent substitute to visits onsite.
Conclusion COVID-19 has abruptly and dramatically changed the way that society functions, including the operation of the health and medical research sector. Our experience shows that it is possible to continue to perform evidence-based, epilepsy-related neuropsychological research while at the same time fully supporting public health strategies aimed at containing and mitigating the effects of COVID-19. In the event that sustaining such policies into the medium or longer-term is necessary, the strategies adopted by the AEP have positioned it to continue to grow and expand, with no impact on the feasibility, integrity or safety of the project. Indeed, this model of telehealth-based operations provides a template for the healthcare of tomorrow, while decreasing the burden on traditional hospital systems. The challenges posed by COVID-19 are immense, and we must respond swiftly and creatively, where possible converting the adversity of the present into opportunities for the future.
Christoph Helmstaedter has received grants from the EU, travel support by Desitin, honoraries for talks, counselling, and advisory boards by GW pharmaceuticals, Eisai, UCB pharma, and Precisis, as well as license fees by EISAI, UCB pharma and Precisis. The remaining authors have no conflicts of interest.
|
Effect of librarian collaboration on otolaryngology systematic review and meta-analysis quality | 7a756f1f-6e74-4f5e-81d6-942e62241b0d | 11412119 | Otolaryngology[mh] | Systematic reviews, including meta-analyses, have become a hallmark of holistically unifying research. For health disciplines, these studies were first established in the early 1990s by the founding of the Cochrane Collaboration . Despite the increase in quantity over the past few years, systematic review quality and adherence to reporting standards have remained highly variable [ - ]. As the quantity of systematic review publications increased and formalized guidelines were established, the services of medical librarians (also known as health information professionals or medical information specialists) have evolved to encompass and facilitate these studies . Oftentimes, the medical librarian's role in research is assumed to be mainly focused on knowledge organization and access. However, librarians have expertise in conducting literature searches, managing citations, creating data extraction and quality assessment forms, peer-reviewing searches, writing or editing portions of manuscripts, performing statistical analyses, or acting as methodology consultants for research teams . In addition to contributing expertise, librarians spend a considerable amount of time on systematic review tasks and do not always receive recognition for their efforts . For example, a study found that librarians spend an average of 26.9 hours (median 18.5 hours) for a single systematic review . Many organizations that guide best practices for systematic reviews recommend involving librarians in the research process. The National Academies of Sciences, Engineering, and Medicine (formally the Institute of Medicine) recommends working with a librarian or other information specialist to plan and peer review the search strategy . Likewise, the Cochrane Collaboration recommends that review authors seek guidance from a medical librarian on the development and documentation of the search strategy . The Medical Library Association (MLA) released a statement, which was cosigned by the Canadian Health Libraries Association/Association des bibliothèques de la santé du Canada (CHLA/ABSC), advocating for librarian co-authorship on evidence synthesis publications, including guidelines and systematic reviews . A strong and comprehensive systematic review search strategy can ameliorate several types of reporting biases, including publication bias, language bias, citation bias, outcome reporting bias, time-lag bias, and location bias . These recommendations for librarian collaboration on systematic reviews aim to increase adherence to reporting guidelines and improve systematic review search quality. In response to these recommendations, several studies have examined the value of including librarians in the systematic review process. These studies found low rates of librarian acknowledgment or co-authorship, yet involvement of librarians yielded improved search quality, better adherence to reporting standards, and lower risk of bias [ , , , - ]. Several of these studies were limited to certain journals within one or a few medical specialties (e.g., dentistry, cardiology, or pediatrics), and none have examined otolaryngology . Additionally, many of these studies were published before 2019, and numerous systematic reviews were conducted after this time. This study addresses the gap in published literature for otolaryngology researchers and clinicians, provides further justification for the inclusion of librarians on otolaryngology systematic review teams, and contributes evidence of quantifiable changes in search strategy quality when medical librarians are involved. Therefore, our study aims to 1) elucidate the systematic review reporting quality and literature search quality of otolaryngology literature, and 2) investigate the effect of librarian involvement on search quality, search reproducibility, and systematic review reporting in otolaryngology systematic reviews and meta-analyses.
Study Design and Participants For this retrospective cross-sectional study, otolaryngology journals were selected using Journal Citation Reports™ . From the journals in the “OTORHINOLARYNGOLOGY – SCIENCE” category, three researchers (MS, TG, TM) independently reviewed and selected journals based on pre-defined eligibility criteria. Inclusion criteria consisted of English language, clinically focused, otolaryngology specific, and indexed in MEDLINE. The librarians (EB, RW) identified the journals that were indexed in MEDLINE, as these journals passed the rigorous, multi-step, quality control process required by the National Library of Medicine . Journals were excluded if they were non-English language, non-clinically focused, non-otolaryngology specific, and not indexed in MEDLINE. Non-English language articles were excluded due to the lack of funding for translation services or reliable translation software. PubMed was queried to identify systematic reviews and meta-analyses in included otolaryngology journals. To identify trends over time, studies from 2010, 2015, and 2021 were included. Due to the number of articles retrieved, each year was limited to a period of six months, beginning January 1 and ending June 30. Publication dates were determined by using the “Custom Range Publication Date” filter, equivalent to using the [dp] or [pdat] field tags, in PubMed. The full search strategy is shown in . Retrieved articles were uploaded to Covidence systematic review software for screening . Two researchers (MS, TG) independently performed title/abstract screening followed by full-text screening using pre-defined inclusion and exclusion criteria. Articles were included if the article title, abstract, or text indicated the study was a systematic review or meta-analysis; the articles were published in the selected otolaryngology journals; and the article was published between 1/1/10 - 6/30/10, 1/1/15 - 6/30/15, 1/1/21 - 6/30/21. Articles were excluded if they discussed a basic science topic, were non-English language or if the full text was irretrievable. Full-text articles were retrieved via library subscriptions, interlibrary loan, and outreach to authors. Data Collection Two researchers (MS, TG) independently extracted data from the selected articles using a customized data extraction form ( ). The two librarians (EB, RW) provided consensus over any disagreements in the original data extraction process. The following data elements were extracted: journal name, publication type, level of librarian involvement, reporting guideline followed, number of databases searched, dates of database searches, database limits and filters, search peer review by a second librarian, flow diagram inclusion, grey literature searched, and citation searching performed. Journal impact factors were collected for 2010, 2015, and 2021 according to Journal Citations Report™ . If supplemental files containing search strategies were missing from the journal website, corresponding authors were contacted in an attempt to obtain those files. In this study, four types of librarian involvement were identified: no acknowledgment, mentioned in text, acknowledgment, and co-authorship. “No acknowledgment” indicated that a librarian was not mentioned in the text, acknowledgments, or author byline. For “librarian mentioned in the text,” authors specified in the text of the article, normally the methods section, that a librarian assisted with search strategy development. “Librarian acknowledgment” was defined as a formal acknowledgment at the end of a manuscript. The final type, “librarian co-authorship,” means a librarian was identified in the author byline. This determination was made by examining author credentials or degrees, departmental affiliations, or by searching author names in institutional directories. Two librarians (EB, RW) independently rated the reproducibility and quality of the search strategy for each included article. A reproducible search strategy was defined as a search strategy that was sufficiently described and could be replicated in the appropriate database with minimal effort. This would include fully described search strategies, or a combination of features of reproducible search strategies. These features included, but were not limited to, PICO tables, keywords, and Boolean operators. For articles that included at least one reproducible search strategy, six elements were rated: 1) Translation of the research question, 2) Boolean & proximity operators, 3) Subject headings, 4) Text word searching, 5) Spelling, syntax, and line numbers, 6) Limits and filters. These six elements were based on The Peer Review of Electronic Search Strategies (PRESS) checklist . PRESS is a validated structured tool for the peer review of electronic literature search strategies. Each of the six elements is rated “no revisions,” “revisions suggested,” or “revisions required.” For our study, this scale was adapted to a Likert scale, ranging from 1 (low quality) to 3 (high quality) . See for the search quality form. Articles where the search was conducted by an author of this study were blinded and sent to two additional librarians (CA, IL) for quality assessment. Data Analysis All data were analyzed using SPSS v27.0.1 (IBM Corporation, Armonk, NY). For Likert scale ratings on search strategy quality, interrater agreement was assessed via Cohen's Kappa statistic. Level of interrater agreement was classified according to Landis and Koch's criteria . The Likert scale scores provided by each librarian were averaged for analyses. All data were assessed for normality via Shapiro-Wilk Tests. Categorical data were presented as counts (% whole) and compared with Chi-Squared tests. For analyses of two groups vs. two groups, and one grouping had fewer than 10 counts, Fisher's Exact test was used instead of Chi-Squared. Continuous variables were presented as median (25-75% interquartile range) and compared via Mann Whitney U Tests for two groups, and Kruskal-Wallis Tests for three or more groups. Because of the low number of studies from 2010, these studies were not separately analyzed in the comparison of search strategy quality.
For this retrospective cross-sectional study, otolaryngology journals were selected using Journal Citation Reports™ . From the journals in the “OTORHINOLARYNGOLOGY – SCIENCE” category, three researchers (MS, TG, TM) independently reviewed and selected journals based on pre-defined eligibility criteria. Inclusion criteria consisted of English language, clinically focused, otolaryngology specific, and indexed in MEDLINE. The librarians (EB, RW) identified the journals that were indexed in MEDLINE, as these journals passed the rigorous, multi-step, quality control process required by the National Library of Medicine . Journals were excluded if they were non-English language, non-clinically focused, non-otolaryngology specific, and not indexed in MEDLINE. Non-English language articles were excluded due to the lack of funding for translation services or reliable translation software. PubMed was queried to identify systematic reviews and meta-analyses in included otolaryngology journals. To identify trends over time, studies from 2010, 2015, and 2021 were included. Due to the number of articles retrieved, each year was limited to a period of six months, beginning January 1 and ending June 30. Publication dates were determined by using the “Custom Range Publication Date” filter, equivalent to using the [dp] or [pdat] field tags, in PubMed. The full search strategy is shown in . Retrieved articles were uploaded to Covidence systematic review software for screening . Two researchers (MS, TG) independently performed title/abstract screening followed by full-text screening using pre-defined inclusion and exclusion criteria. Articles were included if the article title, abstract, or text indicated the study was a systematic review or meta-analysis; the articles were published in the selected otolaryngology journals; and the article was published between 1/1/10 - 6/30/10, 1/1/15 - 6/30/15, 1/1/21 - 6/30/21. Articles were excluded if they discussed a basic science topic, were non-English language or if the full text was irretrievable. Full-text articles were retrieved via library subscriptions, interlibrary loan, and outreach to authors.
Two researchers (MS, TG) independently extracted data from the selected articles using a customized data extraction form ( ). The two librarians (EB, RW) provided consensus over any disagreements in the original data extraction process. The following data elements were extracted: journal name, publication type, level of librarian involvement, reporting guideline followed, number of databases searched, dates of database searches, database limits and filters, search peer review by a second librarian, flow diagram inclusion, grey literature searched, and citation searching performed. Journal impact factors were collected for 2010, 2015, and 2021 according to Journal Citations Report™ . If supplemental files containing search strategies were missing from the journal website, corresponding authors were contacted in an attempt to obtain those files. In this study, four types of librarian involvement were identified: no acknowledgment, mentioned in text, acknowledgment, and co-authorship. “No acknowledgment” indicated that a librarian was not mentioned in the text, acknowledgments, or author byline. For “librarian mentioned in the text,” authors specified in the text of the article, normally the methods section, that a librarian assisted with search strategy development. “Librarian acknowledgment” was defined as a formal acknowledgment at the end of a manuscript. The final type, “librarian co-authorship,” means a librarian was identified in the author byline. This determination was made by examining author credentials or degrees, departmental affiliations, or by searching author names in institutional directories. Two librarians (EB, RW) independently rated the reproducibility and quality of the search strategy for each included article. A reproducible search strategy was defined as a search strategy that was sufficiently described and could be replicated in the appropriate database with minimal effort. This would include fully described search strategies, or a combination of features of reproducible search strategies. These features included, but were not limited to, PICO tables, keywords, and Boolean operators. For articles that included at least one reproducible search strategy, six elements were rated: 1) Translation of the research question, 2) Boolean & proximity operators, 3) Subject headings, 4) Text word searching, 5) Spelling, syntax, and line numbers, 6) Limits and filters. These six elements were based on The Peer Review of Electronic Search Strategies (PRESS) checklist . PRESS is a validated structured tool for the peer review of electronic literature search strategies. Each of the six elements is rated “no revisions,” “revisions suggested,” or “revisions required.” For our study, this scale was adapted to a Likert scale, ranging from 1 (low quality) to 3 (high quality) . See for the search quality form. Articles where the search was conducted by an author of this study were blinded and sent to two additional librarians (CA, IL) for quality assessment.
All data were analyzed using SPSS v27.0.1 (IBM Corporation, Armonk, NY). For Likert scale ratings on search strategy quality, interrater agreement was assessed via Cohen's Kappa statistic. Level of interrater agreement was classified according to Landis and Koch's criteria . The Likert scale scores provided by each librarian were averaged for analyses. All data were assessed for normality via Shapiro-Wilk Tests. Categorical data were presented as counts (% whole) and compared with Chi-Squared tests. For analyses of two groups vs. two groups, and one grouping had fewer than 10 counts, Fisher's Exact test was used instead of Chi-Squared. Continuous variables were presented as median (25-75% interquartile range) and compared via Mann Whitney U Tests for two groups, and Kruskal-Wallis Tests for three or more groups. Because of the low number of studies from 2010, these studies were not separately analyzed in the comparison of search strategy quality.
The flow diagrams for journal and article inclusion are shown in . Of 59 journals, 33 were included for article retrieval from PubMed. Of 559 articles retrieved, 505 were included for data extraction and analysis. All data collected did not exhibit normality. Temporal Changes in Reporting of Systematic Reviews and Meta-Analyses compares the reporting quality of systematic reviews and meta-analyses by year. Significantly more librarians were co-authors in 2021 (n=34, 9.8%) compared to 2015 (n=2, 1.5%) and 2010 (n=0, 0.0%) (p=0.04). Conversely, significantly fewer studies were unclear or did not mention librarian involvement in 2021 (n=276, 79.3%) compared to 2015 (n=121, 89.6%) and 2010 (n=20, 90.9%) (p=0.04). Systematic review and meta-analysis reporting quality improved in 2021 compared to prior years in using a reporting tool (p < 0.001), number of databases queried (p < 0.001), describing the date of database searches (p < 0.001), and including a flow diagram (p < 0.001). No significant difference was seen for mentioning a peer reviewer for search strategies, searching grey literature, performing citation searching, and providing reproducible search strategies. The Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) reporting checklist was the most frequently used tool. Librarian Involvement and Study Reporting Quality compares the reporting of systematic reviews and meta-analyses with vs. without librarian involvement. There were statistically significant differences with regards to using a reporting tool (p < 0.001), number of databases queried (p < 0.001), describing the date of database search (p = 0.002), mentioning of a search strategy peer reviewer (p = 0.02), and reproducibility of search strategies (p < 0.001). No significant difference was seen for querying at least three databases, describing limits/filters, including a flow diagram, searching grey literature, and performing citation searching. When comparing librarians as co-authors vs. librarians involved without co-authorship, the only statistically significant difference seen was that studies involving librarian coauthors more frequently reported grey literature searching with details provided. Librarian Involvement and Search Strategy Quality shows the descriptive statistics and interrater agreement for Likert scale ratings of search strategy quality. The greatest agreement was in “subject headings” (Kappa value 0.90 [0.89–0.94]), and the weakest was seen in “spelling/syntax/line numbers” (Kappa value 0.47 [0.38–0.56]). shows differences in search strategy quality grouped by levels of librarian involvement. Librarian involvement was associated with significant improvements in “Boolean & proximity operators” (p=0.004), “subject headings” (p < 0.001), “text word searching” (p < 0.001), and “spelling/syntax/line numbers” (p < 0.001). When comparing librarians as co-authors to librarian involvement without co-authorship, there were no statistically significant differences in search quality. However, trends toward significance were seen, such as in “text word searching” (p=0.06). Analyses examining studies published in 2021 found improvements in search quality that mirrored analyses examining all years aggregately. These improvements again included “Boolean & proximity operators” (p=0.01), “subject headings” (p < 0.001), “text word searching” (p < 0.001), and “spelling/syntax/line numbers” (p = 0.002). When comparing librarians as co-authors to librarian involvement without co-authorship, statistically significant difference was seen with librarian co-authors for “text word searching” (p = 0.03). shows box plots illustrating the quality of search strategies comparing any librarian involvement (regardless of co-authorship) versus no librarian involvement. Librarian Involvement and Publication Metrics shows differences in Journal Impact Factor when comparing different levels of librarian involvement. Journal impact factors were higher for articles with librarian involvement in 2015 (p = 0.003) and 2021 (p < 0.001). Impact factors were not higher with for articles published with librarian co-authors as compared to librarian involvement without co-authorship (2010 p = 0.03, 2015 p = 0.3, 2021 p = 0.9).
compares the reporting quality of systematic reviews and meta-analyses by year. Significantly more librarians were co-authors in 2021 (n=34, 9.8%) compared to 2015 (n=2, 1.5%) and 2010 (n=0, 0.0%) (p=0.04). Conversely, significantly fewer studies were unclear or did not mention librarian involvement in 2021 (n=276, 79.3%) compared to 2015 (n=121, 89.6%) and 2010 (n=20, 90.9%) (p=0.04). Systematic review and meta-analysis reporting quality improved in 2021 compared to prior years in using a reporting tool (p < 0.001), number of databases queried (p < 0.001), describing the date of database searches (p < 0.001), and including a flow diagram (p < 0.001). No significant difference was seen for mentioning a peer reviewer for search strategies, searching grey literature, performing citation searching, and providing reproducible search strategies. The Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) reporting checklist was the most frequently used tool.
compares the reporting of systematic reviews and meta-analyses with vs. without librarian involvement. There were statistically significant differences with regards to using a reporting tool (p < 0.001), number of databases queried (p < 0.001), describing the date of database search (p = 0.002), mentioning of a search strategy peer reviewer (p = 0.02), and reproducibility of search strategies (p < 0.001). No significant difference was seen for querying at least three databases, describing limits/filters, including a flow diagram, searching grey literature, and performing citation searching. When comparing librarians as co-authors vs. librarians involved without co-authorship, the only statistically significant difference seen was that studies involving librarian coauthors more frequently reported grey literature searching with details provided.
shows the descriptive statistics and interrater agreement for Likert scale ratings of search strategy quality. The greatest agreement was in “subject headings” (Kappa value 0.90 [0.89–0.94]), and the weakest was seen in “spelling/syntax/line numbers” (Kappa value 0.47 [0.38–0.56]). shows differences in search strategy quality grouped by levels of librarian involvement. Librarian involvement was associated with significant improvements in “Boolean & proximity operators” (p=0.004), “subject headings” (p < 0.001), “text word searching” (p < 0.001), and “spelling/syntax/line numbers” (p < 0.001). When comparing librarians as co-authors to librarian involvement without co-authorship, there were no statistically significant differences in search quality. However, trends toward significance were seen, such as in “text word searching” (p=0.06). Analyses examining studies published in 2021 found improvements in search quality that mirrored analyses examining all years aggregately. These improvements again included “Boolean & proximity operators” (p=0.01), “subject headings” (p < 0.001), “text word searching” (p < 0.001), and “spelling/syntax/line numbers” (p = 0.002). When comparing librarians as co-authors to librarian involvement without co-authorship, statistically significant difference was seen with librarian co-authors for “text word searching” (p = 0.03). shows box plots illustrating the quality of search strategies comparing any librarian involvement (regardless of co-authorship) versus no librarian involvement. Librarian Involvement and Publication Metrics shows differences in Journal Impact Factor when comparing different levels of librarian involvement. Journal impact factors were higher for articles with librarian involvement in 2015 (p = 0.003) and 2021 (p < 0.001). Impact factors were not higher with for articles published with librarian co-authors as compared to librarian involvement without co-authorship (2010 p = 0.03, 2015 p = 0.3, 2021 p = 0.9).
Our study provided the first investigation of librarian collaboration and temporal changes in otolaryngology systematic reviews and meta-analyses. In comparing studies published in 2010, 2015, and 2021, there were significant improvements in adherence to reporting standards. However, several deficits in consistent reporting were still noted in 2021. There was significantly more collaboration with librarians, which could in part account for some improvements noted between years. Librarian involvement was associated with several statistically significant improvements in reporting quality of systematic reviews, as well as search strategy quality. Studies involving librarians were generally published in journals with higher impact factors. There was no statistically significant difference in journal impact factors between studies published with librarian co-authors compared to studies with librarians involved but not co-authors. However, some p-values approached but did not reach 0.05, which suggested that additional data and greater power could have led to statistically significant differences in journal impact factors between studies with librarians involved as co-authors and those with librarians in other roles. Reporting Quality of Systematic Reviews in Otolaryngology Reporting tools were designed to facilitate transparent, complete, and accurate reporting of systematic reviews and meta-analyses. shows that the use of a reporting tool increased from 28.9% in 2015 to 82.5% in 2021. Despite this increase, our study found significant discrepancies in proper adherence. Based on , adherence to the PRISMA checklist and PRISMA for Searching (PRISMA-S) extension was inadequate in 2021, as 36.8% did not provide a reproducible search strategy, and 52.0% did not provide the exact date that database searching was conducted. Some areas of reporting were adequately addressed by 2021, including describing limits and filters (97.1%) and using a flow diagram (96.0%). However, 61.2% did not report whether they searched grey literature, 37.6% did not report whether they performed citation searching, and 20.4% did not query at least 3 databases. Another consideration when comparing the reporting quality in systematic reviews is the major revision that was made to PRISMA in 2020 . Articles published in 2015 would have utilized the PRISMA 2009 reporting guidelines, while articles published in 2021 may have used either the original 2009 guidelines, or the revisions released in 2020. The major revisions to PRISMA in 2020 should not have impacted whether a systematic review reported the utilization of a reporting tool. The reporting recommendations by PRISMA and the Cochrane Handbook for Systematic Reviews of Interventions provide structured guidance on the methodology and reporting of comprehensive literature searches [ , - ]. As such, lack of adherence to reporting tools in systematic reviews and meta-analyses may result in the omission of potentially relevant articles due to a lower quality search strategy. Regardless of librarian involvement, the conduct and reporting standards of systematic reviews in otolaryngology could benefit from stricter publication criteria and pre-publication screening. Librarian Collaboration and Reporting Quality Concordant with the literature for other medical specialties, our study found that otolaryngology systematic reviews with better reporting quality were associated with librarian involvement. Several studies have demonstrated the benefits of including librarians in systematic review teams. Specifically, librarian collaboration was associated with a higher likelihood of: searching at least three databases, providing at least one reproducible search strategy, including more search terms in search strategies, better reporting scores in methodology sections, better search quality, presenting flow diagrams, and searching grey literature [ , , , , ]. However, our study found that librarians were only involved in 20.7% of otolaryngology systematic reviews published in 2021, and less than half of these (9.8%) included librarian co-authorship. This low level of involvement could be attributed to lack of recognition towards librarian contributions, sometimes referred to as invisible labor . One potential influence on systematic review search quality were the revisions to PRISMA in 2020, specifically whether a reproducible search strategy was provided for more than one database . This was because the original 2009 PRISMA guidelines only required reviews to “present the full electronic search strategy for at least one database” while the 2020 update required reviews to “present the full search strategies for all databases, registers, and websites” . Considering that librarian services are not uniformly available for researchers, mandating the inclusion of librarians in systematic reviews would likely widen disparities in academic publishing. However, librarians have the expertise to peer review search methodology, and their inclusion on peer review teams would help to ensure adherence to reporting tools and reduce the risk of bias. Otolaryngology journals could consider incorporating librarians as reviewers of systematic reviews to further ensure scientific reporting integrity. This suggestion is further supported by a study indicating a high level of interest by medical librarians to serve as peer reviewers for academic journals which led to the creation of a Librarian Peer Review Database . Librarian Collaboration and Search Strategy Quality To quantify search strategy quality, our study adapted the PRESS checklist. Current methods of grading search strategy quality are affected by a grader's expertise and skill. Furthermore, a level of subjectivity is introduced by using the PRESS checklist . Additional studies are needed to better develop objective and quantitative search strategy grading tools. Our study demonstrated moderate to substantial interrater agreement which was determined to be sufficient for further analyses. Concordant with the literature, our study found that librarian collaboration was associated with improved search strategy quality . With librarian involvement the median score almost always was “3,” which indicated “no revisions necessary.” In contrast, most studies without librarian involvement required significant revisions for “subject headings,” and revisions suggested for “text word searching” and “spelling/syntax/line numbers.” Furthermore, conclusions were consistent when examining only studies published in 2021. Altogether, our study again supports the incorporation of librarians on systematic review and journal editing teams. We initially hoped to examine the value of peer reviewing search strategies and its association with search strategy quality. However, analysis was not appropriate considering the inadequate number of articles indicating search strategy peer reviewing. Instead, we used librarian co-authorship as a surrogate indicator for greater involvement and investment in search strategy development. Previous literature has noted that librarians as co-authors are associated with improved search strategy quality compared to librarians only mentioned in the text or acknowledged . For our study, search strategy quality mostly did not differ between studies involving librarians as co-authors vs librarians without co-authorship. However, grading scores were generally higher for studies involving librarian co-authors, and a statistically significant increase was seen for studies published in 2021 for “text word searching.” These trends suggest that additional data and greater power could lead to more statistically significant differences. Librarian Collaboration and Publication Metrics Our study found that systematic reviews published in 2021 with librarian involvement were associated with publication in otolaryngology journals with higher impact factors. There was no statistically significant association between librarian involvement and high impact factor journals for the systematic reviews published in 2015, but this may be because the study was underpowered. It must be noted that impact factors are not synonymous with the prestige or reputation of a journal. Additionally, access to librarian services varies between researchers, and thus the observed differences may be due to other factors related to resource availability. As our study found that librarian involvement in systematic reviews was associated with higher search quality, this finding may indicate that librarian collaboration may be associated with publication acceptance in a higher impact journal but further research to confirm this is required. Nonetheless, our study supports the collaboration with librarians for systematic reviews and meta-analyses when possible. Study Limitations Because of our strict inclusion and exclusion criteria, our study findings are not generalizable to other databases, languages, or non-clinically focused articles. Non-English language articles were not included in our study and may have excluded additional systematic reviews involving librarian involvement. Additionally, a very specific strategy was used in PubMed to identify systematic reviews and meta-analyses. This method may have excluded articles that were not yet indexed in MeSH or did not self-identify in their title. This study was dependent on the information published in the articles we reviewed. If a librarian created a search strategy but that information was not stated in the article, the article would have been miscategorized. Librarians are not always credited as authors even if their contributions are in accordance with the International Health Library Associations to International Committee of Medical Journal Editors (ICMJE) authorship criteria . These types of omissions could have led to the underestimating the level of librarian collaboration and the influence of librarian contributions on otolaryngology systematic reviews. It is important to acknowledge that the PRISMA reporting guidelines originally published in 2009 were revised and updated in 2020 . Updates to the checklist included changes to incorporate more inclusive language, clarifying wording, and requiring a full search strategy for all databases . These changes may have impacted the number of reproducible searches in 2021 to be higher than in previous years. Additionally, the PRESS checklist was published in 2016 and may have indirectly led to an improvement in search quality as institutions may have utilized an internal librarian peer review process that was not mentioned in the text . Many of our statistical comparisons did not show statistical significance. It is important to note that our study did not conduct a power analysis, and thus the minimum number of studies that had to be examined to achieve statistical significance was not pre-determined. As such, a lack of statistical significance may not mean that the relationship does not exist, but rather that another study of greater power may be needed.
Reporting tools were designed to facilitate transparent, complete, and accurate reporting of systematic reviews and meta-analyses. shows that the use of a reporting tool increased from 28.9% in 2015 to 82.5% in 2021. Despite this increase, our study found significant discrepancies in proper adherence. Based on , adherence to the PRISMA checklist and PRISMA for Searching (PRISMA-S) extension was inadequate in 2021, as 36.8% did not provide a reproducible search strategy, and 52.0% did not provide the exact date that database searching was conducted. Some areas of reporting were adequately addressed by 2021, including describing limits and filters (97.1%) and using a flow diagram (96.0%). However, 61.2% did not report whether they searched grey literature, 37.6% did not report whether they performed citation searching, and 20.4% did not query at least 3 databases. Another consideration when comparing the reporting quality in systematic reviews is the major revision that was made to PRISMA in 2020 . Articles published in 2015 would have utilized the PRISMA 2009 reporting guidelines, while articles published in 2021 may have used either the original 2009 guidelines, or the revisions released in 2020. The major revisions to PRISMA in 2020 should not have impacted whether a systematic review reported the utilization of a reporting tool. The reporting recommendations by PRISMA and the Cochrane Handbook for Systematic Reviews of Interventions provide structured guidance on the methodology and reporting of comprehensive literature searches [ , - ]. As such, lack of adherence to reporting tools in systematic reviews and meta-analyses may result in the omission of potentially relevant articles due to a lower quality search strategy. Regardless of librarian involvement, the conduct and reporting standards of systematic reviews in otolaryngology could benefit from stricter publication criteria and pre-publication screening.
Concordant with the literature for other medical specialties, our study found that otolaryngology systematic reviews with better reporting quality were associated with librarian involvement. Several studies have demonstrated the benefits of including librarians in systematic review teams. Specifically, librarian collaboration was associated with a higher likelihood of: searching at least three databases, providing at least one reproducible search strategy, including more search terms in search strategies, better reporting scores in methodology sections, better search quality, presenting flow diagrams, and searching grey literature [ , , , , ]. However, our study found that librarians were only involved in 20.7% of otolaryngology systematic reviews published in 2021, and less than half of these (9.8%) included librarian co-authorship. This low level of involvement could be attributed to lack of recognition towards librarian contributions, sometimes referred to as invisible labor . One potential influence on systematic review search quality were the revisions to PRISMA in 2020, specifically whether a reproducible search strategy was provided for more than one database . This was because the original 2009 PRISMA guidelines only required reviews to “present the full electronic search strategy for at least one database” while the 2020 update required reviews to “present the full search strategies for all databases, registers, and websites” . Considering that librarian services are not uniformly available for researchers, mandating the inclusion of librarians in systematic reviews would likely widen disparities in academic publishing. However, librarians have the expertise to peer review search methodology, and their inclusion on peer review teams would help to ensure adherence to reporting tools and reduce the risk of bias. Otolaryngology journals could consider incorporating librarians as reviewers of systematic reviews to further ensure scientific reporting integrity. This suggestion is further supported by a study indicating a high level of interest by medical librarians to serve as peer reviewers for academic journals which led to the creation of a Librarian Peer Review Database .
To quantify search strategy quality, our study adapted the PRESS checklist. Current methods of grading search strategy quality are affected by a grader's expertise and skill. Furthermore, a level of subjectivity is introduced by using the PRESS checklist . Additional studies are needed to better develop objective and quantitative search strategy grading tools. Our study demonstrated moderate to substantial interrater agreement which was determined to be sufficient for further analyses. Concordant with the literature, our study found that librarian collaboration was associated with improved search strategy quality . With librarian involvement the median score almost always was “3,” which indicated “no revisions necessary.” In contrast, most studies without librarian involvement required significant revisions for “subject headings,” and revisions suggested for “text word searching” and “spelling/syntax/line numbers.” Furthermore, conclusions were consistent when examining only studies published in 2021. Altogether, our study again supports the incorporation of librarians on systematic review and journal editing teams. We initially hoped to examine the value of peer reviewing search strategies and its association with search strategy quality. However, analysis was not appropriate considering the inadequate number of articles indicating search strategy peer reviewing. Instead, we used librarian co-authorship as a surrogate indicator for greater involvement and investment in search strategy development. Previous literature has noted that librarians as co-authors are associated with improved search strategy quality compared to librarians only mentioned in the text or acknowledged . For our study, search strategy quality mostly did not differ between studies involving librarians as co-authors vs librarians without co-authorship. However, grading scores were generally higher for studies involving librarian co-authors, and a statistically significant increase was seen for studies published in 2021 for “text word searching.” These trends suggest that additional data and greater power could lead to more statistically significant differences.
Our study found that systematic reviews published in 2021 with librarian involvement were associated with publication in otolaryngology journals with higher impact factors. There was no statistically significant association between librarian involvement and high impact factor journals for the systematic reviews published in 2015, but this may be because the study was underpowered. It must be noted that impact factors are not synonymous with the prestige or reputation of a journal. Additionally, access to librarian services varies between researchers, and thus the observed differences may be due to other factors related to resource availability. As our study found that librarian involvement in systematic reviews was associated with higher search quality, this finding may indicate that librarian collaboration may be associated with publication acceptance in a higher impact journal but further research to confirm this is required. Nonetheless, our study supports the collaboration with librarians for systematic reviews and meta-analyses when possible.
Because of our strict inclusion and exclusion criteria, our study findings are not generalizable to other databases, languages, or non-clinically focused articles. Non-English language articles were not included in our study and may have excluded additional systematic reviews involving librarian involvement. Additionally, a very specific strategy was used in PubMed to identify systematic reviews and meta-analyses. This method may have excluded articles that were not yet indexed in MeSH or did not self-identify in their title. This study was dependent on the information published in the articles we reviewed. If a librarian created a search strategy but that information was not stated in the article, the article would have been miscategorized. Librarians are not always credited as authors even if their contributions are in accordance with the International Health Library Associations to International Committee of Medical Journal Editors (ICMJE) authorship criteria . These types of omissions could have led to the underestimating the level of librarian collaboration and the influence of librarian contributions on otolaryngology systematic reviews. It is important to acknowledge that the PRISMA reporting guidelines originally published in 2009 were revised and updated in 2020 . Updates to the checklist included changes to incorporate more inclusive language, clarifying wording, and requiring a full search strategy for all databases . These changes may have impacted the number of reproducible searches in 2021 to be higher than in previous years. Additionally, the PRESS checklist was published in 2016 and may have indirectly led to an improvement in search quality as institutions may have utilized an internal librarian peer review process that was not mentioned in the text . Many of our statistical comparisons did not show statistical significance. It is important to note that our study did not conduct a power analysis, and thus the minimum number of studies that had to be examined to achieve statistical significance was not pre-determined. As such, a lack of statistical significance may not mean that the relationship does not exist, but rather that another study of greater power may be needed.
Our study provided the first investigation of temporal changes and librarian activity in otolaryngology systematic reviews and meta-analyses. Despite the frequent indication of using reporting tools, several deficits in adequate reporting were still noted in 2021. Librarian collaboration remains sparse in otolaryngology systematic reviews and meta-analyses. However, librarian involvement was associated with improved reporting quality and search strategy quality. Studies involving librarians were also published in journals with higher impact factors. As the landscape begins to shift towards embracing librarian involvement on systematic reviews through the support of leading systematic review entities (e.g., Cochrane and Johanna Briggs Institute) and national organizations (e.g., MLA and CHLA/ABSC), we are hopeful that librarians will be invited to systematic review teams and as a part of the journal peer review process. The publication and growing awareness of additional structured guidance on systematic reviews, such as the PRISMA-S extension and the validated PRESS checklist, provides an opportunity to further increase search quality and reproducibility. Future research should include studies more directly examining the quality of recent systematic reviews with librarian co-authors compared to librarian involvement without co-authorship. Additionally, similar studies of systematic review quality and librarian involvement are needed in other disciplines.
|
Perioperative Echocardiography During the Coronavirus Crisis: Considerations in Pediatrics and Congenital Heart Disease | dfeb8bda-648a-4cf2-8adf-133b6e186b08 | 7165086 | Pediatrics[mh] | Pediatric echocardiography, including transthoracic, transesophageal, and fetal imaging, has established indications and procedures. , Based on published appropriate-use criteria for pediatric echocardiography, the indication for an echocardiographic examination is considered appropriate when the expected incremental information, combined with clinical judgment, exceed the expected risks to an acceptable and reasonable degree. , Furthermore, an indication for echocardiographic imaging has been classified into 1 of the following 3 categories: generally appropriate (as reflected by a median panel score of 7-9), may be appropriate (as reflected by a median panel score of 4-6), and rarely appropriate (as reflected by a median panel score of 1-3). , , The goal of this scale of appropriate-use criteria has been to minimize echocardiography examinations in pediatric practice for rarely appropriate criteria. , , With the advent of the coronavirus crisis and the potentially life-threatening risks of infection, the imaging indication in pediatric echocardiography should be screened carefully, with a preference for delaying examinations that are either elective or rarely appropriate in accordance with institutional practice. , , , , Emergency examinations in pediatric echocardiography with strong indications therefore have a high priority to proceed. Given that the intensity of the coronavirus crisis is variable and dynamic, the triage of echocardiography examinations must remain agile and responsive to local conditions. This management process also should focus on strict infection control. , , Fetal echocardiography also should be triaged based on published levels of risk. A fetal echocardiogram for a low-risk patient typically will have a low-risk referral indication in the setting of a normal cardiac screening examination and as such has a low priority for additional consideration during the peak of the coronavirus crisis. , A fetal echocardiogram for a moderate-risk patient typically will be indicated by a moderate-to-high risk referral indication with a gestational age greater than 24 weeks or by confirmed congenital heart disease with a gestational age less than 34 weeks. , These examinations typically can be rescheduled after the peak of the crisis has passed. A fetal echocardiogram for high-risk patients typically will include an urgent clinical indication, or a moderate-to-high risk referral indication with a gestational age less than 24 weeks, or confirmed congenital heart disease with a gestational age more than 34 weeks. , The examinations in this category should be scheduled as soon as possible. The details of this management process have been fully covered in the provided references and are beyond the scope of this editorial. , , , , , Transesophageal imaging is considered high risk because it is associated with viral aerosolization and consequent increased risk of transmission. , Consequently, the threshold for this imaging modality in pediatric practice should be high during the coronavirus crisis. These examinations have low priority in the setting of a weak indication, borderline clinical effect, or if an alternative imaging modality could be diagnostic, according to published consensus and guidelines. , , , ,
Echocardiographic examinations may be possible at the point of care by the clinicians already taking care of these children with suspected or proven coronavirus infection. , , This approach is advantageous not only for patient convenience but also for infection control. The final location for an echocardiographic examination often will require thoughtful consideration of the following: risk of viral transmission, including pregnant women; monitoring capabilities; and staffing requirements. , , , , A complicating factor in pediatric practice is that children with this infection often may be asymptomatic. , In certain circumstances, such as the peak of the coronavirus crisis, it may be reasonable to test new pediatric hospital admissions for this infection to guide the choice of appropriate measures, including infection control. , , In the operating room environment, transesophageal echocardiography often is performed in the setting of a secure airway. This approach to airway management can minimize aerosolization of viral particles and contain viral spread. , The conduct of transesophageal imaging in the setting of pediatric coronavirus infection should consider current recommendations indexed to institutional practice and the intensity of the coronavirus crisis. , , , , There may be dedicated probes and machines in this pediatric setting, depending on local factors. , , , ,
The conduct of the echocardiographic examination in children with suspected or confirmed coronavirus infection should be tailored to address the clinical question. , , , , The cardiac manifestations of COVID-19, such as pericarditis and myocarditis, should be considered during this focused examination. Prolonged echocardiographic examinations should be minimized to limit exposure, given that infectious risks likely are present in asymptomatic children during the crisis phase of COVID-19. , , , Consequently, an experienced practitioner should complete the examination in a focused, time-efficient but comprehensive fashion. , , , , Even though this strategy may erode the educational environment, the safety of learners and trainees is more important, as outlined clearly by the Accreditation Council for Graduate Medical Education (full details available at www.acgme.org/covid-19 ). Apart from the imaging protocol, the conduct of the pediatric echocardiographic examination should take place according to institutional standards for infection control during the crisis, including adequate barrier techniques. , , , , The degree of personal protective equipment will depend on level of infectious risk as defined by specific testing, institutional protocol, and the level of the pandemic at a given hospital. , , Clinical symptoms in infected children often may be absent, prompting interim strategies such as testing all hospitalized children as needed or raising the index of suspicion for active infection. Airborne precautions against viral droplet infection include N95 and N99 masks and powered air purifying respirators. , Transesophageal imaging in suspected or confirmed coronavirus-infected patients carries a heightened risk of viral transmission because of the increased load from viral aerosolization. , , , , It may be reasonable during the height of the crisis to assume that all children who require transesophageal examinations are positive for the infection. In the setting of a protocol for disease testing, a documented negative test within 48- to- 72 hours may be considered adequate at some institutions to conduct the examination with standard precautions such as eye protection, mask, and gloves rather than the enhanced standards with full personal protective equipment. In pediatric patients for whom testing results are unknown and who have an endotracheal tube before arrival in the operating room or interventional suite, the risk of viral transmission from aerosolization is considered low. , , , , How should the risks of viral aerosolization be managed in asymptomatic untested children who have not undergone tracheal intubation and who require transesophageal imaging in the operating room or interventional suite? In this scenario, it is reasonable to expect that the infectious risk is high, assuming that these children may be positive for infection and that endotracheal intubation generates a high load of aerosolized viral particles. In this setting, airway management and probe placement likely should proceed with maximal barrier precautions, including personal protective equipment and consideration for air turnover in the given space. It also is reasonable that the transesophageal probe be placed and positioned by the airway team during aerosol precautions to minimize operator and infectious risks. , , , , In the setting of children with known positive infection, full isolation and aerosol precautions should apply not only for the conduct of the echocardiographic examination but also the overall care of those pediatric patients. , , , The intensity of the coronavirus crisis at a given institution challenges in many ways not only the imaging protocols but also the infectious control procedures for pediatric perioperative echocardiography. An additional consideration for infection control concerns the appropriate care of echocardiographic equipment (the “hardware”) to minimize the risks of viral transmission. , , , , The relevant probes and machine consoles may be covered with disposable plastic. Depending on institutional circumstances, certain hardware can be specifically designated for imaging of suspected or confirmed pediatric cases of coronavirus infection. , , , , Although most disinfectant solutions are virucidal, all echocardiographic equipment should be processed thoroughly for the goals of viral clearance and hardware functionality with maximal protection of patients and ultrasound providers. , , Despite the variations in sanitation protocols, these standards should comply with the recommendations from the American Institute for Ultrasound in Medicine to balance infectious risks with imaging performance. , , , , Education and teaching in pediatric echocardiography are important. , During the coronavirus crisis, however, learner well-being has a higher priority. In this stressful clinical learning environment, it is reasonable to cancel elective rotations and to restrict trainee exposure. Furthermore, education in echocardiography can be transitioned to distance-based learning, including remote conferencing technology. The protection of echocardiography personnel can be enhanced further by thoughtful assignments for staff with risk factors for severe infection such as advanced age, chronic conditions, immunosuppression, and pregnancy.
The coronavirus pandemic has significantly affected the conduct of pediatric echocardiography in the perioperative setting. Careful consideration of the indications, venues, and approaches for echocardiographic imaging will both optimize patient care and infection control during the crisis.
None.
|
Glypican-3-targeted precision diagnosis of hepatocellular carcinoma on clinical sections with a supramolecular 2D imaging probe | 35833fa3-df59-4daa-90be-3d443af4ae0f | 6010994 | Pathology[mh] | Hepatocellular carcinoma (HCC) is the fifth most prevalent malignancy and the second leading cause of cancer-related deaths worldwide. It is estimated that there are more than 782,000 new cases and 746,000 resultant deaths annually . Liver resection is the cornerstone of HCC treatment, and the thorough removal of HCC tissues is vital to improve the overall survival rate of patients . The invasion level of cancer is assessed pre-operatively with imaging techniques such as magnetic resonant imaging and computed tomography. However, assessment of tumor invasion during operation is crucial for surgeons to precisely remove the malignant tissues. There is always a trade-off between the removal of malignant tissues and preservation of relatively normal structures. As a consequence, intra-operative pathology diagnosis based on effective immunohistochemistry, which largely depends on HCC-specific biomarkers, is crucial for HCC treatment. Effective immunohistochemical methods have been developed mainly as a support technique for the diagnosis of various tumors. However, problems remain in terms of the instability of specimens during the preparation and immunostaining of frozen sections, as well as the long time and complex procedures required for current staining techniques. Glypican-3 (GPC-3) is a heparan sulfate proteoglycan responsible for regulating embryonal cell growth . GPC-3 is an ideal biomarker for HCC because it is overexpressed in up to 80% of HCC patients, and is absent in normal tissues, cirrhotic liver, and benign lesions . As a membrane-bound proteoglycan, GPC-3 can be easily stained by the antibody-based technique, and the immunohistochemical staining of GPC-3 has demonstrated a 97% specificity for HCC diagnosis . The rise of graphene has led to an ever-growing interest in academia to develop 2D materials for a number of practical applications - . In recent years, water soluble, biocompatible 2D materials and composites have also been extensively constructed for biomedical applications - . Among the 2D materials reported, 2D molybdenum disulfide (MoS 2 ) has been proven to be of low cytotoxicity, good water solubility, and high biocompatibility by several in vivo studies - . They have also been successfully used as a material substrate for fluorogenic biosensing, bioimaging , - and in vivo theranostics (e.g., multimodal imaging and photodynamic and photothermal therapy) . Despite the rapidly increasing biomedical studies of 2D materials, their potential in diagnosis of pathological sections has not been fully explored . Here, we developed a 2D imaging probe for the effective diagnosis of frozen sections removed from HCC patients. A fluorophore-tagged peptide ligand for GPC-3 was used for the self-assembly with 2D MoS 2 , producing a 2D probe with minimal fluorescence and high affinity for GPC-3. This probe has been shown to be capable of 1) imaging HCC cells over a number of control cells without GPC-3 expression and 2) rapidly imaging HCC pathological sections over para-carcinoma tissues.
Written informed consent was obtained from patients, and the protocol was approved by the Review Board of the Eastern Hepatobiliary Surgery Hospital. For the complete Methods section, see .
Construction and characterization of the 2D imaging probe A known peptide ligand (RLNVGGTYFLTTRQ) for GPC-3 was used for the synthesis of the peptide probe. To increase the water solubility, three arginine groups (K) were grafted to the peptide , followed by the introduction of 5-TAMRA (5-carboxytetramethylrhodamine) as the fluorescence reporter ( Figure A ). Then, the peptide probe (P-probe) dissolved in water was mixed with a homogenous 2D MoS 2 solution for supramolecular self-assembly, producing the 2D imaging probe (2D probe). A series of techniques were subsequently used for the characterization of the 2D probe. We determined that the morphology of P-probe was particle-like ( Figure B , inset), which might be the result of an amphiphilic self-assembly of the probe possessing three arginine groups. Upon further assembly with 2D MoS 2 , the peptide particles were observed to be adhered onto the material surface ( Figure B ) without interrupting the crystal structure of 2D MoS 2 ( Figure B , the enlarged areas display the (1 0 0) facet of 2D MoS 2 , for example) . Adhesion of the spherical aggregates of P-probe to the surface of 2D MoS 2 was also corroborated by atomic force microscopy ( Figure ). Energy dispersive X-ray spectrometry mapping analysis also showed the presence of P-probe on 2D MoS 2 since additional C, N and O signals belonging to the peptide were observed for the 2D probe ( Figure C ). Next, Raman spectroscopy was used to characterize the assembly between the material and peptide. The increased E 1 2g (~379 nm) / A 1g (~405 nm) ratio of the 2D probe (0.62) with respect to 2D MoS 2 (0.57) implies a perturbed in-plane motion between S and Mo , suggesting the coating of P-probe to the material surface ( Figure A ). In addition, the zeta potential of the 2D probe increased with respect to that of the 2D material alone upon assembly with the positively charged TAMRA probe ( Figure B ), and dynamic light scattering also showed that the particle size of 2D probe (PDI = 0.301, average size = 534.7 nm) was larger than that of 2D MoS 2 (PDI = 0.164, average size = 207.0 nm) ( Figure C ). Since it has been reported that 2D MoS 2 can quench the fluorescence of closely attached fluorophores, fluorescence microscopy was used to measure the optical properties of the 2D probe. It was determined that the fluorescence of P-probe was gradually quenched with increasing 2D MoS 2 ( Figure D ), suggesting the assembly between the two species , , . In a subsequent titration assay, we further determined that the 2D probe showed a gradually enhanced fluorescence with increasing GPC-3 in a homogenous buffer solution ( Figure E ), and the fluorescence was hardly changed in the presence of a range of unselective proteins and ions ( Figure F ). These data suggest that the peptide ligand on the surface of the 2D probe can be removed from the 2D MoS 2 surface because of a selective binding with GPC-3, thereby leading to the enhancement of fluorescence , . The subnanomolar limit of detection (0.16 nM) suggests the good sensitivity of the 2D probe for GPC-3, setting the basis for imaging endogenous GPC-3 on the surface of cancer cells. GPC-3-targeted imaging of HCC cells Having successfully constructed the 2D probe, we further tested its ability for GPC-3-targeted cell imaging. A human hepatoma cell line (Hep-G2) was used. In previous studies, we have shown that the presence of 2D materials can drastically enhance the imaging capacity of fluorophore-tagged bioligands probably through the enhancement of the avidity between the ligand and its transmembrane receptor , , . Therefore, we examined the imaging efficiency of the P-probe with increasing 2D MoS 2 . The result showed that the fluorescence of P-probe was gradually enhanced in Hep-G2 with 2D MoS 2 ( Figure A, C ); with 20 μg mL -1 of 2D MoS 2 , the fluorescence intensity was almost 7-fold larger than that of P-probe alone ( Figure C ). This suggests that the supramolecular 2D probe could enhance the binding of the peptide probe with GPC-3, which is in agreement with our previous observations , . We then used a normal human liver cell line (L02) to test the imaging specificity of the probe. We observed that the fluorescence of the 2D probe produced in the HCC cell line (Hep-G2) was much larger than that in the normal cell line (L02) ( Figure A-B ), and the difference in fluorescence intensity was similar to that of the GPC-3 expression level of the two cell lines as determined by real-time quantitative polymerase chain reaction ( Figure C ). This preliminarily verifies the HCC selectivity of the developed 2D probe. To further corroborate that the imaging is based on the recognition between 2D probe and GPC-3 rather than other proteins that are overexpressed in HCC or structurally close to GPC3, a transfection assay was carried out. GPC-3 and a range of other proteins expressed in HCC cells (including glypican-1 [GPC-1], glypican-5 [GPC-5], golgi membrane protein-1 [GOLM-1] and dickkopf-like-1 protein [DKK-1]) were over-expressed in a human embryonic kidney cell line (HEK293T) through the transfection assay. The expression level of the proteins in HEK293T was determined to be similar using western blotting ( Figure E ). Then, the cells were imaged by the 2D probe. We determined that the fluorescence of the 2D probe was only produced in the HEK293T cells that overexpress GPC-3 rather than the control cells expressing other proteins ( Figure B, D ), suggesting the good selectivity of the probe for GPC-3. In addition, we observed that the pre-incubation of a free GPC-3 ligand (KKKRLNVGGTYFLTTRQ) with Hep-G2 largely suppressed the fluorescence of the 2 D probe ( Figure ), which also suggests that the imaging is based on the selective interaction between P-probe and GPC-3. The 2D probe was also shown to hardly interrupt the proliferation of Hep-G2 ( Figure F ), L02 ( Figure G ), and HEK293T cells without transfection and those with an overexpressed GPC-3 level ( Figure ). GPC-3-targeted imaging of HCC sections With the promising imaging results obtained at the cellular level, we finally turned our attention to the examination of the diagnostic potential of the 2D probe for clinical tissue sections. Freshly prepared frozen sections removed from patients pathologically confirmed as HCC were used. We first used an HCC-positive section to optimize the incubation time of the 2D probe. It was determined that the probe's fluorescence became appreciable using fluorescence microscope only 1 min after incubation, and the fluorescence reached equilibrium in about 5 min ( Figure A ). This suggests the ability of the 2D probe to rapidly diagnose HCC-positive tissues. We also compared the imaging effect of the 2D probe with that of P-probe alone by time-dependent fluorescence imaging. We determined that while the P-probe produced an insufficient level of fluorescence on the HCC tissue slide after 15 min of incubation, the use of the 2D probe gave rise to a 6-fold enhanced fluorescence (with respect to the control slide without treatment of probes) ( Figure ). This is in agreement with the imaging result at the cellular level, highlighting the ability of 2D MoS 2 to remarkably enhance the binding of the peptide probe with GPC-3 on HCC tissues. To further test the specificity of the probe, we used different tumor and para-carcinoma sections for the imaging experiment. The result showed that whereas all the tumor sections were clearly imaged by the 2D probe ( Figure B ), almost no fluorescence was produced in the non-tumor slides ( Figure C ). Subsequently, the imaging precision of 2D probe was compared with traditionally used clinical staining methods (i.e., nuclei staining with Hoechst 33342 and the Hematoxylin-Eosin (H&E) staining; Figure D ) in a double-blind manner. We determined that the fluorescently imaged area of HCC tumor sections (nos. 194260 and 193933) were in accordance with both Hoechst 33342 (the cell nuclei of HCC cells are larger than those of healthy cells) and H&E staining, whereas two control slides (cavernous hemangioma, nos. 192513 and 192544) were not falsely imaged by the probe. This implies the good specificity of the imaging probe for HCC over benign liver tumors. Interestingly, the subsequent fluorescence imaging using two independent frozen sections (nos. 194665 and 194404) containing both HCC-positive and para-carcinoma regions gave rise to the evident differentiation of the two regions (as divided by the dashed curve) ( Figure D ). The result is in good accordance with the reference staining methods (Hoechst 33342 and H&E staining). These data clearly suggest the effectiveness of the developed 2D probe facilitating the clinical diagnosis of HCC on frozen sections during surgery. In order to obtain a cut-off value of the imaging method developed, we used six additional frozen sections containing both cancer and para-carcinoma regions. The sections were treated with the 2D probe, and the fluorescence intensities of the HCC-positive and para-carcinoma (negative) regions, as differentiated by H&E staining ( Figure A ), were quantified. Notably, the fluorescence quantification from a single tissue section containing both cancer and para-carcinoma tissues might be more accurate than that from pure positive and negative tissues separately because artificial errors during section preparation and staining could be avoided. On the basis of the statistical relative fluorescence intensity, the positive diagnostic result (gold standard) and the sample size, an ROC curve was drawn. Then, the cut-off value, which is the coordinate closest to the upper left corner of the ROC curve that effectively differentiates the fluorescence intensity of para-carcinoma from that of HCC-positive tissues, was obtained ( Figure B ) .
A known peptide ligand (RLNVGGTYFLTTRQ) for GPC-3 was used for the synthesis of the peptide probe. To increase the water solubility, three arginine groups (K) were grafted to the peptide , followed by the introduction of 5-TAMRA (5-carboxytetramethylrhodamine) as the fluorescence reporter ( Figure A ). Then, the peptide probe (P-probe) dissolved in water was mixed with a homogenous 2D MoS 2 solution for supramolecular self-assembly, producing the 2D imaging probe (2D probe). A series of techniques were subsequently used for the characterization of the 2D probe. We determined that the morphology of P-probe was particle-like ( Figure B , inset), which might be the result of an amphiphilic self-assembly of the probe possessing three arginine groups. Upon further assembly with 2D MoS 2 , the peptide particles were observed to be adhered onto the material surface ( Figure B ) without interrupting the crystal structure of 2D MoS 2 ( Figure B , the enlarged areas display the (1 0 0) facet of 2D MoS 2 , for example) . Adhesion of the spherical aggregates of P-probe to the surface of 2D MoS 2 was also corroborated by atomic force microscopy ( Figure ). Energy dispersive X-ray spectrometry mapping analysis also showed the presence of P-probe on 2D MoS 2 since additional C, N and O signals belonging to the peptide were observed for the 2D probe ( Figure C ). Next, Raman spectroscopy was used to characterize the assembly between the material and peptide. The increased E 1 2g (~379 nm) / A 1g (~405 nm) ratio of the 2D probe (0.62) with respect to 2D MoS 2 (0.57) implies a perturbed in-plane motion between S and Mo , suggesting the coating of P-probe to the material surface ( Figure A ). In addition, the zeta potential of the 2D probe increased with respect to that of the 2D material alone upon assembly with the positively charged TAMRA probe ( Figure B ), and dynamic light scattering also showed that the particle size of 2D probe (PDI = 0.301, average size = 534.7 nm) was larger than that of 2D MoS 2 (PDI = 0.164, average size = 207.0 nm) ( Figure C ). Since it has been reported that 2D MoS 2 can quench the fluorescence of closely attached fluorophores, fluorescence microscopy was used to measure the optical properties of the 2D probe. It was determined that the fluorescence of P-probe was gradually quenched with increasing 2D MoS 2 ( Figure D ), suggesting the assembly between the two species , , . In a subsequent titration assay, we further determined that the 2D probe showed a gradually enhanced fluorescence with increasing GPC-3 in a homogenous buffer solution ( Figure E ), and the fluorescence was hardly changed in the presence of a range of unselective proteins and ions ( Figure F ). These data suggest that the peptide ligand on the surface of the 2D probe can be removed from the 2D MoS 2 surface because of a selective binding with GPC-3, thereby leading to the enhancement of fluorescence , . The subnanomolar limit of detection (0.16 nM) suggests the good sensitivity of the 2D probe for GPC-3, setting the basis for imaging endogenous GPC-3 on the surface of cancer cells.
Having successfully constructed the 2D probe, we further tested its ability for GPC-3-targeted cell imaging. A human hepatoma cell line (Hep-G2) was used. In previous studies, we have shown that the presence of 2D materials can drastically enhance the imaging capacity of fluorophore-tagged bioligands probably through the enhancement of the avidity between the ligand and its transmembrane receptor , , . Therefore, we examined the imaging efficiency of the P-probe with increasing 2D MoS 2 . The result showed that the fluorescence of P-probe was gradually enhanced in Hep-G2 with 2D MoS 2 ( Figure A, C ); with 20 μg mL -1 of 2D MoS 2 , the fluorescence intensity was almost 7-fold larger than that of P-probe alone ( Figure C ). This suggests that the supramolecular 2D probe could enhance the binding of the peptide probe with GPC-3, which is in agreement with our previous observations , . We then used a normal human liver cell line (L02) to test the imaging specificity of the probe. We observed that the fluorescence of the 2D probe produced in the HCC cell line (Hep-G2) was much larger than that in the normal cell line (L02) ( Figure A-B ), and the difference in fluorescence intensity was similar to that of the GPC-3 expression level of the two cell lines as determined by real-time quantitative polymerase chain reaction ( Figure C ). This preliminarily verifies the HCC selectivity of the developed 2D probe. To further corroborate that the imaging is based on the recognition between 2D probe and GPC-3 rather than other proteins that are overexpressed in HCC or structurally close to GPC3, a transfection assay was carried out. GPC-3 and a range of other proteins expressed in HCC cells (including glypican-1 [GPC-1], glypican-5 [GPC-5], golgi membrane protein-1 [GOLM-1] and dickkopf-like-1 protein [DKK-1]) were over-expressed in a human embryonic kidney cell line (HEK293T) through the transfection assay. The expression level of the proteins in HEK293T was determined to be similar using western blotting ( Figure E ). Then, the cells were imaged by the 2D probe. We determined that the fluorescence of the 2D probe was only produced in the HEK293T cells that overexpress GPC-3 rather than the control cells expressing other proteins ( Figure B, D ), suggesting the good selectivity of the probe for GPC-3. In addition, we observed that the pre-incubation of a free GPC-3 ligand (KKKRLNVGGTYFLTTRQ) with Hep-G2 largely suppressed the fluorescence of the 2 D probe ( Figure ), which also suggests that the imaging is based on the selective interaction between P-probe and GPC-3. The 2D probe was also shown to hardly interrupt the proliferation of Hep-G2 ( Figure F ), L02 ( Figure G ), and HEK293T cells without transfection and those with an overexpressed GPC-3 level ( Figure ).
With the promising imaging results obtained at the cellular level, we finally turned our attention to the examination of the diagnostic potential of the 2D probe for clinical tissue sections. Freshly prepared frozen sections removed from patients pathologically confirmed as HCC were used. We first used an HCC-positive section to optimize the incubation time of the 2D probe. It was determined that the probe's fluorescence became appreciable using fluorescence microscope only 1 min after incubation, and the fluorescence reached equilibrium in about 5 min ( Figure A ). This suggests the ability of the 2D probe to rapidly diagnose HCC-positive tissues. We also compared the imaging effect of the 2D probe with that of P-probe alone by time-dependent fluorescence imaging. We determined that while the P-probe produced an insufficient level of fluorescence on the HCC tissue slide after 15 min of incubation, the use of the 2D probe gave rise to a 6-fold enhanced fluorescence (with respect to the control slide without treatment of probes) ( Figure ). This is in agreement with the imaging result at the cellular level, highlighting the ability of 2D MoS 2 to remarkably enhance the binding of the peptide probe with GPC-3 on HCC tissues. To further test the specificity of the probe, we used different tumor and para-carcinoma sections for the imaging experiment. The result showed that whereas all the tumor sections were clearly imaged by the 2D probe ( Figure B ), almost no fluorescence was produced in the non-tumor slides ( Figure C ). Subsequently, the imaging precision of 2D probe was compared with traditionally used clinical staining methods (i.e., nuclei staining with Hoechst 33342 and the Hematoxylin-Eosin (H&E) staining; Figure D ) in a double-blind manner. We determined that the fluorescently imaged area of HCC tumor sections (nos. 194260 and 193933) were in accordance with both Hoechst 33342 (the cell nuclei of HCC cells are larger than those of healthy cells) and H&E staining, whereas two control slides (cavernous hemangioma, nos. 192513 and 192544) were not falsely imaged by the probe. This implies the good specificity of the imaging probe for HCC over benign liver tumors. Interestingly, the subsequent fluorescence imaging using two independent frozen sections (nos. 194665 and 194404) containing both HCC-positive and para-carcinoma regions gave rise to the evident differentiation of the two regions (as divided by the dashed curve) ( Figure D ). The result is in good accordance with the reference staining methods (Hoechst 33342 and H&E staining). These data clearly suggest the effectiveness of the developed 2D probe facilitating the clinical diagnosis of HCC on frozen sections during surgery. In order to obtain a cut-off value of the imaging method developed, we used six additional frozen sections containing both cancer and para-carcinoma regions. The sections were treated with the 2D probe, and the fluorescence intensities of the HCC-positive and para-carcinoma (negative) regions, as differentiated by H&E staining ( Figure A ), were quantified. Notably, the fluorescence quantification from a single tissue section containing both cancer and para-carcinoma tissues might be more accurate than that from pure positive and negative tissues separately because artificial errors during section preparation and staining could be avoided. On the basis of the statistical relative fluorescence intensity, the positive diagnostic result (gold standard) and the sample size, an ROC curve was drawn. Then, the cut-off value, which is the coordinate closest to the upper left corner of the ROC curve that effectively differentiates the fluorescence intensity of para-carcinoma from that of HCC-positive tissues, was obtained ( Figure B ) .
We have developed a simple 2D imaging probe based on the supramolecular assembly between a fluorescent peptide ligand for GPC-3 and 2D MoS 2 . The probe with minimal background fluorescence and high binding affinity for GPC-3 has been shown to be capable of sensitively and selectively imaging HCC cells and a normal cell line overexpressing GPC-3. We have also demonstrated the applicability of the 2D probe for the effective imaging of frozen sections removed from HCC patients in a rapid and precise manner. The fact that it can effectively differentiate between HCC-positive and para-carcinoma tissue regions on a single pathological section makes the 2D probe also potentially applicable for fluorescence imaging-guided surgery. We believe that this research offers new insights into the precise and effective clinical diagnosis of HCC using simple and economic supramolecular 2D imaging probes.
Additional figures including atomic force microscopy images, cell viability and additional cell and tissue imaging results, and the complete Methods section are provided in the Supplementary material associated with this article. Supplementary figures. Click here for additional data file.
|
Impact of unilateral periacetabular osteotomy on bony birth canal in female hip dysplasia patients with gynecoid and anthropoid type pelvis: a pelvic three-dimensional computed tomography measurement using maximum- inscribed-sphere method | 77ee9d1a-cbd7-40f5-997e-b985c6841bb5 | 11938640 | Musculoskeletal System[mh] | Hip dysplasia is one of major etiology to cause the hip osteoarthritis . In order to correct the deformities and prolonged the development of hip osteoarthritis, different surgical methods have been designed, such as Chiari innominate osteotomy , rotational acetabular osteotomy (RAO) , and periacetabular osteotomy (PAO) . Ganz et al. first introduced PAO for the treatment of adult hip dysplasia in 1984, which characterized by deformity correction and keeping pelvic ring intact . The mid and long-term follow-up results of PAO were well recognized . Ganz R , Loder , Flückiger , and Trousdale believed that the true pelvis shape and bony birth canal diameters after PAO didn’t change not to affect normal delivery. The main location of the pelvic inner wall changes after PAO is in the quadrilateral and its surroundings. It does not involve the traditional bony markers used to measure the bony birth canal in traditional obstetrics. Because there is no clear bony marker in the inner wall of the quadrilateral, the traditional two-dimensional measurement method can’t be used to measure bony birth canal after PAO. Loder RT et al. reported 4 main different pelvis types divided according to the pelvic inlet (gynecoid, android, platypelloid, and anthropoid). The gynecoid pelvis is also the normal female type, occurring in over 40% of women, and is ideal for childbirth. Anthropoid pelvis, found in 38.1% of general population, accounts for a much larger proportion of 72.5% in DDH patients . The gynecoid and anthropoid type pelvis are the top two common type among DDH patients, and the effect of unilateral PAO on bony birth canal in these two types is not clear yet. We hypothesis that unilateral PAO induces birth canal narrowing via acetabular fragment medialization, with distinct patterns between gynecoid and anthropoid pelvis. This study aims to use maximum-inscribed-sphere method in female DDH patients with gynecoid and anthropoid type pelvis on measuring the narrowing and narrowest part of the bony birth canal before and after unilateral PAO.
The fetus head was considered as a “dynamic pelvimeter” by Thiery M . Based on this theory, we developed the maximum-inscribed-sphere method using a sphere that was inscribed to the inner wall of bony birth canal, simulating the fetus head in measurement of the bony birth canal during labor. The size of the maximum-inscribed-sphere would represent the width of the bony birth canal. According to common sense, there is a unique central path in any bony birth canal. We defined this central path as the optimal bony birth canal path. Finding the optimal bony birth canal path and the maximum-inscribed-sphere at each point on the path was the task of our three-dimensional measurement of the bony birth canal. Cooperating with the School of optics and photonics of Beijing Institute of Technology, we developed a program based on the Medical Imaging Interaction Toolkit (MITK 2016.11, German Cancer Research Center, Heidelburg, Germany) platform to perform automatic measurement on the bony birth canal, obtaining 201 maximum-inscribed-spheres, and realizing the full-scale automatic identification and measurement of the bony birth canal in three dimensions . We documented positions of the centers of these maximum-inscribed-spheres (connecting which would yield optimal bony birth canal path), the diameters of these spheres, and their inscription points. The results were visualized in Fig. . These patients were admitted to our department as candidates to treat hip diseases from March 2011 to May 2019. We included patients to meet the following criteria: female patients with childbearing age (14–50 years) treated by unilateral PAO, available pre- and post-operative anteroposterior pelvic radiographs and bilateral hip false-profile radiographs in standing position, available pre- and post-operative thin slice CT, and no previous pelvic osteotomy surgery. Exclusion criteria included: an age of ≤ 14 years and ≥ 50 years, unavailable radiography of pre- or post-operative anteroposterior pelvic radiographs or bilateral hip false-profile radiographs in standing position, unavailable radiography of pre- or post-operative thin slice CT, and previous pelvic osteotomy surgery (Tables and ). Pre- and post-operative anteroposterior pelvic radiographs and bilateral hip false-profile radiographs in standing position were obtained in standard procedure using Definium-6000 digital X-ray system (GE Healthcare Pittsburgh USA), Viewerand were stored in imaging system (PACS, UniWeb Viewer, Edition 4.0, EBM Technologies Taipei Taiwan ) for succeeding study and measurements. The Definium-6000 digital X-ray system (GE Medical) was used to obtain the pelvic standing anteroposterior and false-profile radiographs before and after surgery according to the standard X-ray procedure . The X-rays are stored in the image storage system (PACS, UniWeb Viewer Version 4.0; EBM Corporation) for observation and measurement. Four measurements were taken on pre- and post-operative X-rays: lateral CE angle (LCE) , anterior CE angle (ACE) , Tönnis angle , and distance between the medial margin of the femoral head and Kohler’s line (Fig. a and b). Pelvic thin slice CT with window width 1.25 mm (Aquilion TSX-101/HA Toshiba Medical Systems Co. Ltd. Tokyo Japan) was obtained from each patient from 50 mm inferior to lesser trochanter to whole iliac wings. The DICOM data from pelvic CT was transformed to three-dimensional pelvic images using MITK platform (Fig. ). According to the shape of the pelvic inlet, the patients were divided into two groups: female pelvis and ape-shaped pelvis Fig. a and b. Due to the individual differences between different pelvises and the large number of maximum-inscribed-spheres obtained by the measurement, it was impossible to locate each maximum-inscribed-sphere on the inner wall of the pelvis, and to analyze the position and degree of the deformation of the bony birth canal. In order to solve this problem, we divided the inner wall of the pelvis into 25 layers, and selected the corresponding maximum-inscribed-sphere for comparison and analysis. Our measurements were based on real pre- and post-operative CT data from patients, not theoretical 3D planning projections. This approach allowed us to analyze the true anatomical changes that occurred following PAO surgery. The specific steps were as follow: Pelvic anatomical landmarks were chosen on pelvic 3D-CT images using MITK platform to determine the 1st layer (bilateral pubic tubercles, midpoint of sacral promontory), 20th layer (midpoint between the bilateral 4th sacral foramen, bilateral ischial spines), and 25th layer (midpoint of the sacrococcygeal junction, bilateral sciatic tubercles). Next, the pelvis between the 1st and 20th layer was evenly divided into 20 layers, while 5 additional layers were yielded between the 20th and 25th layer. The positions of these 25 layers on the inner pelvic wall were shown in Table ; Fig. a and b. The method of selection was that the 25-layer plane passed through the centers of the 25 maximum-inscribed-spheres, and the diameter of the maximum-inscribed-sphere was documented. The anterior margin of the acetabular fragment was divided into two parts: the pubic ramus, and lateral margin of the obturator foramen (Fig. b). The study was approved by our hospital ethics committee review board (number 2023KY002-KS001). Statistics SPSS version-22.0 statistical software (IBM) was used to perform the statistical analysis. Categorical variables were reported as percentages and frequencies, and continuous variables were reported as the mean and standard deviation if the data followed a normal distribution; otherwise, the median and interquartile range were used. A paired T test was used to compare continuous variables between pre- and post-operative continuous variables of normal distribution on outcomes. The comparison of two different pelvic types was analyzed by independent sample T test. Between preoperative and postoperative normally distributed continuous variables Pearson correlation analysis was used in the correlation analysis. P < 0.05 was considered to be significant.
SPSS version-22.0 statistical software (IBM) was used to perform the statistical analysis. Categorical variables were reported as percentages and frequencies, and continuous variables were reported as the mean and standard deviation if the data followed a normal distribution; otherwise, the median and interquartile range were used. A paired T test was used to compare continuous variables between pre- and post-operative continuous variables of normal distribution on outcomes. The comparison of two different pelvic types was analyzed by independent sample T test. Between preoperative and postoperative normally distributed continuous variables Pearson correlation analysis was used in the correlation analysis. P < 0.05 was considered to be significant.
In the gynecoid type group, the narrowest parts of the bony birth canal before and after the surgery both occurred at the 21st layer (before surgery: 107.39 mm ± 4.33 mm; after surgery: 106.27 mm ± 4.76 mm). In the anthropoid type group, the narrowest part of the bony birth canal was found at the 21st layer before surgery (105.89 mm ± 6.58 mm), and at the 19th layer after surgery (103.48 mm ± 7.36 mm). The most significant narrowing of the bony birth canal was found at the 8th layer (4.25 mm ± 3.51 mm p < 0.05) in those with gynecoid type pelvis, and at the 9th layer (5.8 mm ± 3.15 mm p < 0.001) in those with anthropoid type pelvis after unilateral PAO (Fig. a and b). The difference between the four pre- and post-operative radiographic measurements reflected the deformity correction extent of PAO (Table ). Figure clearly show the surgical outcomes and the quality of acetabular reorientation by PAO procedure. Table shows the difference in diameters of maximum-inscribed-spheres (reflexing the narrowing of the bony birth canal) along the 25th layers. The correlation between these two differences would demonstrate the influence mechanism of PAO on bony birth canals in two different pelvis types (Tables and ). Unilateral PAO had different impact on the acetabulum and bony birth canal in gynecoid and anthropoid type pelvis. The LCE angle increased ( p < 0.05) and the 13th to 19th bony birth canal became smaller ( p < 0.05) after unilateral PAO in anthropoid type pelvis. The shortened mid-pelvic diameters in anthropoid type pelvis after unilateral PAO are shown in Table . Correlation was seen between the difference of ACE and LCE angles ( r = 0.864, p < 0.05) after unilateral PAO in anthropoid type pelvis. The impact of unilateral PAO on the gynecoid type pelvis is shown in Table , with a positive correlation between the distance difference between the medial margin of the femoral head and Kohler’s line, and the difference between the pre- and post-operative largest-inscribed-spheres diameters on the level of 1st to 12th layer ( r = 0.765, 0.806, 0.861, 0.833, 0.825, 0.742, 0.766, 0.701, 0.729, 0.716, 0.728, and 0.652; p < 0.05). PAO resulted in greater medial shift of the acetabular fragment, leading to narrowing of the bony birth canal. In the gynecoid type group, the pre- and post-operative differences of the narrowest part of the 21st layer of bony birth canal, and those of the Tönnis angle, LCE angle, ACE angle, and distance between the medial margin of the femoral head and Kohler’s line showed no correlation ( p > 0.05). Unilateral PAO had no impact on the narrowest part of the bony birth canal in patients with gynecoid type pelvis (Table ). In the anthropoid type group, the pre- and post-operative differences of the narrowest part of the 19th layer of bony birth canal, and those of the Tönnis angle, LCE angle, ACE angle, and distance between the medial margin of the femoral head and Kohler’s line showed no correlation ( p > 0.05). Unilateral PAO had no impact on the narrowest part of the bony birth canal in patients with anthropoid type pelvis (Table ). Correlation coefficient was 0.90 (95% confidence interval (CI) 0.87 to 0.96) for intraobserver agreement and 0.91 (95% CI 0.88 to 0.99) for interobserver agreement in all of the manual measurements, suggesting good agreement between the two independent observers in the manual outcomes recorded.
The impact of pelvic osteotomy on the birth canal has been concerned by orthopedic surgeons. Loder et al. used a pelvic model to study the effects of six pelvic osteotomies on the 3 sagittal diameters and 3 transverse diameters of the bony birth canal. They found that none of the osteotomies affected the pelvic inlet, and Salter, Sutherland, and Steel significantly affected the pelvic outlet. Flückiger et al. . evaluated the pelvic inlet, middle pelvis, and pelvic outlet of 17 female patients after PAO by using 3 transverse diameters of the birth canal on the pelvic X-ray. They found that PAO does not affect the anatomical diameter of the birth canal, so there is no indication for cesarean section. Trousdale et al. measured the 4 birth canal diameters (transverse diameter and sagittal diameter of the pelvic inlet, sagittal diameter and transverse diameter of the middle pelvis) on the MRI. The PAO osteotomy site does not involve the pubic symphysis, iliac spine or posterior arcuate line, so it has nothing to do with these diameters. Ishimatsu et al. . measured the 5 diameters of bony birth canal on three-dimensional pelvic CT, and proposed the transverse diameter of the pelvic expansion and tear drop (distance between the bilateral centers of the internal surfaces of the acetabula, and distance between the bilateral most inferior part of the medial walls of the acetabula, respectively) for pelvic measurement. It was believed that the pelvic teardrop transverse diameter is the narrowest part of the bony birth canal after bilateral CPO. The researcher substituted the pelvic teardrop diameter on 3D-CT as the teardrop ratio (postoperative/preoperative ratio for teardrop distance) on plain radiographs, and investigated the correlation between preoperative radiographic parameters and the teardrop ratio. As a result, the teardrop ratio was significantly correlated with the total bilateral CE angle and the total bilateral ARO (acetabular roof obliquity) preoperatively. Therefore, the postoperative narrowing of the pelvic teardrop diameter was correlated with the degree of rotation of the osteomized acetabular fragment laterally to be required preoperatively. The orthopedic surgeons not only measured the changes in the diameter of the birth canal after pelvic osteotomy, but also tried to clarify the correlation between the pelvic osteotomy and the changes in the diameter of the birth canal. Our research found that the location of unilateral PAO osteotomy was mainly in the quadrilateral area without obvious anatomical landmarks. After unilatreal PAO, the acetabular fragment protruded into the pelvic cavity. Most of these traditional diameters can’t be used to measure the acetabular fragment protruding into the pelvic cavity (Figs. b and ). Traditional two-dimensional pelvic measurement proposed by Thoms. H a century ago was not suitable for analyzing the influence of PAO on the bony birth canal. In order to solve the above problems, we designed the maximum-inscribed-sphere method to measure the bony birth canal and realized the automatic three-dimensional measurement of the bony birth canal by the maximum-inscribed-sphere. The results of this study can provide valuable guidance for obstetricians dealing with patients who have undergone unilateral PAO in terms of measuring pelvic indicators before delivery. This information will also assist hip preservation surgeons in providing better responses to patients’ inquiries regarding fertility-related issues. Patients who have undergone PAO should be assessed with three-dimensional CT of the bony birth canal before pregnancy. A delivery plan should be developed considering the fetal head circumference and the dimensions of the bony birth canal. From the statistical results, the anthropoid type pelvis had oval inlets while the gynecoid type pelvis had round ones. The first half of the bony birth canal was of higher width in gynecoid type pelvis than in the anthropoid type pelvis (Fig. a and b), but no statistically significant difference was obtained possibly due to small sample size. Unilateral PAO did result in the narrowing of bony birth canal superior to the ischial spine in both gynecoid and anthropoid pelvis. Unilateral PAO aims to enhance the stability of the hip joint by increasing acetabular coverage and containment of the femur head. For DDH patients with a laterally shifted hip joint center, Unilateral PAO helps to move the center inward and increase the lateral and anterior bony coverage of the femur head, resulting in a narrowed bony birth canal because of the protruding acetabular fragment (Fig. ). During the operation, elevated pubic bone is partially removed to prevent over-elevation of the pubic ramus, possibly resulting in the narrowest part at the inferior margin of pubic ramus and the region superior to the lateral margin of obturator foramen, instead of the pubic ramus of the acetabular fragment. Unilateral PAO leads to a greater LCE angle in the anthropoid pelvis than in the gynecoid one, which increases the external rotation of the acetabular fragment and causes the anterior-inferior edge of the acetabular fragment to protrude more into bony birth canal. Unilateral PAO significantly reduces the size of the middle part of bony birth canal in the anthropoid pelvis. A transverse diameter shorter than 95 mm in middle pelvis is normally regarded as the threshold for cesarean Sects. . The narrowest part of bony birth canal in those with gynecoid type pelvis was recorded at the 21st level of ischial spine and S4 both before and after surgery. The narrowest part of bony birth canal in those with anthropoid type pelvis was found at the 21st level of ischial spine and S4 before surgery, and above the level of ischial spine after surgery. The narrowest part of bony birth canal after PAO was determined by itself before the procedure, but not unilateral PAO. Kojima et al. reported risks in cesarean section in patients shorter than 150 cm, which was coherent with our findings. Statistic results showed that the only individual shorter than 150 cm had a narrowest diameter less than 95 mm, resulting in higher possibility for cesarean section. This study had some limitations. Firstly, the sample size (19 cases) in our study was relatively small. Secondly, no pelvic controls were set using healthy subjects. Additionally, we mainly focused on pre- and post-operative pelvic morphological characteristics, lacking subsequent data on clinical delivery, which would require further follow-ups to confirm our prediction. Besides, this study serves as a preliminary analysis, with future work planned to include bilateral PAO cases. Lastly, the maximum-inscribed-sphere by our design cannot fully imitate actual fetus head during labor.
Unilateral PAO had no influence on the narrowest part of bony birth canal in both gynecoid and anthropoid type pelvis, but resulted in mild narrowing superior to the ischial spine. The unilateral PAO has significantly decreased the size of middle part of bony birth canal in female DDH patients with anthropoid type pelvis, and narrowed the distance between the medial margin of the femoral head and Kohler’s line resulting in the narrowing of bony birth canal in DDH patients with gynecoid type pelvis.
|
Socioeconomic Determinants of Willingness to Pay for Emergency Public Dental Services in Saudi Arabia: A Contingent Valuation Approach | 495aadb8-f08b-4efc-a527-64a729e82d46 | 9690303 | Dental[mh] | Although oral or dental diseases can be largely prevented, they remain major health problems for several countries, causing disfigurement, discomfort, pain, and death. Approximately 3.5 billion people are affected by oral diseases worldwide; however, treating conditions of oral health is expensive and normally not included in universal health coverage . Estimates show that, globally, the direct treatment cost of dental diseases each year is USD 298 billion (corresponding to 4.6% of the global health expenditure, on average) and the indirect cost per year is USD 144 billion . In the Kingdom of Saudi Arabia (KSA), the prevalence of dental diseases remains very high among children and adults. For instance, among children, the prevalence of dental caries and its associated severity is roughly 80% for primary dentition . Among adults aged 30–45 years, a dental caries prevalence of up to 98% has been reported . Although basic dental care services are provided for free by the public (government) healthcare sector in the KSA , these services are characterized by overload, overburden, overcrowding, shortages of providers, long waiting times, and limited treatment choices . These factors can lead to limited access, which can delay discomfort relief and effective pain management, especially during emergencies, hence exacerbating the suffering of patients. There is therefore a need for a new treatment approach as an alternative to the services provided by the public healthcare sector that would make it possible for people to access dental care at any time, based on need. However, such a treatment approach would come at a cost; hence, its success is dependent on the willingness of people to pay for such services, especially in emergency situations. Thus, willingness to pay (WTP) is one of the best means of assessing the acceptability of novel treatment approaches . To this end, several studies have been conducted to evaluate the WTP for dental services [ , , , , , , , , , , , ]. Among these studies, to the best of our knowledge, only a few were devoted to the KSA [ , , ]. The study by Al Garni et al. sampled 100 respondents who had at least one missing tooth and found that 67% of them were willing to pay the median price (SAR 3000) for an implant. In addition, the study by Fatani and Al-Yousef examined the WTP for orthodontic treatment of children by sampling 171 parents. While they found that 71.6% of the parents disagreed with the fairness of the median WTP amount of SAR 10,000, among those who agreed that the median amount was fair, 71.4% indicated their WTP for more advanced dental treatment for their children . Moreover, the study of 200 patients in the KSA by Linjawi et al. found that 47.5% of the respondents indicated their ability to pay for additional procedures to lessen the treatment time for orthodontic procedures. However, none of the studies on the KSA considered WTP for dental care due to unexpected dental pain occurring outside of normal service hours or days. Examining WTP for urgent dental pain is important because the 74th session of the World Health Assembly held in 2021 approved an oral health resolution that, among other proposals, recommends timely and inclusive oral care . Based on this background and gaps in the literature, this study examined the socioeconomic determinants of WTP for immediate public dental services in the case of a sudden dental pain occurring outside normal services hours or days among 549 adults in Makkah City of the KSA, using a contingent valuation approach. This approach facilitates using surveys to examine the economic value of goods and services not ordinarily found on the market, and thus aids in revealing the willingness of individuals to pay for such goods and services, as well as the amount they are willing to pay . Therefore, the findings of this study can offer important guidance in informing the acceptability and success of new treatment models aimed at providing immediate dental services for patients outside regular services hours or days, ultimately reducing the delays associated with managing urgent dental pain and discomfort.
2.1. Study Design, Setting, and Sample A cross-sectional design was employed from 15 July to 10 August 2021 to obtain data from adult citizens of Saudi Arabia who were residents of Makkah City in the Makkah region, one of the two holy sites in the KSA. Makkah City was selected because of its cosmopolitan nature, encompassing a sample of people with diverse backgrounds. This study included participants who were Saudi citizens, residents of Makkah city, and aged 18 years and above. The study was limited to only Saudi citizens because non-Saudi nationals are not eligible to access free public health services . According to the latest KSA census in 2019, Makkah had a total population of 855,805 Saudi citizens . The minimum sample size necessary for the study was calculated to be 384 (margin of error of 5%, confidence level of 95%, and 50% response distribution, based on the total 855,805 Saudi citizens in the region). However, to curtail the rate of nonresponses and unusable data, over 600 respondents were targeted. The questionnaire was developed based on previous studies . The questionnaire was initially drafted in English by H.S.H. and was translated from English to Arabic by M.K.A. The questionnaire was then back translated to English by another colleague to ensure the intended meaning of the content. Moreover, the questionnaire was pre-tested, and all errors were rectified before final data collection. Given the restrictions instituted in the country during the study period to curtail the spread of COVID-19 in the midst of the pandemic, the data were collected (in the Arabic language) via an online self-reported questionnaire using Google Forms. A snowball sampling technique was used in selecting the potential respondents. This sampling technique was chosen because of the restrictions on movement imposed by the government during the COVID-19 pandemic, which would have made it very difficult to use a probability sampling technique. To this end, we obtained phone contacts of dental patients from major health facilities in Makkah City. We then contacted these individuals via WhatsApp to participate in the study by sending them the Google Forms link. The patients were also kindly requested to forward the Google Forms link to their WhatsApp contacts. A total of 652 participants completed the questionnaire. However, due to incomplete responses and missing data, the study restricted the analysis to the sample of 549 respondents who provided complete information. 2.2. Variables and Data Analysis Techniques The dependent variables used in this study were WTP and the highest amount respondents were willing to pay for immediate public dental service. These variables were obtained using a contingent valuation approach by posing the scenario below to the respondents: Imagine that you or one of your family members suddenly complain of tooth pain and you are seeking to see the doctor in the public dental services, but it is a weekend or a holiday and you have to wait for two days or more; the pain gets worse, and you need to visit the dentist urgently. So, you go to a public dental facility and find that the service will be provided to you at a fee. Would you be willing to pay to see the dentist immediately, or, at the latest, the day after you first noticed the pain? If yes, what would be the highest amount you would be willing to pay? With regard to the contingent valuation approach, we minimized hypothetical bias using the familiarity approach by first contacting patients who were familiar with dental care. Thus, according to Mitchell and Carson , the more familiar individuals are with a good or service, the lower the likelihood of hypothetical bias in a contingent valuation response. Moreover, the collection of data via a non-face-to-face (online) approach and the anonymity of responses provided by the respondents helped in reducing the guilt effect since responses could not be linked to the respondents. Additionally, the scenario narrated to the respondents has the potential to reduce any indignation effect because it does not suggest scrapping the free public dental services, but rather, inquiries about the respondents’ WTP for immediate public dental service during a weekend or holiday. The response to the WTP question was binary (yes or no). Similarly, the response to the highest amount respondents would be willing to pay was categorized as less than SAR 500 or SAR 500 and above. The independent variables used in this study were age, gender, marital status, educational level, employment status, monthly income level, and private health insurance status. All the independent variables were categorical and hence, were treated as dummy variables. The respondents were not restricted to household heads; therefore, the WTP, the maximum amount the respondent was willing to pay, and monthly income of the respondent were all requested at the individual level. With regard to the estimation techniques, simple frequencies and percentages were used to analyze the WTP, as well as the maximum amount the respondent was willing to pay, and the Pearson’s chi-square (χ 2 ) test was used to determine the extent of the associations between the socioeconomic (independent) variables and WTP, as well as the maximum amount the respondents were willing to pay. The binary probit regression model was used to examine the socioeconomic determinants of WTP and the highest amount respondents were willing to pay for immediate public dental services. The binary probit regression model was selected for this analysis given the binary nature of the dependent variables. The ordinary least-squares regression was not used because it is unable to capture the discrete nature of the dependent variables . To facilitate a more intuitive interpretation of the regression results, we presented the average marginal effects (AMEs) of the probit regression , in addition to the estimated coefficients. All analyses were conducted using STATA software version 14.0 (StataCorp LP, College Station, TX, USA).
A cross-sectional design was employed from 15 July to 10 August 2021 to obtain data from adult citizens of Saudi Arabia who were residents of Makkah City in the Makkah region, one of the two holy sites in the KSA. Makkah City was selected because of its cosmopolitan nature, encompassing a sample of people with diverse backgrounds. This study included participants who were Saudi citizens, residents of Makkah city, and aged 18 years and above. The study was limited to only Saudi citizens because non-Saudi nationals are not eligible to access free public health services . According to the latest KSA census in 2019, Makkah had a total population of 855,805 Saudi citizens . The minimum sample size necessary for the study was calculated to be 384 (margin of error of 5%, confidence level of 95%, and 50% response distribution, based on the total 855,805 Saudi citizens in the region). However, to curtail the rate of nonresponses and unusable data, over 600 respondents were targeted. The questionnaire was developed based on previous studies . The questionnaire was initially drafted in English by H.S.H. and was translated from English to Arabic by M.K.A. The questionnaire was then back translated to English by another colleague to ensure the intended meaning of the content. Moreover, the questionnaire was pre-tested, and all errors were rectified before final data collection. Given the restrictions instituted in the country during the study period to curtail the spread of COVID-19 in the midst of the pandemic, the data were collected (in the Arabic language) via an online self-reported questionnaire using Google Forms. A snowball sampling technique was used in selecting the potential respondents. This sampling technique was chosen because of the restrictions on movement imposed by the government during the COVID-19 pandemic, which would have made it very difficult to use a probability sampling technique. To this end, we obtained phone contacts of dental patients from major health facilities in Makkah City. We then contacted these individuals via WhatsApp to participate in the study by sending them the Google Forms link. The patients were also kindly requested to forward the Google Forms link to their WhatsApp contacts. A total of 652 participants completed the questionnaire. However, due to incomplete responses and missing data, the study restricted the analysis to the sample of 549 respondents who provided complete information.
The dependent variables used in this study were WTP and the highest amount respondents were willing to pay for immediate public dental service. These variables were obtained using a contingent valuation approach by posing the scenario below to the respondents: Imagine that you or one of your family members suddenly complain of tooth pain and you are seeking to see the doctor in the public dental services, but it is a weekend or a holiday and you have to wait for two days or more; the pain gets worse, and you need to visit the dentist urgently. So, you go to a public dental facility and find that the service will be provided to you at a fee. Would you be willing to pay to see the dentist immediately, or, at the latest, the day after you first noticed the pain? If yes, what would be the highest amount you would be willing to pay? With regard to the contingent valuation approach, we minimized hypothetical bias using the familiarity approach by first contacting patients who were familiar with dental care. Thus, according to Mitchell and Carson , the more familiar individuals are with a good or service, the lower the likelihood of hypothetical bias in a contingent valuation response. Moreover, the collection of data via a non-face-to-face (online) approach and the anonymity of responses provided by the respondents helped in reducing the guilt effect since responses could not be linked to the respondents. Additionally, the scenario narrated to the respondents has the potential to reduce any indignation effect because it does not suggest scrapping the free public dental services, but rather, inquiries about the respondents’ WTP for immediate public dental service during a weekend or holiday. The response to the WTP question was binary (yes or no). Similarly, the response to the highest amount respondents would be willing to pay was categorized as less than SAR 500 or SAR 500 and above. The independent variables used in this study were age, gender, marital status, educational level, employment status, monthly income level, and private health insurance status. All the independent variables were categorical and hence, were treated as dummy variables. The respondents were not restricted to household heads; therefore, the WTP, the maximum amount the respondent was willing to pay, and monthly income of the respondent were all requested at the individual level. With regard to the estimation techniques, simple frequencies and percentages were used to analyze the WTP, as well as the maximum amount the respondent was willing to pay, and the Pearson’s chi-square (χ 2 ) test was used to determine the extent of the associations between the socioeconomic (independent) variables and WTP, as well as the maximum amount the respondents were willing to pay. The binary probit regression model was used to examine the socioeconomic determinants of WTP and the highest amount respondents were willing to pay for immediate public dental services. The binary probit regression model was selected for this analysis given the binary nature of the dependent variables. The ordinary least-squares regression was not used because it is unable to capture the discrete nature of the dependent variables . To facilitate a more intuitive interpretation of the regression results, we presented the average marginal effects (AMEs) of the probit regression , in addition to the estimated coefficients. All analyses were conducted using STATA software version 14.0 (StataCorp LP, College Station, TX, USA).
3.1. Socio-Economic Characteristics of the Study Participants With regard to the background of the sampled respondents, 32.6% were aged 30–39 years, followed by those aged 18–29 years (24.2%). The majority of the sampled respondents were women (61.6%), married (63.8%), and college/university degree holders (63.0%) ( ). Government and non-government employees formed the majority of the respondents (63.8%). Only 6.6% of the respondents had income levels of SAR 20,000 or more, and the majority of the respondents did not have private health insurance (69.6%). Concerning the dependent variables, the majority of the respondents (79.4%) were willing to pay for immediate public dental services, with 86% of them willing to pay less than SAR 500 ( ). 3.2. Bivariate Analysis of Socioeconomic Factors, WTP, and the Amount Respondents Were Willing to Pay for Immediate Public Dental Services in an Emergency Among the age groups, those 30–39 years old had the highest percentage of respondents who indicated that they were willing to pay for immediate public dental services (32.1%). The majority of those who were willing to pay for immediate public dental services were women (59.6%); however, 69.0% of those who were not willing to pay were also women. Similarly, the majority of those who were willing to pay for immediate public dental services were married (65.1%) and had a college/university degree (62.6%) ( ). There was a statistically significant association between educational level and WTP for immediate public dental services (χ 2 = 7.67, p = 0.02). Moreover, concerning the association between employment status and WTP for immediate public dental services, government employees included the highest percentage of people who were willing to pay (44.5%), followed by non-government employees (22.0%). Hence, we found a statistically significant difference among the various employment categories with regard to WTP for immediate public dental services (χ 2 = 21.17, p = 0.001). The majority of those who were willing to pay for immediate public dental service were those not having private health insurance (66.7%), showing a statistically significant association (χ 2 = 8.06, p = 0.005) ( ). As shown in , among the socioeconomic variables, only marital status, educational level, and employment status showed significant differences between categories with respect to the maximum amount respondents were willing to pay. 3.3. Regression Results and present the results of the binary probit regression estimates of the socioeconomic determinants of WTP for immediate public dental services and the maximum amount respondents were willing to pay, respectively. As shown in , the respondents with a college/university degree or a postgraduate degree were more willing to pay for immediate public dental services relative to those with a high school level of education or below. Specifically, respondents with a college/university degree and those with a postgraduate degree were 11% (AME: 0.11, p < 0.1) and 19% (AME: 0.19, p < 0.01) more willing to pay for immediate public dental services, respectively, relative to their counterparts with a high school level of education or below. With regard to employment status, those who were students, government and non-government employees, and retirees were found to be more willing to pay for immediate public dental services relative to their unemployed counterparts. Specifically, respondents who were students, government employees, non-government employees, and retirees were 21% (AME: 0.21, p < 0.01), 17% (AME: 0.17, p < 0.05), 18% (AME: 0.18, p < 0.05), and 20% (AME: 0.20, p < 0.05) more willing to pay for immediate public dental services, respectively, relative to their unemployed counterparts. Private health insurance was also found to have a positive association with WTP for immediate public dental services. Specifically, respondents with private health insurance were 9% (AME: 0.09, p < 0.05) more willing to pay for immediate public dental services relative to those without private health insurance ( ). Turning to the socioeconomic determinants of the maximum amount respondents were willing to pay for immediate public dental service ( ), we found that respondents aged 50 years and above were 12% (AME: −0.12, p < 0.1) less willing to pay SAR 500 and above, relative to those who were 18–29 years old. Moreover, married respondents were found to be 9% (AME: −0.09, p < 0.05) less willing to pay SAR 500 and above for immediate public dental service relative to their unmarried counterparts. Similarly, respondents with a college/university degree and those with a postgraduate degree were 15% (AME: −0.15, p < 0.05) and 13% (AME: −0.13, p < 0.1) less willing to pay SAR 500 and above for immediate public dental services, respectively, relative to their counterparts with a high school level of education or below ( ).
With regard to the background of the sampled respondents, 32.6% were aged 30–39 years, followed by those aged 18–29 years (24.2%). The majority of the sampled respondents were women (61.6%), married (63.8%), and college/university degree holders (63.0%) ( ). Government and non-government employees formed the majority of the respondents (63.8%). Only 6.6% of the respondents had income levels of SAR 20,000 or more, and the majority of the respondents did not have private health insurance (69.6%). Concerning the dependent variables, the majority of the respondents (79.4%) were willing to pay for immediate public dental services, with 86% of them willing to pay less than SAR 500 ( ).
Among the age groups, those 30–39 years old had the highest percentage of respondents who indicated that they were willing to pay for immediate public dental services (32.1%). The majority of those who were willing to pay for immediate public dental services were women (59.6%); however, 69.0% of those who were not willing to pay were also women. Similarly, the majority of those who were willing to pay for immediate public dental services were married (65.1%) and had a college/university degree (62.6%) ( ). There was a statistically significant association between educational level and WTP for immediate public dental services (χ 2 = 7.67, p = 0.02). Moreover, concerning the association between employment status and WTP for immediate public dental services, government employees included the highest percentage of people who were willing to pay (44.5%), followed by non-government employees (22.0%). Hence, we found a statistically significant difference among the various employment categories with regard to WTP for immediate public dental services (χ 2 = 21.17, p = 0.001). The majority of those who were willing to pay for immediate public dental service were those not having private health insurance (66.7%), showing a statistically significant association (χ 2 = 8.06, p = 0.005) ( ). As shown in , among the socioeconomic variables, only marital status, educational level, and employment status showed significant differences between categories with respect to the maximum amount respondents were willing to pay.
and present the results of the binary probit regression estimates of the socioeconomic determinants of WTP for immediate public dental services and the maximum amount respondents were willing to pay, respectively. As shown in , the respondents with a college/university degree or a postgraduate degree were more willing to pay for immediate public dental services relative to those with a high school level of education or below. Specifically, respondents with a college/university degree and those with a postgraduate degree were 11% (AME: 0.11, p < 0.1) and 19% (AME: 0.19, p < 0.01) more willing to pay for immediate public dental services, respectively, relative to their counterparts with a high school level of education or below. With regard to employment status, those who were students, government and non-government employees, and retirees were found to be more willing to pay for immediate public dental services relative to their unemployed counterparts. Specifically, respondents who were students, government employees, non-government employees, and retirees were 21% (AME: 0.21, p < 0.01), 17% (AME: 0.17, p < 0.05), 18% (AME: 0.18, p < 0.05), and 20% (AME: 0.20, p < 0.05) more willing to pay for immediate public dental services, respectively, relative to their unemployed counterparts. Private health insurance was also found to have a positive association with WTP for immediate public dental services. Specifically, respondents with private health insurance were 9% (AME: 0.09, p < 0.05) more willing to pay for immediate public dental services relative to those without private health insurance ( ). Turning to the socioeconomic determinants of the maximum amount respondents were willing to pay for immediate public dental service ( ), we found that respondents aged 50 years and above were 12% (AME: −0.12, p < 0.1) less willing to pay SAR 500 and above, relative to those who were 18–29 years old. Moreover, married respondents were found to be 9% (AME: −0.09, p < 0.05) less willing to pay SAR 500 and above for immediate public dental service relative to their unmarried counterparts. Similarly, respondents with a college/university degree and those with a postgraduate degree were 15% (AME: −0.15, p < 0.05) and 13% (AME: −0.13, p < 0.1) less willing to pay SAR 500 and above for immediate public dental services, respectively, relative to their counterparts with a high school level of education or below ( ).
This study examined the socioeconomic determinants of WTP for immediate public dental services in the face of dental emergencies, as well as the maximum amount individuals were willing to pay among residents of Makkah City in the KSA. Identifying the factors that determine WTP for immediate public dental services is important to guide the introduction and acceptability of new treatment approaches aimed at dealing with urgent dental cases outside the regular working hours and days of public dental services. We found that the majority of the respondents were willing to pay for public dental services in the case of an emergency (79.4%), which is very encouraging. This finding is similar to those of prior studies [ , , , ]. For instance, Al Garni et al. found that 67% of respondents were willing to pay for implants in Riyadh. Similarly, Fatani and Al-Yousef found that 71.4% of parents in Saudi Arabia were willing to pay for more advanced dental treatment for their children. Educational level was found to play a role in WTP for immediate public dental services in both the bivariate and regression analyses. Since those with a college/university degree and postgraduate degrees are more likely to appreciate the importance of healthcare, as well as to be gainfully employed relative to those with a high school level of education or below , it is not implausible that they would be more willing to pay for public dental services. In fact, the willingness of those with a postgraduate education to pay for public dental services (19%) was even greater than that of respondents with a college/university degree (11%), which highlights the role of a rising level of education with regard to WTP for immediate public dental services. Therefore, initiatives aimed at enhancing WTP for dental services should target those with a high school level of education or below. Our findings are in line with those of Srivastava et al. , who found higher education (university graduate or higher) to be positively associated with WTP for dental care among individuals in Canada. Al-Hanawi et al. also found that individuals with more years of education were more likely to have higher WTP for most indicators regarding improved public health services in Saudi Arabia. Notwithstanding, among those who were willing to pay for immediate public dental services, respondents with college/university and postgraduate degrees were less willing to pay SAR 500 and above for immediate public dental services relative to their counterparts with a high school level of education or below. This could be due to the fact that respondents with higher level of education are more likely to have subscribed to health insurance schemes, and hence, they expect that medical expenses will be covered by such plans. In fact, our data confirm this speculation, since 127 of the 167 respondents with private health insurance (76.05%) had college/university or postgraduate degrees. This is further supported by the negative relationship, albeit statistically insignificant, between the willingness to pay SAR 500 and above and holding private health insurance. Given that people without employment are more likely to have lower incomes, it was not surprising that they were less likely to be willing to pay for immediate public dental services relative to those who were employed (government and non-government employees). In fact, higher income has been found to be positively associated with high ability to pay for unexpected medical expenses among adults in Finland . A positive association between higher income and WTP for dental care was also reported in Canada . Moreover, those employed would have a higher opportunity cost if they are unable to go to work as a result of ill health, leading to a loss in earnings . This would increase the WTP for immediate public dental services among those employed relative to those who are unemployed. Our findings point to the likelihood of widening inequities between those with high socioeconomic status (the employed and those with a higher level of education) and those with low socioeconomic status (the unemployed and the less educated) regarding the WTP for immediate public dental services. Therefore, there is a need to institute measures that would ensure that only those who are capable of paying for such immediate dental services are required to pay, while the less advantaged patients are offered such services at subsidized rates or for free. Last but not the least, since those with private health insurance may perceive healthcare to be more valuable as compared with those without private health insurance, it is not surprising that the former group was more willing to pay for immediate public dental services. In addition, those with private health insurance may also be more likely to have the financial resources to pay for immediate public dental services than those without private health insurance, given that private health insurance policies are usually expensive. Our findings highlight the importance of socioeconomic factors in determining WTP for immediate public dental services during an emergency. Specifically, our results indicate that policies aimed at enhancing the WTP for immediate public dental services should focus on those with a high school level of education or below, the unemployed, and those without private health insurance. However, our study is not without limitations. Since we used only respondents from Makkah City and WhatsApp alone as a means to collect data, caution must be exercised in extending the findings to represent the whole country. In particular, the use of WhatsApp excludes the perceptions of individuals who do not use this social networking platform and those who are illiterate; hence, the findings may not be a true reflection of the WTP for immediate public dental services and the maximum amount patients are willing to pay among more vulnerable groups, such as the illiterate. Similarly, the use of self-reported data is likely to be influenced by the biases of respondents, and the snowball sampling technique adopted by this study is likely to lead to a homogenous and less representative sample. In addition, given that this study was a one-time survey with a limited scenario, we were unable to determine the embedding effect, hence the consistency of the contingent valuation estimates using either an internal or external scope test. Moreover, our study does not reveal how respondents perceived their oral health, which could have been an important factor in determining the WTP for immediate public dental services. Additionally, we did not distinguish between respondents who were dental patients at the time of the study and those who were not in order to determine the effect of this condition on WTP. Nonetheless, being a current dental patient could be a potential source of bias. We therefore recommend that future studies focus on addressing some of these limitations.
Public dental services in the KSA are characterized by overload, overburden, overcrowding, shortages of providers, long waiting times, and limited treatment choices, which together can lead to limited access, thereby extending the period of discomfort while delaying pain management. Therefore, there is a need for alternative models that provide dental care services anytime they are needed. Nonetheless, such alternative models come at a personal cost. We therefore examined the socioeconomic determinants of WTP for immediate public dental care in the face of dental emergencies among adult citizens of Saudi Arabia residing in Makkah. Our findings show that respondents with a high school level of education or below, the unemployed, and those without private health insurance were less willing to pay for immediate public dental services in an emergency situation. These findings suggest that policies and initiatives aimed at enhancing WTP for immediate public dental services should target the unemployed, those with a high school level of education or below, and people without private health insurance.
|
Effectiveness of a 3D-printed mask fitter in an Ophthalmology setting during COVID-19 | 43a12928-6fb5-4eb4-affe-8c08bca79495 | 7972674 | Ophthalmology[mh] | This was a proof-of-concept study to test the feasibility of a new type of PPE. Institutional Review Board approval was obtained from Trillium Health Partners and the protocol adhered to the tenets of the Declaration of Helsinki. Informed consent was obtained from each participant in the study. Statistical analysis was conducted on SPSS version 24 (IBM, Armonk, NY). Participants were recruited from Prism Eye Institute, a tertiary ophthalmology centre in Oakville, Ontario. Participants included technicians, front-facing administrative staff (such as front desk staff and surgical booking assistants), optometrists, ophthalmologists, and medical students. The Bellus3D (Campbell, CA) smartphone application was used on an iPhone XS (Apple, Cupertino, CA) smartphone to capture a 3D image of each person's face. The standard size mask fitter frame was selected, and custom designed to the 3D shape of each participant's face. Each mask fitter design was exported as an STL file and printed by a Prusa i3 MK3S 3D Printer (Prusa Research, Prague, Czech Republic) using polyactic acid filament, a plant-based biodegradable plastic. Perforated 2-cm elastic sewing bands were used as straps to secure the mask fitter over the regular face mask. To simulate the N95 respirator, the custom mask fitter was placed on top of an American Society for Testing and Materials (ASTM) Level 3 fluid-resistant face mask with ear loops (Halyard, Alpharetta, GA; ). Each participant had only one mask fitter created, which was printed from the first successful 3D image capture. To assess the mask fitter efficacy compared with N95 respirators, each participant underwent a standardized N95 Qualitative Fit Test (QLFT), consistent with protocols from the Occupational Safety and Health Administration standard 1910.134 App A (3M, St. Paul, MN; ). According to the test protocols, if an individual passes either on their first or second try, it is considered a pass overall. If they fail on their second try, then they have failed the test overall. For this study, each participant wore their custom mask fitter over a Level 3 mask and underwent the QLFT. Additionally, all participants who passed the QLFT on the first try underwent the QLFT with only the Level 3 face mask and no custom mask fitter, to assess whether wearing the mask fitter made any difference. The same individual (J.L.) administered the QLFT to almost all participants to ensure consistency between administrations. One participant had their N95 QLFT administered at a hospital N95 testing centre. At the conclusion of the QLFT, all participants answered a confidential Likert-scale questionnaire on comfort, ease of use, and everyday feasibility of the custom mask fitter, along with a question that allowed participants to leave comments and suggestions about the custom mask fitter. The questionnaire is shown in .
outlines the participants who were enrolled in the study along with comments about the comfort level of the mask fitter. Of 20 participants, 18 (90%) passed the QLFT, 13 of whom (65%) passed on the first attempt . One female ophthalmologist passed the QLFT on the first try but also passed the QLFT when it was re-administered with only the fluid-resistant face mask. Among the 5 individuals who failed on the first try, 4 were female. A higher proportion of male participants passed on their first try (7 of 8) compared with female participants (6 of 12), although this was not statistically significant on χ 2 test ( p = 0.085). There was no statistically significant association between age and the likelihood of passing the QLFT on the first try. Among all questionnaire responses , the median scores on the Likert scales for comfort, ease of use, and everyday feasibility were 3.5, 4.5, and 3, respectively. A regular face mask had a median comfort score of 8.5, representing a median difference of 5 between the mask fitter and a regular face mask. None of the participants rated the mask fitter as more comfortable than the regular face mask, although 1 participant rated them at the same level of comfort. When stratifying by sex, there was a noticeable difference for everyday feasibility, with male participants assigning a higher median score for feasibility than female participants (7 vs 3), although this was not significant on Mann–Whitney U test ( p = 0.135). No significant associations were found with questionnaire scores and age.
In the context of a global pandemic and limited PPE supplies, health care workers have been encouraged to reuse disposable N95 respirators. However, given N95 respirators are designed for 8 hours of use with the failure rate reported at 46% after 4 days of use, this poses a significant risk to both health care providers and patients. There has also been discussion about the reuse of N95 respirators after sterilization with ionizing radiation. However, doing so causes their measured particular filtering efficiency to decline significantly, and is therefore not recommended. Ideally, health care workers in true need of N95 respirators should be using them as they are designed and disposing of them when appropriate. A 3D-printed custom mask fitter presents a possible PPE resource for health care providers who are in regular close contact with patient's faces but are not involved with aerosol-generating procedures or do not come into contact with confirmed or suspected COVID-19–positive patients. Ophthalmologists and other ECPs are examples of this population niche. Use of the custom mask fitter by this population can preserve the limited N95 respirator supply for higher-risk health care workers while maintaining adequate protection for ECPs. A mask fitter or brace is also mentioned by the CDC as a way to prevent air from leaking around the edges of the mask. Our proof-of-concept study demonstrates promising results for the implementation of a 3D mask fitter in an outpatient setting. Ninety percent of participants’ custom mask fitters performed at the appropriate standard of an N95 respirator based on the QLFT. Sixty percent of participants passed on the first try, whereas an additional 25% passed on their second attempt after modifying their seal. A study with N95 respirators similarly demonstrated that 44.2% of untrained health care workers passed the QLFT on their first try, with an additional 30.2% passing on a subsequent attempt after receiving proper donning instructions. The role of 3D printing to enhance PPE supplies amidst the COVID-19 pandemic has been discussed previously. Custom mask fitters printed using the Bellus3D smartphone application have been informally demonstrated to enhance a peripheral seal on Level 2 and Level 3 surgical masks. However, to the best of our knowledge, this was the first study to formally investigate the effectiveness of 3D-printed custom mask fitters compared with N95 respirators while exploring its feasibility. There has also been formal investigation of the use of 3D-printed mask frames to prolong the life span of N95 respirators. However, unlike the present study, those mask frames were not custom-fitted to an individual's face. The influence of facial features on the fit of N95 respirators has been previously studied; the unique contours of a face alter how a face mask or respirator creates a seal. Because the custom mask fitter is specifically designed for an individual's face, it eliminates issues associated with a “one-size-fits-all” solution. Additionally, the custom mask fitter is secured using adjustable, notched elastic straps, which allows for customization of the fit . Five of the 7 participants who failed the Fit Test on the first try passed the test after the mask fitter was readjusted and tightened using the elastic straps. N95 respirators do not offer the same type of customization as there are only a few different models, which come in a single size and have nonadjustable elastic bands. However, participants in this study did comment on the difficulty of “fiddling around” with the straps and described them as “cumbersome.” Although this was not a factor directly measured in our study, spectacle-wearing participants mentioned how wearing the mask fitter prevented fogging of their glasses. Furthermore, ophthalmologists noted that the mask reduced fogging while using the slit lamp and microscope oculars. A higher proportion of women failed the QLFT on their first try compared with men in this study. This finding is similar to previous studies demonstrating that females were more likely to fail the Fit Test. One recurring issue seen in female participants was the difficulty in securing the elastic straps when their hair was tied back. This was resolved by looping the lower elastic strap from the mask fitter on top of their ponytail . The most common comment regarding the mask fitter was discomfort in wearing it. Twenty-five percent of participants rated the mask fitter as a 1 out of 10 on comfort, and 60% rated it as 4 out of 10 or less. The most common area of discomfort was the mask fitter pushing up against the wearer's eyes, affecting one participant's ability to read. The discomfort was also felt where the frame overlay bony structures, such as the nasal bridge. Although loosening the straps could increase comfort, this also could compromise the seal. Improvements in design could include a thinner plastic frame around the sensitive area under the eyes, or a stronger hold around the chin to prevent the mask fitter from sliding up to the eyes. Although N95 respirators have generally been shown to be comfortable, a direct comparison of the custom mask fitter and N95 respirator comfort level would be useful to study in the future. The mask fitter's discomfort may be its most significant limitation as it effects feasibility. On the questionnaire, everyday feasibility was the only question to receive scores on both extremes of the spectrum although most scores were low. Two participants rated feasibility as 10 out of 10, while 13 participants rated it as 4 out of 10 or less. The 2 individuals who rated feasibility as 10 out of 10 were already using their custom mask fitters on a day-to-day basis for several months. Both acknowledged that the mask fitter was initially very uncomfortable, but they became accustomed to wearing it every day. An additional limitation to using the custom mask fitter involves the type of regular face mask required. Although the seal around a mask is an important aspect of preventing aerosol transmission, the mask's physical material is another factor. In this study, the mask fitter was placed over an ASTM Level 3 facemask. The specific ASTM Level 3 facemask used in this study has an advertised particle filtration efficiency (PFE) of ≥99% at 0.1 microns and bacterial filtration efficiency (BFE) of ≥99% at 3 microns. A National Institute for Occupational Safety and Health-approved N95 respirator sold by the same manufacturer has the exact same PFE and BFE. Before commencing the study, the authors tested the mask fitter over an ASTM Level 1 mask and found it would fail the QLFT. Thus, ASTM Level 3 masks are required to successfully mimic an N95 respirator with the mask fitter, which are more costly than ASTM Level 1 masks. ASTM Level 2 masks were not tested and present an opportunity for further investigation. Finally, there are imitations of the QLFT itself, which has not been shown to be an accurate representation of the respirator's ability to filter all particles. Furthermore, as participants were aware of the goals of this study, there could be an element of participant bias. Additionally, the Likert scales provided in this study ranged from 1 to 10. A 0 to 10 scale would have been more useful, as Likert scales typically have an odd number of selections to allow a midpoint selection. Finally, with its small sample size of only 20 participants, these results are not generalizable to large populations and should be interpreted within its scope as a proof-of-concept study. Overall, the 3D-printed custom mask fitter is a potential option for ECPs seeking more robust PPE given the current limited N95 respirator supply. As it is reusable, cost-effective, and custom-designed to each individual, it offers advantages over N95 respirators and may even be further investigated a new form of PPE after the COVID-19 pandemic resolves. However, the custom mask fitter requires further investigation to test its effectiveness through quantitative means, and further design adjustments to improve its comfort, user-friendliness, and everyday feasibility. In its current state, it cannot replace the N95 respirator, but may provide an alternative PPE solution when N95 supplies are limited.
|
The utility of immunohistochemistry for detecting mycobacterial infections in bronchoalveolar lavage & bronchial washings | 74aca68e-6359-4196-a87b-44cc6207f6f8 | 10284363 | Anatomy[mh] | The study was carried in the department of Microbiology, Apollo Hospital, Hyderabad, Telangana, India. All consecutive BAL and bronchial washings (BW) specimen submitted to the department, over a period of one year (from May 2017 to April 2018) for which AFB cultures were available, were included in the study. Samples with a diagnosis other than inflammatory pathology like malignancies or inadequate samples were excluded. The study was approved by the Institutional Ethics Committee. A total of 203 BAL and BW specimens were analysed from patients with age ranging from 14 to 86 yr. Giemsa, Papanicolaou and Ziehl-Neelsen (ZN) stains were performed on smears as per standard procedures . ZN stain was done after destaining the Giemsa-stained smears in retrospective cases, while in prospective cases; it was done on fresh smears using the standard procedure. Hematoxylin and eosin (H and E) staining and IHC were done on cell blocks to determine the presence of mycobacteria. Immunohistochemistry (IHC) was performed on cell blocks prepared from the respective BAL and BW using M. tuberculosis concentrated (catalogue no. CP 140 A; Biocare Medical, USA) and pre-diluted (1:200) rabbit polyclonal antibody (catalogue no. PP 140 AA; Biocare Medical) in an automated slide stainer, Ventana BenchMark ® XT (Roche, USA) by following manufacturer’s instructions. This antibody is reactive with mycobacteria species including M. tuberculosis, M. avium, M. phlei and M. parafortuitum . BD BACTEC MGIT (mycobacteria growth indicator tube) 960 automated mycobacterial detection system (Becton Dickinson, USA) was used for AFB culture. It was used as a gold standard for comparing the efficacy of ZN and IHC stains.
To the best of our knowledge, this is one of the few studies with such a large sample size comparing the efficacy of ZN stain and IHC for detecting mycobacteria on fluids from BAL and BW. Out of the 203 samples studied, 141 (69.50%) were male and 62 (30.50%) were female. Of these, 112 were BAL samples (55.17%) and 91 were BW (44.83%). In the present study, 175 cases were reported as smears on cytology showing an inflammatory pathology and were negative for malignancy. These showed inflammation composed of neutrophils and lymphocytes admixed with benign squamous cells and ciliated columnar cells with alveolar macrophages. Ten of the 203 specimens were reported as smears showing inflammation with tuberculous etiology based on the presence of AFB on ZN stain. Thirteen were reported as negative for malignancy; three showed inflammation with fungal hyphae and two showed inflammation along with numerous macrophages. Out of the 203 specimens, 21 were AFB culture positive, accounting for 10.3 per cent of the total number of cases. This included both typical and atypical mycobacteria. On ZN staining, the smears showing beaded, bright pink rods were reported as AFB positive . Twelve out of 203 (5.91%) cases showed positivity on ZN staining. None of the samples showed false positivity . The lowest ZN positivity of mycobacterial strains (0%) was reported by Radhakrishnan et al and by Padmavathy et al , whereas the highest positivity of the same (50%) was reported by Purohit et al . Typically, ZN staining is positive only when the bacillary count is more than 10,000 organisms/ml of the specimen . Other reasons for false negativity could be technical problems leading to suboptimal staining. Antimycobacterial therapy can also alter capsule integrity to render organism non-acid-fast ; hence, partially treated smears may also be negative on ZN stain. As the AFB gets engulfed and phagocytosed by the macrophages, only fragments of bacilli are left in the lesion which are not identified by ZN stain . Due to intense phagocytic activity of the macrophages in tuberculous granulomas, the morphological characteristics of AFB often get distorted. This may account for the low detectability on ZN staining similar to the finding in the present study. The IHC staining of bacilli showing different staining patterns is shown in . IHC positivity was reported as intracellular as well as extracellular fine brown granularity or as slender rods . In the present study, 50 out of 203 (24.63%) showed positive staining on IHC, 17 (8.37%) were true positive i.e . these cases were culture positive. However, 33 cases were false positive (16.25%) i.e . these cases were IHC positive but culture negative (Fig and ). The bacilli were mostly seen in clusters (Figs and ), however, a few were also seen as singly scattered . Two fluids which showed histiocytes and neutrophils on smears, particularly displayed intracellular positivity in histiocytes or within aggregates of histiocytes and neutrophils (Figs and ). In these smears, AFB were seen on ZN staining. The contaminants and dark brown course granularity resulting in false positive interpretation are shown in Figures and . The lowest IHC positivity of 64 per cent was reported by Mustafa et al , whereas the highest positivity of 100 per cent was reported by Barbolini et al and Goel and Budhawar on tissue sections. The possible reasons for high false positivity in this study could be contamination due to various reasons such as unsterile sampling technique, unsterile containers, long storage time of sample or contamination during processing. For example, the use of egg albumin stored for a long time for preparing cell blocks is a good medium for the growth of contaminants. The contaminants were recognized as short, stubby rods rather than slender beaded rods or fine granularity. Usually, these were seen in groups or big clusters, often extracellular. The other reason for false positivity could be the use of the polyclonal antibody, which might have resulted in the uptake of the stain by mycobacteria other than MTB. Morphologically, typical and atypical mycobacteria cannot be distinguished on immunohistochemical stain . This indicates that the use of polyclonal antibody for IHC may increase the chances of false positivity. IHC stains the dead as well as fragmented bacilli. It has the capacity to stain antigenic dust as well, which is not stained by ZN stain . This may be the cause of the higher sensitivity of IHC. In this study, differentiating IHC stain from intracellular anthracotic pigment and haemosiderin pigment in the macrophages was a challenge. It was found that the anthracotic pigment appeared darker and blackish as compared to the positive IHCstain which is brown, and haemosiderin pigment showed coarse staining pattern only within macrophages. Three cases were positive for AFB culture as well as ZN stain; however, these did not show IHC positivity. The reason for this false negativity cannot really be ascertained but could possibly be due to technical issues. The result of IHC staining (81% true positivity) in the current study is similar to the studies by Baba et al , and Purohit et al . The results of studies using monoclonal antibodies however, appear to be superior, with 100 per cent IHC positivity , . As reported by Purohit et al , the overall sensitivity and specificity of IHC with monoclonal antibodies such as anti-MPT64 were 92 and 97 per cent, respectively, while the corresponding values for anti-BCG were 88 and 85 per cent. Comparison with various studies on IHC for MTB is shown in . The sensitivity and specificity of ZN stain was 57.1 and 100 per cent, respectively, whereas for IHC staining these were 81 and 81.9 per cent, respectively. The combined ZN stain and IHC results had a sensitivity of 69 per cent and specificity of 90.1 per cent. The likelihood ratio positive and likelihood ratio negative for IHC and ZN stains are included in . The sensitivity of IHC depends on various factors such as distribution of mycobacterial antigen within the granuloma, the clinical stage of disease, duration of anti-tubercular treatment received prior to sampling and specificity of the primary antibody . The reason for 100 per cent positivity in few previous studies could be the use of monoclonal antibody. Most of the studies published in literature have done studies on tissue sections, where PCR was used for tuberculosis as a standard for comparison, unlike the present study in which fluids were used as samples to compare ZN and IHC results with AFB culture, which is considered a gold standard in the diagnosis of tuberculosis. Culture with the new and robust BACTEC machine is not only considered reliable but also quick as compared to conventional culture methods. Studies comparing the utility of IHC and ZN stain in detecting mycobacteria in tissue sections are included in . According to literature, IHC staining gives superior results as compared to the ZN staining, with a positivity of 100 per cent using monoclonal antibody . In our study also, the sensitivity of IHC (81%) was found to be better than ZN staining (57.10%). But as the morphology of mycobacteria on IHC stain is varied, distinguishing these from contaminants was a challenge and the main limitation of this study. The false-positive results were high, thus limiting the use of this method alone for the detection of mycobacteria in fluids. Overall, based on the study findings, AFB culture as a gold standard, neither IHC nor ZN stain appears to be useful as independent methods for the detection of mycobacteria, but these can be used as adjuncts to each other along with other diagnostic tests. A combination of tests can give better and more accurate results for the detection of mycobacteria in bronchoscopic fluids and aid in timely institution of therapy. Further studies with a larger sample size are, however, needed to validate the findings of this study and also to validate the use of IHC as an independent diagnostic test in the diagnosis of tuberculosis.
|
The Delivery of Person-Centered Care for People Living With Dementia in Residential Aged Care: A Systematic Review and Meta-Analysis | f7e7e799-cfb0-47b1-8375-b39e1e122aac | 11020247 | Patient-Centered Care[mh] | Design A systematic review and meta-analysis were undertaken (PROSPERO International Prospective Register of Systematic Reviews registration number 106919). The review is reported according to the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) statement ( ). Deviations From Protocol Because registering this review on PROSPERO, some changes have been made to the research objectives and methods. The three research questions listed on PROSPERO have been replaced with the four objectives listed earlier. We removed Organization for Economic Co-operation and Development (OECD) countries from the inclusion criteria. Review Manager V.5.0 was used for data synthesis, and not STATA V16. Search Strategy An electronic literature search was undertaken in Medline, PsycINFO, Embase, and CINAHL databases (up to December 2021, ). The reference lists of key literature and systematic reviews identified in the initial search yield were reviewed for additional primary studies. The search strategy was limited to the English language with no timeframe limit placed on published studies. The search strategy did not include gray literature, nonprimary data, or systematic reviews to ensure that the review was manageable and that studies addressed the research objectives. Study Selection Eligible studies were primary quantitative, qualitative, or mixed-methods design studies that reported on person-centered care for people living with dementia in RAC. All dementia subtypes were eligible for inclusion. Diagnoses such as cognitive impairment were also eligible for inclusion, as long as it was specified that this impairment stemmed from a dementia diagnosis. Studies were excluded if they did not address person-centered care, outcomes reported did not occur in RAC, diagnostic criteria were not clear, or the full text was not available in English or unavailable in its entirety. Two reviewers (D. Berkovic, A. Macrae) independently screened titles and abstracts of retrieved studies using Covidence (Veritas Health Innovation Ltd, Melbourne, Australia)—a web-based software that assists researchers to screen references and undertake data extraction for systematic reviews and meta-analyses—to determine eligibility. All potentially eligible studies were reviewed independently at the full-text stage (D. Berkovic, A. Macrae). At each review stage, discordance regarding eligibility was discussed and resolved through consensus. Data Extraction Three reviewers (D. Berkovic, A. Macrae, H. Gulline) independently extracted data using a customized template. Data extracted included the study design, country, diagnosis, gender, age, and relevant outcomes concerning the delivery of person-centered care to people living with dementia in RAC. Outcome Measures and Qualitative Themes As there are multiple definitions of person-centered care, all person-centered care definitions, models of care, and frameworks were included. Qualitative results emerged through second-order author-derived themes and were categorized through first-order examination of direct quotes. Risk of Bias Assessment Two reviewers (D. Berkovic, H. Gulline) assessed the quality of included studies using validated critical appraisal tools from the Joanna Briggs Institute (JBI; ). The Mixed Methods Appraisal Tool (MMAT) was used to assess mixed-methods studies ( ). The JBI critical appraisal tools included 8–13 items depending on the study design. Scores were converted to percentages to allow for the comparison of evidence quality scores across different study types (the higher the score of the study, the less bias present). The JBI advises that studies should not be included in the analysis if they are of low quality (score ≤50%). We included all moderate (51%–79%) and good-quality studies (80%–100%). Moderate studies are likely to have moderate but justifiable levels of bias, and good-quality studies are likely to contain little bias. The MMAT is a critical appraisal tool designed for use across mixed-methods studies. The MMAT asks five questions relating to the rationale for using mixed methods to address the research aims. Response options to each question include “yes,” “no,” or “can’t tell.” If a study achieved a “yes” for four to five questions, it was rated as high quality; if a study achieved a “yes” for three questions, it was rated as medium quality, and a study that achieved less than three yes, it was classified as low quality. The MMAT user guide does not state that studies rated as low quality should be excluded. As such, all studies assessed using the MMAT were included ( ). Two reviewers (D. Berkovic, H. Gulline) independently conducted the quality assessment. For scores with ≤10% difference, the average score between the reviewers was used. Where there was stronger disagreement, the study was assessed in tandem, and a consensus score was derived. Where agreement counts not be reached, a third reviewer (D. Ayton) was consulted. Data Synthesis and Meta-Analysis of Quantitative Studies Study characteristics and demographic data were reported using mean (standard deviation [ SD ]), median (interquartile range [IQR]), or frequency. Meta-analysis was conducted where more than three studies measured the same outcome ( ). Experimental and quasi-experimental studies that contained a control group, and aimed to determine the effect of person-centered care interventions on all potential resident outcomes, were eligible for meta-analysis inclusion ( ). Sensitivity analysis was conducted where studies were removed based on their design to ensure that those with a high risk of bias did not affect meta-analysis results. The effect estimate reported for each intervention group was included separately within the meta-analysis to include studies with more than one intervention arm but only one control group. In this instance, the number of participants in the control group was divided equally between the comparisons as per the method described by Cochrane ( ). All outcomes included in the meta-analysis reported continuous data but used varying outcome measures. For example, agitation was measured using the Cohen–Mansfield Agitation Inventory and the Behavioral Activity Rating Scale, and quality of life was measured using the Quality of Life in Late Stage Dementia (QUALID) Scale, and the Alzheimer’s Disease-Related Quality of Life (ADRQL) Instrument. Hence, we used the standardized mean difference as our summary statistic. Mean difference and 95% confidence interval (CI) were used to describe the treatment effect. Statistical heterogeneity was assessed using the I 2 statistic, and a high degree of statistical heterogeneity was present if I 2 values were greater than 50%. A random-effects model was applied in this instance. It was anticipated that a random-effects model would be required across all meta-analyses due to the diverse RAC residents. All analyses were conducted with the use of Review Manager V.5.0 (Revman, The Cochrane Collaboration; Oxford, UK). Data Synthesis of Qualitative Studies A narrative meta-synthesis approach was undertaken to categorize verbatim participant quotes into themes to facilitate an examination of outcomes based on primary data ( ).
A systematic review and meta-analysis were undertaken (PROSPERO International Prospective Register of Systematic Reviews registration number 106919). The review is reported according to the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) statement ( ).
Because registering this review on PROSPERO, some changes have been made to the research objectives and methods. The three research questions listed on PROSPERO have been replaced with the four objectives listed earlier. We removed Organization for Economic Co-operation and Development (OECD) countries from the inclusion criteria. Review Manager V.5.0 was used for data synthesis, and not STATA V16.
An electronic literature search was undertaken in Medline, PsycINFO, Embase, and CINAHL databases (up to December 2021, ). The reference lists of key literature and systematic reviews identified in the initial search yield were reviewed for additional primary studies. The search strategy was limited to the English language with no timeframe limit placed on published studies. The search strategy did not include gray literature, nonprimary data, or systematic reviews to ensure that the review was manageable and that studies addressed the research objectives.
Eligible studies were primary quantitative, qualitative, or mixed-methods design studies that reported on person-centered care for people living with dementia in RAC. All dementia subtypes were eligible for inclusion. Diagnoses such as cognitive impairment were also eligible for inclusion, as long as it was specified that this impairment stemmed from a dementia diagnosis. Studies were excluded if they did not address person-centered care, outcomes reported did not occur in RAC, diagnostic criteria were not clear, or the full text was not available in English or unavailable in its entirety. Two reviewers (D. Berkovic, A. Macrae) independently screened titles and abstracts of retrieved studies using Covidence (Veritas Health Innovation Ltd, Melbourne, Australia)—a web-based software that assists researchers to screen references and undertake data extraction for systematic reviews and meta-analyses—to determine eligibility. All potentially eligible studies were reviewed independently at the full-text stage (D. Berkovic, A. Macrae). At each review stage, discordance regarding eligibility was discussed and resolved through consensus.
Three reviewers (D. Berkovic, A. Macrae, H. Gulline) independently extracted data using a customized template. Data extracted included the study design, country, diagnosis, gender, age, and relevant outcomes concerning the delivery of person-centered care to people living with dementia in RAC.
As there are multiple definitions of person-centered care, all person-centered care definitions, models of care, and frameworks were included. Qualitative results emerged through second-order author-derived themes and were categorized through first-order examination of direct quotes.
Two reviewers (D. Berkovic, H. Gulline) assessed the quality of included studies using validated critical appraisal tools from the Joanna Briggs Institute (JBI; ). The Mixed Methods Appraisal Tool (MMAT) was used to assess mixed-methods studies ( ). The JBI critical appraisal tools included 8–13 items depending on the study design. Scores were converted to percentages to allow for the comparison of evidence quality scores across different study types (the higher the score of the study, the less bias present). The JBI advises that studies should not be included in the analysis if they are of low quality (score ≤50%). We included all moderate (51%–79%) and good-quality studies (80%–100%). Moderate studies are likely to have moderate but justifiable levels of bias, and good-quality studies are likely to contain little bias. The MMAT is a critical appraisal tool designed for use across mixed-methods studies. The MMAT asks five questions relating to the rationale for using mixed methods to address the research aims. Response options to each question include “yes,” “no,” or “can’t tell.” If a study achieved a “yes” for four to five questions, it was rated as high quality; if a study achieved a “yes” for three questions, it was rated as medium quality, and a study that achieved less than three yes, it was classified as low quality. The MMAT user guide does not state that studies rated as low quality should be excluded. As such, all studies assessed using the MMAT were included ( ). Two reviewers (D. Berkovic, H. Gulline) independently conducted the quality assessment. For scores with ≤10% difference, the average score between the reviewers was used. Where there was stronger disagreement, the study was assessed in tandem, and a consensus score was derived. Where agreement counts not be reached, a third reviewer (D. Ayton) was consulted.
Study characteristics and demographic data were reported using mean (standard deviation [ SD ]), median (interquartile range [IQR]), or frequency. Meta-analysis was conducted where more than three studies measured the same outcome ( ). Experimental and quasi-experimental studies that contained a control group, and aimed to determine the effect of person-centered care interventions on all potential resident outcomes, were eligible for meta-analysis inclusion ( ). Sensitivity analysis was conducted where studies were removed based on their design to ensure that those with a high risk of bias did not affect meta-analysis results. The effect estimate reported for each intervention group was included separately within the meta-analysis to include studies with more than one intervention arm but only one control group. In this instance, the number of participants in the control group was divided equally between the comparisons as per the method described by Cochrane ( ). All outcomes included in the meta-analysis reported continuous data but used varying outcome measures. For example, agitation was measured using the Cohen–Mansfield Agitation Inventory and the Behavioral Activity Rating Scale, and quality of life was measured using the Quality of Life in Late Stage Dementia (QUALID) Scale, and the Alzheimer’s Disease-Related Quality of Life (ADRQL) Instrument. Hence, we used the standardized mean difference as our summary statistic. Mean difference and 95% confidence interval (CI) were used to describe the treatment effect. Statistical heterogeneity was assessed using the I 2 statistic, and a high degree of statistical heterogeneity was present if I 2 values were greater than 50%. A random-effects model was applied in this instance. It was anticipated that a random-effects model would be required across all meta-analyses due to the diverse RAC residents. All analyses were conducted with the use of Review Manager V.5.0 (Revman, The Cochrane Collaboration; Oxford, UK).
A narrative meta-synthesis approach was undertaken to categorize verbatim participant quotes into themes to facilitate an examination of outcomes based on primary data ( ).
Study Selection and Inclusion There were 1,099 articles identified (67 duplicates). After reviewing titles and abstracts, 898 articles were excluded, leaving 133 articles for full-text review. The full-text review process yielded 51 articles for quality and risk of bias assessment. Ten articles were deemed to be of low methodological quality and were excluded from the review. The full study selection and inclusion process are shown in . Study Characteristics Forty-one studies from 13 different countries: 7 from Australia, 2 from Belgium, 5 from Canada, 1 from Denmark, 3 from Germany, 1 from Japan, 2 from the Netherlands, 3 from Norway, 4 from Portugal, 1 from Sweden, 3 from the United Kingdom, 8 from the United States; and 1 study conducted across Belgium, Norway, Portugal, and Romania with a wide range of person-centered care-related interventions and resident outcomes were included. The included studies were published from 1996 to 2021. Of the 41 studies, 15 were randomized-controlled trials (RCTs), 10 adopted a quasi-experimental methodology, 8 adopted a mixed-methods design, and 8 adopted a qualitative methodology. All but one study reported on the number of included RAC sites, ranging from 1 (where a qualitative case study was conducted) to 69. All studies also reported the number of residents or staff, depending on the target population group. The smallest sample was 4 staff members (where the same qualitative case study was conducted), but one mixed-methods study included 452 staff members. The smallest number of residents included in the studies, where residents were the target population, was 16; the highest number of residents was 847. Across the majority of studies, participants were primarily female (64%–100%). Three studies did not report on participants’ sex, and two studies reported majority male participants. Almost all studies included participants with a dementia diagnosis; five included participants with Alzheimer’s Disease, one study included participants with cognitive impairment, and one study included participants with either moderate to severe dementia, Alzheimer’s, vascular dementia, mixed vascular dementia and Alzheimer’s, and Lewy body dementia. Study characteristics for all included studies can be found in . Risk of Bias Assessment Results Seventeen studies were rated as high quality ( , ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ), and 23 studies were rated as medium quality ( ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ). Ten studies were rated as low quality and subsequently excluded. was graded as low quality, but as stated in the MMAT user guide, low-quality mixed-methods studies should be not excluded from a review. As a result, this study was still included in this review. JBI quality and risk of bias results can be found in . What are the different frameworks, models of care, and programs currently used in the delivery of person-centered care in RAC? The included studies contained 34 different frameworks, models of care, and programs to deliver person-centered care to people in RAC. The frameworks were broadly categorized into two subtypes: validated methods of delivering person-centered care that have been trialed across multiple studies and by multiple researchers, and fit-for-purpose methods specific to one study. There were 23 validated frameworks, models of care, or programs used to deliver person-centered care across the included studies ( , ; ; ; ; , ; ; ; ; ; ; ; ; , ; ; ; ; ; ; ; ; , ; ). Dementia Care Mapping was the most frequently adopted method, found across five studies ( ; ; ; ; ). The Montessori for Dementia and Ageing model of care was the next most frequently adopted model, employed across four studies ( ; ; ; ). Booth et al. used the Positive Interaction Engagement Program based on Montessori principles ( ). Roberts et al. used the ABLE Model of Care based on Montessori’s principle of acknowledging people with dementia as autonomous adults ( ). The VIPS Practice Model was used across two studies ( ; ), as was Well-being and Health for People with Dementia (WHELD; , ). There were 11 fit-for-purpose, researcher-designed methods of delivering person-centered care found in this review ( , ; ; ; ; ; ; ; ; ; ; ; ; ). Three of these methods focused on communication between staff and residents ( ; ; ) and music therapy ( ; ; ), two focused on activities of daily living ( ; ) and improving quality of life ( ; ). The four Barbosa et al. studies used the same intervention which focused on psychoeducation and multisensory stimulation ( , ; ; ). describes these studies in more detail. What types of person-centered care, and outcomes achieved, were delivered within the identified frameworks, models of care, and programs? The included studies contained 14 various elements of person-centered care. Outcomes include agitation, antipsychotic medication use, mobility, activities of daily living, quality of life, understanding dementia, understanding person-centered care, neuropsychiatric symptoms, person-centered care and dementia-friendly environments, multisensory and motor stimulation, staff burnout, staff training and caregiving, and staff job satisfaction. Eleven quantitative studies assessed agitation ( ; ; ; ; ; ; ; ; ; ; ). Agitation was reported to improve among residents in almost all studies using validated models (results ranging from p = .05 to p < .001; ; ), yet in one cluster RCT, where residents were assigned to the dementia care mapping group, agitation did not improve ( p = .77; ). Agitation also did not improve across studies utilizing researcher-designed, fit-for-purpose interventions. For example, Burack et al., who implemented a culture change program, found no significant change in verbal agitation among residents ( p = .061). Ten quantitative studies assessed the quality of life ( ; ; ; ; , ; ; ; ; ). Descriptively, quality of life was reported to improve among residents in all but one study that included pre-existing, validated modes of delivering person-centered care (results ranging from p = .04 to p < .001; ; ). Only in Halek et al. were there no reported differences in quality of life between participants in the intervention and control groups ( ). Ten quantitative studies assessed neuropsychiatric symptoms ( , ; ; ; ; ; ; ; ; ). Five studies reported improvements in neuropsychiatric symptoms among residents, particularly for those participating in the WHELD intervention, with symptoms improving significantly in both Ballard et al. studies ( p = .05; , and p < .0001; ). Elder clowning had a similarly significant effect on reducing residents’ neuropsychiatric symptoms ( p = .01; ). Both Rokstad et al. and Rosvik et al. using the VIPS Practice Model demonstrated a significant reduction in neuropsychiatric symptoms ( p = .04; , 2014). Two studies found no significant change in residents’ neuropsychiatric symptoms despite using validated models ( ; ). Eleven quantitative studies assessed shared decision making and communication ( , ; ; ; ; ; ; ; ; ; ). In both Barbosa et al. studies, interventions led to increased staff engagement with residents, such as laughing with residents ( p < .001; ; ). Staff were also statistically more likely to participate in social conversations with residents as a result of the intervention. Staff being taught dementia-friendly communication strategies were more likely to submit reports detailing concerns about residents’ emotional well-being ( p = .027; ). Passalacqua and Harwood also measured dementia-friendly communication components of asking yes/no questions ( p < .05), giving choice ( p < .05), and using humor ( p < .01) with residents, all of which produced significant results ( ). Williams et al. reduced the mean percentage of time that staff used elderspeak to communicate with residents from 28.5 to 19.6 ( p = .002; ). The various quantitative studies delivered person-centered care over different periods, ranging from 4 weeks to 2 years. Fourteen studies collected data only from people living with dementia in RAC, nine studies collected data only from staff working in RAC, six studies collected data from both people living with dementia in RAC and staff working in RAC, and two studies collected data via other methods, for example, family members. describes these findings in more detail. Meta-analysis Three outcomes were included in the meta-analysis: agitation, neuropsychiatric symptoms, and quality of life. Very high I 2 values were observed for all three outcomes indicating a high degree of statistical heterogeneity. A random-effects model was, therefore, adopted for all meta-analyses based on the assumption that the effects of person-centered care vary between studies. John Fletcher’s four questions to consider for heterogeneity ( ) were discussed among the authorship team before progressing with the meta-analysis. Based on the pressing need to better deliver person-centered care to people living with dementia in RAC, and to try and understand what currently does and does not work in this setting, the authorship team agreed to proceed with the meta-analysis. The delivery of person-centered care was not found to have a statistically significant impact on agitation, neuropsychiatric symptoms, or quality of life for people living with dementia in RAC. All studies included in agitation and neuropsychiatric symptoms meta-analyses were RCTs. As such, sensitivity analysis was not performed. One of the studies included in the quality-of-life meta-analysis was a nonrandomized quasi-experimental time series. We, therefore, performed a sensitivity analysis by removing this study to examine the impact of assumptions or unmeasured variables in the data. Quality of Life. Data were only able to be pooled for five studies (1,375 residents, 108 RAC facilities). Two of these studies contained two intervention arms but only one control group. As described in the methods, due to multiple intervention arms, results are available for seven subgroups. Although four of these indicated that person-centered care interventions significantly improved the quality of life for people living with dementia in RAC, meta-analysis results, overall, were statistically insignificant for quality of life (standardized mean difference −0.63, 95% CI: −1.95, 0.70, I 2 99%). Results were still statistically insignificant after removing the quasi-experimental study from the meta-analysis (standardized mean difference −0.80, 95% CI: −2.26, 0.66, I 2 99%) ( ). Neuropsychiatric Symptoms. Data were only able to be pooled for four studies (1,349 residents, 106 RAC facilities). Similarly, due to multiple intervention arms across two studies, results are available for six groups. Half of the studies indicated that person-centered care interventions improved neuropsychiatric symptoms in people living with dementia in RAC, but these findings were not statistically significant overall (mean difference −1.06, 95% CI: −2.16, 0.05, I 2 74%; ). Agitation. Data were pooled for six studies that measured agitation (1,968 residents, 180 RAC facilities). Again, due to multiple intervention arms across three studies, results are available for nine groups. Five studies indicated that person-centered care interventions significantly improved agitation in people living with dementia in RAC, but these findings were not statistically significant overall (standardized mean difference −0.27, 95% CI: −0.58, 0.03, I 2 83%; ). Qualitative meta-synthesis Barriers to providing person-centered care. Six qualitative or mixed-methods studies explored barriers to providing person-centered care. Additional supporting quotes are provided in . Three studies explored time as a barrier to providing person-centered care ( ; ; ). Staff lamented that “sometimes you have to rush rush rush,” but recognized that rushing “affects the residents’ mood” ( ). Documentation of toileting and bathing schedules were brought up as examples of “how fast things have to get done” ( ). One study acknowledged the role that staff shortages play in trying to manage time: “our dementia unit is quite active right at the moment and at this point it’s attention span and staffing levels” ( ). The same study recognized that “the very task-oriented work environment” of RAC contributes to time constraints. Two studies identified challenging resident behaviors as a barrier to providing person-centered care, citing “aggressive stuff on display” ( ; ). Staff revealed that they have had residents try to “spit on you, hit you, get aggressive” ( ), but others showed an understanding of why residents may display this type of anger: “if their needs are not met, they are frustrated, they feel bored, and they have no control” ( ). One study described that “residents’ psychosocial health needs are ignored” ( ), and that “the psychological aspects are not taken very well care of… this causes a lot of problems” ( ). Two studies had staff members describe family expectations as a barrier to delivering person-centered care ( ; ). Staff lamented that they “receive a lot of resistance from certain families because they expect to have everyone singing and dancing all the time” but that this view is unrealistic: “do you do that in your own life?” ( ). Different staff reported similar frustrations but recognized that may be because family members are “worn out” ( ). Enablers to providing person-centered care.— Twelve qualitative or mixed-methods studies explored enablers to providing person-centered care. Further supporting quotes are provided in . Getting to know the resident was highlighted across six studies as the biggest enabler, of person-centered care ( ; ; ; ; ; ). Staff explained that “it’s important to know the residents’ life stories, otherwise you don’t know how to help them” ( ). This sentiment was summed up across other studies, with participants stating that “knowing the interests of the residents prior to them having dementia does help” ( ), and that it’s important to get to know residents so that their RAC facility “feels like home” ( ). Staff collaboration was identified across seven studies as an enabler to providing person-centered care ( ; ; ; ; ; ; ). Teamwork was a particularly important component of providing care, with staff acknowledging that “if colleagues get along with each other and collaborate well, this has a positive effect on the residents” ( ). In mixed-methods studies, when commenting on the success of an intervention, staff commented that “we did it together” ( ) which made the results even more satisfying. Some stated that it is not just important to get along with your colleagues, but “to listen to your colleagues’ experiences and learn from them” to support best practice ( ).
There were 1,099 articles identified (67 duplicates). After reviewing titles and abstracts, 898 articles were excluded, leaving 133 articles for full-text review. The full-text review process yielded 51 articles for quality and risk of bias assessment. Ten articles were deemed to be of low methodological quality and were excluded from the review. The full study selection and inclusion process are shown in .
Forty-one studies from 13 different countries: 7 from Australia, 2 from Belgium, 5 from Canada, 1 from Denmark, 3 from Germany, 1 from Japan, 2 from the Netherlands, 3 from Norway, 4 from Portugal, 1 from Sweden, 3 from the United Kingdom, 8 from the United States; and 1 study conducted across Belgium, Norway, Portugal, and Romania with a wide range of person-centered care-related interventions and resident outcomes were included. The included studies were published from 1996 to 2021. Of the 41 studies, 15 were randomized-controlled trials (RCTs), 10 adopted a quasi-experimental methodology, 8 adopted a mixed-methods design, and 8 adopted a qualitative methodology. All but one study reported on the number of included RAC sites, ranging from 1 (where a qualitative case study was conducted) to 69. All studies also reported the number of residents or staff, depending on the target population group. The smallest sample was 4 staff members (where the same qualitative case study was conducted), but one mixed-methods study included 452 staff members. The smallest number of residents included in the studies, where residents were the target population, was 16; the highest number of residents was 847. Across the majority of studies, participants were primarily female (64%–100%). Three studies did not report on participants’ sex, and two studies reported majority male participants. Almost all studies included participants with a dementia diagnosis; five included participants with Alzheimer’s Disease, one study included participants with cognitive impairment, and one study included participants with either moderate to severe dementia, Alzheimer’s, vascular dementia, mixed vascular dementia and Alzheimer’s, and Lewy body dementia. Study characteristics for all included studies can be found in .
Seventeen studies were rated as high quality ( , ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ), and 23 studies were rated as medium quality ( ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ; ). Ten studies were rated as low quality and subsequently excluded. was graded as low quality, but as stated in the MMAT user guide, low-quality mixed-methods studies should be not excluded from a review. As a result, this study was still included in this review. JBI quality and risk of bias results can be found in . What are the different frameworks, models of care, and programs currently used in the delivery of person-centered care in RAC? The included studies contained 34 different frameworks, models of care, and programs to deliver person-centered care to people in RAC. The frameworks were broadly categorized into two subtypes: validated methods of delivering person-centered care that have been trialed across multiple studies and by multiple researchers, and fit-for-purpose methods specific to one study. There were 23 validated frameworks, models of care, or programs used to deliver person-centered care across the included studies ( , ; ; ; ; , ; ; ; ; ; ; ; ; , ; ; ; ; ; ; ; ; , ; ). Dementia Care Mapping was the most frequently adopted method, found across five studies ( ; ; ; ; ). The Montessori for Dementia and Ageing model of care was the next most frequently adopted model, employed across four studies ( ; ; ; ). Booth et al. used the Positive Interaction Engagement Program based on Montessori principles ( ). Roberts et al. used the ABLE Model of Care based on Montessori’s principle of acknowledging people with dementia as autonomous adults ( ). The VIPS Practice Model was used across two studies ( ; ), as was Well-being and Health for People with Dementia (WHELD; , ). There were 11 fit-for-purpose, researcher-designed methods of delivering person-centered care found in this review ( , ; ; ; ; ; ; ; ; ; ; ; ; ). Three of these methods focused on communication between staff and residents ( ; ; ) and music therapy ( ; ; ), two focused on activities of daily living ( ; ) and improving quality of life ( ; ). The four Barbosa et al. studies used the same intervention which focused on psychoeducation and multisensory stimulation ( , ; ; ). describes these studies in more detail. What types of person-centered care, and outcomes achieved, were delivered within the identified frameworks, models of care, and programs? The included studies contained 14 various elements of person-centered care. Outcomes include agitation, antipsychotic medication use, mobility, activities of daily living, quality of life, understanding dementia, understanding person-centered care, neuropsychiatric symptoms, person-centered care and dementia-friendly environments, multisensory and motor stimulation, staff burnout, staff training and caregiving, and staff job satisfaction. Eleven quantitative studies assessed agitation ( ; ; ; ; ; ; ; ; ; ; ). Agitation was reported to improve among residents in almost all studies using validated models (results ranging from p = .05 to p < .001; ; ), yet in one cluster RCT, where residents were assigned to the dementia care mapping group, agitation did not improve ( p = .77; ). Agitation also did not improve across studies utilizing researcher-designed, fit-for-purpose interventions. For example, Burack et al., who implemented a culture change program, found no significant change in verbal agitation among residents ( p = .061). Ten quantitative studies assessed the quality of life ( ; ; ; ; , ; ; ; ; ). Descriptively, quality of life was reported to improve among residents in all but one study that included pre-existing, validated modes of delivering person-centered care (results ranging from p = .04 to p < .001; ; ). Only in Halek et al. were there no reported differences in quality of life between participants in the intervention and control groups ( ). Ten quantitative studies assessed neuropsychiatric symptoms ( , ; ; ; ; ; ; ; ; ). Five studies reported improvements in neuropsychiatric symptoms among residents, particularly for those participating in the WHELD intervention, with symptoms improving significantly in both Ballard et al. studies ( p = .05; , and p < .0001; ). Elder clowning had a similarly significant effect on reducing residents’ neuropsychiatric symptoms ( p = .01; ). Both Rokstad et al. and Rosvik et al. using the VIPS Practice Model demonstrated a significant reduction in neuropsychiatric symptoms ( p = .04; , 2014). Two studies found no significant change in residents’ neuropsychiatric symptoms despite using validated models ( ; ). Eleven quantitative studies assessed shared decision making and communication ( , ; ; ; ; ; ; ; ; ; ). In both Barbosa et al. studies, interventions led to increased staff engagement with residents, such as laughing with residents ( p < .001; ; ). Staff were also statistically more likely to participate in social conversations with residents as a result of the intervention. Staff being taught dementia-friendly communication strategies were more likely to submit reports detailing concerns about residents’ emotional well-being ( p = .027; ). Passalacqua and Harwood also measured dementia-friendly communication components of asking yes/no questions ( p < .05), giving choice ( p < .05), and using humor ( p < .01) with residents, all of which produced significant results ( ). Williams et al. reduced the mean percentage of time that staff used elderspeak to communicate with residents from 28.5 to 19.6 ( p = .002; ). The various quantitative studies delivered person-centered care over different periods, ranging from 4 weeks to 2 years. Fourteen studies collected data only from people living with dementia in RAC, nine studies collected data only from staff working in RAC, six studies collected data from both people living with dementia in RAC and staff working in RAC, and two studies collected data via other methods, for example, family members. describes these findings in more detail. Meta-analysis Three outcomes were included in the meta-analysis: agitation, neuropsychiatric symptoms, and quality of life. Very high I 2 values were observed for all three outcomes indicating a high degree of statistical heterogeneity. A random-effects model was, therefore, adopted for all meta-analyses based on the assumption that the effects of person-centered care vary between studies. John Fletcher’s four questions to consider for heterogeneity ( ) were discussed among the authorship team before progressing with the meta-analysis. Based on the pressing need to better deliver person-centered care to people living with dementia in RAC, and to try and understand what currently does and does not work in this setting, the authorship team agreed to proceed with the meta-analysis. The delivery of person-centered care was not found to have a statistically significant impact on agitation, neuropsychiatric symptoms, or quality of life for people living with dementia in RAC. All studies included in agitation and neuropsychiatric symptoms meta-analyses were RCTs. As such, sensitivity analysis was not performed. One of the studies included in the quality-of-life meta-analysis was a nonrandomized quasi-experimental time series. We, therefore, performed a sensitivity analysis by removing this study to examine the impact of assumptions or unmeasured variables in the data. Quality of Life. Data were only able to be pooled for five studies (1,375 residents, 108 RAC facilities). Two of these studies contained two intervention arms but only one control group. As described in the methods, due to multiple intervention arms, results are available for seven subgroups. Although four of these indicated that person-centered care interventions significantly improved the quality of life for people living with dementia in RAC, meta-analysis results, overall, were statistically insignificant for quality of life (standardized mean difference −0.63, 95% CI: −1.95, 0.70, I 2 99%). Results were still statistically insignificant after removing the quasi-experimental study from the meta-analysis (standardized mean difference −0.80, 95% CI: −2.26, 0.66, I 2 99%) ( ). Neuropsychiatric Symptoms. Data were only able to be pooled for four studies (1,349 residents, 106 RAC facilities). Similarly, due to multiple intervention arms across two studies, results are available for six groups. Half of the studies indicated that person-centered care interventions improved neuropsychiatric symptoms in people living with dementia in RAC, but these findings were not statistically significant overall (mean difference −1.06, 95% CI: −2.16, 0.05, I 2 74%; ). Agitation. Data were pooled for six studies that measured agitation (1,968 residents, 180 RAC facilities). Again, due to multiple intervention arms across three studies, results are available for nine groups. Five studies indicated that person-centered care interventions significantly improved agitation in people living with dementia in RAC, but these findings were not statistically significant overall (standardized mean difference −0.27, 95% CI: −0.58, 0.03, I 2 83%; ). Qualitative meta-synthesis Barriers to providing person-centered care. Six qualitative or mixed-methods studies explored barriers to providing person-centered care. Additional supporting quotes are provided in . Three studies explored time as a barrier to providing person-centered care ( ; ; ). Staff lamented that “sometimes you have to rush rush rush,” but recognized that rushing “affects the residents’ mood” ( ). Documentation of toileting and bathing schedules were brought up as examples of “how fast things have to get done” ( ). One study acknowledged the role that staff shortages play in trying to manage time: “our dementia unit is quite active right at the moment and at this point it’s attention span and staffing levels” ( ). The same study recognized that “the very task-oriented work environment” of RAC contributes to time constraints. Two studies identified challenging resident behaviors as a barrier to providing person-centered care, citing “aggressive stuff on display” ( ; ). Staff revealed that they have had residents try to “spit on you, hit you, get aggressive” ( ), but others showed an understanding of why residents may display this type of anger: “if their needs are not met, they are frustrated, they feel bored, and they have no control” ( ). One study described that “residents’ psychosocial health needs are ignored” ( ), and that “the psychological aspects are not taken very well care of… this causes a lot of problems” ( ). Two studies had staff members describe family expectations as a barrier to delivering person-centered care ( ; ). Staff lamented that they “receive a lot of resistance from certain families because they expect to have everyone singing and dancing all the time” but that this view is unrealistic: “do you do that in your own life?” ( ). Different staff reported similar frustrations but recognized that may be because family members are “worn out” ( ). Enablers to providing person-centered care.— Twelve qualitative or mixed-methods studies explored enablers to providing person-centered care. Further supporting quotes are provided in . Getting to know the resident was highlighted across six studies as the biggest enabler, of person-centered care ( ; ; ; ; ; ). Staff explained that “it’s important to know the residents’ life stories, otherwise you don’t know how to help them” ( ). This sentiment was summed up across other studies, with participants stating that “knowing the interests of the residents prior to them having dementia does help” ( ), and that it’s important to get to know residents so that their RAC facility “feels like home” ( ). Staff collaboration was identified across seven studies as an enabler to providing person-centered care ( ; ; ; ; ; ; ). Teamwork was a particularly important component of providing care, with staff acknowledging that “if colleagues get along with each other and collaborate well, this has a positive effect on the residents” ( ). In mixed-methods studies, when commenting on the success of an intervention, staff commented that “we did it together” ( ) which made the results even more satisfying. Some stated that it is not just important to get along with your colleagues, but “to listen to your colleagues’ experiences and learn from them” to support best practice ( ).
The included studies contained 34 different frameworks, models of care, and programs to deliver person-centered care to people in RAC. The frameworks were broadly categorized into two subtypes: validated methods of delivering person-centered care that have been trialed across multiple studies and by multiple researchers, and fit-for-purpose methods specific to one study. There were 23 validated frameworks, models of care, or programs used to deliver person-centered care across the included studies ( , ; ; ; ; , ; ; ; ; ; ; ; ; , ; ; ; ; ; ; ; ; , ; ). Dementia Care Mapping was the most frequently adopted method, found across five studies ( ; ; ; ; ). The Montessori for Dementia and Ageing model of care was the next most frequently adopted model, employed across four studies ( ; ; ; ). Booth et al. used the Positive Interaction Engagement Program based on Montessori principles ( ). Roberts et al. used the ABLE Model of Care based on Montessori’s principle of acknowledging people with dementia as autonomous adults ( ). The VIPS Practice Model was used across two studies ( ; ), as was Well-being and Health for People with Dementia (WHELD; , ). There were 11 fit-for-purpose, researcher-designed methods of delivering person-centered care found in this review ( , ; ; ; ; ; ; ; ; ; ; ; ; ). Three of these methods focused on communication between staff and residents ( ; ; ) and music therapy ( ; ; ), two focused on activities of daily living ( ; ) and improving quality of life ( ; ). The four Barbosa et al. studies used the same intervention which focused on psychoeducation and multisensory stimulation ( , ; ; ). describes these studies in more detail.
The included studies contained 14 various elements of person-centered care. Outcomes include agitation, antipsychotic medication use, mobility, activities of daily living, quality of life, understanding dementia, understanding person-centered care, neuropsychiatric symptoms, person-centered care and dementia-friendly environments, multisensory and motor stimulation, staff burnout, staff training and caregiving, and staff job satisfaction. Eleven quantitative studies assessed agitation ( ; ; ; ; ; ; ; ; ; ; ). Agitation was reported to improve among residents in almost all studies using validated models (results ranging from p = .05 to p < .001; ; ), yet in one cluster RCT, where residents were assigned to the dementia care mapping group, agitation did not improve ( p = .77; ). Agitation also did not improve across studies utilizing researcher-designed, fit-for-purpose interventions. For example, Burack et al., who implemented a culture change program, found no significant change in verbal agitation among residents ( p = .061). Ten quantitative studies assessed the quality of life ( ; ; ; ; , ; ; ; ; ). Descriptively, quality of life was reported to improve among residents in all but one study that included pre-existing, validated modes of delivering person-centered care (results ranging from p = .04 to p < .001; ; ). Only in Halek et al. were there no reported differences in quality of life between participants in the intervention and control groups ( ). Ten quantitative studies assessed neuropsychiatric symptoms ( , ; ; ; ; ; ; ; ; ). Five studies reported improvements in neuropsychiatric symptoms among residents, particularly for those participating in the WHELD intervention, with symptoms improving significantly in both Ballard et al. studies ( p = .05; , and p < .0001; ). Elder clowning had a similarly significant effect on reducing residents’ neuropsychiatric symptoms ( p = .01; ). Both Rokstad et al. and Rosvik et al. using the VIPS Practice Model demonstrated a significant reduction in neuropsychiatric symptoms ( p = .04; , 2014). Two studies found no significant change in residents’ neuropsychiatric symptoms despite using validated models ( ; ). Eleven quantitative studies assessed shared decision making and communication ( , ; ; ; ; ; ; ; ; ; ). In both Barbosa et al. studies, interventions led to increased staff engagement with residents, such as laughing with residents ( p < .001; ; ). Staff were also statistically more likely to participate in social conversations with residents as a result of the intervention. Staff being taught dementia-friendly communication strategies were more likely to submit reports detailing concerns about residents’ emotional well-being ( p = .027; ). Passalacqua and Harwood also measured dementia-friendly communication components of asking yes/no questions ( p < .05), giving choice ( p < .05), and using humor ( p < .01) with residents, all of which produced significant results ( ). Williams et al. reduced the mean percentage of time that staff used elderspeak to communicate with residents from 28.5 to 19.6 ( p = .002; ). The various quantitative studies delivered person-centered care over different periods, ranging from 4 weeks to 2 years. Fourteen studies collected data only from people living with dementia in RAC, nine studies collected data only from staff working in RAC, six studies collected data from both people living with dementia in RAC and staff working in RAC, and two studies collected data via other methods, for example, family members. describes these findings in more detail.
Three outcomes were included in the meta-analysis: agitation, neuropsychiatric symptoms, and quality of life. Very high I 2 values were observed for all three outcomes indicating a high degree of statistical heterogeneity. A random-effects model was, therefore, adopted for all meta-analyses based on the assumption that the effects of person-centered care vary between studies. John Fletcher’s four questions to consider for heterogeneity ( ) were discussed among the authorship team before progressing with the meta-analysis. Based on the pressing need to better deliver person-centered care to people living with dementia in RAC, and to try and understand what currently does and does not work in this setting, the authorship team agreed to proceed with the meta-analysis. The delivery of person-centered care was not found to have a statistically significant impact on agitation, neuropsychiatric symptoms, or quality of life for people living with dementia in RAC. All studies included in agitation and neuropsychiatric symptoms meta-analyses were RCTs. As such, sensitivity analysis was not performed. One of the studies included in the quality-of-life meta-analysis was a nonrandomized quasi-experimental time series. We, therefore, performed a sensitivity analysis by removing this study to examine the impact of assumptions or unmeasured variables in the data. Quality of Life. Data were only able to be pooled for five studies (1,375 residents, 108 RAC facilities). Two of these studies contained two intervention arms but only one control group. As described in the methods, due to multiple intervention arms, results are available for seven subgroups. Although four of these indicated that person-centered care interventions significantly improved the quality of life for people living with dementia in RAC, meta-analysis results, overall, were statistically insignificant for quality of life (standardized mean difference −0.63, 95% CI: −1.95, 0.70, I 2 99%). Results were still statistically insignificant after removing the quasi-experimental study from the meta-analysis (standardized mean difference −0.80, 95% CI: −2.26, 0.66, I 2 99%) ( ). Neuropsychiatric Symptoms. Data were only able to be pooled for four studies (1,349 residents, 106 RAC facilities). Similarly, due to multiple intervention arms across two studies, results are available for six groups. Half of the studies indicated that person-centered care interventions improved neuropsychiatric symptoms in people living with dementia in RAC, but these findings were not statistically significant overall (mean difference −1.06, 95% CI: −2.16, 0.05, I 2 74%; ). Agitation. Data were pooled for six studies that measured agitation (1,968 residents, 180 RAC facilities). Again, due to multiple intervention arms across three studies, results are available for nine groups. Five studies indicated that person-centered care interventions significantly improved agitation in people living with dementia in RAC, but these findings were not statistically significant overall (standardized mean difference −0.27, 95% CI: −0.58, 0.03, I 2 83%; ).
Data were only able to be pooled for five studies (1,375 residents, 108 RAC facilities). Two of these studies contained two intervention arms but only one control group. As described in the methods, due to multiple intervention arms, results are available for seven subgroups. Although four of these indicated that person-centered care interventions significantly improved the quality of life for people living with dementia in RAC, meta-analysis results, overall, were statistically insignificant for quality of life (standardized mean difference −0.63, 95% CI: −1.95, 0.70, I 2 99%). Results were still statistically insignificant after removing the quasi-experimental study from the meta-analysis (standardized mean difference −0.80, 95% CI: −2.26, 0.66, I 2 99%) ( ).
Data were only able to be pooled for four studies (1,349 residents, 106 RAC facilities). Similarly, due to multiple intervention arms across two studies, results are available for six groups. Half of the studies indicated that person-centered care interventions improved neuropsychiatric symptoms in people living with dementia in RAC, but these findings were not statistically significant overall (mean difference −1.06, 95% CI: −2.16, 0.05, I 2 74%; ).
Data were pooled for six studies that measured agitation (1,968 residents, 180 RAC facilities). Again, due to multiple intervention arms across three studies, results are available for nine groups. Five studies indicated that person-centered care interventions significantly improved agitation in people living with dementia in RAC, but these findings were not statistically significant overall (standardized mean difference −0.27, 95% CI: −0.58, 0.03, I 2 83%; ).
Barriers to providing person-centered care. Six qualitative or mixed-methods studies explored barriers to providing person-centered care. Additional supporting quotes are provided in . Three studies explored time as a barrier to providing person-centered care ( ; ; ). Staff lamented that “sometimes you have to rush rush rush,” but recognized that rushing “affects the residents’ mood” ( ). Documentation of toileting and bathing schedules were brought up as examples of “how fast things have to get done” ( ). One study acknowledged the role that staff shortages play in trying to manage time: “our dementia unit is quite active right at the moment and at this point it’s attention span and staffing levels” ( ). The same study recognized that “the very task-oriented work environment” of RAC contributes to time constraints. Two studies identified challenging resident behaviors as a barrier to providing person-centered care, citing “aggressive stuff on display” ( ; ). Staff revealed that they have had residents try to “spit on you, hit you, get aggressive” ( ), but others showed an understanding of why residents may display this type of anger: “if their needs are not met, they are frustrated, they feel bored, and they have no control” ( ). One study described that “residents’ psychosocial health needs are ignored” ( ), and that “the psychological aspects are not taken very well care of… this causes a lot of problems” ( ). Two studies had staff members describe family expectations as a barrier to delivering person-centered care ( ; ). Staff lamented that they “receive a lot of resistance from certain families because they expect to have everyone singing and dancing all the time” but that this view is unrealistic: “do you do that in your own life?” ( ). Different staff reported similar frustrations but recognized that may be because family members are “worn out” ( ). Enablers to providing person-centered care.— Twelve qualitative or mixed-methods studies explored enablers to providing person-centered care. Further supporting quotes are provided in . Getting to know the resident was highlighted across six studies as the biggest enabler, of person-centered care ( ; ; ; ; ; ). Staff explained that “it’s important to know the residents’ life stories, otherwise you don’t know how to help them” ( ). This sentiment was summed up across other studies, with participants stating that “knowing the interests of the residents prior to them having dementia does help” ( ), and that it’s important to get to know residents so that their RAC facility “feels like home” ( ). Staff collaboration was identified across seven studies as an enabler to providing person-centered care ( ; ; ; ; ; ; ). Teamwork was a particularly important component of providing care, with staff acknowledging that “if colleagues get along with each other and collaborate well, this has a positive effect on the residents” ( ). In mixed-methods studies, when commenting on the success of an intervention, staff commented that “we did it together” ( ) which made the results even more satisfying. Some stated that it is not just important to get along with your colleagues, but “to listen to your colleagues’ experiences and learn from them” to support best practice ( ).
Six qualitative or mixed-methods studies explored barriers to providing person-centered care. Additional supporting quotes are provided in . Three studies explored time as a barrier to providing person-centered care ( ; ; ). Staff lamented that “sometimes you have to rush rush rush,” but recognized that rushing “affects the residents’ mood” ( ). Documentation of toileting and bathing schedules were brought up as examples of “how fast things have to get done” ( ). One study acknowledged the role that staff shortages play in trying to manage time: “our dementia unit is quite active right at the moment and at this point it’s attention span and staffing levels” ( ). The same study recognized that “the very task-oriented work environment” of RAC contributes to time constraints. Two studies identified challenging resident behaviors as a barrier to providing person-centered care, citing “aggressive stuff on display” ( ; ). Staff revealed that they have had residents try to “spit on you, hit you, get aggressive” ( ), but others showed an understanding of why residents may display this type of anger: “if their needs are not met, they are frustrated, they feel bored, and they have no control” ( ). One study described that “residents’ psychosocial health needs are ignored” ( ), and that “the psychological aspects are not taken very well care of… this causes a lot of problems” ( ). Two studies had staff members describe family expectations as a barrier to delivering person-centered care ( ; ). Staff lamented that they “receive a lot of resistance from certain families because they expect to have everyone singing and dancing all the time” but that this view is unrealistic: “do you do that in your own life?” ( ). Different staff reported similar frustrations but recognized that may be because family members are “worn out” ( ).
Twelve qualitative or mixed-methods studies explored enablers to providing person-centered care. Further supporting quotes are provided in . Getting to know the resident was highlighted across six studies as the biggest enabler, of person-centered care ( ; ; ; ; ; ). Staff explained that “it’s important to know the residents’ life stories, otherwise you don’t know how to help them” ( ). This sentiment was summed up across other studies, with participants stating that “knowing the interests of the residents prior to them having dementia does help” ( ), and that it’s important to get to know residents so that their RAC facility “feels like home” ( ). Staff collaboration was identified across seven studies as an enabler to providing person-centered care ( ; ; ; ; ; ; ). Teamwork was a particularly important component of providing care, with staff acknowledging that “if colleagues get along with each other and collaborate well, this has a positive effect on the residents” ( ). In mixed-methods studies, when commenting on the success of an intervention, staff commented that “we did it together” ( ) which made the results even more satisfying. Some stated that it is not just important to get along with your colleagues, but “to listen to your colleagues’ experiences and learn from them” to support best practice ( ).
To the best of our knowledge, this systematic review and meta-analysis is the first to report available mixed-methods evidence on the delivery of person-centered care to people living with dementia in RAC. Our findings indicate that outcomes are highly variable. The evidence for person-centered care and its association with improved outcomes is limited. There were some signals in the descriptive quantitative data that person-centered care is positively associated with reduced agitation but the evidence for this was limited to five studies. All meta-analysis results were insignificant, but with narrow CIs and results trending toward significance. This creates challenges in making a recommendation regarding RAC practice and policy, yet provides a starting point to consider subjective resident concerns within person-centered care for people living with dementia. Thirty-four frameworks, models of care, and programs were used to deliver person-centered care. Because of the number of different ways person-centered care was implemented (both within and across studies), it is challenging to interpret the consistency of results. For example, five studies provided person-centered care via Dementia Care Mapping, yet each of these studies yielded different results. Similarly, three of the studies that used the Montessori for Dementia and Ageing model of care were mixed methods; two focused on agitation ( ; ), and one focused on activities of daily living ( ). The challenges of implementing person-centered care in RAC are recognized, with calls for a more robust instrument to measure the success of person-centered care interventions from the perspective of those affected ( ). The new Implementation Framework for Aged Care was recently codesigned as a fit-for-purpose framework for embedding evidence into practice in aged care ( ). The Review of Innovative Models of Aged Care similarly acknowledges that the various methods of delivering person-centered care complicates evaluation ( ). Agitation results varied greatly, reported to improve across four RCTs ( ; ; ; ), but not in two other RCTs ( ; ). In fact, agitation results remained insignificant after meta-analysis. Other reviews have attempted to explain the prevalence and varied nature of agitation in people living with dementia, and why it can be complex to manage. One systematic review suggests that because nonpharmacologic interventions can take longer to be effective, they are difficult to implement in real-world settings ( ). A 2019 systematic review found that researchers are yet to quantify the impact of agitation on people living with dementia, and suggests that until this is clearly established, few interventions may be associated with a reduction in agitation ( ). Qualitative research has found that aged care staff find it challenging to manage agitation behaviors amongst residents with dementia due to impaired communication or comprehension of instructions ( ), which may explain why staff did not perceive agitation levels of residents to change in the mixed-methods and qualitative studies ( ; ). Neuropsychiatric symptoms were reported to improve across four RCTs ( , ; ; ), but mixed-methods findings described residents spitting and hitting staff during the interventions ( ). The Neuropsychiatric Inventory Questionnaire, which was used to measure neuropsychiatric symptoms across the majority of studies in this review, has been criticized for being unable to accurately measure heterogeneity within neuropsychiatric symptoms, perhaps contributing to varied results ( ). Meta-analysis results for quality of life were also insignificant. A review of the availability and appropriateness of tools used in RAC settings found that the majority of quality of life measures do not have items relevant to RAC residents, such as control and autonomy. Instead, the quality-of-life measures used focused on physical strength and work, which may be less applicable ( ). A recently published systematic review further found that depression, functional impairment, and polypharmacy affect the quality of life of people living with dementia ( ). These three elements are prevalent among older people living in RAC ( ), and may explain the lack of improvement in some studies. Mixed-methods research has been suggested as the best approach to assess the quality of life in RAC ( ). Qualitative findings of this review provided complementary data to the descriptive findings and meta-analysis. Time constraints were identified as a salient pattern to providing person-centered care; getting to know the resident and working within a collaborative staff environment were identified as enablers to providing person-centered care, all in line with existing literature ( ; ). These findings provide important insight, but people living with dementia should also be included in qualitative research ( ). Qualitative research is important in this field, especially as the voice of the person with dementia is still absent from most quantitative tools ( )
This systematic review and meta-analysis incorporate quantitative and qualitative evidence focusing on person-centered care for people with dementia living in RAC. Further strengths include a comprehensive and systematic search of the literature, an examination of study design, quality of evidence, and outcome measures to compile the best-evidence base for this group. We also acknowledge the review limitations. Meta-analysis results were remarkably heterogeneous and likely speak to the real-world challenges of implementing research in RAC, with different types of residents and staff members. The length of time that studies were conducted is a potential confounder, and we did not examine results relative to the length of time that studies were implemented. The relationship between resident outcomes and person-centered care may be influenced by factors that were not measured or reported, including the temporal relationship between dementia progression and comorbid conditions. Although this did not form part of our exclusion criteria, generalizability of results is potentially limited due to all studies being conducted in high-income countries. We also did not use search terms related to the subtypes of dementia, potentially limiting the number of retrieved articles. Results may not be transferable to specific RAC facilities, or low- and middle-income countries, where the prevalence of dementia is also high ( ). In addition to potential limitations associated with the implementation of this research, meta-analysis outcomes may be variable due to the clinometric differences between measurement tools employed across the various studies. Articles published in a language other than English were excluded from this review; as such, we may have missed results written in another language that address the research objectives.
With an aging population and projected increase in the incidence and prevalence of dementia, we must build a strong evidence and understanding of how to best employ person-centered care to assist people living with dementia in RAC. This is especially important today, where health and social care systems globally are underfunded and understaffed. Directing limited resources to best-evidence care is likely to assist, as is including people living with dementia in shared decision making about their care.
Although current evidence varies greatly in how person-centered care is implemented and affects people living with dementia in RAC, there are signals in the data to suggest that person-centered care is associated with some improved resident outcomes, particularly agitation and quality of life. Qualitative data provide individual staff perspectives and augment our understanding of what it means to deliver person-centered care in RAC. Additional research is required to inform tailored interventions that maximize the quality of life and improve resident outcomes more broadly.
gnad052_suppl_Supplementary_Material
|
Serum neuron-specific enolase, magnetic resonance imaging, and electrophysiology for predicting neurodevelopmental outcomes of neonates with hypoxic-ischemic encephalopathy: a prospective study | 04bef2e9-1279-4bae-ba21-9d94b63c7e2e | 9112575 | Physiology[mh] | Neonatal hypoxic-ischemic encephalopathy (HIE) is a type of encephalopathy caused by oxygen deprivation to the brain . It is mainly reported and studied in infants > 35 weeks gestation . Of note, the definition of HIE is more difficult in preterm infants, and preterm infants are not eligible for cooling; therefore, they have not been studied thoroughly. The reported incidence of neonatal HIE is 2–8 per 1000 live births in developed countries and 26 per 1000 live births in underdeveloped countries . It is most likely the result of an acute peripartum (most common) or chronic (during pregnancy) event . It can be caused by sentinel hypoxic or ischemic events such as severe placental abruption, umbilical cord prolapse, or shoulder dystocia, but most cases are idiopathic . Common neurologic sequelae of HIE include neurodevelopmental disabilities such as cerebral palsy, seizures, hearing loss, blindness, learning disabilities, and behavioral disabilities . Predicting the neurological prognosis would ensure optimal management and treatment and avoiding exposure to potentially toxic therapies . In the last decades, the Apgar score and Sarnat staging have been applied to predict prognosis , but follow-up studies demonstrated a low predictive value of Sarnat grading for the outcome of infants with HIE, especially moderate (Sarnat grade II) encephalopathy . The infant’s condition after therapeutic hypothermia, including neurophysiologic tests, amplitude-integrated electroencephalography (aEEG), biochemical markers, and neuroimaging, have been used to assess the severity of HIE and predict long-term outcomes . Indeed, the tests above are recommended by guidelines and by the literature . In addition, cranial ultrasound, computed tomography (CT), magnetic resonance imaging (MRI) are valuable tools to determine the extent of ischemia and brain damage , with MRI being the most sensitive modality for determining the patterns of brain injury, without the use of radiations . Unfortunately, the results are influenced by the timing of the monitoring among studies or the different timing of the different examinations within a given study . Neuron-specific enolase (NSE) is specific to central and peripheral neurons and neuroendocrine cells . High blood levels of NSE are associated with poor outcomes after cardiac arrest, stroke, and pediatric traumatic brain injury . Celtik et al. showed that NSE could be used as a predictor of HIE severity. Hypothermia treatment for HIE might be associated with decreased NSE levels and possibly with neurodevelopmental outcomes , but the results need to be confirmed. Still, whether a combination of different methods could improve the prognostication of HIE remains to be examined. Some recent studies examined NSE levels, MRI, and aEEG on the outcomes of HIE within the same study, but without combining the different modalities . A study suggested that the combination of cerebrospinal fluid NSE, MRI, and EEG could predict the outcomes of HIE , but sampling cerebrospinal fluid in neonates is riskier than blood. Therefore, this study aimed to examine the predictive value of serum NSE, aEEG, and MRI, alone and in combination, for the neurological outcomes in neonates with HIE.
Patients All consecutive surviving full-term newborns with HIE born and treated at the Neonatal Intensive Care Unit of the Third Affiliated Hospital of An-Hui Medical University and subsequently followed at the same hospital were included in this prospective cohort study (June 2013 to December 2020). Ethical approval was obtained from the clinical research and ethics committee of the Third Affiliated Hospital of An-Hui Medical University (#2013–021-01). All parents gave their informed consent. The inclusion criteria were 1) born after a gestational age of ≥37 and < 42 weeks , 2) birth weight of 2500–4000 g , 3) initial umbilical artery blood pH < 7.0, 4) Apgar score 0–3 for more than 5 minutes , 4) abnormal neurological manifestation, including included altered consciousness (irritation, drowsiness, or coma), muscle tone (increased or decreased), primal reflexes (decreased or disappeared from sucking or embrace reflexes), convulsive seizures, brainstem symptoms (altered respiratory rhythm, altered pupil, retardation or disappearance of light reflexes), and increased fontanelle tension, and 5) multiple organ dysfunction occurred shortly after birth (two or more organ dysfunction, including cardiovascular, gastrointestinal, hematologic, pulmonary, or renal systems) . The exclusion criteria were 1) congenital malformations or major dysmorphic features, 2) congenital viral infections, and 3) defined metabolic syndromes. Respiratory dysfunction was defined as 1) respiratory rate > 90 breaths/min, 2) PaO 2 < 5.3 kPa in the absence of cyanotic disease, 3) PaCO 2 > 8.7 kPa, PaO 2 /FiO 2 < 200 Torr in the absence of cyanotic disease, or 4) mechanical ventilation . Cardiovascular dysfunction was defined as 1) systolic blood pressure < 40 mmHg, 2) heart rate < 50 or > 220 bpm, 3) cardiac arrest, 4) pH < 7.2 with normal PaCO 2 , or 5) continuous hemodynamic support . Hematologic dysfunction was defined as 1) hemoglobin < 5 g/L, white blood cells < 3 × 10 9 /L, 3) platelet counts < 20 × 10 9 /L, or 4) disseminated intravascular coagulation . Neurologic dysfunction was defined as 1) Glasgow coma score < 5 or 2) fixed dilated pupils . Hepatic dysfunction was defined as total bilirubin > 60 μmol/L . Gastrointestinal dysfunction was defined as 1) upper gastrointestinal bleeding or 2) any of hemoglobin by > 20 g/L, blood transfusion, hypotension <3rd percentile, or gastric or duodenal surgery . Renal dysfunction was defined as 1) blood urea nitrogen > 36 mmol/L, 2) serum creatinine > 177 μmol/L, or 3) dialysis/hemofiltration . Hypothermia was not used in this study, ruling out the influence of hypothermia on many test results. Mild hypothermia is the HIE grade A treatment evidence, but many hospitals in China do not have mild hypothermia treatment equipment. Different from Western countries, hospitals without such treatment equipment will transfer children to hospitals with medical equipment; but in China the medical system is continuing to improve and transport systems are not yet fully developed, such hospitals are rare, often far away, and the children are not transported in time. Still, all patients received appropriate ventilation and perfusion, maintaining blood glucose levels, and controlling convulsions and cerebral edema. Amplitude-integrated electroencephalography According to the hospital protocol after admission, two-channel monitoring of 8 electrodes aEEG (Bio-logic Ceegraphscan Netlink Monitor, NATUS Medical Incorporated, Mundelein, IL, USA) within 6 h after birth was recorded for 4 h . The aEEGs were assessed by two independent researchers blinded to the clinical data. The representative background pattern was determined using the scoring system suggested by Hellstrom-Westas et al. . The aEEG results were divided into three categories according to background activity of the neonatal aEEG: 1) normal amplitude: the boundary of the amplitude spectrum was > 10 μV, and the lower boundary was > 5 μV; 2) mildly abnormal amplitude: boundary > 10 μV and lower boundary ≤5 μV in the spectral band; 3) severe amplitude anomaly: the boundary of the spectral band was < 10 μV, and the lower boundary was ≤5 μV. All three categories might be associated with an epileptic activity. According to aEEG background activity and the presence or absence of epileptic activity, the aEEG results were divided into four categories: 1) normal aEEG with normal amplitude; 2) mildly abnormal aEEG with normal amplitude and epileptic activity; 3) mildly abnormal aEEG with normal amplitude; 4) all other combinations were severely abnormal aEEG. According to the needs of this study, the results were divided into two categories, normal and abnormal (including mild and severe abnormalities). Serum NSE levels NSE was detected at 1 day (1-day NSE) and 3 days (3-day NSE) after birth. NSE was measured on a Cobas Core II Immunoanalyzer with the NSE EIA II kit (Roche Molecular Biochemicals, Mannheim, Germany), a one-step sandwich-type enzyme immunoassay that uses monoclonal mouse antibodies. The manufacturers claim a low detection limit of 0.1 μg/L for NSE. Brain MRI MRI (T1- and T2-weighed imaging) scans were obtained within the 4–7 days after birth when the vital sign of the children was relatively stable. All infants were scanned with a 1.5-T Magnetom Avanto MRI system (Siemens, Erlangen, Germany), including MRI plain0 T1WI and T2WI scans. T1WI sequence parameters: RF pulse repetition time 585 ms, echo time 15 ms. T2WI sequence parameters: RF pulse repetition time 4000 ms, echo time 102 ms, a layer thickness of 4 mm, layer spacing of 0.4 mm, and field of view of 20 × 20 cm. All MRIs in this study were read independently by two radiologists who specialized in neuroimaging, and they discussed their findings together. The basal ganglia/watershed scores were determined as proposed by Barkovich et al. : 0 = no abnormalities in the basal ganglia or cortex; 1 = an abnormal signal in the basal ganglia or thalamus; 2 = an abnormal signal in the cortex; 3 = an abnormal signal in the cortex and basal nuclei (basal ganglia or thalami); 4 = an abnormal signal in the entire cortex and basal nuclei. All images and qualitative scores were assessed by an experienced radiologist, who was blinded to the serum NSE levels and clinical data. Follow-up and neurodevelopment outcomes Periodical follow-ups of the neurodevelopment outcomes were performed once a month before 6 months of age, every 2 months between 6 months and 1 year, and every 6 months after 1 year. At 18 months of age, the children were assessed in the neurological rehabilitation clinic according to the study protocol. All children received standardized assessment at 18 months. None of the children were lost to follow-up. If the evaluation could not be made according to the appointment time, telephone communication, or picking up the children at home to the hospital for evaluation, to avoid the loss of follow-up and data loss. The neurodevelopmental function was assessed with the Bayley Scales of Infant Development 2nd version (BSID-II) and neurological examination by a pediatric neurodevelopmental specialist. Motor testing at 1 year improves the prediction of motor and mental outcomes at 2 years after perinatal HIE . The pediatrist was not involved in the neonatal and postnatal care of the children under investigation and was blinded for the clinical status, histories, NSE levels, and neonatal brain MRI scores of the children. This examination took place at a time chosen by the mother as the most optimal time for the child to cooperate, mostly around 10 a.m. The children were classified into three groups depending on the examination results; normal (MDI and PDI > 79), moderate delay (MDI or PDI: 70–79), and severe delay (MDI or PDI < 70) . When normal and abnormal outcomes were predicted, excepting for MDI and PDI in BSID (both mild and severe abnormalities were classified as abnormal) were used to determine the neurodevelopment outcome, a diagnosis of cerebral palsy, visual impairment and hearing loss or epilepsy was also considered an abnormal outcome. Statistical analysis Statistical analysis was conducted using SPSS 17.0 (SPSS, NY, USA). All values are presented as mean ± standard deviation or median (range). The distributions of the continuous data were assessed using the Shapiro-Wilk tests. ANOVA and Tukey’s HSD post hoc test were used to analyze normally distributed data, while the Kruskal-Wallis test was used for non-parametric analyses. Categorical data are presented as n and were analyzed using the chi-square test. A logistic regression model was fit to the binary outcome using aEEG, NSE, and brain MRI together as predictors for neurodevelopment delay. Receiver operating characteristic (ROC) analysis was used to evaluate the predictive accuracy of the logistic regression model, as well as aEEG, NSE, and brain MRI, separately and together. Sensitivity, specificity, PPV, and NPV of aEEG, NSE, and brain MRI, separately and together, were calculated using binomial exact method. P -values < 0.05 were considered statistically significant.
All consecutive surviving full-term newborns with HIE born and treated at the Neonatal Intensive Care Unit of the Third Affiliated Hospital of An-Hui Medical University and subsequently followed at the same hospital were included in this prospective cohort study (June 2013 to December 2020). Ethical approval was obtained from the clinical research and ethics committee of the Third Affiliated Hospital of An-Hui Medical University (#2013–021-01). All parents gave their informed consent. The inclusion criteria were 1) born after a gestational age of ≥37 and < 42 weeks , 2) birth weight of 2500–4000 g , 3) initial umbilical artery blood pH < 7.0, 4) Apgar score 0–3 for more than 5 minutes , 4) abnormal neurological manifestation, including included altered consciousness (irritation, drowsiness, or coma), muscle tone (increased or decreased), primal reflexes (decreased or disappeared from sucking or embrace reflexes), convulsive seizures, brainstem symptoms (altered respiratory rhythm, altered pupil, retardation or disappearance of light reflexes), and increased fontanelle tension, and 5) multiple organ dysfunction occurred shortly after birth (two or more organ dysfunction, including cardiovascular, gastrointestinal, hematologic, pulmonary, or renal systems) . The exclusion criteria were 1) congenital malformations or major dysmorphic features, 2) congenital viral infections, and 3) defined metabolic syndromes. Respiratory dysfunction was defined as 1) respiratory rate > 90 breaths/min, 2) PaO 2 < 5.3 kPa in the absence of cyanotic disease, 3) PaCO 2 > 8.7 kPa, PaO 2 /FiO 2 < 200 Torr in the absence of cyanotic disease, or 4) mechanical ventilation . Cardiovascular dysfunction was defined as 1) systolic blood pressure < 40 mmHg, 2) heart rate < 50 or > 220 bpm, 3) cardiac arrest, 4) pH < 7.2 with normal PaCO 2 , or 5) continuous hemodynamic support . Hematologic dysfunction was defined as 1) hemoglobin < 5 g/L, white blood cells < 3 × 10 9 /L, 3) platelet counts < 20 × 10 9 /L, or 4) disseminated intravascular coagulation . Neurologic dysfunction was defined as 1) Glasgow coma score < 5 or 2) fixed dilated pupils . Hepatic dysfunction was defined as total bilirubin > 60 μmol/L . Gastrointestinal dysfunction was defined as 1) upper gastrointestinal bleeding or 2) any of hemoglobin by > 20 g/L, blood transfusion, hypotension <3rd percentile, or gastric or duodenal surgery . Renal dysfunction was defined as 1) blood urea nitrogen > 36 mmol/L, 2) serum creatinine > 177 μmol/L, or 3) dialysis/hemofiltration . Hypothermia was not used in this study, ruling out the influence of hypothermia on many test results. Mild hypothermia is the HIE grade A treatment evidence, but many hospitals in China do not have mild hypothermia treatment equipment. Different from Western countries, hospitals without such treatment equipment will transfer children to hospitals with medical equipment; but in China the medical system is continuing to improve and transport systems are not yet fully developed, such hospitals are rare, often far away, and the children are not transported in time. Still, all patients received appropriate ventilation and perfusion, maintaining blood glucose levels, and controlling convulsions and cerebral edema.
According to the hospital protocol after admission, two-channel monitoring of 8 electrodes aEEG (Bio-logic Ceegraphscan Netlink Monitor, NATUS Medical Incorporated, Mundelein, IL, USA) within 6 h after birth was recorded for 4 h . The aEEGs were assessed by two independent researchers blinded to the clinical data. The representative background pattern was determined using the scoring system suggested by Hellstrom-Westas et al. . The aEEG results were divided into three categories according to background activity of the neonatal aEEG: 1) normal amplitude: the boundary of the amplitude spectrum was > 10 μV, and the lower boundary was > 5 μV; 2) mildly abnormal amplitude: boundary > 10 μV and lower boundary ≤5 μV in the spectral band; 3) severe amplitude anomaly: the boundary of the spectral band was < 10 μV, and the lower boundary was ≤5 μV. All three categories might be associated with an epileptic activity. According to aEEG background activity and the presence or absence of epileptic activity, the aEEG results were divided into four categories: 1) normal aEEG with normal amplitude; 2) mildly abnormal aEEG with normal amplitude and epileptic activity; 3) mildly abnormal aEEG with normal amplitude; 4) all other combinations were severely abnormal aEEG. According to the needs of this study, the results were divided into two categories, normal and abnormal (including mild and severe abnormalities).
NSE was detected at 1 day (1-day NSE) and 3 days (3-day NSE) after birth. NSE was measured on a Cobas Core II Immunoanalyzer with the NSE EIA II kit (Roche Molecular Biochemicals, Mannheim, Germany), a one-step sandwich-type enzyme immunoassay that uses monoclonal mouse antibodies. The manufacturers claim a low detection limit of 0.1 μg/L for NSE.
MRI (T1- and T2-weighed imaging) scans were obtained within the 4–7 days after birth when the vital sign of the children was relatively stable. All infants were scanned with a 1.5-T Magnetom Avanto MRI system (Siemens, Erlangen, Germany), including MRI plain0 T1WI and T2WI scans. T1WI sequence parameters: RF pulse repetition time 585 ms, echo time 15 ms. T2WI sequence parameters: RF pulse repetition time 4000 ms, echo time 102 ms, a layer thickness of 4 mm, layer spacing of 0.4 mm, and field of view of 20 × 20 cm. All MRIs in this study were read independently by two radiologists who specialized in neuroimaging, and they discussed their findings together. The basal ganglia/watershed scores were determined as proposed by Barkovich et al. : 0 = no abnormalities in the basal ganglia or cortex; 1 = an abnormal signal in the basal ganglia or thalamus; 2 = an abnormal signal in the cortex; 3 = an abnormal signal in the cortex and basal nuclei (basal ganglia or thalami); 4 = an abnormal signal in the entire cortex and basal nuclei. All images and qualitative scores were assessed by an experienced radiologist, who was blinded to the serum NSE levels and clinical data.
Periodical follow-ups of the neurodevelopment outcomes were performed once a month before 6 months of age, every 2 months between 6 months and 1 year, and every 6 months after 1 year. At 18 months of age, the children were assessed in the neurological rehabilitation clinic according to the study protocol. All children received standardized assessment at 18 months. None of the children were lost to follow-up. If the evaluation could not be made according to the appointment time, telephone communication, or picking up the children at home to the hospital for evaluation, to avoid the loss of follow-up and data loss. The neurodevelopmental function was assessed with the Bayley Scales of Infant Development 2nd version (BSID-II) and neurological examination by a pediatric neurodevelopmental specialist. Motor testing at 1 year improves the prediction of motor and mental outcomes at 2 years after perinatal HIE . The pediatrist was not involved in the neonatal and postnatal care of the children under investigation and was blinded for the clinical status, histories, NSE levels, and neonatal brain MRI scores of the children. This examination took place at a time chosen by the mother as the most optimal time for the child to cooperate, mostly around 10 a.m. The children were classified into three groups depending on the examination results; normal (MDI and PDI > 79), moderate delay (MDI or PDI: 70–79), and severe delay (MDI or PDI < 70) . When normal and abnormal outcomes were predicted, excepting for MDI and PDI in BSID (both mild and severe abnormalities were classified as abnormal) were used to determine the neurodevelopment outcome, a diagnosis of cerebral palsy, visual impairment and hearing loss or epilepsy was also considered an abnormal outcome.
Statistical analysis was conducted using SPSS 17.0 (SPSS, NY, USA). All values are presented as mean ± standard deviation or median (range). The distributions of the continuous data were assessed using the Shapiro-Wilk tests. ANOVA and Tukey’s HSD post hoc test were used to analyze normally distributed data, while the Kruskal-Wallis test was used for non-parametric analyses. Categorical data are presented as n and were analyzed using the chi-square test. A logistic regression model was fit to the binary outcome using aEEG, NSE, and brain MRI together as predictors for neurodevelopment delay. Receiver operating characteristic (ROC) analysis was used to evaluate the predictive accuracy of the logistic regression model, as well as aEEG, NSE, and brain MRI, separately and together. Sensitivity, specificity, PPV, and NPV of aEEG, NSE, and brain MRI, separately and together, were calculated using binomial exact method. P -values < 0.05 were considered statistically significant.
Characteristics of the patients A total of 50 HIE neonates were recruited and followed during the study period (Fig. ). The neonates were divided into three groups based on according to Bayley II scores: 32 children (64.0%) were in the normal group; 5 (10.0%) and 13 (26.0%) children were in the moderate and severe neurodevelopmental delay group, respectively. The characteristics of the study population are shown in Table . There were no significant differences in sex, birth weight, gestational age, Apgar scores, and HIE severity among the children with three different neurodevelopment outcomes. Six neonates had recurrent seizures, which were significantly associated with poor prognosis ( P < 0.001), and 12 had single clinical or subclinical seizures (six had poor outcomes, and six had good outcomes). The antiepileptic drugs used were mainly phenobarbital sodium and midazolam. More infants with severe delay were intubated or ventilated ( P = 0.046) or received epinephrine ( P = 0.030) in the delivery room. The BSID-II MDI and PDI scores decreased with increasing severity of delay (both P < 0.001) (Table ). aEEG, NSE levels, and MRI of the neonates with HIE Of the 50 neonates with HIE, 42 (84.0%) had abnormal aEEG, and eight (16.0%) had normal aEEG. Serum 1-day NSE levels were not different among the groups ( P = 0.929), but serum 3-day NSE levels increased with worsening neurodevelopment outcomes at 18 months of age (normal: 20.52 ± 6.42 μg/L vs. moderate: 39.82 ± 5.92 μg/L vs. severe: 44.60 ± 9.01 μg/L, P < 0.001; P < 0.01 normal vs. moderate, P < 0.01 normal vs. severe, P = 0.147 moderate vs. severe; 22 were above the cutoff value and 28 were below) (Table and Fig. A). The MRI findings at 4–7 days after birth were significantly different among the three groups ( P < 0.001) (Table ). The predictive values of aEEG, 3-day NSE levels, and brain MRI for neurodevelopmental outcomes As shown in Table , aEEG had high sensitivity (100%) but poor specificity (25.00%) for the neurodevelopmental outcomes at 18 months. MRI had good sensitivity (94.44%) and fair specificity (71.88%), while 3-day NSE had high sensitivity (100%) and specificity (87.50%). 3-day NSE levels were evaluated by ROC curve analysis, and the cut-off value that maximized sensitivity and specificity for the moderate or severe developmental delay was 27.3 μg/L (Fig. B). The combination of the three variables (high NSE, abnormal aEEG, and MRI abnormalities together) had a sensitivity of 100%, specificity of 97.70%, PPV of 98.25%, and NPV of 99.98% (Table ).
A total of 50 HIE neonates were recruited and followed during the study period (Fig. ). The neonates were divided into three groups based on according to Bayley II scores: 32 children (64.0%) were in the normal group; 5 (10.0%) and 13 (26.0%) children were in the moderate and severe neurodevelopmental delay group, respectively. The characteristics of the study population are shown in Table . There were no significant differences in sex, birth weight, gestational age, Apgar scores, and HIE severity among the children with three different neurodevelopment outcomes. Six neonates had recurrent seizures, which were significantly associated with poor prognosis ( P < 0.001), and 12 had single clinical or subclinical seizures (six had poor outcomes, and six had good outcomes). The antiepileptic drugs used were mainly phenobarbital sodium and midazolam. More infants with severe delay were intubated or ventilated ( P = 0.046) or received epinephrine ( P = 0.030) in the delivery room. The BSID-II MDI and PDI scores decreased with increasing severity of delay (both P < 0.001) (Table ).
Of the 50 neonates with HIE, 42 (84.0%) had abnormal aEEG, and eight (16.0%) had normal aEEG. Serum 1-day NSE levels were not different among the groups ( P = 0.929), but serum 3-day NSE levels increased with worsening neurodevelopment outcomes at 18 months of age (normal: 20.52 ± 6.42 μg/L vs. moderate: 39.82 ± 5.92 μg/L vs. severe: 44.60 ± 9.01 μg/L, P < 0.001; P < 0.01 normal vs. moderate, P < 0.01 normal vs. severe, P = 0.147 moderate vs. severe; 22 were above the cutoff value and 28 were below) (Table and Fig. A). The MRI findings at 4–7 days after birth were significantly different among the three groups ( P < 0.001) (Table ).
As shown in Table , aEEG had high sensitivity (100%) but poor specificity (25.00%) for the neurodevelopmental outcomes at 18 months. MRI had good sensitivity (94.44%) and fair specificity (71.88%), while 3-day NSE had high sensitivity (100%) and specificity (87.50%). 3-day NSE levels were evaluated by ROC curve analysis, and the cut-off value that maximized sensitivity and specificity for the moderate or severe developmental delay was 27.3 μg/L (Fig. B). The combination of the three variables (high NSE, abnormal aEEG, and MRI abnormalities together) had a sensitivity of 100%, specificity of 97.70%, PPV of 98.25%, and NPV of 99.98% (Table ).
Neonatal HIE is an important cause of mortality and morbidity , but effective indicators for the early diagnosis of brain injury after asphyxia and prognosis are lacking . Therefore, this study aimed to examine the predictive value of serum NSE, aEEG, and MRI, alone and in combination, for the neurological outcomes in neonates with HIE. The results suggest that the combination of MRI, aEEG and 3-day NSE can predict the neurological prognosis of newborns with HIE. The combination of the three methods can improve the predictive ability for long-term neurobehavioral prognosis. Previous studies examined the value of the Sarnat score that could be used to determine the prognosis of the infants; Polat et al. and Liauw et al. showed that the PPV and NPV of the Sarnat staging for the adverse outcome (disability and death) were 52 and 100%, while the PPV and NPV were respectively 45 and 0% in children with HIE grade II . In the present study, the Sarnat staging had difficulty in assessing neurodevelopment at 18 months due to the multifactor dependency of HIE, as supported by the above studies. Therefore, more reliable methods are needed. The Thompson score has been shown to be of prognostic value in infants with HIE but was not assessed in this study. In this study, HIE grading was based on Sarnat grading but could be affected by a variety of clinical factors. Sarnat grading is not good in predicting long-term outcomes, and this study discusses other predictive indicators. This is consistent with Shankaran et al. . Elevated NSE levels have been reported after stroke, brain injury, cardiac surgery, cardiopulmonary arrest, and perinatal asphyxia . Kunda et al. showed that important increases in NSE levels could be observed after seizures in patients with temporal lobe epilepsy, suggesting neuronal damage, but the cause-to-effect relationship could not be established. Lee et al. suggested that NSE levels could be used for the differential diagnosis of seizures over syncope. In addition, Sheik et al. showed that NSE is elevated in patients with acute diseases of the central nervous system, and that it was elevated in patients with seizures vs. those without. Therefore, NSE is associated with brain damage and seizures, but whether NSE levels increases before or after seizures remain to be determined. HIE is a fetal or neonatal brain injury caused by decreased or suspended cerebral blood flow, with partial or complete asphyxia resulting from perinatal hypoxia. At the cellular level, decreased blood flow and oxygen delivery initiates a series of harmful biochemical reactions. Oxygen consumption interferes with oxidative phosphorylation, which leads to the conversion to anaerobic metabolism, rapid depletion of high-energy phosphate reserves, accumulation of lactic acid, and failure to maintain cellular function . Cell membranes are damaged, and intracellular components of neurons, including NSE, are released into the cerebrospinal fluid and blood. In this study, increased NSE levels were observed at 3 days in neonates who eventually showed neurodevelopmental delay. It is supported by studies that suggested the value of NSE for HIE . Recent studies in asphyxiated neonates have correlated NSE with developmental outcomes, but the conclusions were conflicting because of different detection timing because it takes some time for the damaged cells to die and release their content , supporting the significant difference at 3 days but not at 1 day in the present study. Kecskes et al. also reported that NSE levels at 3 days had the best predictive value. Still, it is unknown whether the higher NSE levels at 3 days compared with 1 day is the result of delayed NSE release in the blood compartment or the result of the ongoing injury process. This will have to be examined in future studies. The abnormal electrical activity in brain regions can be diagnosed by aEEG. Nakamura et al. reported that electrical changes could occur in damaged brain regions within 10–20 min after a hypoxic-ischemic attack, which can be detected by aEEG monitoring. The present study suggested that a persistent abnormal aEEG over 4 h and within 6 h after birth was associated with adverse neurodevelopmental outcomes. Still, the PPV of 6-h aEEG was poor (PPV of 42.9%), and good outcomes might occur despite an abnormal aEEG. Therefore, an abnormal aEEG within 6 h after birth might have no clinical value if it is used alone. Conversely, a normal 6-h aEEG has a good NPV (100%). These results are supported by a previous study , but Vasiljevic et al. showed higher specificity and PPV than in the present study. Their study showed the pattern of abnormal aEEG correlated with the severity of HIE ( P < 0.0001) and subsequent neurodevelopmental outcome ( P < 0.001) . They found that abnormal aEEG background patterns exhibited superior prediction of abnormal developmental outcomes at 12 months of age. Still, the discrepancies between the present and previous studies could be due to differences in monitoring time (within the first 72 h of life in their case ). aEEG is a useful tool for the monitoring and detection of seizures . In neonates with seizures, more than two seizures detected by aEEG is predictive of progression to early-onset epileptic encephalopathy . Still, the exact relationship between seizures and aEEG abnormalities could not be determined in the present study. Future studies will have to examine the exact role of seizures, duration of seizures, and seizure burden in infants with HIE. Conventional MRI is an essential tool for assessing the neonatal brain injury and degree of injury, but sometimes T1- and T2-weighted images can appear normal during the few days after birth or show only subtle findings that might be difficult to interpret or to distinguish from normal maturational phenomena in the neonatal brain. The most appropriate time for examination is 4–7 days after birth . The present study showed that in term infants with HIE, conventional MRI detection 4–7 days after birth can be used for the relatively accurate prediction of neurodevelopment outcomes at 18 months old (sensitivity, specificity, PPV, and NPV, respectively were 94.44, 71.88, 65.38, and 95.83%). Still, whether a combination of NSE, MRI, and EEG could improve the prognostication of HIE remains to be examined. Indeed, Leon-Lozano et al. and Shim et al. recently reported NSE levels, MRI, and aEEG (all assessed within the same study) on the outcomes of HIE, but without combining the different modalities. A previous study published by Ezgu et al. in 2002 suggested that the combination of NSE, MRI, and EEG could be used to determine the prognosis of HIE, but they measured NSE in the cerebrospinal fluid, which is less practical than from the blood. In the present study, consistent relationships between the outcomes at 18 months and the injury degree on conventional MRI, aEEG results, and serum NSE levels of 3 days after birth (r s = 0.719, P < 0.001) were observed. When dichotomizing the NSE levels based on the best cut-off value and when considering MRI and aEEG as normal/abnormal, combining the three test results in better sensitivity, specificity, PPV, and NPV than for each test alone. This possible predictive tool should be validated in a larger group of patients. In this study, 27% of the children with mild HIE had a severe delay. Most previous studies used the traditional clinical Sarnat grading system, in which infants with mild Sarnat grading had a good prognosis and no long-term disability. For this reason, many studies did not carry out a systematic examination and long-term follow-up on neonates with mild HIE. Still, more and more studies are finding that they can experience major disability, and when followed up to 18 months, about 25% of infants with mild HIE have abnormal outcomes, defined as death or movement and/or developmental delay . Nevertheless, it is not clear why a minor injury would lead to such severe adverse outcomes. It might be that early hypoxia injury will change, worsen, and expand over time. Since a significant proportion of children with mild encephalopathy have been reported to have significant long-term neurological sequelae, more measures are needed to evaluate brain injury and better predict prognosis in addition to Sarnat grading. There are limitations of this study. The sample size was small and from a single center. Long-term outcomes were not followed. A larger sample size, multiple centers, and longer clinical follow-up are needed to confirm the findings. In addition, additional indicators of electrophysiological and biochemical changes of neurons after hypoxia and ischemia should be explored. Of note, the MRI examinations were read by a general radiologist, not a pediatric radiologist or a neuroradiologist. Due to copyright issues, Bayley III cannot be translated, standardized, and validated in Chinese for the time being; therefore, Bayley II was used. Finally, in this study,the mortality among moderate/severe HIE infants was much lower than seen in the RCTs of therapeutic hypothermia and in subsequent prospective studies, part of the reason is that the special and complex doctor-patient relationship in China. When there is subjective evaluation for disease severity, Chinese doctors tend to classify them as severe cases, which is relatively common and may be different from other countries, may have had some effect on the results. In conclusion, in the earlier stages after birth, aEEG, MRI, and 3-day NSE levels can be used to predict the 18-month neurodevelopmental outcomes of children with HIE not treated with hypothermia treatment. These tests might have a use in predicting prognosis in full-term infants with HIE, providing them with more accurate management strategies.
|
Strategic allocation of working memory resource | a49a7b0c-cd8e-4716-98f8-17343809ec3c | 6212458 | Physiology[mh] | One of the hallmarks of VWM is that it is supported by a limited resource. In natural environments, where objects vary in how relevant they are, the process by which our memory resource is allocated appears flexible and strategic. Indeed, experiments demonstrate that increasing the behavioral relevance of a set of items results in better memory for those items – . Yet, it is still unknown how people decide how much resource to allocate to the encoding and storing of items with different behavioral relevancies. Here, our overall objective is to use computational models of VWM performance to understand the strategy by which memory resource is flexibly allocated when items vary in behavioral relevance. To do so, we first established that the amount of allocated resource is monotonically related to the behavioral relevance, or priority, of memorized items. We used a memory-guided saccade task in which, on each trial, participants remembered the location of four dots, one in each visual quadrant (Fig. ). To operationalize resource prioritization, we used a precue to indicate the probability that each dot would be later probed. On each trial, the probe probabilities were 0.6 (“high”), 0.3 (“medium”), 0.1 (“low”), and 0.0. After a variable delay period, one of the quadrants was cued and the participant made a saccade to the remembered location of the dot within that quadrant. For every trial, we computed the Euclidean distance, in degrees of visual angle, between the true and reported target location. We conducted a repeated-measures ANOVA with priority condition as the within-subject variable. In line with our hypothesis, error decreased monotonically with increasing priority ( F (1.18,15.37) = 10.95, p = 0.003), reflecting the intuition that people allocate more resource to a more behaviorally relevant target (Fig. ). Next, we asked what strategy people use to allocate resource in response to unequal relevance. Emrich et al . proposed that resource is allocated in approximate proportion to the probe probabilities. Bays proposed that resource is allocated such that the expected squared error is minimized. Sims proposed more generally that resource is allocated to minimize loss. Thus, our second goal was to compare computational models of resource allocation. We tested variable-precision models of estimation errors , augmented with different resource allocation strategies. Memory precision for a given item is a random variable whose mean depends on the item’s priority (middle and bottom panels of Fig. ; see Supplementary for more a detailed description of the model). In the Proportional model, the amount allocated to an item is proportional to the item’s probe probability. This model provided a poor fit to the data (left panel of Fig. ), suggesting that people do not allocate resource in proportion to probe probability. Perhaps this model was too constrained, so we tested the Flexible model, in which the proportions allocated to each priority condition were free parameters. We compared models using the corrected Akaike Information Criterion (AICc) and the Bayesian Information Criterion (BIC). Both AICc and BIC penalize models with more parameters, but BIC is more conservative. The Flexible model fit the data well (middle panel of Fig. ) and formal model comparison showed that it outperformed the Proportional model (median ΔAICc [bootstrapped 95% CI]: 63 [37, 107], ΔBIC: 54 [29, 99]). The proportions allocated to the high-, medium-, and low-priority targets were estimated as 0.49 ± 0.04 ( M ± SEM ), 0.28 ± 0.02, and 0.23 ± 0.03, respectively (Fig. ), suggesting that the brain underallocates resource to high-priority targets and overallocates resource to low-priority targets, relative to the experimental probe probabilities. The Flexible model offered a good explanation for how much participants were allocating to each item, but not why . We considered that resource was allocated to minimize expected loss, where loss is defined as estimation error to a power , , , . In this Minimizing Error model, resource allocation differs substantially from the Proportional model . An observer with limited resource should allocate their resource more equally than proportional (Supplementary contains a section showing how optimal allocation changes with total resource). Such a strategy would lower the probability of very large errors for low-priority targets, at a small expense of the high-priority targets (top panel of Fig. ). The exponent on the error serves as a “sensitivity to error” parameter: an observer with a large exponent will experience a large error as much more costly than an observer with a lower exponent, and will adjust their strategy accordingly to avoid those errors. The Minimizing Error model fits better than the Proportional model (median ΔAICc [bootstrapped 95% CI]: 49 [21, 99], ΔBIC: 44 [17, 94]. Figure ) and comparably to the Flexible model (ΔAICc: −7 [−30, −1], ΔBIC: −3 [−26, 3]). Additionally, the model estimated an allocation of resource similar to the Flexible model (0.46 ± 0.02, 0.32 ± 0.01, and 0.22 ± 0.02 for high-, medium-, and low-priority targets, respectively). This suggests that the under- and over-allocation of resources relative to probe probabilities may be rational, stemming from an attempt to minimize error across the experiment. The first experiment showed that prioritizing items affects memory representations, and that people allocate memory resource in an error-minimizing way. However, this experiment, along with much of the VWM literature, overlooks other information available in VWM: memory uncertainty. Indeed, people can successfully report on the quality of their memory-based decisions , suggesting a representation and use of uncertainty over the memorized stimulus – . We conducted a second experiment to investigate how, if at all, priority affects working memory uncertainty. We tested this with a very similar memory-guided saccade task with an addition wager to measure uncertainty. After the participant made a saccade, a circle appeared centered at the endpoint of the saccade (Fig. ). Participants made a wager by adjusting the size of the circle with the goal of enclosing the true target location within the circle. If successful, they received points based on the size of the circle, such that a smaller circle corresponded to more points. In unsuccessful, they received no points. This procedure served as a measure of memory uncertainty because participants were incentivized to make smaller circles when their memory was more certain. Our predictions for this experiment were the following: (a) estimation error decreases with increasing priority, (b) circle size decreases with increasing priority, and c) estimation error correlates positively with circle size within each priority level. To test the first two predictions, we conducted a repeated-measures ANOVA with priority condition as the within-subject variable. The ANOVA for circle size violated the assumption of sphericity, so we implemented a Greenhouse-Geisser correction. For the third prediction, we conducted a Spearman correlation for each priority condition, computing correlations across participants as well as for individual participants. For the correlation across participants, we removed any participant-specific main effects by standardizing the data ( M = 0, SD = 1) for each participant before aggregating data for each priority condition. We confirmed all three predictions. First, estimation error decreased monotonically with increasing priority ( F (2,20) = 12.5, p < 0.001, η 2 = 0.55; left panel of Fig. ), indicating that participants allocated more resource to higher priority targets. Second, the radii of the circle wagers decreased monotonically with increasing priority ( F (1.3,12.9) = 10.60, p < 0.005, η 2 = 0.51; middle panel of Fig. ), indicating that participants had higher memory certainty in higher priority trials. Third, estimation error and circle size were correlated within each priority level across participants ( ρ 0.6 = 0.22, p < 0.001; ρ 0.3 = 0.28, p < 0.001; ρ 0.1 = 0.18, p < 0.001; right panel of Fig. ). Correlations at the individual level resulted in similar correlation values ( M ± SEM : ρ 0.6 = 0.22 ± 0.04, ρ 0.3 = 0.27 ± 0.03, ρ 0.1 = 0.16 ± 0.04), though not all correlations were significant. These positive correlations indicate that people have a single-trial representation of their uncertainty independent of the priority manipulation, as suggested by earlier work , . This correlation, however, could be driven by some other factor, such as stimulus location or trial delay time. Perhaps stimuli closer to cardinal axes are remembered more precisely than those presented obliquely – , and knowledge of this effect could be driving the measured within-priority correlation. An effect of delay on error, and knowledge of this, could also be driving the correlation. To test these hypotheses, we conducted two permutations test for each participant and priority level (details in Supplementary). We found that the actual correlations ( M ± SEM: 0.29 ± 0.04) were significantly higher than the median of the correlations obtained in the null distribution when permuting based on stimulus location ( M ± SEM: −0.007 ± 0.006; Wilcoxon signed-rank test, z = −4.69, p < 1e-5) or delay time ( M ± SEM: −0.004 ± 0.004; z = −4.53, p < 1e-5), suggesting that the correlation within each priority condition was driven by knowledge of internal fluctuations in the quality of the memory representation above and beyond any location- or delay-dependent variation. We extended the computational models from the first experiment to account for the additional wager data. The observer uses trial-to-trial knowledge of memory quality to calculate the probability that the target lies within the circle of a proposed size. The observer multiplies this value by the utility of that circle size to calculate the expected utility (Fig. ). We assume the observer noisily chooses the circle size that maximizes the expected utility (Fig. ; details in Supplementary). We again tested the Proportional model and Flexible model, jointly fitting the estimation data and the post-estimation wager data. We again found that the Proportional model did not provide a good fit to human data and the Flexible model provided an excellent fit to the data (left two columns of Fig. ). As before, the Flexible model suggests that the brain underallocates resource to high-priority targets and overallocates resource to low-priority targets relative to experimental probe probabilities. The proportion allocated to the high-, medium-, and low-priority targets were estimated as 0.44 ± 0.02, 0.31 ± 0.02, and 0.25 ± 0.02, respectively (Fig. ). Given the analogous results, we again asked if we could describe a normative model for the resource allocation strategy. Unlike in the first experiment, optimal performance in this experiment requires maximizing points. This Maximizing Points model has qualitatively different properties from the Minimizing Error model. An observer that maximizes points would receive more points by ignoring the low-priority targets completely in order to remember the high-priority targets better, while an observer that minimizes error would allocate it more evenly across targets. Because these two strategies conflict, we are able to test whether the intrinsically-driven, error-minimizing strategy that people seem to be using in the absence of reward can withstand being put in conflict with an external incentive. To our surprise, the Maximizing Points model fit very poorly (right column Fig. ), indicating that participants were not allocating resource in order to earn the most points (Proportional model: median ΔAICc: −75 [−109, −26], ΔBIC: −75 [−109, −26]; Flexible model: ΔAICc: −156 [−308, −94], ΔBIC: −148 [−300, −86]). Perhaps participants were still acting in accordance with the Minimizing Error model. In this experiment, this strategy is myopic: the observer allocates resource to minimize error in the estimation without considering how this allocation may affect the points of the wager. Nonetheless, the Minimizing Error model fit the data substantially better than the Proportional model (median ΔAICc: 55 [20, 106], ΔBIC: 50 [17, 102]) and the Maximizing Points model (ΔAICc: 140 [85, 249], ΔBIC: 137 [81, 245]) and about as well as the Flexible model (ΔAICc: −16 [−44, −5], ΔBIC: −12 [−40, 0]; third column Fig. ). The Minimizing Error model fitted the proportions of resource allocated to high-, medium-, and low-priority targets as 0.52 ± 0.02, 0.32 ± 0.01, and 0.16 ± 0.01, respectively, similar to the allocation estimated in the Flexible model. In this work, we examined how people allocate resource in a task with items of varying behavioral relevance. First, we found that people flexibly allocate resource according to behavioral relevance. This is comforting: we remember more important things better. Second, we found accurate knowledge of both priority-driven and spontaneous trial-to-trial fluctuations in memory quality, even when controlling for spatial location and delay time. This is also comforting: to some extent, we can trust our confidence in our memories. This knowledge is useful when deciding whether to use or resample information. Third, we explained not just how people allocate VWM resource, but what strategy they may be using. We find that people minimize estimation error, a strategy whose consequence is an overallocation to low- and underallocation to high-priority targets relative to probe probabilities. Fourth, this strategy persists even when presented with a conflicting external reward. Perhaps we find minimizing the errors of our memory intrinsically rewarding. Rewards influence the metrics of saccades in humans and monkeys , . For instance, extrinsic rewards affect both the velocity of saccades as well as neural activity in dopamine-associated reward circuits , and they modulate neural activity in cortical areas that represent the goals of saccade plans . Perhaps the intrinsic reward associated with veridical memory eclipses the extrinsic reward associated with gaining more points, which would explain why people minimized error instead of maximized points in the second experiment. Additionally, minimizing memory error might be computationally easier than maximizing points because it does not require the observer to think and optimize performance two steps ahead. In other words, the amount that performance may improve from maximizing points may not be worth the computational and metabolic cost. Future studies should investigate if resource allocation abides by the Minimizing Error model across a variety of experimental probe probabilities (see Supplementary for how optimal resource allocation should change) and reward contexts (e.g., monetary reward ). Our results identify a single, simple model of how the resource that supports VWM is allocated despite the large variability in WM abilities across individuals and age , . For example, electrophysiological signals measured at the scalp predict individual differences in WM capacity as well as trial-to-trial variation in the precision of WM , . Individual differences in WM can also be explained by differences in control processes, such as inhibition of irrelevant distractors . Our model accounts for these individual differences by explicitly assuming inter-trial variability, as well as having parameters that account for participants’ differences in total amount of resource as well as sensitivity to error. Some participants prefer making a few large errors in order to maximize the number of extremely precise memory guided saccades, while others prefer avoiding large errors at the expense of those precise saccades. While we use the Variable Precision model to study resource allocation in this paper, we do not make any claims about whether this model is the best computational model to capture the data. Its assumptions regarding resource allocation are less constrained, but alternative models such as an Interference or the Slots Plus Averaging model may still provide a good fit. Future research could investigate these models as well. In everyday life, we are bombarded with information constantly, and we have to decide what to look at, pay attention to, and remember. This study finds that not only do people remember more important items and their associated uncertainty, but they also do so in a way that minimizes the overall magnitude of memory errors. Perhaps this strategic allocation is how we are able to function so well despite such limited working memory resource.
Participants Fourteen participants (5 males, mean age = 30.3, SD = 7.2) participated in Experiment 1 and eleven (5 males, age = 28.6, SD = 3.03) in Experiment 2. Everyone had normal or corrected-to-normal vision and no history of neurological disorders. Participants were naive to the study hypotheses and were paid $10/hour. We obtained informed, written consent from all participants. The study was in accordance with the Declaration of Helsinki and was approved by the Institutional Review Board of New York University. Apparatus Participants were placed 56 cm from the monitor (19 inches, 60 Hz), with their heads in a chinrest. Eye movements were calibrated using the 9-point calibration and recorded at a frequency of 1000 Hz (Eyelink 1000, SR Research). Target stimuli were programmed in MATLAB (MathWorks) using the MGL toolbox (Gardner Lab, Stanford) and were displayed against a uniform grey background. In Experiment 2, participants made behavioral responses using a space bar with their left hand and a circular knob (PowerMate, Griffin Technology) with their right hand. For eye-tracking, we applied an online drift correction when the recorded location of center of fixation exceeded 1 degree of visual angle from the center of the fixation cross. This correction was implemented to remove the participants’ discomfort from discrepancies between the actual and measured location of fixation, because this experiment provided live visual feedback of the participants’ current fixation. The online drift correction was not implemented in Experiment 1 because there was no feedback; the corrective saccade provided a measure of drift, which we used offline to correct position. Trial procedure Each trial (Fig. ) began with a 300 ms increase in the size of the fixation symbol, an encircled fixation cross. This was followed by a 400 ms endogenous precue, consisting of three colored wedges presented within the fixation symbol, each of which angularly filled one quadrant. The radial sizes and colors (pink, yellow, and blue) of the wedges corresponded to probe probabilities of 0.6, 0.3, and 0.1, respectively. The quadrant with a probe probability of 0.0 did not have a wedge. The precue was followed by a 700 ms interstimulus interval, then by the targets, presented for 100 ms. The targets were four dots, each in separate visual quadrants. The dots were presented at approximately 10 degrees of visual angle from fixation, with random jitter of 1 degree of visual angle to each location. The location of the targets in polar coordinates were pseudo-randomly sampled from every 10 degrees, avoiding cardinal axes. This was followed by a variable delay, chosen with equal probability from the range between 1000 and 4000 ms in 500 ms increments. A response cue appeared afterward, which was a white wedge that filled an entire quadrant of the fixation symbol. Participants were instructed to saccade to the remembered dot location within the corresponding quadrant of the screen. If participants took shorter than 100 ms or longer than 1200 ms to make the saccade, the trial was discarded. In Experiment 1, after the saccade, the actual dot location was presented as feedback and the participant made a corrective saccade to that location. After 500 ms, the feedback disappeared, participants returned their gaze to the central fixation cross, and a 1500 ms inter-trial interval began. In Experiment 2, simultaneously with the response cue, a red dot appeared at the location of the participants’ fixation as measured online by the eye tracker. Because of eyetracker noise, the red dot occasionally appeared in a slightly different location than where the participant was fixating. In these cases, participants were instructed to adjust their gaze such that the red dot was at the remembered location, and press the space bar to indicate that this was their intended saccade endpoint. After completing this response, participants performed a post-estimation wager. A circle appeared, centered at the saccade endpoint. Participants received points based on the size of the circle, such that a smaller circle corresponded to more points. However, participants were only rewarded points if the true target was within the circle. The number of points awarded was 120 e −0.4 r , in which r was the radius of the circle. Data Processing Processing and manual scoring of eye movement data was performed in an in-house MATLAB function-graphing toolbox (iEye). Eye position and saccadic reaction time (SRT) were extracted from iEye. Statistical analyses were performed in MATLAB (Mathworks) and SPSS (IBM). We excluded trials in which (a) participants were not fixating in the middle of the screen during stimulus presentation, (b) saccades were initiated before 100 ms or after 1200 ms after the response cue onset, (c) pupil data during the response period were missing, or (d) participants made a saccade to the wrong quadrant, ignoring the response cue. This resulted in removing between 1% and 7% of trials per subject. Raw gaze positions were transformed offline into degrees of visual angle using a third order polynomial algorithm that fit eye positions to known spatial locations and we used gaze velocity trace to determine the onset and offset of saccades with a 30°/s threshold. Priority effects were not significantly different between initial and final saccade position, so we report the results for the final saccades. Model fitting, Parameter Recovery, and Model Recovery For each participant and each model, we estimated the parameters using maximum-likelihood estimation. The likelihood of the parameters are defined as p (data|model, θ), in which θ is a vector of the model parameters. To calculate the parameter likelihood, we use numerical integration to marginalize over the internal variables x and J (Supplementary Information). To find the maximum-likelihood parameter estimate, we used the optimization algorithm Bayesian Adaptive Direct Search in MATLAB, which combines mesh grid and Bayesian optimization methods. We completed 50 optimizations with different starting values for each participant and model, to ensure the obtained estimates were not a result of a local minimum. We took the maximum of all the runs as our estimate of the maximum-likelihood, and the corresponding parameter combination as our ML parameter estimates. To validate the data-generating and model-fitting code, we performed parameter and model recovery. We simulated data from each model then fit each model to the simulated data. Successful parameter recovery occurs when the estimated parameters for the model that generated the data are equivalent or close to the true parameters. Parameter recovery is necessary for the interpretability of the parameter estimates. Successful model recovery occurs when the model which generated the data also fits the data better than any other model. Model recovery is necessary to ensure the models are distinguishable in a psychologically plausible model space. We successfully recovered both the parameters and models for each model. Code Availability All analyses were conducted in MATLAB. Code used to fit and compare models and generate figures are available on the following github repository: github.com/aspenyoo/WM_resource_allocation. Code used to process, visualize, and score eye-tracking data are available on the Curtislab github repository: github.com/clayspacelab/iEye.
Fourteen participants (5 males, mean age = 30.3, SD = 7.2) participated in Experiment 1 and eleven (5 males, age = 28.6, SD = 3.03) in Experiment 2. Everyone had normal or corrected-to-normal vision and no history of neurological disorders. Participants were naive to the study hypotheses and were paid $10/hour. We obtained informed, written consent from all participants. The study was in accordance with the Declaration of Helsinki and was approved by the Institutional Review Board of New York University.
Participants were placed 56 cm from the monitor (19 inches, 60 Hz), with their heads in a chinrest. Eye movements were calibrated using the 9-point calibration and recorded at a frequency of 1000 Hz (Eyelink 1000, SR Research). Target stimuli were programmed in MATLAB (MathWorks) using the MGL toolbox (Gardner Lab, Stanford) and were displayed against a uniform grey background. In Experiment 2, participants made behavioral responses using a space bar with their left hand and a circular knob (PowerMate, Griffin Technology) with their right hand. For eye-tracking, we applied an online drift correction when the recorded location of center of fixation exceeded 1 degree of visual angle from the center of the fixation cross. This correction was implemented to remove the participants’ discomfort from discrepancies between the actual and measured location of fixation, because this experiment provided live visual feedback of the participants’ current fixation. The online drift correction was not implemented in Experiment 1 because there was no feedback; the corrective saccade provided a measure of drift, which we used offline to correct position.
Each trial (Fig. ) began with a 300 ms increase in the size of the fixation symbol, an encircled fixation cross. This was followed by a 400 ms endogenous precue, consisting of three colored wedges presented within the fixation symbol, each of which angularly filled one quadrant. The radial sizes and colors (pink, yellow, and blue) of the wedges corresponded to probe probabilities of 0.6, 0.3, and 0.1, respectively. The quadrant with a probe probability of 0.0 did not have a wedge. The precue was followed by a 700 ms interstimulus interval, then by the targets, presented for 100 ms. The targets were four dots, each in separate visual quadrants. The dots were presented at approximately 10 degrees of visual angle from fixation, with random jitter of 1 degree of visual angle to each location. The location of the targets in polar coordinates were pseudo-randomly sampled from every 10 degrees, avoiding cardinal axes. This was followed by a variable delay, chosen with equal probability from the range between 1000 and 4000 ms in 500 ms increments. A response cue appeared afterward, which was a white wedge that filled an entire quadrant of the fixation symbol. Participants were instructed to saccade to the remembered dot location within the corresponding quadrant of the screen. If participants took shorter than 100 ms or longer than 1200 ms to make the saccade, the trial was discarded. In Experiment 1, after the saccade, the actual dot location was presented as feedback and the participant made a corrective saccade to that location. After 500 ms, the feedback disappeared, participants returned their gaze to the central fixation cross, and a 1500 ms inter-trial interval began. In Experiment 2, simultaneously with the response cue, a red dot appeared at the location of the participants’ fixation as measured online by the eye tracker. Because of eyetracker noise, the red dot occasionally appeared in a slightly different location than where the participant was fixating. In these cases, participants were instructed to adjust their gaze such that the red dot was at the remembered location, and press the space bar to indicate that this was their intended saccade endpoint. After completing this response, participants performed a post-estimation wager. A circle appeared, centered at the saccade endpoint. Participants received points based on the size of the circle, such that a smaller circle corresponded to more points. However, participants were only rewarded points if the true target was within the circle. The number of points awarded was 120 e −0.4 r , in which r was the radius of the circle.
Processing and manual scoring of eye movement data was performed in an in-house MATLAB function-graphing toolbox (iEye). Eye position and saccadic reaction time (SRT) were extracted from iEye. Statistical analyses were performed in MATLAB (Mathworks) and SPSS (IBM). We excluded trials in which (a) participants were not fixating in the middle of the screen during stimulus presentation, (b) saccades were initiated before 100 ms or after 1200 ms after the response cue onset, (c) pupil data during the response period were missing, or (d) participants made a saccade to the wrong quadrant, ignoring the response cue. This resulted in removing between 1% and 7% of trials per subject. Raw gaze positions were transformed offline into degrees of visual angle using a third order polynomial algorithm that fit eye positions to known spatial locations and we used gaze velocity trace to determine the onset and offset of saccades with a 30°/s threshold. Priority effects were not significantly different between initial and final saccade position, so we report the results for the final saccades.
For each participant and each model, we estimated the parameters using maximum-likelihood estimation. The likelihood of the parameters are defined as p (data|model, θ), in which θ is a vector of the model parameters. To calculate the parameter likelihood, we use numerical integration to marginalize over the internal variables x and J (Supplementary Information). To find the maximum-likelihood parameter estimate, we used the optimization algorithm Bayesian Adaptive Direct Search in MATLAB, which combines mesh grid and Bayesian optimization methods. We completed 50 optimizations with different starting values for each participant and model, to ensure the obtained estimates were not a result of a local minimum. We took the maximum of all the runs as our estimate of the maximum-likelihood, and the corresponding parameter combination as our ML parameter estimates. To validate the data-generating and model-fitting code, we performed parameter and model recovery. We simulated data from each model then fit each model to the simulated data. Successful parameter recovery occurs when the estimated parameters for the model that generated the data are equivalent or close to the true parameters. Parameter recovery is necessary for the interpretability of the parameter estimates. Successful model recovery occurs when the model which generated the data also fits the data better than any other model. Model recovery is necessary to ensure the models are distinguishable in a psychologically plausible model space. We successfully recovered both the parameters and models for each model.
All analyses were conducted in MATLAB. Code used to fit and compare models and generate figures are available on the following github repository: github.com/aspenyoo/WM_resource_allocation. Code used to process, visualize, and score eye-tracking data are available on the Curtislab github repository: github.com/clayspacelab/iEye.
Supplementary Information
|
Universal protection of allogeneic T-cell therapies from natural killer cells via CD300a agonism | 244d8410-c602-4e1f-a805-175850c3f6b7 | 11782830 | Surgical Procedures, Operative[mh] | Off-the-shelf T and chimeric antigen receptor (CAR) T-cell therapies promise to build on the success of its autologous counterparts while providing patients a well-defined, scalable, and cost-efficient drug product that can be administered on demand. However, host immune rejection is a major obstacle that limits their persistence, efficacy, and redosing potential. Although elimination of donor HLA class I (HLA-I) is a common strategy to evade T-cell alloreactivity and HLA-specific anti-donor antibodies, this strategy also unleashes natural killer (NK) alloreactivity owing to loss of inhibitory signaling from killer cell immunoglobulin-like receptors (KIRs) binding to HLA-I that normally restrains the NK cell response. Current approaches to reinstate NK cell inhibition rely on the expression of natural cloaking ligands such as HLA-E, , , HLA-A, HLA-C, and CD47 to agonize natural killer group 2A (NKG2A), KIR, and signal regulatory protein α (SIRPα) inhibitory receptors. However, these targets are only expressed on limited subsets of NK cells or upon sustained cytokine stimulation, rendering the ligands ineffective against hosts with low frequencies of the corresponding NK subset. , In this study, we discover and assess a universal solution against NK alloreactivity to enhance the persistence and efficacy of allogeneic T-cell therapies against all hosts.
T-cell engineering Human primary T-cell targets are engineered for knockout (KO) by CRISPR/Cas9 and knockin (KI) by nonviral homology-directed repair (HDR) in accordance with previous protocols. HLA-A2 alloreactive T (allo-T) cell clones are generated by orthotopic T-cell receptor α and β (TCRαβ) replacement as previously described. For messenger RNA (mRNA)–based screening, 1 to 2 million activated T cells with beta-2 microglobulin (B2M) KO were electroporated in 20 μL P3 buffer containing 1.5 μg of mRNA using code CM-137 and rested overnight. Transgene expression was verified by staining with relevant antibody and/or recombinant protein ligand and acquired on flow cytometry. See for detailed T-cell engineering, culture, and staining protocols. Allo-T and NK cell challenge assay Cryopreserved allo-T cells are thawed, stimulated with Immunocult CD2/CD3/CD28 activator (STEMCELL 10990) and cultured for 6 to 7 days in Rh10p (RPMI, 10% volume-to-volume ratio of human serum, 1× penicillin/streptomycin) supplemented with 40 ng/mL interleukin-2 [IL-2], 10 ng/mL interleukin-7 [IL-7], and 10 ng/mL interleukin-15 [IL-15]). Cryopreserved NK cells are thawed and cultured in Rh10p + 10 ng/mL IL-15 for 3 to 5 days before the start of coculture. For cocultures, 30 000 target T cells are plated with NK and/or allo-T cells at various effector-to-target ratios (E:T ratios) in a 96-well round-bottom plate for 20 hours in R10p (RPMI, 10% v/v fetal bovine serum, 1× penicillin/streptomycin) + 10 ng/mL IL-15 and then subjected to fluorescent barcoding flow cytometry. Amine-reactive dyes AF647 (Thermo Fisher Scientific A20006), IF700 (AAT Bioquest 71514), and IR800CW (LI-COR 929-70021) are resuspended in dimethyl sulfoxide and mixed in 8 combinations to make 100× working stocks of 0.2 mg/mL, 1 mg/mL, and 1.5 mg/mL, respectively, and stored in −80°C. The coculture plate was resuspended to 50 μL phosphate buffer saline (PBS) by centrifugation at 500 g for 2 minutes to decant supernatant. Fluorescent barcode working stocks were thawed and diluted 1:50 with PBS, and 50 μL of the appropriate barcode combination was immediately transferred to the corresponding wells of the coculture plate using a multichannel pipette and incubated in the dark for 15 minutes at room temperature. Afterwards, 100 μL of R10p was added to each well and incubated for 10 minutes to quench the reaction, followed by 2 washes in staining buffer (PBS, 5 mM EDTA, and 2% v/v fetal bovine serum) by centrifugation at 1200 g for 2 minutes. Wells are mixed column-wise and transferred into a new 96-well round-bottom plate, spun 1200 g 2 minutes, and then resuspended in the following antibody staining mix in staining buffer: BV421 CD56, BV510 CD3, BV605 HLA-I, 7-AAD, and PE Qbend10. The plate is washed twice in staining buffer by centrifugation at 1200 g for 2 minutes and then acquired at equal volumes on flow cytometry. PBMC challenge assay Cryopreserved peripheral blood mononuclear cells (PBMCs) are thawed, assessed for phenotype by flow cytometry, and rested overnight in Rh10p + 40 ng/mL IL-2. Notably, 30 000 HDR-edited T cells are plated with PBMCs at various E:T ratios in a 96-well flat-bottom plate for 3 days in R10p + 40 ng/mL IL-2. Cells are transferred to a 96-well round-bottom plate, and T-cell survival is determined by fluorescent barcoding flow cytometry using the protocol outlined above using the following antibody staining mix: PE Qbend10, BV605 HLA-I, BV650 CD56, BV785 CD3, and 7-AAD. Allo-T + NK cell competition assay HDR-edited T cells are pooled together at 1:1 ratio, seeded at 200 000 per well, and cocultured with allo-T and NK cells at various E:T ratios in a 96-well flat-bottom plate for 20 hours in R10p + 10 ng/mL IL-15; 10 μL of counting beads (BioLegend 424902) are added to each well and transferred to a new 96-well round-bottom plate. The plate is spun 500 g 3 minutes and resuspended into 50 μL of antibody staining mix, consisting of the following reagents in staining buffer: BV421 Qbend10, BV510 CD3, BV605 HLA-I, BV785 CD56, PE 218 Linker, 7-AAD, PE-Cy7 HLA-E, AF647 G4S Linker, and APC-Cy7 CD47. After staining, cells were washed twice with FACS buffer and acquired on flow cytometry at equal volumes per well. B-cell depletion assay Cryopreserved PBMCs are thawed and cultured in Rh10p + 40 ng/mL IL-2 for 3 days. PBMCs are seeded at 250 000 per well and cocultured with CAR T cells at various CAR T cell to PBMC ratios in a 96-well flat-bottom plate for 3 days in R10p + 40 ng/mL IL-2. Cells are transferred to a 96-well round-bottom plate spun by centrifugation at 500 g for 3 minutes into the following antibody staining mix in 30 μL of staining buffer: BV421 CD56, BV510 CD3, PE CD19, AF647 G4S Linker, 7-AAD are first added, and then 20 μL of an antibody-based fluorescent barcoding mix is added. Barcode staining mix consists of a combination of anti-CD45 antibodies on the BV711, BV785, PE-Cy7, and APC-Cy7 channels suspended in staining buffer that are unique to each row. The plate is washed twice in FACS buffer by centrifugation at 500 g for 3 minutes. Wells are mixed column-wise and transferred into a 96-well deep well plate and then acquired at equal volumes on flow cytometry. All primary cells are obtained from commercial sources from deidentified donors in adherence with the relevant institutional review board and with a written informed consent.
Human primary T-cell targets are engineered for knockout (KO) by CRISPR/Cas9 and knockin (KI) by nonviral homology-directed repair (HDR) in accordance with previous protocols. HLA-A2 alloreactive T (allo-T) cell clones are generated by orthotopic T-cell receptor α and β (TCRαβ) replacement as previously described. For messenger RNA (mRNA)–based screening, 1 to 2 million activated T cells with beta-2 microglobulin (B2M) KO were electroporated in 20 μL P3 buffer containing 1.5 μg of mRNA using code CM-137 and rested overnight. Transgene expression was verified by staining with relevant antibody and/or recombinant protein ligand and acquired on flow cytometry. See for detailed T-cell engineering, culture, and staining protocols.
Cryopreserved allo-T cells are thawed, stimulated with Immunocult CD2/CD3/CD28 activator (STEMCELL 10990) and cultured for 6 to 7 days in Rh10p (RPMI, 10% volume-to-volume ratio of human serum, 1× penicillin/streptomycin) supplemented with 40 ng/mL interleukin-2 [IL-2], 10 ng/mL interleukin-7 [IL-7], and 10 ng/mL interleukin-15 [IL-15]). Cryopreserved NK cells are thawed and cultured in Rh10p + 10 ng/mL IL-15 for 3 to 5 days before the start of coculture. For cocultures, 30 000 target T cells are plated with NK and/or allo-T cells at various effector-to-target ratios (E:T ratios) in a 96-well round-bottom plate for 20 hours in R10p (RPMI, 10% v/v fetal bovine serum, 1× penicillin/streptomycin) + 10 ng/mL IL-15 and then subjected to fluorescent barcoding flow cytometry. Amine-reactive dyes AF647 (Thermo Fisher Scientific A20006), IF700 (AAT Bioquest 71514), and IR800CW (LI-COR 929-70021) are resuspended in dimethyl sulfoxide and mixed in 8 combinations to make 100× working stocks of 0.2 mg/mL, 1 mg/mL, and 1.5 mg/mL, respectively, and stored in −80°C. The coculture plate was resuspended to 50 μL phosphate buffer saline (PBS) by centrifugation at 500 g for 2 minutes to decant supernatant. Fluorescent barcode working stocks were thawed and diluted 1:50 with PBS, and 50 μL of the appropriate barcode combination was immediately transferred to the corresponding wells of the coculture plate using a multichannel pipette and incubated in the dark for 15 minutes at room temperature. Afterwards, 100 μL of R10p was added to each well and incubated for 10 minutes to quench the reaction, followed by 2 washes in staining buffer (PBS, 5 mM EDTA, and 2% v/v fetal bovine serum) by centrifugation at 1200 g for 2 minutes. Wells are mixed column-wise and transferred into a new 96-well round-bottom plate, spun 1200 g 2 minutes, and then resuspended in the following antibody staining mix in staining buffer: BV421 CD56, BV510 CD3, BV605 HLA-I, 7-AAD, and PE Qbend10. The plate is washed twice in staining buffer by centrifugation at 1200 g for 2 minutes and then acquired at equal volumes on flow cytometry.
Cryopreserved peripheral blood mononuclear cells (PBMCs) are thawed, assessed for phenotype by flow cytometry, and rested overnight in Rh10p + 40 ng/mL IL-2. Notably, 30 000 HDR-edited T cells are plated with PBMCs at various E:T ratios in a 96-well flat-bottom plate for 3 days in R10p + 40 ng/mL IL-2. Cells are transferred to a 96-well round-bottom plate, and T-cell survival is determined by fluorescent barcoding flow cytometry using the protocol outlined above using the following antibody staining mix: PE Qbend10, BV605 HLA-I, BV650 CD56, BV785 CD3, and 7-AAD.
HDR-edited T cells are pooled together at 1:1 ratio, seeded at 200 000 per well, and cocultured with allo-T and NK cells at various E:T ratios in a 96-well flat-bottom plate for 20 hours in R10p + 10 ng/mL IL-15; 10 μL of counting beads (BioLegend 424902) are added to each well and transferred to a new 96-well round-bottom plate. The plate is spun 500 g 3 minutes and resuspended into 50 μL of antibody staining mix, consisting of the following reagents in staining buffer: BV421 Qbend10, BV510 CD3, BV605 HLA-I, BV785 CD56, PE 218 Linker, 7-AAD, PE-Cy7 HLA-E, AF647 G4S Linker, and APC-Cy7 CD47. After staining, cells were washed twice with FACS buffer and acquired on flow cytometry at equal volumes per well.
Cryopreserved PBMCs are thawed and cultured in Rh10p + 40 ng/mL IL-2 for 3 days. PBMCs are seeded at 250 000 per well and cocultured with CAR T cells at various CAR T cell to PBMC ratios in a 96-well flat-bottom plate for 3 days in R10p + 40 ng/mL IL-2. Cells are transferred to a 96-well round-bottom plate spun by centrifugation at 500 g for 3 minutes into the following antibody staining mix in 30 μL of staining buffer: BV421 CD56, BV510 CD3, PE CD19, AF647 G4S Linker, 7-AAD are first added, and then 20 μL of an antibody-based fluorescent barcoding mix is added. Barcode staining mix consists of a combination of anti-CD45 antibodies on the BV711, BV785, PE-Cy7, and APC-Cy7 channels suspended in staining buffer that are unique to each row. The plate is washed twice in FACS buffer by centrifugation at 500 g for 3 minutes. Wells are mixed column-wise and transferred into a 96-well deep well plate and then acquired at equal volumes on flow cytometry. All primary cells are obtained from commercial sources from deidentified donors in adherence with the relevant institutional review board and with a written informed consent.
Modeling cellular rejection of allogeneic T cells We first developed an ex-vivo model of cellular rejection by measuring the survival of human allogeneic T-cell targets cocultured with human primary cell effectors using a modified fluorescently barcoded flow cytometry readout ( A). This method enabled sensitive and high-throughput measurement of target T-cell survival by direct counting across multiple E:T ratios with high reproducibility in the value of half maximal inhibitory concentration (IC50) ( B-C). Analogous to dose-response curves, we define the IC50 value as the E:T ratio corresponding to 50% of target T-cell survival, with higher values indicating enhanced T-cell survival ability. To confirm that disruption of HLA-I eliminates T-cell alloreactivity, we cocultured target T cells from HLA-A2 + serotyped donors with HLA-A2 − effector alloreactive T cells engineered to express a known HLA-A2 reactive TCR ( A). Target T cells were rejected in a dose-dependent manner but were rescued by disruption of HLA-I expression via CRISPR/Cas9-mediated KO of the B2M gene ( A-B). However, B2M KO sensitized target T cells to NK allorecognition, and neither HLA-I + nor B2M KO target T cells survived when both NK and alloreactive T cells were added as effectors ( B). Live cell counts of allo-T and NK cell effectors at assay end point were similar when added alone or together across all E:T ratios against both B2M KO and HLA-I + targets, demonstrating that cytotoxicity was restricted to target cells and negligible between effector cells ( A). B2M KO–mediated NK cell alloreactivity was consistent across multiple donor replicates, confirming missing-self mechanism of NK recognition in this context ( B). , Screening and discovery of NKG2A and CD300a TASR We applied this experimental system to rapidly screen previously published “cloaking” strategies known to inhibit NK cell cytotoxicity ( C; A). One set of strategies involves the removal of ligands that can agonize activating or adhesion receptors on NK cells. , , , A second set involves the expression of ligands that agonize inhibitory receptors on NK cells or reduce the effect of cytotoxic granules. , , , , We challenged B2M KO T cells engineered with each strategy against human primary NK cells from a single donor at various E:T ratios and measured the fold change in T-cell survival relative to noncloaked negative control using the IC50 value. Genetic ablation of CD48, CD54, CD58, and CD155 via CRISPR/Cas9 did not protect against NK cell challenge. This result was reproducible across 2 T-cell lines per target, each engineered with a separate guide RNA with confirmed knockdown of surface expression. Although all natural NK inhibitory ligands were successfully and uniformly expressed via mRNA electroporation, only the well-studied HLA-E enhanced the survival of B2M KO T cells against NK challenge, in line with previous works. , Constrained by the limited targeting repertoire of natural ligand-based approaches, we hypothesized that surface expression of single-chain antigen-binding fragments (scFVs) on an inert scaffold, herein termed trans-antigen signaling receptors (TASRs), can agonize a wider repertoire of NK inhibitory receptors than is currently possible ( D). We constructed TASRs using an initial CD8 hinge and transmembrane scaffold using scFV constructs of well-studied antibody clones derived from literature and patents to target inhibitory receptors known to be expressed on NK cells. We screened TASRs via mRNA electroporation of B2M KO primary T cells for expression and functional potency against human primary NK cells derived from 1 donor ( A). From this screen, we identified a NKG2A and a CD300a-binding TASR that functioned in a dose-dependent manner ( C; A-C). NKG2A TASR agonizes the same target as HLA-E, whereas CD300a is an unexplored target for this application. Optimization of CD300a TASR We set out to optimize the expression and function of CD300a TASR version 1 by testing several mutant constructs using our mRNA screening platform. We discovered a mouse B7-1 transmembrane and cytoplasmic domain that enhanced expression and an inverse correlation between TASR hinge length and the function of CD300a TASR but not NKG2A TASR ( E-G; ). An optimized version 2, containing no hinge but stabilized by the mouse B7-1 cytoplasmic tail and hereafter termed CD300a TASR, exhibited significantly enhanced function while maintaining a fully human extracellular domain ( H-J). Validation of CD300a TASR To model cloaking ligands under constitutive and physiologically relevant levels of expression, we inserted transgenes via CRISPR/Cas9–mediated nonviral HDR of human T cells. Transgenes were integrated into either the B2M locus using the endogenous B2M promoter concomitant with B2M KO or AAVS1 locus using the exogenous elongation factor 1 alpha (EF1α) promoter ( A; A). CD300a TASR expression was confirmed in both integration approaches by staining with either cognate CD300a protein ligand or an antibody specific to scFV linker sequence ( B; B). CD300a TASR enhanced T-cell survival in our standard 20-hour challenge assay with IL-15–treated NK cell effectors from multiple donors relative to noncloaked B2M KO T cells ( C). This result was reproducible with noncytokine-treated NK cells and in long-term 7-day cocultures, with protection on par with HLA-I + control. CD300a TASR also reduced the secretion of interferon gamma and tumor necrosis factor α release after NK challenge relative to negative control B2M KO T cells and to levels on par with HLA-I + T-cell positive control ( C-D). To confirm utilization of the CD300a pathway, we performed NK challenge in the presence of blocking antibodies. The addition of anti-CD300a blocking antibody but not isotype control or anti-CD300c, a close paralog of CD300a, completely abrogated the protective effect of CD300a TASR to noncloaked levels ( D). Anti-CD300a did not affect the survival of noncloaked T cells. Thus, CD300a-mediated inhibition is not functionally used during endogenous T cell and NK cell interactions and the protective effect of CD300a TASR relies on the CD300a signaling axis. CD300a TASR outperforms alternative strategies CD300a TASR outperformed the current best-in-class ligands HLA-E and CD47, as well as the recently published SIRPα engager in NK protection. , , , , To provide head-to-head comparisons, we inserted each ligand into the B2M locus of human T cells concomitant with B2M KO and confirmed surface expression ( A). Challenged with NK cells, CD300a TASR and HLA-E outperformed CD47 and SIRPα engager, whereas CD300a TASR was the only protective ligand against a donor with high frequencies of an NKG2A − NKG2C + NK cell phenotype, hereby termed 2C dom ( B). The dominance of CD300a TASR over CD47 was observed under all conditions tested, including when overexpressed via mRNA, inserted into the AAVS1 locus via HDR under EF1α promoter, and challenged with NK cells cultured in high-dose IL-2, which upregulates SIRPα . , Using mRNA EP, we further determined that CD300a TASR outcompeted TIM3 engager against 4/4 NK donors under comparable levels of expression . As an orthogonal readout, we pooled HDR-edited T cells from the previous experiment along with an HLA-I + control and challenged with allo-T and NK cells in a competition assay. T cells expressing the indicated ligands were pooled in equal ratios and challenged with NK cells and HLA-A2 alloreactive T cells ( A,C). The identity and frequency of each T-cell member can be discerned based on antibody staining of the cloaking ligand and measured by conventional multicolor flow cytometry ( A-E). CD300a TASR emerged as the dominant survivor when challenged with both NK and allo-T cells ( C). This result was reproducible across a 2A dom and 2C dom NK cell donor and 2 alloreactive T-cell clones specific to HA-2 or endoplasmic reticulum membrane protein complex subunit 7 (EMC7) antigen, as well as any combination thereof. In this mixed T-cell model, active T-cell alloreactivity can be observed with both allo–T-cell effectors given that the fraction of HLA-I + T cells remain significantly lower relative to conditions without allo-T cells. Likewise, active rejection of noncloaked B2M KO T cells can be observed with both NK cell donors in this model relative to conditions without NK cells. Competition with additional NK donors in the absence of HLA-I + control cells further corroborated dominance of CD300a TASR in 2C dom NK donors ( F). CD300a TASR is a pan-NK inhibitor We next assessed CD300a TASR and HLA-E, the most potent alternative, against a large human cohort to assess the heterogeneity of responses across donors and to identify correlates of protection. To model the physiological context of intravenously injected T-cell therapies, we measured the survival of cloaked T cells challenged with PBMCs ( A; ). We selected 45 PBMC donors for diversity in ethnicity, age, gender, and cytomegalovirus (CMV) seropositivity due to their known effects on NK cell function and phenotype ( A). , , , Notably, although large variations in CD57, KIR, NKG2A, and NKG2C NK marker expression were observed, CD300a was uniformly expressed on NK cells in all donors ( B; ). CD300a TASR and HLA-E were expressed at comparable levels from the AAVS1 locus based on RQR8 expression ( C). Within this cohort, CD300a TASR protected B2M KO T cells against all PBMC donors tested (45/45), indicating universal protection against NK cell reactivity ( D; ). CMV serostatus emerged as the most significant demographic determinant of reduced protection by HLA-E and correlated with the frequency of the well-known NKG2C + NKG2A − CD57 + CD16 + CD56 hi adaptive NK cell subset ( E-F). Although HLA-E lost protection with increasing frequencies of adaptive NK cells, CD300a TASR maintained protection ( G). CD300a TASR enhances anti-CD19 CAR T-cell killing under allogeneic conditions Finally, we integrated recent innovations to produce a model allogeneic CAR T-cell therapy coexpressing a clinically relevant anti-CD19 CAR into the TRAC locus and cloaking ligand into the B2M locus via multiplex nonviral HDR ( A; A-B). , , Expression of CD300a TASR, HLA-E, and RQR8 control cloaking ligand in CAR T cells was confirmed ( B). To model CAR T-cell function under a physiological allogeneic environment, we dosed our engineered CAR T cells into PBMCs and measured killing of CD19-expressing B cells under NK alloreactivity ( C; C,G). CD300a TASR enhanced B-cell killing potency in all PBMC donors (5/5) and outcompeted HLA-E in all donors with 2C dom NK phenotype ( C). This result was not due to intrinsic differences, given that all CAR T cells had similar potency against CD19-expressing Raji cells in the absence of allogeneic pressure ( D). In addition, TASR-mediated protection against NK cells was maintained in the presence of CAR ( E).
We first developed an ex-vivo model of cellular rejection by measuring the survival of human allogeneic T-cell targets cocultured with human primary cell effectors using a modified fluorescently barcoded flow cytometry readout ( A). This method enabled sensitive and high-throughput measurement of target T-cell survival by direct counting across multiple E:T ratios with high reproducibility in the value of half maximal inhibitory concentration (IC50) ( B-C). Analogous to dose-response curves, we define the IC50 value as the E:T ratio corresponding to 50% of target T-cell survival, with higher values indicating enhanced T-cell survival ability. To confirm that disruption of HLA-I eliminates T-cell alloreactivity, we cocultured target T cells from HLA-A2 + serotyped donors with HLA-A2 − effector alloreactive T cells engineered to express a known HLA-A2 reactive TCR ( A). Target T cells were rejected in a dose-dependent manner but were rescued by disruption of HLA-I expression via CRISPR/Cas9-mediated KO of the B2M gene ( A-B). However, B2M KO sensitized target T cells to NK allorecognition, and neither HLA-I + nor B2M KO target T cells survived when both NK and alloreactive T cells were added as effectors ( B). Live cell counts of allo-T and NK cell effectors at assay end point were similar when added alone or together across all E:T ratios against both B2M KO and HLA-I + targets, demonstrating that cytotoxicity was restricted to target cells and negligible between effector cells ( A). B2M KO–mediated NK cell alloreactivity was consistent across multiple donor replicates, confirming missing-self mechanism of NK recognition in this context ( B). ,
We applied this experimental system to rapidly screen previously published “cloaking” strategies known to inhibit NK cell cytotoxicity ( C; A). One set of strategies involves the removal of ligands that can agonize activating or adhesion receptors on NK cells. , , , A second set involves the expression of ligands that agonize inhibitory receptors on NK cells or reduce the effect of cytotoxic granules. , , , , We challenged B2M KO T cells engineered with each strategy against human primary NK cells from a single donor at various E:T ratios and measured the fold change in T-cell survival relative to noncloaked negative control using the IC50 value. Genetic ablation of CD48, CD54, CD58, and CD155 via CRISPR/Cas9 did not protect against NK cell challenge. This result was reproducible across 2 T-cell lines per target, each engineered with a separate guide RNA with confirmed knockdown of surface expression. Although all natural NK inhibitory ligands were successfully and uniformly expressed via mRNA electroporation, only the well-studied HLA-E enhanced the survival of B2M KO T cells against NK challenge, in line with previous works. , Constrained by the limited targeting repertoire of natural ligand-based approaches, we hypothesized that surface expression of single-chain antigen-binding fragments (scFVs) on an inert scaffold, herein termed trans-antigen signaling receptors (TASRs), can agonize a wider repertoire of NK inhibitory receptors than is currently possible ( D). We constructed TASRs using an initial CD8 hinge and transmembrane scaffold using scFV constructs of well-studied antibody clones derived from literature and patents to target inhibitory receptors known to be expressed on NK cells. We screened TASRs via mRNA electroporation of B2M KO primary T cells for expression and functional potency against human primary NK cells derived from 1 donor ( A). From this screen, we identified a NKG2A and a CD300a-binding TASR that functioned in a dose-dependent manner ( C; A-C). NKG2A TASR agonizes the same target as HLA-E, whereas CD300a is an unexplored target for this application.
We set out to optimize the expression and function of CD300a TASR version 1 by testing several mutant constructs using our mRNA screening platform. We discovered a mouse B7-1 transmembrane and cytoplasmic domain that enhanced expression and an inverse correlation between TASR hinge length and the function of CD300a TASR but not NKG2A TASR ( E-G; ). An optimized version 2, containing no hinge but stabilized by the mouse B7-1 cytoplasmic tail and hereafter termed CD300a TASR, exhibited significantly enhanced function while maintaining a fully human extracellular domain ( H-J).
To model cloaking ligands under constitutive and physiologically relevant levels of expression, we inserted transgenes via CRISPR/Cas9–mediated nonviral HDR of human T cells. Transgenes were integrated into either the B2M locus using the endogenous B2M promoter concomitant with B2M KO or AAVS1 locus using the exogenous elongation factor 1 alpha (EF1α) promoter ( A; A). CD300a TASR expression was confirmed in both integration approaches by staining with either cognate CD300a protein ligand or an antibody specific to scFV linker sequence ( B; B). CD300a TASR enhanced T-cell survival in our standard 20-hour challenge assay with IL-15–treated NK cell effectors from multiple donors relative to noncloaked B2M KO T cells ( C). This result was reproducible with noncytokine-treated NK cells and in long-term 7-day cocultures, with protection on par with HLA-I + control. CD300a TASR also reduced the secretion of interferon gamma and tumor necrosis factor α release after NK challenge relative to negative control B2M KO T cells and to levels on par with HLA-I + T-cell positive control ( C-D). To confirm utilization of the CD300a pathway, we performed NK challenge in the presence of blocking antibodies. The addition of anti-CD300a blocking antibody but not isotype control or anti-CD300c, a close paralog of CD300a, completely abrogated the protective effect of CD300a TASR to noncloaked levels ( D). Anti-CD300a did not affect the survival of noncloaked T cells. Thus, CD300a-mediated inhibition is not functionally used during endogenous T cell and NK cell interactions and the protective effect of CD300a TASR relies on the CD300a signaling axis.
CD300a TASR outperformed the current best-in-class ligands HLA-E and CD47, as well as the recently published SIRPα engager in NK protection. , , , , To provide head-to-head comparisons, we inserted each ligand into the B2M locus of human T cells concomitant with B2M KO and confirmed surface expression ( A). Challenged with NK cells, CD300a TASR and HLA-E outperformed CD47 and SIRPα engager, whereas CD300a TASR was the only protective ligand against a donor with high frequencies of an NKG2A − NKG2C + NK cell phenotype, hereby termed 2C dom ( B). The dominance of CD300a TASR over CD47 was observed under all conditions tested, including when overexpressed via mRNA, inserted into the AAVS1 locus via HDR under EF1α promoter, and challenged with NK cells cultured in high-dose IL-2, which upregulates SIRPα . , Using mRNA EP, we further determined that CD300a TASR outcompeted TIM3 engager against 4/4 NK donors under comparable levels of expression . As an orthogonal readout, we pooled HDR-edited T cells from the previous experiment along with an HLA-I + control and challenged with allo-T and NK cells in a competition assay. T cells expressing the indicated ligands were pooled in equal ratios and challenged with NK cells and HLA-A2 alloreactive T cells ( A,C). The identity and frequency of each T-cell member can be discerned based on antibody staining of the cloaking ligand and measured by conventional multicolor flow cytometry ( A-E). CD300a TASR emerged as the dominant survivor when challenged with both NK and allo-T cells ( C). This result was reproducible across a 2A dom and 2C dom NK cell donor and 2 alloreactive T-cell clones specific to HA-2 or endoplasmic reticulum membrane protein complex subunit 7 (EMC7) antigen, as well as any combination thereof. In this mixed T-cell model, active T-cell alloreactivity can be observed with both allo–T-cell effectors given that the fraction of HLA-I + T cells remain significantly lower relative to conditions without allo-T cells. Likewise, active rejection of noncloaked B2M KO T cells can be observed with both NK cell donors in this model relative to conditions without NK cells. Competition with additional NK donors in the absence of HLA-I + control cells further corroborated dominance of CD300a TASR in 2C dom NK donors ( F).
We next assessed CD300a TASR and HLA-E, the most potent alternative, against a large human cohort to assess the heterogeneity of responses across donors and to identify correlates of protection. To model the physiological context of intravenously injected T-cell therapies, we measured the survival of cloaked T cells challenged with PBMCs ( A; ). We selected 45 PBMC donors for diversity in ethnicity, age, gender, and cytomegalovirus (CMV) seropositivity due to their known effects on NK cell function and phenotype ( A). , , , Notably, although large variations in CD57, KIR, NKG2A, and NKG2C NK marker expression were observed, CD300a was uniformly expressed on NK cells in all donors ( B; ). CD300a TASR and HLA-E were expressed at comparable levels from the AAVS1 locus based on RQR8 expression ( C). Within this cohort, CD300a TASR protected B2M KO T cells against all PBMC donors tested (45/45), indicating universal protection against NK cell reactivity ( D; ). CMV serostatus emerged as the most significant demographic determinant of reduced protection by HLA-E and correlated with the frequency of the well-known NKG2C + NKG2A − CD57 + CD16 + CD56 hi adaptive NK cell subset ( E-F). Although HLA-E lost protection with increasing frequencies of adaptive NK cells, CD300a TASR maintained protection ( G).
Finally, we integrated recent innovations to produce a model allogeneic CAR T-cell therapy coexpressing a clinically relevant anti-CD19 CAR into the TRAC locus and cloaking ligand into the B2M locus via multiplex nonviral HDR ( A; A-B). , , Expression of CD300a TASR, HLA-E, and RQR8 control cloaking ligand in CAR T cells was confirmed ( B). To model CAR T-cell function under a physiological allogeneic environment, we dosed our engineered CAR T cells into PBMCs and measured killing of CD19-expressing B cells under NK alloreactivity ( C; C,G). CD300a TASR enhanced B-cell killing potency in all PBMC donors (5/5) and outcompeted HLA-E in all donors with 2C dom NK phenotype ( C). This result was not due to intrinsic differences, given that all CAR T cells had similar potency against CD19-expressing Raji cells in the absence of allogeneic pressure ( D). In addition, TASR-mediated protection against NK cells was maintained in the presence of CAR ( E).
Immunological rejection is a major barrier to the persistence and efficacy of allogeneic cell therapies and transplants. In this study, we describe the discovery, engineering, and assessment of TASRs as a new modality to inhibit the alloreactive NK cell response. We demonstrate that an optimized CD300a TASR in combination with HLA-I ablation protected against both T and NK cell alloreactivity, exhibited superior function relative to alternative strategies, and possessed broad NK inhibitory potential against a large human cohort. Finally, we show that CD300a TASR can be integrated into a next-generation allogeneic CAR T-cell product via multiplexed, nonviral targeted integration and that the resulting product enhanced B-cell killing potency under allogeneic immune pressure. We expect the B-cell lysis model to be relevant for allogeneic CAR T-cell therapy against systemic lupus erythematosus, where B-cell depletion is the main mechanism of action for dampening the autoreactive antibody response. CD300a is a new and unexplored target for NK cloaking. CD300a is a classical inhibitory receptor expressed on all NK cells and signals through immunoreceptor tyrosine–based inhibitory motifs upon binding its endogenous ligands phosphatidylserine and phosphatidylethanolamine (PS/PE) on the surface of apoptotic cells. , , Antibody cross-linking studies have also demonstrated Src homology region 2 domain-containing phosphatase 1 (SHP-1) and 2 (SHP-2) recruitment to immunoreceptor tyrosine–based inhibitory motifs to be the primary mechanism of inhibition by CD300a. , , , Mechanistically, CD300a TASR mirrors the inhibitory effect of anti-CD300a antibody in redirected killing assays with NK effectors against P815 target cells. CD300a TASR leaves the host immune system intact and functional, given that noncloaked T cells were still depleted in the presence of CD300a TASR expressing T cells in the competition assay. Thus, CD300a TASR could offer an enhanced safety profile over CAR or antibody-mediated approaches that ablate the host immune system and also leave the endogenous T and NK antitumor response intact. The functionality of TASR is determined by both the antigen-binding fragment and the scaffold structure. Despite attempts to reinstate KIR agonism after HLA-I ablation, none of the KIR-targeting TASR clones in our screen demonstrated efficacy. It is possible that our anti-KIR scFVs lack intrinsic agonistic activity or cross reacts with both inhibitory and activating KIRs, which can be difficult to distinguish with TASRs due to high sequence homology between KIR members. The beneficial effect of hinge removal for CD300a TASR contrasts with CAR designs, where hinges can enhance CAR T-cell function and signaling. , Given that CD300a TASR is a mimic to the small phospholipids PS/PE, we suspect that hinge removal enhances CD300a agonism by reducing the interaction distance between TASR and CD300a to be comparable with the endogenous interaction of PS/PE and CD300a. We expect the use of CD300a TASR in allogeneic cell therapies to significantly expand the addressable patient population. By comparing CD300a TASR and HLA-E with a large human cohort, we highlight how NK alloreactivity can exhibit large donor-to-donor variations that should be considered. Although previous work has reported HLA-E’s lack of efficacy against the 2C dom NK phenotype, our study provides the first comprehensive link among cloaking ligand functional efficacy, NK phenotype, and patient demographics. With these data, we were able to robustly show that HLA-E has diminished protection against NK cells isolated from CMV + individuals whereas CD300a TASR retains function in all hosts independent of demographic determinants. Notably, 50% of CMV + individuals (12/24) in our cohort harbored substantial adaptive NK cell frequency >5% of total NK cells, which is in line with past estimates and corresponds to the point where HLA-E begins to lose potency in the PBMC challenge assay. Given global CMV seroprevalence rates of 43% to 96% by country, we expect CD300a TASR to expand the addressable patient population by 22% to 48% over HLA-E and have the greatest impact in the older individuals aged > 60 years, which have 80%+ CMV seropositivity rate, as is the case with CAR T-cell therapy for most cancer malignancies. , We note several limitations to our study design and scope. First, it remains to be seen whether NKG2A and CD300a TASR can maintain NK protection across a broader range of therapeutic cell types beyond T cells. Second, although CD300a TASR enhanced the persistence and function of an allogeneic CAR T cell produced by multiplexed nonviral integration, additional testing will be required to assess the feasibility of this drug product with respect to genomic integrity, cell phenotype, long-term function, and comparison to other production methods. Finally, additional studies are warranted to investigate the efficacy of TASRs in vivo. In summary, we have demonstrated that CD300a TASR acts as a universal ligand against NK cell alloreactivity. Its combination with HLA and TCRαβ ablation to prevent T-cell alloreactivity, anti-donor antibodies, and graft-versus-host disease paves the way for fully immunoevasive CAR T cells that are effective for anyone in need of treatment, from cancer to autoimmunity and beyond.
|
Acute myeloid leukemia with T lymphoblastic lymphoma: a case report and literature review | 21ce5602-1f27-4689-9d95-144037d4adfb | 8161867 | Histology[mh] | Clinically, both acute myelocytic leukemia (AML) and T-lymphoblastic lymphoma (T-LBL) are common hematological malignancies. , Because AML and T-LBL have different cell origins, their onsets are usually isolated, and it is rare for both diseases to develop concurrently in the same patient. , We report an AML patient complicated with T-LBL, and we describe the patient’s disease evolution, with a literature review.
General information and physical examination findings The patient was a 54-year-old man who first visited our hospital on 30 November 2016 because of “fatigue for 2 months and fever for 1 week”. Physical examination revealed signs of anemia and five enlarged lymph nodes in the left submandibular region and left posterior triangle of the neck. The maximum lymph node diameter was 3 cm, and the nodes were firm but not tender. The patient had poor range of motion in his neck, and no tenderness in the sternum, and the liver and spleen were not palpable below his ribs. Laboratory testing, and imaging and immunophenotyping findings Peripheral blood evaluation revealed the following: white blood cell (WBC) count: 3.9 × 10 9 cells/L, neutrophil (NEU) count: 0.07 × 10 9 cells/L, hemoglobin (Hb): 83.00 g/L, and platelet (PLT) count: 77.00 × 10 9 cells/L. Computed tomography (CT) revealed small lymph nodes in bilateral axillae, and multiple enlarged retroperitoneal lymph nodes were noted. The lymph nodes were obviously enlarged in the bilateral iliac vessel distribution areas. Bone marrow cell morphology revealed that archeocytes accounted for 84% of the cells. Immunophenotyping revealed expression of myeloperoxidase (MPO), cluster of differentiation (CD)34, CD38, CD33, and CD7; CD13 and CD117 were weakly expressed, and CD3 was not expressed. Bone marrow biopsy revealed hyperreactivity (>90%) and diffusely hyperplastic immature cells. Immunohistochemistry revealed CD34 (+), CD117 (+), transmission disequilibrium test (TDT) (+), PAX5 (−), CD3 (−), slight MPO (+), lysozyme (Lys) (−), and CD10 (+). Bone marrow gene tests for platelet-derived growth factor receptor-alpha (PDGFRA), platelet-derived growth factor receptor-beta (PDGFRB), platelet-derived growth factor-1 (PDGF1), and Janus kinase-2 (JAK2) were negative. The patient’s chromosome number was 46 (XY), and the leukemia fusion genes, retinoic acid receptor alpha (RARA), eight-twenty-one (ETO), MYH11, nucleophosmin (NPM), and mixed-lineage leukemia (MLL) were negative. Leukemia mutation gene WT1 quantification was 8.53%, and C-KIT and FMS-like tyrosine kinase 3 (FLT3) were negative. The patient was initially diagnosed with acute myelocytic leukemia (WT1 positive). On 1 December 2016, the standard-dose cytarabine + idarubicin (IA) regimen was given as induction chemotherapy. After 4 weeks of chemotherapy, bone marrow evaluation revealed obvious hyperplasia, with 5% archeocytes. However, no significant changes were found in the enlarged lymph nodes on repeat imaging; therefore, left cervical lymph node biopsy was performed on 14 January 2017. Histopathology showed that the lymph node structure was destroyed, and that tumor cells were densely infiltrated in the lymph node capsule and parenchyma. The cell volume was medium to large, with scant cytoplasm and no obvious cytoplasmic inclusions ( ). Immunohistochemistry revealed CD3 (+), CD34 (+), MPO (−), Ki-67 (+) and approximately 80%, TDT (+), CD5 partially (+), PAX5 (−), and CD7 (+). On 11 February 2017, repeat bone marrow evaluation revealed 20.5% archeocytes; CD33, CD34, CD7, and human leukocyte antigen DR (HLA-DR) were expressed, while MPO, CD3, CD19, CD10, CD10, and CD79a were not expressed. Combined with the bone marrow examination findings at the first visit, the patient was finally diagnosed with AML complicated with T-LBL. Treatment regimen After the final diagnosis of AML complicated with T-LBL, the IA + vinovelbine, cyclophosphamide, pegaspargase, and prednisone (VCLP) regimen was initiated as induction therapy on 15 February 2017. The lymph nodes shrank significantly after this chemotherapy, and repeat bone marrow evaluation revealed 1.5% archeocytes. The previous regimen (IA+VCLP regimen) was given as consolidation chemotherapy on 24 March 2017. Post-chemotherapy imaging revealed that the enlarged lymph nodes had disappeared, and WT1 was 0.21%. High-dose cytarabine (3 g every 12 hours on days 1, 3, and 5) + pegaspargase was given for four courses as consolidation chemotherapy. Clinical follow-up At the end of the chemotherapy, bone marrow evaluation revealed 6% archeocytes, and CT revealed multiple enlarged lymph nodes in the neck, axillae, and mediastinum, indicating recurrent disease. Chemotherapy with cyclophosphamide, vinorelbine, prednisone, etoposide, pegaspargase, and idarubicin (COPEL + IDA) and cyclophosphamide, vinorelbine, prednisone, etoposide, pegaspargase, and mitoxantrone (COPEL +MTZ) were given on 31 October 2017 and 30 November 2017, respectively. Bone marrow evaluation on 8 January 2018 revealed 15.5% archeocytes, and immunophenotyping revealed that HLA-DR, CD7, CD38, CD33, and CD11b were expressed, cytoplasmic (cy)MPO was weakly expressed, and CD117, CD13, CD64, CD14, CD3, and CD79a were not expressed. Positron emission tomography (PET)-CT revealed that multiple superficial and deep lymph nodes were enlarged, and the maximum standardized uptake value (SUV max ) was approximately 13.3. Cervical lymph node biopsy was repeated and revealed destruction of the normal structure and diffuse immature cell proliferation. Immunohistochemistry revealed the following: CD5 (+), CD7 (+), TDT partial (+), CD99 (+), CD33 (+), CD117 partial (+), B cell chronic lymphocytic leukemia/lymphoma-2 (Bcl-2) (+), and CD3 partial weak (+); TCRγδ gene rearrangement was positive. Following decitabine + cyclophosphamide + vindesine (HOAP) chemotherapy and two courses of gemcitabine, dexamethasone, and cisplatin (GDP) + hyaluronic acid (HA), the enlarged lymph nodes disappeared. In June 2018, bone marrow evaluation revealed 38% archeocytes, and immunophenotyping revealed two groups of abnormal archeocytes, accounting for 34.2% of all cells; 24.1% of these cells expressed CD3, CD5, CD7, and CD34, and 10.1% of the cells expressed CD34dim, CD7, CD5, and CD38. The following were not expressed: CD33, cyMPO, CD3, CD79a, CD13, and HLA-DR. The patient underwent successive fludarabine + cytarabine + recombinant human granulocyte colony-stimulating factor (G-CSF) (FLAG) and HOAP chemotherapy regimens, after which, bone marrow evaluation revealed 47% archeocytes. Immunophenotyping revealed CD34 (+), CD38 (+), CD7 (+), and CD45dim (+), and weak expression of CD33, CD3, and CD79a. Subsequently, three courses of chemotherapy with vincristine, cyclophosphamide, daunorubicin/idarubicin, and prednisone (VDCP) was administered. In addition, the patient’s lactate level fluctuated between 10.0 and 20.0 mmol/L from 30 January 2018. During the treatment, the patient’s condition continued to progress, and he died in January 2019.
The patient was a 54-year-old man who first visited our hospital on 30 November 2016 because of “fatigue for 2 months and fever for 1 week”. Physical examination revealed signs of anemia and five enlarged lymph nodes in the left submandibular region and left posterior triangle of the neck. The maximum lymph node diameter was 3 cm, and the nodes were firm but not tender. The patient had poor range of motion in his neck, and no tenderness in the sternum, and the liver and spleen were not palpable below his ribs.
Peripheral blood evaluation revealed the following: white blood cell (WBC) count: 3.9 × 10 9 cells/L, neutrophil (NEU) count: 0.07 × 10 9 cells/L, hemoglobin (Hb): 83.00 g/L, and platelet (PLT) count: 77.00 × 10 9 cells/L. Computed tomography (CT) revealed small lymph nodes in bilateral axillae, and multiple enlarged retroperitoneal lymph nodes were noted. The lymph nodes were obviously enlarged in the bilateral iliac vessel distribution areas. Bone marrow cell morphology revealed that archeocytes accounted for 84% of the cells. Immunophenotyping revealed expression of myeloperoxidase (MPO), cluster of differentiation (CD)34, CD38, CD33, and CD7; CD13 and CD117 were weakly expressed, and CD3 was not expressed. Bone marrow biopsy revealed hyperreactivity (>90%) and diffusely hyperplastic immature cells. Immunohistochemistry revealed CD34 (+), CD117 (+), transmission disequilibrium test (TDT) (+), PAX5 (−), CD3 (−), slight MPO (+), lysozyme (Lys) (−), and CD10 (+). Bone marrow gene tests for platelet-derived growth factor receptor-alpha (PDGFRA), platelet-derived growth factor receptor-beta (PDGFRB), platelet-derived growth factor-1 (PDGF1), and Janus kinase-2 (JAK2) were negative. The patient’s chromosome number was 46 (XY), and the leukemia fusion genes, retinoic acid receptor alpha (RARA), eight-twenty-one (ETO), MYH11, nucleophosmin (NPM), and mixed-lineage leukemia (MLL) were negative. Leukemia mutation gene WT1 quantification was 8.53%, and C-KIT and FMS-like tyrosine kinase 3 (FLT3) were negative. The patient was initially diagnosed with acute myelocytic leukemia (WT1 positive). On 1 December 2016, the standard-dose cytarabine + idarubicin (IA) regimen was given as induction chemotherapy. After 4 weeks of chemotherapy, bone marrow evaluation revealed obvious hyperplasia, with 5% archeocytes. However, no significant changes were found in the enlarged lymph nodes on repeat imaging; therefore, left cervical lymph node biopsy was performed on 14 January 2017. Histopathology showed that the lymph node structure was destroyed, and that tumor cells were densely infiltrated in the lymph node capsule and parenchyma. The cell volume was medium to large, with scant cytoplasm and no obvious cytoplasmic inclusions ( ). Immunohistochemistry revealed CD3 (+), CD34 (+), MPO (−), Ki-67 (+) and approximately 80%, TDT (+), CD5 partially (+), PAX5 (−), and CD7 (+). On 11 February 2017, repeat bone marrow evaluation revealed 20.5% archeocytes; CD33, CD34, CD7, and human leukocyte antigen DR (HLA-DR) were expressed, while MPO, CD3, CD19, CD10, CD10, and CD79a were not expressed. Combined with the bone marrow examination findings at the first visit, the patient was finally diagnosed with AML complicated with T-LBL.
After the final diagnosis of AML complicated with T-LBL, the IA + vinovelbine, cyclophosphamide, pegaspargase, and prednisone (VCLP) regimen was initiated as induction therapy on 15 February 2017. The lymph nodes shrank significantly after this chemotherapy, and repeat bone marrow evaluation revealed 1.5% archeocytes. The previous regimen (IA+VCLP regimen) was given as consolidation chemotherapy on 24 March 2017. Post-chemotherapy imaging revealed that the enlarged lymph nodes had disappeared, and WT1 was 0.21%. High-dose cytarabine (3 g every 12 hours on days 1, 3, and 5) + pegaspargase was given for four courses as consolidation chemotherapy.
At the end of the chemotherapy, bone marrow evaluation revealed 6% archeocytes, and CT revealed multiple enlarged lymph nodes in the neck, axillae, and mediastinum, indicating recurrent disease. Chemotherapy with cyclophosphamide, vinorelbine, prednisone, etoposide, pegaspargase, and idarubicin (COPEL + IDA) and cyclophosphamide, vinorelbine, prednisone, etoposide, pegaspargase, and mitoxantrone (COPEL +MTZ) were given on 31 October 2017 and 30 November 2017, respectively. Bone marrow evaluation on 8 January 2018 revealed 15.5% archeocytes, and immunophenotyping revealed that HLA-DR, CD7, CD38, CD33, and CD11b were expressed, cytoplasmic (cy)MPO was weakly expressed, and CD117, CD13, CD64, CD14, CD3, and CD79a were not expressed. Positron emission tomography (PET)-CT revealed that multiple superficial and deep lymph nodes were enlarged, and the maximum standardized uptake value (SUV max ) was approximately 13.3. Cervical lymph node biopsy was repeated and revealed destruction of the normal structure and diffuse immature cell proliferation. Immunohistochemistry revealed the following: CD5 (+), CD7 (+), TDT partial (+), CD99 (+), CD33 (+), CD117 partial (+), B cell chronic lymphocytic leukemia/lymphoma-2 (Bcl-2) (+), and CD3 partial weak (+); TCRγδ gene rearrangement was positive. Following decitabine + cyclophosphamide + vindesine (HOAP) chemotherapy and two courses of gemcitabine, dexamethasone, and cisplatin (GDP) + hyaluronic acid (HA), the enlarged lymph nodes disappeared. In June 2018, bone marrow evaluation revealed 38% archeocytes, and immunophenotyping revealed two groups of abnormal archeocytes, accounting for 34.2% of all cells; 24.1% of these cells expressed CD3, CD5, CD7, and CD34, and 10.1% of the cells expressed CD34dim, CD7, CD5, and CD38. The following were not expressed: CD33, cyMPO, CD3, CD79a, CD13, and HLA-DR. The patient underwent successive fludarabine + cytarabine + recombinant human granulocyte colony-stimulating factor (G-CSF) (FLAG) and HOAP chemotherapy regimens, after which, bone marrow evaluation revealed 47% archeocytes. Immunophenotyping revealed CD34 (+), CD38 (+), CD7 (+), and CD45dim (+), and weak expression of CD33, CD3, and CD79a. Subsequently, three courses of chemotherapy with vincristine, cyclophosphamide, daunorubicin/idarubicin, and prednisone (VDCP) was administered. In addition, the patient’s lactate level fluctuated between 10.0 and 20.0 mmol/L from 30 January 2018. During the treatment, the patient’s condition continued to progress, and he died in January 2019.
We reported a rare case of AML complicated with T-LBL. According to the flow cytometry and bone marrow biopsy results at the first visit, the patient was confirmed as having single myeloid clones instead of mixed-phenotype acute leukemia (MPAL) in accordance with the 1998 Immunologic Classification of Leukemias (EGIL) diagnostic score system and the 2016 diagnostic classification standard of the World Health Organization (WHO). The confirmation of AML was clear in our patient; however, because there was no significant change in the superficial and deep lymph nodes, we performed lymph node biopsy, and the morphological and immunohistochemical results indicated T-LBL. Hence, we diagnosed AML complicated with T-LBL. Differential diagnosis is particularly important for patients with these diseases concurrently. Wang et al. reported an AML patient with T-LBL who was finally diagnosed with MPAL according to in-depth examination. Therefore, it was proposed that full attention should be given to immunohistochemical examination for patients whose bone marrow immunophenotypes are inconsistent with peripheral tissues at the preliminary diagnosis. Evaluating myeloid markers, such as CD117, CD33, Lys, and MPO, should be performed with lymph node biopsy. We repeatedly evaluated our patient’s recurrently enlarged lymph nodes compared with the initial diagnosis, and also evaluated myeloid molecular markers. Our findings confirmed a clear diagnosis of T-LBL. In addition to MPAL, myeloid and T-lymphoid expression often occurs simultaneously in early T-cell precursor-acute lymphoblastic leukemia (ETP-ALL). ETP-ALL is a new tentative T-ALL/LBL subtype added in the 2016 WHO classification. ETPs are derived from bone marrow and thymus precursor cells; therefore, ETP-ALL may present with myeloid/T-lymphoid expression. However, the important point differentiating ETP-ALL from AML and MPAL is that patients with ETP-ALL are MPO-negative. Hence, ETP-ALL was also excluded in our patient. Furthermore, myeloid/lymphoid leukemia with eosinophils and abnormalities in PDGFRA, PDGFRB, PDGF1, or pericentriolar material 1 (PCM1)-JAK2 are also present with ETP-ALL. Eosinophil counts were not increased in our patient, and bone marrow gene testing for PDGFRA, PDGFRB, PDGF1, and JAK2 were negative. Therefore, a diagnosis of myeloid/lymphoid leukemia was excluded. The most remarkable feature in the present case is the clonal evolution of marrow archeocytes during the disease course. In the early stage of the disease, the bone marrow had a single myeloid clone and expression. In later stages, two groups of abnormal cells appeared; one group expressed early myeloid markers, while the other group expressed T-lymphoid markers. By the end of the disease course, all bone marrow archeocytes had evolved lymphoid expression. The clinical incidence of this cell line transformation is very low. Rossi et al. statistically analyzed 1482 children with acute leukemia; only nine (0.6%) had line transformation. Furthermore, seven of nine patients had 11q23 chromosome/MLL genetic abnormalities. Therefore, we consider that an abnormal 11q23 chromosome/MLL gene may be correlated with the clone transformation in our patient. Pedigree transformation is rarer in adult patients. In the present case, there was a clonal evolution in the later course of the disease; abnormal cells were transformed from myeloid to lymphoid. We assumed that this immune phenotype was unstable and suggests that the abnormal cells in these patients originate from more primitive hemopoietic progenitor cells, even from the pluripotent stem cell phase. However, this hypothesis requires more clinical cases for further analysis. Another characteristic of the present case was the persistent hyperlactatemia late in the disease course. This was considered a clinical manifestation of aggravation of the primary disease. Hyperlactatemia is more common in solid tumors and malignant hematopathy, although this has been reported only in single cases. – The mechanism of hyperlactatemia is considered overexpression of hexokinase and insulin-like growth factors (IGFs) in tumor cells. Hexokinase catalyzes the initial rate-limiting steps in glycolysis. IGFs are an activator of hexokinase, which can promote glycolysis metabolism even in the presence of oxygen, when both compounds are present at high concentrations. During later treatment, our patient’s lactate levels always decreased after chemotherapy, and subsequently increased. The characteristics of our case show that hyperlactatemia is correlated with hematologic tumors. The treatment plan for AML with T-LBL is still being researched ( ). – In the present case, the IA regimen was introduced as the initial treatment. Bone marrow findings improved, but the lymph nodes did not shrink. The patient had short-term recurrence, but he responded to the combination VCLP + IA regimen. A variety of combined regimens were used during the second recurrence, but all failed to achieve remission, suggesting that the prognosis of AML with T-LBL is poor, and patients are prone to relapse. Currently, allogeneic hematopoietic stem cell therapy (allo-HSCT) may be the only option to cure this disease. Considering that the present study was a case report without a control group, our conclusions would be more robust if supported in multicenter clinical research.
There is currently no treatment regimen recommended for AML with T-LBL. Allo-HSCT may be the only option to cure this disease.
|
Maternal nutrient restriction during late gestation reduces vigor and alters blood chemistry and hematology in neonatal beef calves | 7baadcc1-ea98-4d33-8e48-c3a288c3dec1 | 10648570 | Internal Medicine[mh] | Poor maternal nutrition during gestation can alter nutrient delivery to the fetus, ultimately impairing prenatal growth and development as well as pre-weaning health and survival . Survival through the neonatal period is one of the greatest challenges to beef calves, and it depends on a successful transition to extrauterine life . Following birth, calves must be vigorous to quickly stand and ingest colostrum for necessary nutrients and transfer of passive immunity . Despite this, few intensive studies have been conducted in beef cattle to determine effects of poor nutrition during pregnancy on neonatal calf vigor and metabolism. During the last third of gestation, nutrient requirements of beef females increase rapidly to match the increase in demand for nutrients by the fetus ; however, they often may not meet these requirements due to poor forage quality or quantity . Additionally, primiparous dams likely have the greatest challenge while attempting to meet both fetal demands and their own growth requirements . We have previously reported that neonatal calves born to primiparous females have impaired metabolic status compared with calves born to multiparous females , suggesting that calves born to first-parity dams are already at a disadvantage regardless of maternal nutrition. Overall, more data are needed to understand how poor nutrition during late gestation affects neonatal calf physiology and nutrient availability. Data reported here are from a large experiment to determine the effects of nutrient restriction in late pregnant first-parity beef females on prenatal and postnatal nutrient availability and utilization by calves. We hypothesized that late gestational nutrient restriction would impair vigor and metabolic status of neonatal beef calves. Our objectives were to determine the effects of maternal nutrient restriction during late gestation on neonatal calf vigor, passive transfer, maintenance of body temperature, blood chemistry, and hematology.
The University of Missouri Animal Care and Use Committee approved animal care and use (Protocol #9877) in this study which took place at the University of Missouri Beef Research and Teaching Farm (Columbia, MO). Animal management and diets Animal management and diets during pregnancy have been described by . Briefly, 26 single-sired fall-calving Hereford × Simmental-Angus crossbred beef heifers [initial body weight ( BW ) = 451 ± 28 (SD throughout methods) kg, initial body condition score ( BCS ) = 5.4 ± 0.7] bred by artificial insemination to a single Angus sire were allocated by BW, BCS, fetal sex, and expected calving date to 1 of 2 late gestational nutritional planes from day 160 of gestation to parturition. Control ( CON ; n = 13) heifers were individually-fed 100% of metabolizable energy ( ME ) and metabolizable protein ( MP ) requirements for maintenance, pregnancy, and growth, whereas nutrient restricted ( NR ; n = 13) heifers were individually-fed 70% of ME and MP requirements. Heifers were housed in 12 partially-covered 3.7 × 15.8 m pens ( n = 2 to 3 per pen), penned by nutritional plane, and individually-fed via a Calan gate feeding system (Calan Broadbent Feeding System, American Calan, Northwood, NH). Nutrient requirements were estimated using an expected calf birth weight of 34 kg and a projected maternal average daily gain of 0.36 kg/d. Metabolizable energy for maintenance was based on data for heifers in confinement (0.138 Mcal ME/kg non-gravid BW 0.75 ; Freetly and Hales, personal communication). The equation used for ME for conceptus was published previously . Equations from were utilized for ME for gain and MP for maintenance, conceptus, and gain. Nutrient requirements were adjusted weekly using the most recent dam BW (recorded every 21 d) and day of gestation. From days 160 to 265 of gestation, diets were based on chopped sorghum sudan hay [1.74 Mcal ME/kg, 6.69% crude protein ( CP ), 72.0% neutral detergent fiber ( NDF ), 52.8% acid detergent fiber ( ADF ); dry matter ( DM ) basis]. From day 266 of gestation to parturition, diets were based on chopped endophyte-infected tall fescue-based hay (1.90 Mcal ME/kg, 7.22% CP, 65.1% NDF, 43.2% ADF; DM basis). Feeding poor quality forage allowed animals fed both nutritional planes to consume ad libitum hay without exceeding the ME and CP targets at any point. Using expected individual hay intakes (estimated from the previous week’s hay intake), heifers were supplemented daily with whole corn, dried distillers’ grains with solubles, and soyhull pellets to meet their assigned nutritional plane. Dams had ad libitum access to water and a trace mineralized salt block (Big 6 Mineral Salt, Compass Minerals America Inc., Overland Park, KS). Calving monitoring and data collection Heifers remained in partially covered pens during the peripartum period. Beginning on day 274 of gestation, heifers were closely monitored 24 h per day by trained personnel to detect when heifers were in stage 2 of parturition (evident contractions, raised tail, or appearance of amniotic fluid or membranes) by walking down aisles located in the front and back of the pens at least once every 15 to 30 min. Once stage 2 was detected, heifers were continuously monitored to allow for all data collection. Overhead lighting in the barn and outdoor lighting allowed for continuous monitoring of animals throughout the night, and a handheld spotlight was used as necessary. Calving assistance was provided (CON n = 2; NR n = 4) in the pen, or the dam was moved to the chute for delivery assistance if the calf was presenting abnormally, there was a prolonged duration since the first appearance of fetal membranes, or if progress slowed during contractions. A calving difficulty score ranging from 1 to 5 was assigned, with 1 = no assistance, 2 = easy pull (by hand or with obstetrical chains), 3 = mechanically-assisted pull, 4 = abnormal presentation, and 5 = cesarean-section. No calving difficulty score 5 occurred. Trained personnel recorded times for calving and vigor behaviors in real time. This allowed for the determination of duration of parturition, as well as calf vigor using both behavioral latencies and vigor scores. The estimated first appearance of fetal feet was used for 1 CON and 3 NR females based on the last time of monitoring prior to the appearance of feet, but actual times were used for all other measures. Behavioral latencies included time to sternal recumbency, time to first attempt to stand, and time to stand. A vigor score was also assigned to calves at 2, 5, 10, and 20 min of age on a scale of 1 to 5 [1 = very weak (lying flat on side), 2 = weak (flat on side with head up), 3 = active and vigorous (sternal), 4 = very active and vigorous (attempting to stand or has attempted to stand), and 5 = extremely active and vigorous (standing or has stood successfully)] as described in . The number of failed attempts to stand by each calf that were not caused by its dam was also recorded. Following the successful standing of the calf, but prior to suckling, each calf was removed from the dam and processed. Each calf was given an ear tag for visual identification, had their umbilicus sprayed with chlorhexidine solution, and was administered a subcutaneous Bo-Se injection (5.5 mL/100 kg BW; Intervet/Merck Animal Health, Madison, NJ). The average calving date was September 16 ± 32.9 d, and there was no perinatal calf death loss. Neonatal blood collection and analyses Calf jugular blood samples and rectal temperatures were collected at 0 (prior to suckling but after standing), 6, 12, 24, and 48 h of age (0.6 ± 0.3, 6.0 ± 0.2, 12.1 ± 0.2, 24.2 ± 0.2, 48.1 ± 0.3 h, respectively). Blood samples were collected into Vacutainer serum collection tubes containing no additives (2 tubes at 0 and 48 h, 1 tube at other sampling hours; 10 mL draw; Becton Dickinson, Franklin Lakes, NJ) and 1 Monoject plasma collection tube containing 0.10 mL of 15% K 3 EDTA (10 mL draw; Covidien, Mansfield, MA). Following collection, all tubes were inverted several times. Plasma tubes were immediately placed on ice, whereas serum tubes were allowed to clot for approximately 5 to 15 min prior to being placed on ice. Between 1 and 8 h after collection, all samples were centrifuged at 1,500 × g at 4°C for 30 min. Serum and plasma were transferred into multiple 2-mL microcentrifuge tubes and stored at −20 °C until analysis unless they were refrigerated for analysis as described below for blood chemistry. To evaluate the transfer of passive immunity, 48-h calf serum was analyzed for immunoglobulin ( Ig ) G and A concentrations. These were analyzed by a colorimetric sandwich enzyme-linked immunosorbent assay ( ELISA ) using Bovine IgG and IgA ELISA Kits (Bethyl Laboratories, Inc., Montgomery, TX) following the manufacturer’s instructions and methods described in . Serum samples were thawed at 4 °C and diluted 1:250,000 for IgG and 1:7,500 for IgA in sterile polypropylene tubes using the dilution buffer provided with the kit. Samples were analyzed in duplicate, and pooled serum samples served as assay controls. The intraassay and interassay CV were 4.07% and 6.90% for serum IgG, respectively, and the intraassay CV was 3.04% for IgA (single plate). Calf serum non-esterified fatty acids ( NEFA ) were determined as described by , and plasma triglycerides were determined as described by at each sampling hour using commercially-available colorimetric assays. For each assay, samples were analyzed in duplicate, and pooled control samples were used. The intraassay and interassay CV were 3.83% and 4.46% for serum NEFA and 3.30% and 1.05% for plasma triglycerides, respectively. A 1-mL serum aliquot and 1-mL whole blood aliquot (collected with Monoject plasma collection tube containing 0.10 mL of 15% K 3 EDTA; aliquoted after gentle tube inversion prior to centrifugation for plasma) from each sampling time were refrigerated and then transported on ice to the University of Missouri Veterinary Medical Diagnostic Clinical Pathology Laboratory for blood chemistry and complete blood cell analysis. Serum glucose, urea N, creatinine, globulin, albumin, total protein, sodium, potassium, chloride, phosphorus, calcium, magnesium, bicarbonate, anion gap, direct bilirubin, total bilirubin, aspartate aminotransferase ( AST ), gamma-glutamyl transferase ( GGT ), and creatine kinase ( CK ) concentrations were determined using a Beckman Coulter AU480 Chemistry Analyzer (Beckman Coulter, Inc., Brea, CA). White blood cells ( WBC ), red blood cells ( RBC ), hemoglobin, hematocrit, mean corpuscular volume ( MCV ), mean corpuscular hemoglobin ( MCH ), mean corpuscular hemoglobin concentration ( MCHC ), and platelet count were determined using a Sysmex XT-2000iV (Sysmex Nordic Aps Filial Sverige, Landskrona, Sweden). Internal quality control and verification of performance within specific CV were conducted daily. Upon delivery, samples were analyzed at the same time through the instrument’s completely automated process. Samples were analyzed within 24 h of collection. Calf plasma insulin concentration at each sampling hour was analyzed by radioimmunoassay ( RIA ) in duplicate using a commercial kit (Human Insulin Specific RIA, EMD Millipore Corp., Billerica, MA) as previously described . The intra-assay CV was 2.80% (single assay). Calf plasma cortisol concentration at each sampling time was analyzed using a commercial coated-tube RIA kit (ImmunChem Coated Tube Cortisol 125 I RIA Kit, MP Biomedicals, Irvine, CA) in duplicate. Manufacturer instructions were followed, except samples were incubated with Cortisol- 125 I for 4 h at room temperature. The intra-assay CV was 1.09% (single assay). Statistical analyses One heifer (CON) was removed from the study due to late gestational abortion, resulting in CON n = 12 and NR n = 13. Vigor score at 2 min and time to sternal recumbency were not included for 1 NR calf because it was born through the fence into an adjacent pen and required human intervention to be put back into its respective pen. The only vigor measures recorded for calves that had a calving difficulty score of 4 (3 NR) included time to stand and 20 min vigor score because calves were pulled in the chute, which disrupted earlier vigor behaviors. Vigor data from calves with a calving difficulty of 2 or 3 (2 CON and 1 NR) were included in the analysis because assistance was provided in the pen and no further intervention was provided following the birth of the calf. At least 4 of the 5 sampling times were present for all rectal temperature, blood chemistry, hormone, and hematology data (3 animals missing 1 time point for rectal temperature and 1 missed time point for hematology). The final n for each variable is given in tables and figures descriptions. Calf vigor behavioral latencies and serum Ig were analyzed using the MIXED procedure in SAS 9.4 (SAS Institute Inc., Cary, NC) with late gestational nutritional plane as a fixed effect. For rectal temperature, blood chemistry, plasma hormone, and hematology data over time, late gestational nutritional plane, hour, and their interaction were considered fixed effects. These were considered repeated measures using the majority best-fit covariance structure (based on Akaike Information Criterion, Bayesian Information Criterion, and corrected Bayesian Information Criterion) specific for each variable (chosen from unstructured, compound symmetry, heterogeneous compound symmetry, autoregressive, and heterogeneous autoregressive). Calf vigor score data were analyzed using a cumulative logistic regression model with the GLIMMIX procedure in SAS 9.4 with late gestational nutritional plane as a fixed effect. For all measures, animal was considered the experimental unit, Julian date of calving was included as a fixed effect (to account for variation in ambient conditions at calving), and calf sex was included as a fixed effect when P < 0.25. Means were separated using least significant difference and considered different when P ≤ 0.05 and tendencies were considered when 0.05 < P ≤ 0.10. In the absence of interactions, main effects of nutritional plane are reported. Additionally, PROC CORR was used to determine the Pearson correlation coefficients for calf serum IgG with total protein, globulin, and GGT at 48 h, both within nutritional planes and for all data combined.
Animal management and diets during pregnancy have been described by . Briefly, 26 single-sired fall-calving Hereford × Simmental-Angus crossbred beef heifers [initial body weight ( BW ) = 451 ± 28 (SD throughout methods) kg, initial body condition score ( BCS ) = 5.4 ± 0.7] bred by artificial insemination to a single Angus sire were allocated by BW, BCS, fetal sex, and expected calving date to 1 of 2 late gestational nutritional planes from day 160 of gestation to parturition. Control ( CON ; n = 13) heifers were individually-fed 100% of metabolizable energy ( ME ) and metabolizable protein ( MP ) requirements for maintenance, pregnancy, and growth, whereas nutrient restricted ( NR ; n = 13) heifers were individually-fed 70% of ME and MP requirements. Heifers were housed in 12 partially-covered 3.7 × 15.8 m pens ( n = 2 to 3 per pen), penned by nutritional plane, and individually-fed via a Calan gate feeding system (Calan Broadbent Feeding System, American Calan, Northwood, NH). Nutrient requirements were estimated using an expected calf birth weight of 34 kg and a projected maternal average daily gain of 0.36 kg/d. Metabolizable energy for maintenance was based on data for heifers in confinement (0.138 Mcal ME/kg non-gravid BW 0.75 ; Freetly and Hales, personal communication). The equation used for ME for conceptus was published previously . Equations from were utilized for ME for gain and MP for maintenance, conceptus, and gain. Nutrient requirements were adjusted weekly using the most recent dam BW (recorded every 21 d) and day of gestation. From days 160 to 265 of gestation, diets were based on chopped sorghum sudan hay [1.74 Mcal ME/kg, 6.69% crude protein ( CP ), 72.0% neutral detergent fiber ( NDF ), 52.8% acid detergent fiber ( ADF ); dry matter ( DM ) basis]. From day 266 of gestation to parturition, diets were based on chopped endophyte-infected tall fescue-based hay (1.90 Mcal ME/kg, 7.22% CP, 65.1% NDF, 43.2% ADF; DM basis). Feeding poor quality forage allowed animals fed both nutritional planes to consume ad libitum hay without exceeding the ME and CP targets at any point. Using expected individual hay intakes (estimated from the previous week’s hay intake), heifers were supplemented daily with whole corn, dried distillers’ grains with solubles, and soyhull pellets to meet their assigned nutritional plane. Dams had ad libitum access to water and a trace mineralized salt block (Big 6 Mineral Salt, Compass Minerals America Inc., Overland Park, KS).
Heifers remained in partially covered pens during the peripartum period. Beginning on day 274 of gestation, heifers were closely monitored 24 h per day by trained personnel to detect when heifers were in stage 2 of parturition (evident contractions, raised tail, or appearance of amniotic fluid or membranes) by walking down aisles located in the front and back of the pens at least once every 15 to 30 min. Once stage 2 was detected, heifers were continuously monitored to allow for all data collection. Overhead lighting in the barn and outdoor lighting allowed for continuous monitoring of animals throughout the night, and a handheld spotlight was used as necessary. Calving assistance was provided (CON n = 2; NR n = 4) in the pen, or the dam was moved to the chute for delivery assistance if the calf was presenting abnormally, there was a prolonged duration since the first appearance of fetal membranes, or if progress slowed during contractions. A calving difficulty score ranging from 1 to 5 was assigned, with 1 = no assistance, 2 = easy pull (by hand or with obstetrical chains), 3 = mechanically-assisted pull, 4 = abnormal presentation, and 5 = cesarean-section. No calving difficulty score 5 occurred. Trained personnel recorded times for calving and vigor behaviors in real time. This allowed for the determination of duration of parturition, as well as calf vigor using both behavioral latencies and vigor scores. The estimated first appearance of fetal feet was used for 1 CON and 3 NR females based on the last time of monitoring prior to the appearance of feet, but actual times were used for all other measures. Behavioral latencies included time to sternal recumbency, time to first attempt to stand, and time to stand. A vigor score was also assigned to calves at 2, 5, 10, and 20 min of age on a scale of 1 to 5 [1 = very weak (lying flat on side), 2 = weak (flat on side with head up), 3 = active and vigorous (sternal), 4 = very active and vigorous (attempting to stand or has attempted to stand), and 5 = extremely active and vigorous (standing or has stood successfully)] as described in . The number of failed attempts to stand by each calf that were not caused by its dam was also recorded. Following the successful standing of the calf, but prior to suckling, each calf was removed from the dam and processed. Each calf was given an ear tag for visual identification, had their umbilicus sprayed with chlorhexidine solution, and was administered a subcutaneous Bo-Se injection (5.5 mL/100 kg BW; Intervet/Merck Animal Health, Madison, NJ). The average calving date was September 16 ± 32.9 d, and there was no perinatal calf death loss.
Calf jugular blood samples and rectal temperatures were collected at 0 (prior to suckling but after standing), 6, 12, 24, and 48 h of age (0.6 ± 0.3, 6.0 ± 0.2, 12.1 ± 0.2, 24.2 ± 0.2, 48.1 ± 0.3 h, respectively). Blood samples were collected into Vacutainer serum collection tubes containing no additives (2 tubes at 0 and 48 h, 1 tube at other sampling hours; 10 mL draw; Becton Dickinson, Franklin Lakes, NJ) and 1 Monoject plasma collection tube containing 0.10 mL of 15% K 3 EDTA (10 mL draw; Covidien, Mansfield, MA). Following collection, all tubes were inverted several times. Plasma tubes were immediately placed on ice, whereas serum tubes were allowed to clot for approximately 5 to 15 min prior to being placed on ice. Between 1 and 8 h after collection, all samples were centrifuged at 1,500 × g at 4°C for 30 min. Serum and plasma were transferred into multiple 2-mL microcentrifuge tubes and stored at −20 °C until analysis unless they were refrigerated for analysis as described below for blood chemistry. To evaluate the transfer of passive immunity, 48-h calf serum was analyzed for immunoglobulin ( Ig ) G and A concentrations. These were analyzed by a colorimetric sandwich enzyme-linked immunosorbent assay ( ELISA ) using Bovine IgG and IgA ELISA Kits (Bethyl Laboratories, Inc., Montgomery, TX) following the manufacturer’s instructions and methods described in . Serum samples were thawed at 4 °C and diluted 1:250,000 for IgG and 1:7,500 for IgA in sterile polypropylene tubes using the dilution buffer provided with the kit. Samples were analyzed in duplicate, and pooled serum samples served as assay controls. The intraassay and interassay CV were 4.07% and 6.90% for serum IgG, respectively, and the intraassay CV was 3.04% for IgA (single plate). Calf serum non-esterified fatty acids ( NEFA ) were determined as described by , and plasma triglycerides were determined as described by at each sampling hour using commercially-available colorimetric assays. For each assay, samples were analyzed in duplicate, and pooled control samples were used. The intraassay and interassay CV were 3.83% and 4.46% for serum NEFA and 3.30% and 1.05% for plasma triglycerides, respectively. A 1-mL serum aliquot and 1-mL whole blood aliquot (collected with Monoject plasma collection tube containing 0.10 mL of 15% K 3 EDTA; aliquoted after gentle tube inversion prior to centrifugation for plasma) from each sampling time were refrigerated and then transported on ice to the University of Missouri Veterinary Medical Diagnostic Clinical Pathology Laboratory for blood chemistry and complete blood cell analysis. Serum glucose, urea N, creatinine, globulin, albumin, total protein, sodium, potassium, chloride, phosphorus, calcium, magnesium, bicarbonate, anion gap, direct bilirubin, total bilirubin, aspartate aminotransferase ( AST ), gamma-glutamyl transferase ( GGT ), and creatine kinase ( CK ) concentrations were determined using a Beckman Coulter AU480 Chemistry Analyzer (Beckman Coulter, Inc., Brea, CA). White blood cells ( WBC ), red blood cells ( RBC ), hemoglobin, hematocrit, mean corpuscular volume ( MCV ), mean corpuscular hemoglobin ( MCH ), mean corpuscular hemoglobin concentration ( MCHC ), and platelet count were determined using a Sysmex XT-2000iV (Sysmex Nordic Aps Filial Sverige, Landskrona, Sweden). Internal quality control and verification of performance within specific CV were conducted daily. Upon delivery, samples were analyzed at the same time through the instrument’s completely automated process. Samples were analyzed within 24 h of collection. Calf plasma insulin concentration at each sampling hour was analyzed by radioimmunoassay ( RIA ) in duplicate using a commercial kit (Human Insulin Specific RIA, EMD Millipore Corp., Billerica, MA) as previously described . The intra-assay CV was 2.80% (single assay). Calf plasma cortisol concentration at each sampling time was analyzed using a commercial coated-tube RIA kit (ImmunChem Coated Tube Cortisol 125 I RIA Kit, MP Biomedicals, Irvine, CA) in duplicate. Manufacturer instructions were followed, except samples were incubated with Cortisol- 125 I for 4 h at room temperature. The intra-assay CV was 1.09% (single assay).
One heifer (CON) was removed from the study due to late gestational abortion, resulting in CON n = 12 and NR n = 13. Vigor score at 2 min and time to sternal recumbency were not included for 1 NR calf because it was born through the fence into an adjacent pen and required human intervention to be put back into its respective pen. The only vigor measures recorded for calves that had a calving difficulty score of 4 (3 NR) included time to stand and 20 min vigor score because calves were pulled in the chute, which disrupted earlier vigor behaviors. Vigor data from calves with a calving difficulty of 2 or 3 (2 CON and 1 NR) were included in the analysis because assistance was provided in the pen and no further intervention was provided following the birth of the calf. At least 4 of the 5 sampling times were present for all rectal temperature, blood chemistry, hormone, and hematology data (3 animals missing 1 time point for rectal temperature and 1 missed time point for hematology). The final n for each variable is given in tables and figures descriptions. Calf vigor behavioral latencies and serum Ig were analyzed using the MIXED procedure in SAS 9.4 (SAS Institute Inc., Cary, NC) with late gestational nutritional plane as a fixed effect. For rectal temperature, blood chemistry, plasma hormone, and hematology data over time, late gestational nutritional plane, hour, and their interaction were considered fixed effects. These were considered repeated measures using the majority best-fit covariance structure (based on Akaike Information Criterion, Bayesian Information Criterion, and corrected Bayesian Information Criterion) specific for each variable (chosen from unstructured, compound symmetry, heterogeneous compound symmetry, autoregressive, and heterogeneous autoregressive). Calf vigor score data were analyzed using a cumulative logistic regression model with the GLIMMIX procedure in SAS 9.4 with late gestational nutritional plane as a fixed effect. For all measures, animal was considered the experimental unit, Julian date of calving was included as a fixed effect (to account for variation in ambient conditions at calving), and calf sex was included as a fixed effect when P < 0.25. Means were separated using least significant difference and considered different when P ≤ 0.05 and tendencies were considered when 0.05 < P ≤ 0.10. In the absence of interactions, main effects of nutritional plane are reported. Additionally, PROC CORR was used to determine the Pearson correlation coefficients for calf serum IgG with total protein, globulin, and GGT at 48 h, both within nutritional planes and for all data combined.
Vigor and transfer of passive immunity Duration of parturition was not affected ( P = 0.65) by late gestational nutritional plane (CON: 40.6 ± 6.6 min vs. NR: 44.7 ± 6.3 min). Although there was no effect ( P = 0.45) of late gestational nutritional plane on time to sternal recumbency, calves born to NR dams tended to attempt to stand 6.5 min (56%) slower ( P = 0.09) and stood 14 min (67%) slower ( P = 0.02) than calves born to CON dams . Despite this, late gestational nutritional plane did not affect ( P = 0.58) the number of failed attempts to stand. Calves born to NR dams were more likely to have a poorer ( P = 0.05) vigor score at 20 min of age compared with CON calves , but late gestational nutritional plane did not affect ( P ≥ 0.55) vigor scores at 2, 5, and 10 min of age . Calves born to NR dams had greater serum IgG ( P = 0.001) and IgA ( P = 0.03) at 48 h of age compared with calves born to CON . At 48 h, serum IgG was very strongly correlated with serum total protein ( r = 0.86 to 0.92, P < 0.001; ) and globulin ( r = 0.87 to 0.96, P < 0.001) for both nutritional planes, and all data combined. When all data were combined, serum IgG was not correlated with GGT ( r = 0.29, P = 0.16). Serum IgG tended to be moderately correlated ( r = 0.54, P = 0.07) with GGT for calves born to CON, but there was no relationship ( r = −0.08, P = 0.81) for NR. Rectal temperature and blood chemistry The average Julian date of calving (261 ± 10 vs. 260 ± 9 for CON and NR, respectively; P = 0.93) and ambient temperature during the hour of calving (from on-farm weather station; 21.7 ± 2.1 vs. 20.2 ± 2.0 °C for CON and NR, respectively, P = 0.61) did not differ between late gestational nutritional planes. There was an interaction ( P = 0.02) of late gestational nutritional plane × hour for neonatal calf rectal temperature . Calves born to CON dams had greater ( P = 0.02) rectal temperature at 0 h of age; however, at 24 h postnatal, calves from NR dams had greater ( P = 0.04) rectal temperature. Within CON calves, rectal temperatures at 0 and 48 h were greater ( P ≤ 0.05) than at 12 and 24 h postnatal. Rectal temperature of NR calves increased ( P = 0.01) from 0 to 6 h of age, but did not change ( P ≥ 0.45) from 6 to 48 h. Neither the late gestational nutritional plane × hour interaction nor nutritional plane affected ( P ≥ 0.18) neonatal calf serum glucose, serum NEFA, or plasma triglyceride concentrations . Additionally, plasma cortisol and insulin were not affected ( P ≥ 0.46) by the late gestational nutritional plane × hour interaction or nutritional plane . There was an interaction ( P = 0.04) of late gestational nutritional plane × hour for serum urea N concentrations . Although urea N did not differ ( P ≥ 0.17) between nutritional planes at any of the sampling time points, the pattern of change differed between nutritional planes. Within calves born to CON dams, serum urea N increased ( P < 0.001) from 0 to 6 h of age and was less ( P = 0.02) at 48 h than 6 h. Serum urea N increased ( P < 0.001) from 0 to 12 h, but then decreased ( P < 0.001) from 24 to 48 h postnatal in calves born to NR dams. There was also an interaction ( P = 0.03) of late gestational nutritional plane × hour for serum creatinine, where NR calves had greater ( P = 0.03) creatinine concentration at 24 h of age compared with CON calves . Serum creatinine decreased ( P ≤ 0.004) from 0 to 48 h of age in both CON and NR calves. There was an interaction ( P ≤ 0.006) of late gestational nutritional plane × hour for serum total protein and globulin ( and ). Calves born to NR dams had greater ( P ≤ 0.02) serum total protein and globulin than CON at 6, 12, 24, and 48 h of age. Serum total protein increased ( P < 0.001) from 6 to 24 h for both CON and NR. In calves born to NR dams, total protein also increased ( P < 0.001) from 0 to 6 h and decreased ( P < 0.001) from 24 to 48 h of age. Serum globulin had a similar pattern to total protein, increasing ( P < 0.001) from 6 to 24 h of age and then decreasing ( P ≤ 0.01) from 24 to 48 h in both CON and NR calves. Calves born to NR dams also had an increase ( P < 0.001) in globulin from 0 to 6 h. There was no effect ( P ≥ 0.39) of the late gestational nutritional plane × hour interaction or nutritional plane on serum albumin (data not shown). There was an interaction ( P ≤ 0.03) of late gestational nutritional plane × hour for serum AST and GGT activities and a tendency for an interaction ( P = 0.10) of late gestational nutritional plane × hour for serum CK activity . Calves born to NR dams had greater ( P ≤ 0.01) AST at 0, 6, and 12 h and tended to have greater ( P = 0.07) AST at 24 h postnatal compared with CON. In calves born to CON dams, serum AST increased ( P ≤ 0.01) from 0 to 24 h, then decreased ( P < 0.001) from 24 to 48 h of age. Serum AST in NR calves increased ( P < 0.001) between 0 and 12 h and then decreased ( P < 0.001) from 24 to 48 h of age. Serum CK was greater ( P ≤ 0.04) at 6, 12, and 24 h in calves born to NR dams. CK in calves born to CON dams increased ( P = 0.02) from 0 to 6 h and tended to decrease ( P = 0.09) from 12 to 24 h postnatal. In calves born to NR dams, CK increased ( P < 0.001) from 0 to 6 h but then decreased ( P ≤ 0.006) from 12 and 48 h of age. Serum GGT was greater ( P = 0.006) at 6 h and tended to be greater ( P ≤ 0.08) at 12 and 48 h of age in calves born to NR dams than CON. In calves born to CON dams, serum GGT increased ( P < 0.001) from 6 to 12 h and then decreased ( P < 0.001) from 24 to 48 h postnatal. In NR calves, serum GGT increased ( P < 0.001) from 0 to 12 h and then decreased ( P < 0.001) from 24 to 48 h of age. There was a tendency for a late gestational nutritional plane × hour interaction ( P = 0.08) for serum sodium and an interaction ( P = 0.03) of late gestational nutritional plane × hour for serum potassium . Calves born to CON dams tended to have greater ( P = 0.09) serum sodium at 0 h and had greater ( P ≤ 0.004) sodium from 6 to 48 h postnatal. Within calves born to CON dams, sodium concentration decreased ( P < 0.001) from 6 to 24 h of age. Serum sodium of NR calves decreased ( P ≤ 0.003) from 0 to 24 h and then increased ( P = 0.01) from 24 to 48 h of age. Serum potassium was greater ( P = 0.03) in calves born to NR dams at 0 h postnatal when compared with CON. Potassium concentrations in calves born to CON dams tended to increase ( P = 0.06) from 0 to 6 h and tended to decrease ( P = 0.08) from 24 to 48 h. There was a decrease ( P = 0.01) in serum potassium from 0 to 6 h of age in calves born to NR dams. There was no effect ( P = 0.21) of the late gestational nutritional plane × hour interaction for serum chloride, but there tended to be an effect of nutritional plane ( P = 0.08), where calves born to CON dams had greater concentrations compared with NR (100.6 ± 0.5 vs. 99.3 ± 0.5). There was no effect ( P ≥ 0.12) of the late gestational nutritional plane × hour interaction or nutritional plane on serum calcium, phosphorus, or magnesium (data not shown). There tended to be an interaction ( P = 0.10) of late gestational nutritional plane × hour for serum anion gap . Calves born to CON dams had greater ( P = 0.02) anion gap at 6 h of age compared with NR. Anion gap decreased ( P ≤ 0.008) from 6 to 24 h in CON calves and decreased ( P < 0.001) from 12 to 24 h in NR calves. There was no effect ( P ≥ 0.46) of the late gestational nutritional plane × hour interaction or nutritional plane on neonatal calf serum bicarbonate concentrations (data not shown). There was no effect ( P ≥ 0.24) of the late gestational nutritional plane × hour interaction or nutritional plane on neonatal calf serum total or direct bilirubin (data not shown). Hematology There was an interaction ( P ≤ 0.04) of late gestational nutritional plane × hour for RBC, hemoglobin, and hematocrit . Calves born to CON dams tended to have greater ( P ≤ 0.09) RBC and hemoglobin at 6 and 12 h postnatal compared with NR. Red blood cells decreased ( P ≤ 0.02) from 0 to 48 h postnatal in CON. In calves born to NR dams, RBC decreased ( P ≤ 0.04) from 0 to 12 h and then from 24 to 48 h of age. Hemoglobin followed a similar pattern where it decreased ( P ≤ 0.007) from 0 to 48 h of age in CON, and it decreased ( P < 0.001) from 0 to 12 h and tended to decrease ( P = 0.08) from 24 to 48 h in NR. Hematocrit concentrations were greater ( P ≤ 0.04) at 6 and 12 h and tended to be greater ( P = 0.10) at 24 h of age in calves born to CON dams compared with NR. In both CON and NR calves, hematocrit decreased ( P ≤ 0.008) from 0 to 48 h postnatal. Neither the late gestational nutritional plane × hour interaction nor nutritional plane affected neonatal calf WBC ( P ≥ 0.29), MCV ( P ≥ 0.12), MCH ( P ≥ 0.25), MCHC ( P ≥ 0.14), or platelet count ( P ≥ 0.70; ).
Duration of parturition was not affected ( P = 0.65) by late gestational nutritional plane (CON: 40.6 ± 6.6 min vs. NR: 44.7 ± 6.3 min). Although there was no effect ( P = 0.45) of late gestational nutritional plane on time to sternal recumbency, calves born to NR dams tended to attempt to stand 6.5 min (56%) slower ( P = 0.09) and stood 14 min (67%) slower ( P = 0.02) than calves born to CON dams . Despite this, late gestational nutritional plane did not affect ( P = 0.58) the number of failed attempts to stand. Calves born to NR dams were more likely to have a poorer ( P = 0.05) vigor score at 20 min of age compared with CON calves , but late gestational nutritional plane did not affect ( P ≥ 0.55) vigor scores at 2, 5, and 10 min of age . Calves born to NR dams had greater serum IgG ( P = 0.001) and IgA ( P = 0.03) at 48 h of age compared with calves born to CON . At 48 h, serum IgG was very strongly correlated with serum total protein ( r = 0.86 to 0.92, P < 0.001; ) and globulin ( r = 0.87 to 0.96, P < 0.001) for both nutritional planes, and all data combined. When all data were combined, serum IgG was not correlated with GGT ( r = 0.29, P = 0.16). Serum IgG tended to be moderately correlated ( r = 0.54, P = 0.07) with GGT for calves born to CON, but there was no relationship ( r = −0.08, P = 0.81) for NR.
The average Julian date of calving (261 ± 10 vs. 260 ± 9 for CON and NR, respectively; P = 0.93) and ambient temperature during the hour of calving (from on-farm weather station; 21.7 ± 2.1 vs. 20.2 ± 2.0 °C for CON and NR, respectively, P = 0.61) did not differ between late gestational nutritional planes. There was an interaction ( P = 0.02) of late gestational nutritional plane × hour for neonatal calf rectal temperature . Calves born to CON dams had greater ( P = 0.02) rectal temperature at 0 h of age; however, at 24 h postnatal, calves from NR dams had greater ( P = 0.04) rectal temperature. Within CON calves, rectal temperatures at 0 and 48 h were greater ( P ≤ 0.05) than at 12 and 24 h postnatal. Rectal temperature of NR calves increased ( P = 0.01) from 0 to 6 h of age, but did not change ( P ≥ 0.45) from 6 to 48 h. Neither the late gestational nutritional plane × hour interaction nor nutritional plane affected ( P ≥ 0.18) neonatal calf serum glucose, serum NEFA, or plasma triglyceride concentrations . Additionally, plasma cortisol and insulin were not affected ( P ≥ 0.46) by the late gestational nutritional plane × hour interaction or nutritional plane . There was an interaction ( P = 0.04) of late gestational nutritional plane × hour for serum urea N concentrations . Although urea N did not differ ( P ≥ 0.17) between nutritional planes at any of the sampling time points, the pattern of change differed between nutritional planes. Within calves born to CON dams, serum urea N increased ( P < 0.001) from 0 to 6 h of age and was less ( P = 0.02) at 48 h than 6 h. Serum urea N increased ( P < 0.001) from 0 to 12 h, but then decreased ( P < 0.001) from 24 to 48 h postnatal in calves born to NR dams. There was also an interaction ( P = 0.03) of late gestational nutritional plane × hour for serum creatinine, where NR calves had greater ( P = 0.03) creatinine concentration at 24 h of age compared with CON calves . Serum creatinine decreased ( P ≤ 0.004) from 0 to 48 h of age in both CON and NR calves. There was an interaction ( P ≤ 0.006) of late gestational nutritional plane × hour for serum total protein and globulin ( and ). Calves born to NR dams had greater ( P ≤ 0.02) serum total protein and globulin than CON at 6, 12, 24, and 48 h of age. Serum total protein increased ( P < 0.001) from 6 to 24 h for both CON and NR. In calves born to NR dams, total protein also increased ( P < 0.001) from 0 to 6 h and decreased ( P < 0.001) from 24 to 48 h of age. Serum globulin had a similar pattern to total protein, increasing ( P < 0.001) from 6 to 24 h of age and then decreasing ( P ≤ 0.01) from 24 to 48 h in both CON and NR calves. Calves born to NR dams also had an increase ( P < 0.001) in globulin from 0 to 6 h. There was no effect ( P ≥ 0.39) of the late gestational nutritional plane × hour interaction or nutritional plane on serum albumin (data not shown). There was an interaction ( P ≤ 0.03) of late gestational nutritional plane × hour for serum AST and GGT activities and a tendency for an interaction ( P = 0.10) of late gestational nutritional plane × hour for serum CK activity . Calves born to NR dams had greater ( P ≤ 0.01) AST at 0, 6, and 12 h and tended to have greater ( P = 0.07) AST at 24 h postnatal compared with CON. In calves born to CON dams, serum AST increased ( P ≤ 0.01) from 0 to 24 h, then decreased ( P < 0.001) from 24 to 48 h of age. Serum AST in NR calves increased ( P < 0.001) between 0 and 12 h and then decreased ( P < 0.001) from 24 to 48 h of age. Serum CK was greater ( P ≤ 0.04) at 6, 12, and 24 h in calves born to NR dams. CK in calves born to CON dams increased ( P = 0.02) from 0 to 6 h and tended to decrease ( P = 0.09) from 12 to 24 h postnatal. In calves born to NR dams, CK increased ( P < 0.001) from 0 to 6 h but then decreased ( P ≤ 0.006) from 12 and 48 h of age. Serum GGT was greater ( P = 0.006) at 6 h and tended to be greater ( P ≤ 0.08) at 12 and 48 h of age in calves born to NR dams than CON. In calves born to CON dams, serum GGT increased ( P < 0.001) from 6 to 12 h and then decreased ( P < 0.001) from 24 to 48 h postnatal. In NR calves, serum GGT increased ( P < 0.001) from 0 to 12 h and then decreased ( P < 0.001) from 24 to 48 h of age. There was a tendency for a late gestational nutritional plane × hour interaction ( P = 0.08) for serum sodium and an interaction ( P = 0.03) of late gestational nutritional plane × hour for serum potassium . Calves born to CON dams tended to have greater ( P = 0.09) serum sodium at 0 h and had greater ( P ≤ 0.004) sodium from 6 to 48 h postnatal. Within calves born to CON dams, sodium concentration decreased ( P < 0.001) from 6 to 24 h of age. Serum sodium of NR calves decreased ( P ≤ 0.003) from 0 to 24 h and then increased ( P = 0.01) from 24 to 48 h of age. Serum potassium was greater ( P = 0.03) in calves born to NR dams at 0 h postnatal when compared with CON. Potassium concentrations in calves born to CON dams tended to increase ( P = 0.06) from 0 to 6 h and tended to decrease ( P = 0.08) from 24 to 48 h. There was a decrease ( P = 0.01) in serum potassium from 0 to 6 h of age in calves born to NR dams. There was no effect ( P = 0.21) of the late gestational nutritional plane × hour interaction for serum chloride, but there tended to be an effect of nutritional plane ( P = 0.08), where calves born to CON dams had greater concentrations compared with NR (100.6 ± 0.5 vs. 99.3 ± 0.5). There was no effect ( P ≥ 0.12) of the late gestational nutritional plane × hour interaction or nutritional plane on serum calcium, phosphorus, or magnesium (data not shown). There tended to be an interaction ( P = 0.10) of late gestational nutritional plane × hour for serum anion gap . Calves born to CON dams had greater ( P = 0.02) anion gap at 6 h of age compared with NR. Anion gap decreased ( P ≤ 0.008) from 6 to 24 h in CON calves and decreased ( P < 0.001) from 12 to 24 h in NR calves. There was no effect ( P ≥ 0.46) of the late gestational nutritional plane × hour interaction or nutritional plane on neonatal calf serum bicarbonate concentrations (data not shown). There was no effect ( P ≥ 0.24) of the late gestational nutritional plane × hour interaction or nutritional plane on neonatal calf serum total or direct bilirubin (data not shown).
There was an interaction ( P ≤ 0.04) of late gestational nutritional plane × hour for RBC, hemoglobin, and hematocrit . Calves born to CON dams tended to have greater ( P ≤ 0.09) RBC and hemoglobin at 6 and 12 h postnatal compared with NR. Red blood cells decreased ( P ≤ 0.02) from 0 to 48 h postnatal in CON. In calves born to NR dams, RBC decreased ( P ≤ 0.04) from 0 to 12 h and then from 24 to 48 h of age. Hemoglobin followed a similar pattern where it decreased ( P ≤ 0.007) from 0 to 48 h of age in CON, and it decreased ( P < 0.001) from 0 to 12 h and tended to decrease ( P = 0.08) from 24 to 48 h in NR. Hematocrit concentrations were greater ( P ≤ 0.04) at 6 and 12 h and tended to be greater ( P = 0.10) at 24 h of age in calves born to CON dams compared with NR. In both CON and NR calves, hematocrit decreased ( P ≤ 0.008) from 0 to 48 h postnatal. Neither the late gestational nutritional plane × hour interaction nor nutritional plane affected neonatal calf WBC ( P ≥ 0.29), MCV ( P ≥ 0.12), MCH ( P ≥ 0.25), MCHC ( P ≥ 0.14), or platelet count ( P ≥ 0.70; ).
Although the early neonatal period is the most likely time for pre-weaning beef calf death loss , it is an under-studied period for impacts of poor maternal nutrition. We have previously reported that nutrient restricted heifers partitioned nutrients toward their fetuses rather than their own growth and maintenance of body condition, which conserved fetal growth in the current study . Although calf size at birth was not affected, data reported here suggest that developmental or other differences likely occurred to affect calf vigor and physiology. Overall, calves born to nutrient restricted females in the current study were less vigorous at birth, showed more evidence of birth trauma, and had less blood oxygen carrying capacity. Conversely, they had greater serum Ig and minimally affected energy-related metabolites compared with calves born to control dams. Effects of late gestational nutrient restriction on neonatal calves observed in this study, or the lack thereof, were likely caused by a combination of the following factors: 1) fetal development, 2) calving difficulty and duration, 3) colostrum production, and 4) environment at calving and during the neonatal period. Fetal development differences have been demonstrated repeatedly due to maternal nutrient restriction in ruminants . Neonates are not only products of their fetal environment, but they are also greatly impacted by events of parturition . The percentages of heifers assisted at calving and with fetal malpresentations at birth were not different, but 4 of 13 nutrient restricted dams were assisted at calving and 3 of 13 had fetal malpresentations (vs. 2 of 12 and 0 of 12, respectively, for control; ). The current study was powered to have appropriate n for continuous and categorical data collected, but not for binomial data such as these. Previously, nutrient restriction during pregnancy has reduced pelvic growth in heifers , tended to increase duration of stage 2 parturition , and been observed to cause maternal fatigue or decreased effort during calving . We postulate that all of these factors may have also caused more calving difficulty and trauma to calves born to nutrient restricted dams in the current study, regardless of whether visible dystocia or calving assistance occurred. Neonatal metabolism is also affected by nutrient availability and ambient conditions postnatally. Despite 40% less total colostrum yield in nutrient restricted females in the current study, only total lactose was reduced, while total IgG, IgA, protein, and triglyceride yields were conserved compared with controls . Additionally, fall-calving was used in the current study, and the average date of calving and ambient temperatures at birth were similar, allowing calves from both nutritional planes to generally benefit from thermoneutral conditions. Fall-calving allowed for less stress associated with cold, wet conditions altering the early life environment as we have observed previously in spring-born calves . Vigor at birth Using multiple measures, calves born to nutrient restricted dams were less vigorous in the current study. reported similar findings where calves born to heifers fed a low nutritional plane during late gestation took longer to stand and to suckle. Other studies have reported no difference in calf vigor due to late gestational nutrition of beef cows or heifers , but all experiments except that of and used subjective calf vigor scores classifying calves using terms like “very vigorous” or “weak”. These subjective scores without clear definitions may not have been adequate to capture vigor differences, unlike the behavioral latencies or scores based on behaviors used in the current study. Previous studies have associated reduced neonatal calf vigor with dystocia at birth , which can be caused by negative effects on both neonatal physiology and maternal behavior . Rapid intervention when calving difficulty and fetal malpresentation were observed in the current study may have limited additional fetal stress from prolonged births , as after observing fetal feet, duration of parturition was not affected by nutritional plane and was not correlated with any vigor measure (data not shown). Length of prior stages of parturition is unknown, however. Although calving difficulty likely affected calf vigor in this study, removing calves with assisted births did not change trends in the data, where the time to attempt to stand was 50% longer ( P = 0.16), and time to stand was 60% longer ( P = 0.09), for calves born to nutrient restricted dams. This suggests that other underlying impacts of late gestational nutrient restriction also decreased calf vigor at birth, such as fetal development or calving difficulty that did not result in assistance. reported that calf birth weight followed maternal nutritional plane, which may have contributed to poor vigor in calves born to heifers fed a low plane of nutrition. Although observed no effect of nutrient restriction during pregnancy on neonatal lamb vigor behaviors, they reported that nutrient restriction reduced birth weight, and smaller lambs took longer to display vigor behaviors regardless of maternal nutrition. Because fetal growth was not affected in the current study , the reduced vigor of calves born to nutrient restricted dams is not likely related to calf size but may have been due to fetal development instead. Nutrient restriction during late gestation may have reduced fetal neurodevelopment, leading to impairment of complex movement coordination as previously suggested in . In other species, experimentally-induced intrauterine growth restriction has been reported to result in brain network rearrangement in rabbits , reduced neuron numbers in guinea pigs , and impairment of coordination for neurobehaviors in rats . Beef calves are often born into difficult environments and exposed to perinatal stressors that may limit vigor. Ultimately, neonatal vigor is most important to allow for consumption of colostrum shortly after birth. Although time to suckle was not measured due to pre-suckling blood and colostrum sampling in this study, we have previously reported that times to stand and suckle were positively correlated in beef calves . observed that the calves who did not consume colostrum within 4 h of birth were less likely to have optimal passive transfer of immunity and more likely to experience pre-weaning morbidity. Fall-born calves are faster to stand than those born in the spring , so being born into a colder environment would have likely reduced vigor for all calves and may have exacerbated treatment differences. Overall, vigor differences observed here due to late gestational nutrient restriction may decrease calf health and survival in a production setting. Transfer of passive immunity Calves born to nutrient restricted dams had greater indices of passive transfer (serum IgG, IgA, total protein, globulin, and GGT) in this study. All calves in both nutritional planes had serum IgG concentrations greater than the 8 to 9 mg/mL threshold suggested by and the 16 mg/mL threshold given by . Using a 24 mg/mL IgG threshold , 5 calves born to control dams did not have optimal passive transfer. More calves could be classified as inadequate using serum total protein cutoffs of 5.0 g/dL or 5.6 g/dL , as shown in , although this disparity between classifications using IgG and total protein has been reported previously . No neonatal calf morbidity was observed during the first 48 h of age, and no pre-weaning mortality occurred related to infectious disease. Pre-weaning, 1 calf (control) was administered electrolytes for scours, and 4 calves (2 control and 2 nutrient restricted) were treated with antibiotics for either navel infections or fevers (rectal temperature greater than 40.5 °C). Nutrient restriction during pregnancy has often decreased or not affected calf serum Ig concentrations (reviewed by ), but similar results to the current study have been observed previously in calves born to dams after protein restriction or in poor BCS . In the current study, greater serum IgG and IgA for calves born to nutrient restricted dams were likely due to colostrum from these heifers having 68% greater IgG and 72% greater IgA concentrations, despite having no difference in total IgG or IgA content . More concentrated colostral Ig allowed for these calves to consume more Ig during their initial suckling events, while small intestinal Ig absorption was at its greatest . It has previously been reported that lambs born to nutrient restricted ewes and fed artificial colostrum to BW had greater 24-h serum IgG concentrations, which authors hypothesized was due to improved neonatal Ig absorption through increased transport or delayed small intestinal maturation . Fetal and neonatal small intestinal development is affected by maternal nutrition and intrauterine growth restriction may increase small intestinal absorption of macromolecules in neonates ; thus, small intestinal differences may have also contributed to the current study. Assistance at calving has resulted in lower serum Ig concentrations , but the greater incidence of calving assistance for nutrient restricted dams in this study does not appear to have negatively affected transfer of passive immunity. Colostrum contains high GGT concentrations; therefore, neonatal serum GGT concentrations will sharply increase following colostrum intake . Some studies suggest that serum GGT can be used to indicate successful passive transfer in both lambs and calves , but serum IgG and GGT were not correlated in the current study. Because nutrient restricted dams had reduced colostrum yield, circulating GGT in their calves suggests that they had more concentrated GGT in their colostrum. The lack of IgG and GGT correlation in 48-h serum indicates that a poor relationship of colostral IgG and GGT existed or that 48 h was too late to observe the expected relationship in serum. Thermoregulation and metabolism Following birth, thermogenesis is vital in preventing hypothermia as neonates transition to the extrauterine environment . Calves born to nutrient restricted dams had a lower rectal temperature at 0 h, but greater rectal temperature at 24 h. Using 15°C as the lower critical temperature and 25 °C as the upper critical temperature , 2 control and 3 nutrient restricted calves were born in ambient temperatures below this range, whereas 2 control and 4 nutrient restricted calves were born in ambient temperatures above this range. The similar mean ambient temperatures at birth for calves from both nutritional planes suggest that the 0 h difference was not due to ambient temperature. Rectal temperature at 0 h was negatively correlated with age at measurement ( r = −0.53, P = 0.006). Because 0 h data were obtained after calves successfully stood, nutrient restricted calves were 15 min older (29.4 ± 4.2 vs. 44.4 ± 4.2 min; P = 0.02) than control calves at the 0 h sampling. reported that Holstein calves in relatively thermoneutral environments lost approximately 0.4 °C from 30 to 50 min of age; thus, the rectal temperature difference observed pre-suckling may have been solely due to age in the current study. Although dystocia can decrease neonatal calf rectal temperature , similar results were observed when assisted calvings were removed (data not shown). Previously, observed no difference in calf rectal temperature at birth, but 11.4% less heat production between 5 and 13 h postnatal in calves born to protein restricted dams. reported no differences in metabolic rate at 37°C or brown adipose tissue mass or function due to maternal protein restriction in another study. We have previously shown that fall-born calves have more stable rectal temperatures than those born in the spring ; therefore, thermogenesis and rectal temperatures may have been affected differently by late gestational nutrient restriction in colder conditions requiring greater thermogenesis. Neonatal calf energy metabolism is critical during the transition from parenteral to enteral nutrition , especially given the thermoregulation challenges that often exist . Calves are born with limited glycogen that they mobilize quickly after birth, while also increasing gluconeogenesis to make up for the glucose deficit caused by limited lactose intake from colostrum . During this time, fat mobilization also increases to provide energy , and an increase in circulating NEFA is observable by 6 h of age in calves born into both cold and warm environments , which provides an additional non-glucose substrate for brown adipose tissue . Digestion and absorption of colostral lipids increase circulating triglycerides by 6 h of age , but can be variable, likely based on fat concentration of colostrum . Finally, amino acids can also be deaminated to provide carbon skeletons for gluconeogenesis , resulting in elevated circulating urea N. Taken together, the circulating glucose, NEFA, triglycerides, urea N, and insulin of calves in the current study do not show marked contrasts in energy metabolism due to late gestational nutrient restriction. Given that fetal growth was not affected and ambient conditions were not harsh, this lack of differences in energy metabolism likely demonstrates calves from both late gestational nutritional planes had adequate available energy sources in their early life environments. In this study, colostrum triglyceride concentration and yield were not affected by nutrient restriction, and although free glucose was greater in colostrum from control dams, the concentration and total yield of free glucose were too low to provide a major glucose source . Furthermore, protein concentration was decreased by nutrient restriction , but it is unknown how much of this was due to casein differences rather than Ig. Total lactose available in colostrum was greater for control calves, but the concentration was not ; thus, lactose consumption was not affected unless meal size differed between nutritional planes. Colostrum yield was driven by total lactose ; thus, lower lactose production by nutrient restricted dams decreased total colostrum available for calves during a meal. Voluntary colostrum consumption was not measured in this study, but as much as 2.4 kg of colostrum consumption would have been possible on average (up to 8% of BW within first 9 to 12 h of life; ). Only an average of 1.1 kg and 0.7 kg would have been available for control and nutrient restricted calves, respectively, assuming the single rear quarter sampled accounted for 31.2% of total colostrum . Milk yields were not measured in the current study until day 21 of lactation, but we hypothesize that available transition milk was also less for calves born to nutrient restricted dams. Therefore, calves born to nutrient restricted dams may have been more limited in their total nutrients available during the neonatal period due to decreased colostrum and transition milk yields. Colostrum urea N was less for nutrient restricted dams, expressed both as concentration and total content . Given that less urea N was consumed by their neonatal calves without ruminal microbes to utilize it, we hypothesize that similar serum urea N (and an extended initial increase in serum urea N) among treatments indicates more amino acids were being deaminated by calves born to nutrient restricted females. These deaminated amino acids may have provided enough energy substrates for calves from nutrient restricted dams to maintain their glucose status without elevated NEFA. Insulin concentrations in calves born to first-parity dams were greater compared with the current study, but they showed a lack of change over the first 72 h of life similar to the current study . Cold stress of these calves was minimal due to ambient temperature, bedding provided, and shelter from precipitation and wind. Thus, calves in this study likely were able to maintain thermogenesis and glucose status through different metabolic approaches even with differences in colostrum intake. If these calves had experienced cold stress, we hypothesize that calves born to nutrient restricted dams would have had decreased circulating glucose and increased NEFA, as observed in calves born to primiparous dams by , given they had less colostrum available. Previous studies reported that maternal nutrient restriction in beef and dairy × beef had no effect on fetal plasma glucose concentrations or fetal liver gluconeogenic activity, which suggests glucose supply to meet fetal demands is prioritized over maternal demands . Results from the current study are in agreement with the aforementioned studies and support this notion of prioritization of fetal demands as there was no effect of maternal treatment on neonatal energy metabolites. Indicators of stress and trauma Changes in calf blood chemistry over time presented here are similar to our previous observations for healthy, fall-born neonatal calves . Few blood chemistry or hematology measures were divergent at 0 h; thus, the difference in age at the pre-suckling sampling caused by vigor differences may not have influenced data interpretation. Neonatal calf AST, CK, and creatinine were greater for calves born to nutrient restricted dams within the first 24 h of life, and have been associated with trauma (tissue damage, edema, injury) of difficult births. Both AST and CK are important enzymes for amino acid and energy metabolism, respectively, that are commonly found in liver and muscle cells; however, when muscle damage or inflammation occurs these enzymes are released into circulation causing increased concentrations . observed greater AST and CK concentrations in calves whose births were considered a difficult assist compared with unassisted or easy assists. These authors also reported that calves with poor vitality indicators such as pale mucous membranes, incomplete tongue withdrawal, and weak suckling reflex had greater AST and CK concentrations . Therefore, these observations are in good agreement with the reduced neonatal vigor that was displayed by calves born to nutrient restricted dams. Serum creatinine may indicate greater trauma during calving , also suggesting more difficult parturition following nutrient restriction. Creatinine concentrations can also be influenced by muscle mass , but as there were no differences in calf size or shape measures at birth , this is unlikely the cause of the difference observed in the current study. Although CK differences lessened when calves with assisted births were removed from the dataset, AST and creatinine differences persisted (data not shown). Circulating cortisol was not affected by nutrient restriction, but this may be due to the sampling times used in the current study, as we have previously suggested . We have reported similar observations where calves born to first-parity dams had greater AST, CK, and creatinine despite similar cortisol compared with calves born to multiparous females, even though there was limited incidence of calving assistance . Moreover, spring-born neonatal calves had greater AST and CK than fall-born calves, following vigor data, in another study from our lab in which calf data were removed if calving assistance was necessary . Overall, calves born to nutrient restricted dams in the current study appear to have experienced more stressful calving and early life experiences, regardless of visible dystocia. Anion gap is typically used to evaluate metabolic acidosis , which can depress vigor and muscle coordination, leading to increased calf mortality . The greater anion gap at 6 h in calves born to control heifers is likely not due to metabolic acidosis, given they were more vigorous, but may have been caused by elevated sodium concentrations at that sample time point. Sodium concentrations are greatest in colostrum and decrease with production of milk ; thus, sodium intake of calves born to nutrient restricted dams may have been reduced by colostrum yield differences. Concentrations of chloride often follow the pattern of sodium, because renal reabsorption of both occurs simultaneously . Concentrations of albumin are reflective of hydration status , so serum albumin concentrations suggest that hydration status of calves did not differ. Overall, electrolyte status was not greatly affected by maternal nutrient restriction in the current study. Red blood cells, hemoglobin, and hematocrit were not affected at birth but were less in calves born to nutrient restricted dams from 6 to 24 h. The general decrease in RBC measures observed after birth has been observed previously , and it has been suggested that delayed responses to hemorrhage during birth may be the cause . Calves born with assistance due to dystocia had less RBC or hematocrit within the first day of life in other studies . Hematocrit has also generally been lower on the first day of life for calves with meconium staining, pale mucous membranes, incomplete tongue withdrawal, and weak suckling reflex . This suggests that the lower blood oxygen-carrying capacity in nutrient restricted calves was a result of more trauma experienced during birth. These trends became less pronounced when calves with assisted births were removed, but hematocrit results persisted (data not shown). also observed that feed restriction of pregnant dairy females resulted in neonatal calves with less hematocrit and hemoglobin without altered birth weight; thus, our observations may not have been influenced only by calving assistance or dystocia. Elevated RBC, hemoglobin, and hematocrit can be caused by a decrease in plasma volume or following a splenic contraction . An accompanying increase in total protein or albumin was not observed concurrently in calves born to control dams, which excludes dehydration or a decrease in plasma volume as the cause of greater RBC in these calves. Bilirubin was not affected by late gestational nutritional plane, also suggesting that differences in RBC breakdown did not cause these observations. Limited neonatal beef calf hematology data from 0 to 48 h of age are available for comparison, so our data may provide a useful resource for values in very young calves. Hematology changes over time were less pronounced than for most blood chemistry values and had similar patterns to our previous observations in neonatal foals . All RBC, hemoglobin, and MCH values from the current study were within the 90% confidence intervals of the reference intervals proposed by using calves from 1 to 9 d old. Hematocrit (3 above), WBC (1 below, 2 above), and platelet (3 below) values were predominantly within these reference intervals , with all outside values from 0, 6, or 12 h sampling. Neonatal calf MCV (20% above) and MCHC (23% below) were less likely to be within these reference intervals .
Using multiple measures, calves born to nutrient restricted dams were less vigorous in the current study. reported similar findings where calves born to heifers fed a low nutritional plane during late gestation took longer to stand and to suckle. Other studies have reported no difference in calf vigor due to late gestational nutrition of beef cows or heifers , but all experiments except that of and used subjective calf vigor scores classifying calves using terms like “very vigorous” or “weak”. These subjective scores without clear definitions may not have been adequate to capture vigor differences, unlike the behavioral latencies or scores based on behaviors used in the current study. Previous studies have associated reduced neonatal calf vigor with dystocia at birth , which can be caused by negative effects on both neonatal physiology and maternal behavior . Rapid intervention when calving difficulty and fetal malpresentation were observed in the current study may have limited additional fetal stress from prolonged births , as after observing fetal feet, duration of parturition was not affected by nutritional plane and was not correlated with any vigor measure (data not shown). Length of prior stages of parturition is unknown, however. Although calving difficulty likely affected calf vigor in this study, removing calves with assisted births did not change trends in the data, where the time to attempt to stand was 50% longer ( P = 0.16), and time to stand was 60% longer ( P = 0.09), for calves born to nutrient restricted dams. This suggests that other underlying impacts of late gestational nutrient restriction also decreased calf vigor at birth, such as fetal development or calving difficulty that did not result in assistance. reported that calf birth weight followed maternal nutritional plane, which may have contributed to poor vigor in calves born to heifers fed a low plane of nutrition. Although observed no effect of nutrient restriction during pregnancy on neonatal lamb vigor behaviors, they reported that nutrient restriction reduced birth weight, and smaller lambs took longer to display vigor behaviors regardless of maternal nutrition. Because fetal growth was not affected in the current study , the reduced vigor of calves born to nutrient restricted dams is not likely related to calf size but may have been due to fetal development instead. Nutrient restriction during late gestation may have reduced fetal neurodevelopment, leading to impairment of complex movement coordination as previously suggested in . In other species, experimentally-induced intrauterine growth restriction has been reported to result in brain network rearrangement in rabbits , reduced neuron numbers in guinea pigs , and impairment of coordination for neurobehaviors in rats . Beef calves are often born into difficult environments and exposed to perinatal stressors that may limit vigor. Ultimately, neonatal vigor is most important to allow for consumption of colostrum shortly after birth. Although time to suckle was not measured due to pre-suckling blood and colostrum sampling in this study, we have previously reported that times to stand and suckle were positively correlated in beef calves . observed that the calves who did not consume colostrum within 4 h of birth were less likely to have optimal passive transfer of immunity and more likely to experience pre-weaning morbidity. Fall-born calves are faster to stand than those born in the spring , so being born into a colder environment would have likely reduced vigor for all calves and may have exacerbated treatment differences. Overall, vigor differences observed here due to late gestational nutrient restriction may decrease calf health and survival in a production setting.
Calves born to nutrient restricted dams had greater indices of passive transfer (serum IgG, IgA, total protein, globulin, and GGT) in this study. All calves in both nutritional planes had serum IgG concentrations greater than the 8 to 9 mg/mL threshold suggested by and the 16 mg/mL threshold given by . Using a 24 mg/mL IgG threshold , 5 calves born to control dams did not have optimal passive transfer. More calves could be classified as inadequate using serum total protein cutoffs of 5.0 g/dL or 5.6 g/dL , as shown in , although this disparity between classifications using IgG and total protein has been reported previously . No neonatal calf morbidity was observed during the first 48 h of age, and no pre-weaning mortality occurred related to infectious disease. Pre-weaning, 1 calf (control) was administered electrolytes for scours, and 4 calves (2 control and 2 nutrient restricted) were treated with antibiotics for either navel infections or fevers (rectal temperature greater than 40.5 °C). Nutrient restriction during pregnancy has often decreased or not affected calf serum Ig concentrations (reviewed by ), but similar results to the current study have been observed previously in calves born to dams after protein restriction or in poor BCS . In the current study, greater serum IgG and IgA for calves born to nutrient restricted dams were likely due to colostrum from these heifers having 68% greater IgG and 72% greater IgA concentrations, despite having no difference in total IgG or IgA content . More concentrated colostral Ig allowed for these calves to consume more Ig during their initial suckling events, while small intestinal Ig absorption was at its greatest . It has previously been reported that lambs born to nutrient restricted ewes and fed artificial colostrum to BW had greater 24-h serum IgG concentrations, which authors hypothesized was due to improved neonatal Ig absorption through increased transport or delayed small intestinal maturation . Fetal and neonatal small intestinal development is affected by maternal nutrition and intrauterine growth restriction may increase small intestinal absorption of macromolecules in neonates ; thus, small intestinal differences may have also contributed to the current study. Assistance at calving has resulted in lower serum Ig concentrations , but the greater incidence of calving assistance for nutrient restricted dams in this study does not appear to have negatively affected transfer of passive immunity. Colostrum contains high GGT concentrations; therefore, neonatal serum GGT concentrations will sharply increase following colostrum intake . Some studies suggest that serum GGT can be used to indicate successful passive transfer in both lambs and calves , but serum IgG and GGT were not correlated in the current study. Because nutrient restricted dams had reduced colostrum yield, circulating GGT in their calves suggests that they had more concentrated GGT in their colostrum. The lack of IgG and GGT correlation in 48-h serum indicates that a poor relationship of colostral IgG and GGT existed or that 48 h was too late to observe the expected relationship in serum.
Following birth, thermogenesis is vital in preventing hypothermia as neonates transition to the extrauterine environment . Calves born to nutrient restricted dams had a lower rectal temperature at 0 h, but greater rectal temperature at 24 h. Using 15°C as the lower critical temperature and 25 °C as the upper critical temperature , 2 control and 3 nutrient restricted calves were born in ambient temperatures below this range, whereas 2 control and 4 nutrient restricted calves were born in ambient temperatures above this range. The similar mean ambient temperatures at birth for calves from both nutritional planes suggest that the 0 h difference was not due to ambient temperature. Rectal temperature at 0 h was negatively correlated with age at measurement ( r = −0.53, P = 0.006). Because 0 h data were obtained after calves successfully stood, nutrient restricted calves were 15 min older (29.4 ± 4.2 vs. 44.4 ± 4.2 min; P = 0.02) than control calves at the 0 h sampling. reported that Holstein calves in relatively thermoneutral environments lost approximately 0.4 °C from 30 to 50 min of age; thus, the rectal temperature difference observed pre-suckling may have been solely due to age in the current study. Although dystocia can decrease neonatal calf rectal temperature , similar results were observed when assisted calvings were removed (data not shown). Previously, observed no difference in calf rectal temperature at birth, but 11.4% less heat production between 5 and 13 h postnatal in calves born to protein restricted dams. reported no differences in metabolic rate at 37°C or brown adipose tissue mass or function due to maternal protein restriction in another study. We have previously shown that fall-born calves have more stable rectal temperatures than those born in the spring ; therefore, thermogenesis and rectal temperatures may have been affected differently by late gestational nutrient restriction in colder conditions requiring greater thermogenesis. Neonatal calf energy metabolism is critical during the transition from parenteral to enteral nutrition , especially given the thermoregulation challenges that often exist . Calves are born with limited glycogen that they mobilize quickly after birth, while also increasing gluconeogenesis to make up for the glucose deficit caused by limited lactose intake from colostrum . During this time, fat mobilization also increases to provide energy , and an increase in circulating NEFA is observable by 6 h of age in calves born into both cold and warm environments , which provides an additional non-glucose substrate for brown adipose tissue . Digestion and absorption of colostral lipids increase circulating triglycerides by 6 h of age , but can be variable, likely based on fat concentration of colostrum . Finally, amino acids can also be deaminated to provide carbon skeletons for gluconeogenesis , resulting in elevated circulating urea N. Taken together, the circulating glucose, NEFA, triglycerides, urea N, and insulin of calves in the current study do not show marked contrasts in energy metabolism due to late gestational nutrient restriction. Given that fetal growth was not affected and ambient conditions were not harsh, this lack of differences in energy metabolism likely demonstrates calves from both late gestational nutritional planes had adequate available energy sources in their early life environments. In this study, colostrum triglyceride concentration and yield were not affected by nutrient restriction, and although free glucose was greater in colostrum from control dams, the concentration and total yield of free glucose were too low to provide a major glucose source . Furthermore, protein concentration was decreased by nutrient restriction , but it is unknown how much of this was due to casein differences rather than Ig. Total lactose available in colostrum was greater for control calves, but the concentration was not ; thus, lactose consumption was not affected unless meal size differed between nutritional planes. Colostrum yield was driven by total lactose ; thus, lower lactose production by nutrient restricted dams decreased total colostrum available for calves during a meal. Voluntary colostrum consumption was not measured in this study, but as much as 2.4 kg of colostrum consumption would have been possible on average (up to 8% of BW within first 9 to 12 h of life; ). Only an average of 1.1 kg and 0.7 kg would have been available for control and nutrient restricted calves, respectively, assuming the single rear quarter sampled accounted for 31.2% of total colostrum . Milk yields were not measured in the current study until day 21 of lactation, but we hypothesize that available transition milk was also less for calves born to nutrient restricted dams. Therefore, calves born to nutrient restricted dams may have been more limited in their total nutrients available during the neonatal period due to decreased colostrum and transition milk yields. Colostrum urea N was less for nutrient restricted dams, expressed both as concentration and total content . Given that less urea N was consumed by their neonatal calves without ruminal microbes to utilize it, we hypothesize that similar serum urea N (and an extended initial increase in serum urea N) among treatments indicates more amino acids were being deaminated by calves born to nutrient restricted females. These deaminated amino acids may have provided enough energy substrates for calves from nutrient restricted dams to maintain their glucose status without elevated NEFA. Insulin concentrations in calves born to first-parity dams were greater compared with the current study, but they showed a lack of change over the first 72 h of life similar to the current study . Cold stress of these calves was minimal due to ambient temperature, bedding provided, and shelter from precipitation and wind. Thus, calves in this study likely were able to maintain thermogenesis and glucose status through different metabolic approaches even with differences in colostrum intake. If these calves had experienced cold stress, we hypothesize that calves born to nutrient restricted dams would have had decreased circulating glucose and increased NEFA, as observed in calves born to primiparous dams by , given they had less colostrum available. Previous studies reported that maternal nutrient restriction in beef and dairy × beef had no effect on fetal plasma glucose concentrations or fetal liver gluconeogenic activity, which suggests glucose supply to meet fetal demands is prioritized over maternal demands . Results from the current study are in agreement with the aforementioned studies and support this notion of prioritization of fetal demands as there was no effect of maternal treatment on neonatal energy metabolites.
Changes in calf blood chemistry over time presented here are similar to our previous observations for healthy, fall-born neonatal calves . Few blood chemistry or hematology measures were divergent at 0 h; thus, the difference in age at the pre-suckling sampling caused by vigor differences may not have influenced data interpretation. Neonatal calf AST, CK, and creatinine were greater for calves born to nutrient restricted dams within the first 24 h of life, and have been associated with trauma (tissue damage, edema, injury) of difficult births. Both AST and CK are important enzymes for amino acid and energy metabolism, respectively, that are commonly found in liver and muscle cells; however, when muscle damage or inflammation occurs these enzymes are released into circulation causing increased concentrations . observed greater AST and CK concentrations in calves whose births were considered a difficult assist compared with unassisted or easy assists. These authors also reported that calves with poor vitality indicators such as pale mucous membranes, incomplete tongue withdrawal, and weak suckling reflex had greater AST and CK concentrations . Therefore, these observations are in good agreement with the reduced neonatal vigor that was displayed by calves born to nutrient restricted dams. Serum creatinine may indicate greater trauma during calving , also suggesting more difficult parturition following nutrient restriction. Creatinine concentrations can also be influenced by muscle mass , but as there were no differences in calf size or shape measures at birth , this is unlikely the cause of the difference observed in the current study. Although CK differences lessened when calves with assisted births were removed from the dataset, AST and creatinine differences persisted (data not shown). Circulating cortisol was not affected by nutrient restriction, but this may be due to the sampling times used in the current study, as we have previously suggested . We have reported similar observations where calves born to first-parity dams had greater AST, CK, and creatinine despite similar cortisol compared with calves born to multiparous females, even though there was limited incidence of calving assistance . Moreover, spring-born neonatal calves had greater AST and CK than fall-born calves, following vigor data, in another study from our lab in which calf data were removed if calving assistance was necessary . Overall, calves born to nutrient restricted dams in the current study appear to have experienced more stressful calving and early life experiences, regardless of visible dystocia. Anion gap is typically used to evaluate metabolic acidosis , which can depress vigor and muscle coordination, leading to increased calf mortality . The greater anion gap at 6 h in calves born to control heifers is likely not due to metabolic acidosis, given they were more vigorous, but may have been caused by elevated sodium concentrations at that sample time point. Sodium concentrations are greatest in colostrum and decrease with production of milk ; thus, sodium intake of calves born to nutrient restricted dams may have been reduced by colostrum yield differences. Concentrations of chloride often follow the pattern of sodium, because renal reabsorption of both occurs simultaneously . Concentrations of albumin are reflective of hydration status , so serum albumin concentrations suggest that hydration status of calves did not differ. Overall, electrolyte status was not greatly affected by maternal nutrient restriction in the current study. Red blood cells, hemoglobin, and hematocrit were not affected at birth but were less in calves born to nutrient restricted dams from 6 to 24 h. The general decrease in RBC measures observed after birth has been observed previously , and it has been suggested that delayed responses to hemorrhage during birth may be the cause . Calves born with assistance due to dystocia had less RBC or hematocrit within the first day of life in other studies . Hematocrit has also generally been lower on the first day of life for calves with meconium staining, pale mucous membranes, incomplete tongue withdrawal, and weak suckling reflex . This suggests that the lower blood oxygen-carrying capacity in nutrient restricted calves was a result of more trauma experienced during birth. These trends became less pronounced when calves with assisted births were removed, but hematocrit results persisted (data not shown). also observed that feed restriction of pregnant dairy females resulted in neonatal calves with less hematocrit and hemoglobin without altered birth weight; thus, our observations may not have been influenced only by calving assistance or dystocia. Elevated RBC, hemoglobin, and hematocrit can be caused by a decrease in plasma volume or following a splenic contraction . An accompanying increase in total protein or albumin was not observed concurrently in calves born to control dams, which excludes dehydration or a decrease in plasma volume as the cause of greater RBC in these calves. Bilirubin was not affected by late gestational nutritional plane, also suggesting that differences in RBC breakdown did not cause these observations. Limited neonatal beef calf hematology data from 0 to 48 h of age are available for comparison, so our data may provide a useful resource for values in very young calves. Hematology changes over time were less pronounced than for most blood chemistry values and had similar patterns to our previous observations in neonatal foals . All RBC, hemoglobin, and MCH values from the current study were within the 90% confidence intervals of the reference intervals proposed by using calves from 1 to 9 d old. Hematocrit (3 above), WBC (1 below, 2 above), and platelet (3 below) values were predominantly within these reference intervals , with all outside values from 0, 6, or 12 h sampling. Neonatal calf MCV (20% above) and MCHC (23% below) were less likely to be within these reference intervals .
Although no differences were observed in fetal growth in the current study, calves born to nutrient restricted dams were less vigorous after birth and showed more signs of trauma at calving. Despite this, calves born to nutrient restricted dams had successful transfer of passive immunity and minimal differences in energy metabolism. The relatively thermoneutral environment and intensive management in the current study likely decreased overall calf metabolic stress by minimizing the need for thermogenesis; therefore, the effects of nutrient restriction may result in poor pre-weaning health and survival outcomes in many cow-calf production scenarios, especially those involving cold stress. Data from this study suggest that impaired transfer of passive immunity is not the major threat to calves born to nutrient restricted females. Overall, the current study (including previously-published and forthcoming data) allows for integration of many maternal and offspring effects caused by late gestational nutrient restriction, expanding our knowledge of developmental programming in beef cattle and giving greater insight into effects on the neonatal calf. Better understanding of the ramifications of poor maternal nutrition on neonatal beef calves can improve pre-weaning survival and productivity.
skad342_suppl_Supplementary_Figures_1-4 Click here for additional data file.
|
Posterior-only debridement, bone fusion, single-segment versus short-segment instrumentation for mono-segmental lumbar or lumbosacral pyogenic vertebral osteomyelitis: minimum five year follow-up outcomes | ea54c9d5-3d05-4430-8c84-13a181583dfd | 9373455 | Debridement[mh] | Pyogenic vertebral osteomyelitis (PVO) comprises less than 4% of all bone infections , but can be associated with devastating morbidity and mortality, resulting in significant pain, deformity and neurological deficit . The diagnosis and management of PVO remains difficult, not least because of the increasing elderly and immuno-compromised population worldwide. The majority of patients with PVO can be treated nonsurgically with appropriate antibiotics and external immobilization, particularly if the diagnosis is made early and the causative organism can be identified from a closed needle biopsy . As such, indications for surgery have been limited to the unsuccessful conservative therapy, neurologic impairment, epidural abscess formation, intractable pain, or vertebral destruction leading to early or late spinal instability or segmental kyphosis . However, the current surgical treatment of PVO is still controversial, especially the choice of fixed segment length and selection of surgical approach. Although anterior approach have been conventionally preferred, complicated anatomic layers, segmental vessels, occasionally major vessels, and nerves should not be overlooked . Furthermore, some patients often combined with cardiovascular and respiratory disease and single lung ventilation via anterior approach could result in more complications . The combined posterior and anterior procedures lead to increased operating time, prolonged anesthesia, greater blood loss and increased mortality, morbidity and complications. Posterior instrumentation has become popular as a technique for correction of kyphotic deformity and stabilization of the unstable spine in the past decade. In addition, we have applied posterior-only surgery in the treatment for spinal tuberculosis, which have got satisfactory clinical effects [ – ]. In view of this, posterior-only surgery was used in the treatment of mono-segmental lumbar or lumbosacral PVO. There are no comparative studies on mono-segmental lumbar or lumbosacral PVO treated by posterior-only debridement, fusion, and single-segment vs. short-segment fixation. Therefore, the aim of our clinical study was to compare the clinical and radiological outcomes of posterior single-segment and short-segment fixation combined with posterior-only debridement and bone grafting fusion in treating mono-segmental lumbar or lumbosacral PVO.
This study was a retrospective case series (level 4 evidence) and was approved by the ethics board committee of our hospital. Charts of all patients with mono-segmental lumbar or lumbosacral PVO treated from April 2012 to January 2016 by posterior-only debridement, interbody bone graft using titanium mesh cage, posterior instrumentation and fusion were retrospectively reviewed. Patients whose major lesion involves mono-segmental with any of the following conditions were selected: (1) failure of conservative treatment, (2) intractable back pain, (3) vertebral destruction causing spinal instability, (4) neurological deterioration, (5) obvious abscess formation. Exclusion criteria were the following: (1) multi-segmental involvement with severe destruction of vertebral bodies, (2) multi-level large paraspinal abscesses or gravitation abscess, (3) previous lumbar or lumbosacral surgery. According to the length of fixation, the cases were divided into single-segment fixation group (group A, where the fixed/fused range was limited to only one damaged motion segment) and short-segment fixation group (group B, where the fixed/fused range included both the one damaged segment and one normal motion segment located above and below the damaged motion segment, respectively). In group A, there were 31 cases (19 males and 12 females) with a mean age 44.4 years (range 32 to 57 years). In group B, there were 36 cases (22 males and 14 females) with a mean age of 44.8 years (range 35 to 61 years). The main clinical symptoms of patients in two groups included fever, low back pain, lower extremity radiation pain, numbness, weakness, anorexia and weight loss. Hematological inflammatory indices included erythrocyte sedimentation rate (ESR), C-reactive protein (CRP), and total leukocyte count, which were documented preoperatively, postoperatively and at follow-up. Pain severity was assessed by visual analog scale (VAS). The American Spinal Injury Association (ASIA) was used to evaluate neurological function. All patients had imaging evaluations included radiographs, three dimensional computed tomography (CT), and magnetic resonance imaging (MRI) that were taken preoperatively to determine the affected level and detect vertebral body collapse, spinal instability, bone destruction, epidural and paravertebral abscess formation, and narrowing of the intervertebral space (Figs. a–f, a–c). The lumbar lordosis angle is the sagittal Cobb angle measured between the superior end plate of L1 and the superior end plate of S1 on the radiographs. The diagnosis was based upon clinical presentation, imaging findings and laboratory examination, and subsequently confirmed by histopathological analysis of specimens obtained during surgery. Pre-operative management Percutaneous biopsy of the affected vertebral bodies was performed on all patients posteriorly through a transpedicular approach under CT monitoring. All patients were in strict bed rest and were prescribed with continuous intravenous antibiotic therapy for at least 4 weeks before surgery except ten, of whom experienced neurological deterioration after 2–3 weeks of continuous intravenous antibiotic therapy. Surgical procedure All surgical procedures were performed by the same group of surgeons. All patients were operated under general anesthesia in prone position. In group A, through posterior midline approach, the lamina, facet joints, transverse processes were exposed. Exposing the vertebral laminae of involved segments, posterior pedicle screws were installed. Single segment fixation was performed. Transpedicular screws were only placed in the affected vertebrae. During transpedicular screw insertion, the site should be situated at a distance from the foci, but close to the end-plate to avoid exposing the screw after debridement. A temporary rod on the mild side of the focus was stabilized to avoid spinal cord injury induced by instability of the spine during decompression and focal debridement. After removing spinous process of the affected vertebrae, unilateral partial laminectomy or hemilaminectomy at the more severe lesion or more abscess side of the affected vertebrae was performed before debridement of the affected intervertebral discs and vertebrae. If necessary, unilateral partial or total facetectomy was also performed. Then, corpectomy and discectomy were performed, and abscess was evacuated. To achieve adequate debridement, compression wash and negative pressure suction were alternatively performed by inserting a urethral catheter into the abscess cavity. Spinal cord monitoring was also used, including motor-evoked and sensitive-evoked. One or two titanium mesh cages which were filled with allograft bone and autogenous bone, coming from healthy lamina, spinous process were shaped according to the shape and length of bone graft bed (Fig. g). Posterior interbody graft was applied posterolaterally. Interbody compression was performed after placement of interbody titanium mesh cages. Then, autogenous bone or an allograft of the proper size was selected for posterior fusion on the segments that underwent decompression and focal debridement. Tissue specimens obtained from the diseased spinal tissue at operation were sent for microbiological culture and histopathological analysis in all cases. In group B, short segment fixation was performed, whose fixed range included both the one damaged segment and one normal motion segment located above and below the damaged motion segment, respectively. Other procedures were the same with group A. Post-operative management Continuous flushing intervertebral space with 480,000 units of gentamicin and 3000 ml 0.9% sodium chloride solution was performed for 24 h after surgery in cases, which obvious purulent exudate was found intraoperatively. The drain was usually removed when irrigation of drainage fluid was negative for three times after surgery and white blood cell, ESR, CRP decreased obviousely earlier than before, which often lasted for 2–3 weeks. All patients were initially treated with intravenous broad-spectrum antibiotic, and then the antibiotic therapy was adjusted whenever organisms grown in cultures were identified and the sensitivities to antibiotics of these organisms were obtained. Intravenous antibiotic medications were continued for at least 4 weeks (range 4–6 weeks) after surgery or until all laboratory parameters in terms of white blood cell counts, ESR and CRP returned to normal limits. Patients were allowed to ambulate after operation for 14 to 20 days. The postoperative external support was needed for 3–6 months. Follow-up examination was performed during the first year at 4 weeks, 3, 6, 9 months and 1 year. Subsequent follow-ups were at yearly intervals. At each follow-up survey, they were assessed clinically for neurological function and pain and radiologically for spinal alignment and fusion progress. Plain radiographs were obtained at each follow-up measuring time to solid bony fusion (Figs. h, d, e). Successful fusion were defined as absence of local pain and tenderness over the site of fusion, no abnormal motion, no correction loss and hardware failure, presence of trabecular bone bridging between the grafts and the vertebrae, and no lucencies at the bone-cage interfaces. In inconclusive cases, bony fusion was assessed by three dimensional CT scan (Figs. i, j, f, g). Monitoring of laboratory parameters (white blood cell counts, ESR, CRP, hepatic function and renal function) was made regularly until these parameters returned to normal values. Complications were also recorded. Statistical analyses SPSS24.0 statistical software was used for analysis.The clinical information between the two groups were compared using Student’ s t test and Wilcoxon signed-rank test. A rank sum test was used to analyze any discrepancy in normal data distributions. A P -value < 0.05 was considered statistically significant.
Percutaneous biopsy of the affected vertebral bodies was performed on all patients posteriorly through a transpedicular approach under CT monitoring. All patients were in strict bed rest and were prescribed with continuous intravenous antibiotic therapy for at least 4 weeks before surgery except ten, of whom experienced neurological deterioration after 2–3 weeks of continuous intravenous antibiotic therapy.
All surgical procedures were performed by the same group of surgeons. All patients were operated under general anesthesia in prone position. In group A, through posterior midline approach, the lamina, facet joints, transverse processes were exposed. Exposing the vertebral laminae of involved segments, posterior pedicle screws were installed. Single segment fixation was performed. Transpedicular screws were only placed in the affected vertebrae. During transpedicular screw insertion, the site should be situated at a distance from the foci, but close to the end-plate to avoid exposing the screw after debridement. A temporary rod on the mild side of the focus was stabilized to avoid spinal cord injury induced by instability of the spine during decompression and focal debridement. After removing spinous process of the affected vertebrae, unilateral partial laminectomy or hemilaminectomy at the more severe lesion or more abscess side of the affected vertebrae was performed before debridement of the affected intervertebral discs and vertebrae. If necessary, unilateral partial or total facetectomy was also performed. Then, corpectomy and discectomy were performed, and abscess was evacuated. To achieve adequate debridement, compression wash and negative pressure suction were alternatively performed by inserting a urethral catheter into the abscess cavity. Spinal cord monitoring was also used, including motor-evoked and sensitive-evoked. One or two titanium mesh cages which were filled with allograft bone and autogenous bone, coming from healthy lamina, spinous process were shaped according to the shape and length of bone graft bed (Fig. g). Posterior interbody graft was applied posterolaterally. Interbody compression was performed after placement of interbody titanium mesh cages. Then, autogenous bone or an allograft of the proper size was selected for posterior fusion on the segments that underwent decompression and focal debridement. Tissue specimens obtained from the diseased spinal tissue at operation were sent for microbiological culture and histopathological analysis in all cases. In group B, short segment fixation was performed, whose fixed range included both the one damaged segment and one normal motion segment located above and below the damaged motion segment, respectively. Other procedures were the same with group A.
Continuous flushing intervertebral space with 480,000 units of gentamicin and 3000 ml 0.9% sodium chloride solution was performed for 24 h after surgery in cases, which obvious purulent exudate was found intraoperatively. The drain was usually removed when irrigation of drainage fluid was negative for three times after surgery and white blood cell, ESR, CRP decreased obviousely earlier than before, which often lasted for 2–3 weeks. All patients were initially treated with intravenous broad-spectrum antibiotic, and then the antibiotic therapy was adjusted whenever organisms grown in cultures were identified and the sensitivities to antibiotics of these organisms were obtained. Intravenous antibiotic medications were continued for at least 4 weeks (range 4–6 weeks) after surgery or until all laboratory parameters in terms of white blood cell counts, ESR and CRP returned to normal limits. Patients were allowed to ambulate after operation for 14 to 20 days. The postoperative external support was needed for 3–6 months. Follow-up examination was performed during the first year at 4 weeks, 3, 6, 9 months and 1 year. Subsequent follow-ups were at yearly intervals. At each follow-up survey, they were assessed clinically for neurological function and pain and radiologically for spinal alignment and fusion progress. Plain radiographs were obtained at each follow-up measuring time to solid bony fusion (Figs. h, d, e). Successful fusion were defined as absence of local pain and tenderness over the site of fusion, no abnormal motion, no correction loss and hardware failure, presence of trabecular bone bridging between the grafts and the vertebrae, and no lucencies at the bone-cage interfaces. In inconclusive cases, bony fusion was assessed by three dimensional CT scan (Figs. i, j, f, g). Monitoring of laboratory parameters (white blood cell counts, ESR, CRP, hepatic function and renal function) was made regularly until these parameters returned to normal values. Complications were also recorded.
SPSS24.0 statistical software was used for analysis.The clinical information between the two groups were compared using Student’ s t test and Wilcoxon signed-rank test. A rank sum test was used to analyze any discrepancy in normal data distributions. A P -value < 0.05 was considered statistically significant.
All 67 patients were completely cured during the follow-up. In group A, involved levels were observed at 5 cases in L1–2, 6 cases in L2–3, 6 cases in L3–4, 7 cases in L4–5, and 7 cases in L5–S1. In group B, involved levels were observed at 5 cases in L1–2, 6 cases in L2–3, 7 cases in L3–4, 10 cases in L4–5, and 8 cases in L5–S1. Only 25 patients could be presumed for the source of the spinal infection despite careful diagnostic examination: 15 patients had undergone invasive procedures such as lumbar punctures and epidural injections before they experienced infection, of whom 9 patients were in group A, and 6 patients in group B. 4 patients had urinary tract infections, of whom 2 patients were in group A, and 2 patients in group B. In goup A, 2 patients suffered from bacteremia at another hospital, and a blood culture obtained postoperatively at that hospital revealed infection. 4 patients suffered from bacteremia at our hospital before surgery, and a blood culture obtained postoperatively at our hospital revealed infection, of whom one patient was in group A, 3 patients in group B. All patients had no recent (fewer than 12 months) nonspinal procedures or previous surgical interventions to the spine. Medical comorbidities in group A included diabetes mellitus in seven patients, smoking in seven patients, hypertension in three patients, chronic obstructive pulmonary disease (COPD) in two patients, hepatitis B in one patient and alcoholic hepatitis in two patients. Medical comorbidities in group B included diabetes mellitus in nine patients, smoking in nine patients, hypertension in four patients, COPD in one patient, hepatitis B in two patients. All patients underwent blood culture and seven had urine culture done, but infectious organisms were isolated from blood culture in only ten patients, of whom one had positive urine cultures as well. The primary causative organism was Staphylococcus aureus in seven cases and Escherichia coli in three cases. Laboratory examination revealed a leukocytosis in 33 of the 67 patients. All patients had elevation of erythrocyte sedimentation rate (ESR) and C-reactive protein (CRP). In group A, the ESR ranged from 40 to 127 mm/h (average, 84.2 mm/h). The CRP ranged from 31 to 86 mg/L (average, 52.9 mg/L). In group B, the ESR ranged from 45 to 122 mm/h (average, 83.7 mm/h). The CRP ranged from 32 to 87 mg/L (average, 51.4 mg/L). Cultures of biopsy specimens were positive for Staphylococcus aureus in six patients, Staphylococcus epidermidis in two patients, Escherichia coli in two patients, but the diagnosis of PVO was confirmed histologically in all of them. The mean periods of follow-up was 77.8 ± 10.5 months in group A and 80.9 ± 10.0 months in group B. The intra-operative blood loss and operation time in group B were more than that in group A, with a significant difference ( P < 0.05) (Table ). 54 patients of tissue specimens obtained from the diseased spinal tissue at operation had positive cultures. In group A, the most common causative organism was Staphylococcus aureus , which was positive in 10 patients. Other organisms were Staphylococcus epidermidis in 6 patients and Pseudomonas species in 3 patients; Salmonella species in 3 patients and Escherichia coli in 2 patients. In group B, the most common causative organism was also Staphylococcus aureus , which was positive in 12 patients. Other organisms were Staphylococcus epidermidis in 7 patients and Salmonella species in 4 patients; Escherichia coli in 4 patients and Pseudomonas species in 3 patients. Resolution of infection was exhibited in all, as noted by normalization of the ESR and CRP levels. All patients had significant improvement in constitutional symptoms and back pain after surgery. White blood cell, ESR, CRP returned to normal limits in all patients 3 months after surgery (Table ). All patients had pain relief. The VAS was 7.1 ± 0.7 in group A and 7.2 ± 0.6 in group B pre-operatively, which decreased to 2.1 ± 0.6 and 2.0 ± 0.7, respectively, at three months after surgery, then reduced to 0.4 ± 0.5 and 0.5 ± 0.5, respectively, at the final follow-up (Table ). All patients achieved bone fusion. The bony fusion time was 5.4 ± 1.0 months in group A and 5.2 ± 0.9 months in group B (Figs. k, l, h, i) (Table ). No neurological deterioration after surgery was noted in any of the cases. Neurologic deficits improved at final follow-up (Table ). At the last follow-up, the AISA grades of both two groups were significantly improved compared with that before operation ( P < 0.05), and there was no significant difference between the two groups ( P > 0.05) (Table ). Mean preoperative local lordotic angle was similar between the two groups (Table ). There was significant difference between the two groups at postoperative and final follow-up local lordotic angle ( P < 0.05) (Table ), and the local lordotic angle of the two groups was significantly improved compared with that before operation ( P < 0.05) (Table ). There was no obvious loss of correction in both groups (Table ). Complication Postoperative skin infection occurred in two cases (1 in group A and 1 in group B), which were all cured after antibiotic therapy and wound dressing. No complications related to instrumentation occurred. Four patients (2 in group A and 2 in group B) suffered from pneumonia, resolutive with anti-inflammatory and symptomatic supportive treatment during one week. No graft fracture, sliding or resorption was observed.
Postoperative skin infection occurred in two cases (1 in group A and 1 in group B), which were all cured after antibiotic therapy and wound dressing. No complications related to instrumentation occurred. Four patients (2 in group A and 2 in group B) suffered from pneumonia, resolutive with anti-inflammatory and symptomatic supportive treatment during one week. No graft fracture, sliding or resorption was observed.
Our study showed that all the patients in the two groups were cured without relapse, reinfection or long-term pain after surgery. All patients got bone fusion at the final follow-up without complications related to the internal fixation. This observation might indicate that both short-segment fixation and single-segment fixation can achieve the goal of long-term clinical treatment. And single-segment fixation techinique showed less invasive, which also can meets the requirements of reconstruction stability in the treatment of PVO. Single-segment fixation creates a relatively smaller surgical field of exposure and the procedure was performed with limited surgical access, thus causing little damage to the structure and physiological function of the spine, which is therefore less invasive, with reduced operative time and blood loss. Furthermore, in our study, the operation time and intra-operative blood loss in the single-segment fixation group were significantly lower than those in the short-segment fixation group, which brought relatively smaller surgical trauma. From clinical perspective, previous studies have shown that the monosegmental fixation is a safe and effective technique in patients with spinal fractures with a minimum two years follow-up . It can save motion segments in patients with adequate spine stability and good functional outcomes . Biomechanical experiments and finite element analysis have also indicated that single-segment fixation can meet the stability requirements necessary for spinal fracture reconstruction . Single-segment fixation is more hard in the treatment of PVO than that in the treatment of spinal fracture due to the pathological features of PVO. The vertebrae infected by PVO often present new bone formation and bone sclerosis, so the bone mineral density of infected vertebrae is generally higher than normal vertebrae. This results in stronger holding forces in the vertebrae with infection lesions than in those vertebrae with fractures, when performing pedicle scews of placement. In our study, pedicle screws were often inserted close to the endplate, which could get stronger holding forces on the vertebrae. This can achieve good reconstruction of spinal stability, which can get pain relief. In our study, all patients experienced significant relief of low back pain after surgery. The incidence of PVO is increasing, which may be attributable to various factors such as the aging of the society, the abuse of intravenous drugs, the widespread use of immunosuppression therapy for organ implant recipient and the progress in diagnostic methods with higher specificity [ – ]. These diseases often affect the “at-risk” populations, namely the elderly and the immunocomprimised due to malnutrition, diabetes mellitus, chronic smoking, intravenous drug abuse [ – ]. The most common site of infection is the lumbar spine (45–50%) . Spinal infections are often preceded by infections elsewhere in the body; and predisposing conditions include a genitourinary infection, urinary tract intervention, intravenous drug abuse, immunosuppression, indwelling vascular catheter, diabetes mellitus . Despite recent advances in imaging, and microbiological and histopathological techniques, the early detection of vertebral osteomyelitis remains difficult. The onset of the symptom is often insidious and could easily be underestimated (or ignored) by both patients and doctors. The most common symptom is back pain; it may be insidious in onset during the early stages of infection but typically worsen at the advanced stage. Neurologic deficit may not be present until later in the course of disease. Other constitutional symptoms, such as fever, weight loss, chills, and anorexy are non-specific.Timely diagnosis of pyogenic vertebral osteomylitis depends on its consideration in patients with back pain and fever, especially those who are elderly, diabetic or immunocompromised. Delay in diagnosis of results in more severe tissue destruction, spinal instability and worsening neurological deficit . So early diagnosisof PVO, especially identification of etiologic microorganism, become very important in the treatment of patients with PVO. The procedures for this step include blood cultures, percutaneous tissue biopsy and culture, and open biopsy and culture. When clinical history, laboratory values, and radiographic studies suggest spinal infection, blood cultures are firstly performed but show lower positive rates than other procedures . When the result of blood cultures is negative, the procedures for obtaining infected tissue should be considered before antibiotic treatment starts . In order to get enough tissue more accurately from the pathologic area, percutaneous CT-guided needle biopsy is a good option to enhance the culture result . The low positive rate of percutaneous needle biopsy in our study may be due to failure in the procedure, low-virulence organisms, and prior antibiotic use before admission. If biopsy cultures fail to show an organism, histological analysis can often confirm the diagnosis . Surgical treatment of PVO consists of radical debridement, reconstruction of anterior column with or without posterior stabilization aiming for fast postoperative mobilization . There is a broad range of options for the surgical management of spinal infections, which include anterior or posterior approach, combined anteriror and posterior surgery, with or without instrumentation. Because the inflammation is usually anterior to the neural contents, anterior operative approach is usually preferred. An anterior approach allows radical debridement, direct decompression and reconstruction of anterior column. However, this procedure has not been successful in preventing the progression of kyphosis or correcting the pre-existing kyphosis. And surgery performed with an anterior approach could result in higher mortality, especially for patients with combined cardiovascular and respiratory disease . Although anterior-only surgery can be performed, combined anteroposterior surgery provides for a more rigid spinal construct and better kyphosis correction. This combined procedure has a longer operation time, longer healing duration, and higher surgical trauma, especially for aged patients. The decision to perform anteroposterior spinal surgeries in a single-stage or 2-stage fashion is complex and remains controversial. Nowadays, authors have tended to emphasize the importance of tailoring the management options according to patient general medical condition, degree of bony destruction and location of compressive lesions. Also, as posterior instrumentation has become popular as a technique to stabilize the unstable spine and more effective regimens of appropriate antibiotics have become available, posterior-only procedure become an alternative treatment of spinal infection . We have treated contiguous spinal tuberculosis by posterior-only approach surgery and achieved good clinical effects [ – ]. We preferred posterior approach because some patients were in a poor medical condition and it was felt that surgery involving abdominal cavity would subject them to a high anaesthetic risk, with potential severe postoperative anterior complications. This approach far away from the abdominal cavity is characterized as the simple approach. It avoids high anaesthetic risk of anterior procedure with possibility to develop postoperative severe complications. Because there is no advantage of radical surgery over debridement when an extensive spinal lesion is present , we only removed focal tissues and tissues in focal edges, especially the sclerotic walls, caves, dead spaces, and so on, which could result in a incurative or recurrent result for vertebral osteomyelitis and reach the subnormal substance of bones between normal cancellous bones and pathologic bones. In our study, neurological function in patients with paraplegia was significantly improved postoperatively, which was similar to the result after anterior decompression . The choice of the fixation range in lumbar segment remains controversial. Short-segment fixation provided stronger fixation and is still used by most surgeons for treatment of mono-segment lumbar PVO. Longer fixed segment range can distribute the longitudinal stress of the spine, which can significantly maintain the spinal stability and prevent loss of the correction. But short-segment fixation sacrificed two normal motion segments, affecting the activity of the lumbar spine in the long term and leading to aggravation of adjacent segment degeneration (ASD), which may ultimately cause ASD-related complications . However, despite sacrificing more spinal functional units, our study have already demonstrated that compared with single-segment fixation, short-segment fixation can get better correction of kyphosis. Recently, many studies reported that titanium mesh cages have been shown to be effective for reconstructing a deficient anterior column after a corpectomy in the treatment of PVO [ , , ], by providing immediate spinal stability and improving sagittal balance, and thus facilitating bone healing. The use of titanium mesh cages has several benefits over other bone struts. The cage provides immediate stability, is rigid, and can tolerate compression forces well. The significant interface strength between the cage and endplates prevents it from extrusion or displacement. Most of all, the titanium mesh cage is the ideal shape, or it can be tailored to be positioned between adjacent vertebral endplates. It has relatively large weight-bearing surfaces. It is mechanically strong enough that can prevent from leading to discrete loss of height of a fused motion segment, which may be due to osteoporosis of the vertebrae . Compression on the interbody titanium mesh cages allowed correction of the kyphosis. Indications for the single-segment fixation techniques for the treatment of mono-segmental lumbar or lumbosacral PVO include: (1) lesions involved only a single motion segment, (2) ability to implant devices between adjacent vertebrae, (3) kyphosis angle < 20°. Contraindications to single-segment fixation include: (1) patients with severe kyphosis deformity, (2) presence of osteoporosis, (3) patients with vertebral osteomyelitis with bone healing or bone silent oscillation, accompanied by kyphosis deformity; these patients require osteotomy or internal fixation. The limitations of our study include the retrospective nature of the report, the relatively small sample of patients in different groups. A much larger, randomized controlled trial is required to elucidate the benefits and risks of our method. However, the data here may serve as preliminary results that can aid surgeons and patients in decision-making and in the design of future well-designed prospective studies.
Posterior-only debridement, interbody graft using titanium mesh cage, posterior single-segment instrumentation and fusion represents a safe and effective treatment option for selected patients with mono-segmental lumbar and lumbosacral PVO. This approach may preserve more lumbar normal motor units with less blood loss and operation time when compared with that of short-segment fixation. But short-segment fixation was superior to the single-segment fixation in the correction of kyphosis.
|
YKL-40 protein expression in human tumor samples and human tumor cell line xenografts: implications for its use in tumor models | 3d72ddd1-0f64-482d-a205-6052738842fa | 8516773 | Anatomy[mh] | YKL-40, also known as chitinase-3-like protein 1 (CHI3L1) is, as its name implicates, a chitinase-like glycoprotein, but is devoid of chitinase activity caused by mutations present in its active site . It is secreted by a variety of cells including cancer cells, inflammatory cells such as (tumor-associated) macrophages and neutrophils, and by chondrocytes, synovial cells and smooth muscle cells [ – ]. Accordingly, elevated serum levels of YKL-40 have been found in various inflammatory and malignant diseases [ – ]. Some studies have demonstrated a positive correlation of elevated serum levels of YKL-40 and a poor prognosis in solid tumors while others presented negative results . In case a correlation between YKL-40 secretion and malignant progression exists, it has been suggested that YKL-40 exerts its function mainly by stimulation of angiogenesis and regulation of extracellular matrix signaling [ – ]. A key component of this function is thought to be the ability of YKL-40 to bind to heparin and extracellular and cell surface heparan sulfate . Moreover, YKL-40 is known to play a potential role in promoting tumor cell proliferation and survival and to exhibit own growth factor activity [ , , ]. Studies investigating YKL-40 as a therapeutic target, however, yielded opposing results. In one study, the application of a neutralizing anti-YKL-40 antibody reduced tumor growth in a human glioblastoma (U87) xenograft model . In another human melanoma xenograft (LOX) model targeting YKL-40 led to increased tumor growth within hours after injection . Given the multiple and diverse functions attributed to YKL-40 and the reported opposing functions it can execute in malignant progression, its expression may differ depending on tumor entity and tumor subtype. Comprehensive protein expression data in tumor tissues, however, are lacking. Therefore, we set out to assess YKL-40 protein expression immunohistochemically in a large multi-tumor tissue microarray (TMA). In addition, we analyzed YKL-40 expression in widely used tumor cell lines taken directly from 2D cultures and from xenograft tumor tissues of these cell lines.
Xenograft tumors of human cancer cell lines Xenografted primary human tumors grown in immunodeficient scid mice were retrieved from the files of the Institute of Anatomy and Experimental Morphology from previous experiments. Small cell lung cancer: H69AR (RRID:CVCL_3513)*, NCI-H69 (RRID:CVCL_1579)*, NCI-H82 (RRID:CVCL_1591)*, and SW2 (RRID:CVCL_R777)* ; colon cancer: HT-29 (RRID:CVCL_0320)*, SW480 (RRID:CVCL_0546)*, CaCo2 (RRID:CVCL_0025)*, and HCT-116 (RRID:CVCL_0291)* ; breast cancer: DU4475 (RRID:CVCL_1183)*, MCF-7 (RRID:CVCL_0031)*, MDA-MB-231 (RRID:CVCL_0062)*, and T-47D (RRID:CVCL_0553)* [ , , ]; melanoma: FEMX-1 (RRID:CVCL_A011)*, LOX-IMVI (RRID:CVCL_1381)*, MDA-MB-435 (RRID:CVCL_0417)*, MeWo (RRID:CVCL_0445)*, and MV3 (RRID:CVCL_W280)* ; neuroblastoma: LA-N-1 (RRID:CVCL_1827)*, LA-N-5 (RRID:CVCL_0389), SK-N-SH (RRID:CVCL_D044), IMR32 (RRID:CVCL_0346), KELLY (RRID:CVCL_2092), and LS (RRID:CVCL_2105) ; osteosarcoma: HOS (RRID:CVCL_0312)* and U2OS (RRID:CVCL_0042); pancreatic cancer: PaCa5072 (RRID:CVCL_C887)*, PaCa5061 (RRID:CVCL_C886)*, BxPC-3 (RRID:CVCL_0186)*, and PANC-1 (RRID:CVCL_0480)* ; prostate cancer: DU145 (RRID:CVCL_0105)*, LNCaP (RRID:CVCL_0395)*, PC-3 (RRID:CVCL_0035)*, and C5 ; ovarian cancer: OVCAR3 (RRID:CVCL_0465)* and SKOV3 (RRID:CVCL_0532)* ; head and neck squamous cancer: UT-SCC-2 (RRID:CVCL_7820)*, UT-SCC-16A (RRID:CVCL_7812)*, UT-SCC-24A (RRID:CVCL_7826)*, UT-SCC-24B (RRID:CVCL_7827)*, UT-SCC-60A (RRID:CVCL_A089)*, and Carey24 . Cell lines marked with an asterisk have been authenticated by the DSMZ-German Collection of Microorganisms and Cell Cultures GmbH, Braunschweig, Germany using short tandem repeat (STR) profiling. Multi-tumor tissue microarray (TMA) For the production of TMAs, tissue cylinders with a diameter of 0.6 mm were punched from representative tumor or normal areas of each tissue block and transferred to recipient paraffin blocks. All tumor samples were obtained from the archives of the Institute of Pathology of the University Medical Center Hamburg Eppendorf. The use of archived diagnostic left-over tissues for the manufacturing of TMAs and their analysis for research purposes has been approved by local laws (HmbKHG, §12,1) and by the local ethics committee (Ethics commission Hamburg, WF-049/09). All work was carried out in compliance with the Helsinki Declaration. Freshly cut TMA sections were immunostained on one day and in one experiment (see below). Fixation, embedding and sectioning of cancer cells and xenograft tumors Preparation of cancer cells as formalin-fixed paraffin-embedded (FFPE) samples was carried out as previously described . Briefly, cancer cells were collected from culture flasks, fixed in formalin and next embedded into agar pellets. Agar pellets and formalin-fixed xenograft tumors were then subjected to standardized tissue infiltration using a Leica TP1020 tissue processor (Leica Biosystems, Nussloch, Germany). Subsequent paraffin embedding was performed using a Leica EG1160 Paraffin Embedding Center (Leica Biosystems, Nussloch, Germany). FFPE samples were sectioned with a thickness of 4 µm, mounted on HistoBond® glass slides (Paul Marienfeld, Lauda-Königshofen, Germany) and allowed to air-dry, followed by drying in an incubator at 37 °C overnight. Immunohistochemistry (IHC) FFPE sections were de-paraffinized in two changes of xylene replacement (5 min each) and rehydrated in a series of graded ethanol (100, 96, 70 and 50% for 5 min each). Next, sections were rinsed for 5 min each with Tris-buffered saline/0.1% Tween 20 (TBS-T) and TBS (pH 7.6). The subsequent incubation steps were carried out in a moist chamber. Sections were blocked with goat serum (#X0907, Dako, Carpinteria, CA, USA; diluted 1:10) for 30 min at room temperature (RT). Directly afterwards, sections were incubated with an anti-YKL-40 primary monoclonal mouse antibody (MAb 201.F9, 2.1 mg/ml, IgG2b, epitope GAWRGTTGHHS, aa 210–220) diluted 1:100 in antibody diluent (#S0809, Dako, Carpinteria, CA, USA) for 60 min at RT. For isotype control, mouse IgG2b antibody (#16–4732-85, eBioscience, San Diego, CA, USA; diluted 1:50) was used. Staining specificity was validated by preincubation of antibody MAb 201.F9 with recombinant human YKL-40 protein (#2599-CH, R&D Systems, Minneapolis, MN, USA) for 1 h at RT to block YKL-40 binding sites and reveal possible nonspecific staining to the sections (Supplementary Fig. ). After incubation, slides were rinsed twice with TBS-T as well as with TBS for 5 min each. Subsequently, slides were incubated with secondary biotin-conjugated goat anti-mouse antibody (#E0433, Dako, Carpinteria, CA, USA) at a dilution of 1:200 in TBS for 30 min at RT, followed by rinsing twice with TBS-T and once with TBS for 5 min each. Next, sections were treated with Vectastain® ABC-AP Kit (#AK5000, Linaris, Drossenheim, Germany) according to the manufacturer’s recommendations for 30 min at RT and again washed in TBS-T and TBS as described above. Finally, alkaline phosphatase enzyme activity was visualized by incubating the sections with Permanent Red solution (#ZUC001-125, Zytomed Systems GmbH, Berlin, Germany) for 20 min and counterstained with hematoxylin for 4 s, with intermediate washes under running tap water (3 min) and in aqua dest (2 min). Slides were dehydrated in a series of graded ethanol (70% for 15 s, 96 and 100% for 5 min each) and three changes of xylene replacement (5 min each) and, finally, covered with Eukitt® Mounting Medium (#03989, Sigma-Aldrich, Taufkirchen, Germany) and coverslips. Microscopy and image analysis IHC processed sections were first evaluated using a ZEISS Axiophot 2 microscope (Carl Zeiss, Jena, Germany). Digital images were obtained using a ZEISS Axio Scan Z1 slide scanner equipped with a ZEISS EC Plan-Neofluar 20x/0,50 Pol M27 objective (Carl Zeiss, Jena, Germany) and a Hitachi HV-F20SCL camera with 1600 × 1200 pixels (Hitachi Kokusai Electric America Ltd., NY, USA). For image acquisition, ZEISS ZEN 2.3 software was used (Carl Zeiss, Jena, Germany). Images were further processed using netScope Viewer software (Net-Base Software, Freiburg, Germany). YKL-40 staining was observed in the cytoplasm and as fine granular staining in the extracellular matrix. Staining intensity was assessed on a five-step scale (negative, weak, intermediate-low, intermediate-high, high). Further analysis included trichotomization of staining intensities (negative; weak and intermediate-low = ‘low’; intermediate-high and high = ‘high’).
Xenografted primary human tumors grown in immunodeficient scid mice were retrieved from the files of the Institute of Anatomy and Experimental Morphology from previous experiments. Small cell lung cancer: H69AR (RRID:CVCL_3513)*, NCI-H69 (RRID:CVCL_1579)*, NCI-H82 (RRID:CVCL_1591)*, and SW2 (RRID:CVCL_R777)* ; colon cancer: HT-29 (RRID:CVCL_0320)*, SW480 (RRID:CVCL_0546)*, CaCo2 (RRID:CVCL_0025)*, and HCT-116 (RRID:CVCL_0291)* ; breast cancer: DU4475 (RRID:CVCL_1183)*, MCF-7 (RRID:CVCL_0031)*, MDA-MB-231 (RRID:CVCL_0062)*, and T-47D (RRID:CVCL_0553)* [ , , ]; melanoma: FEMX-1 (RRID:CVCL_A011)*, LOX-IMVI (RRID:CVCL_1381)*, MDA-MB-435 (RRID:CVCL_0417)*, MeWo (RRID:CVCL_0445)*, and MV3 (RRID:CVCL_W280)* ; neuroblastoma: LA-N-1 (RRID:CVCL_1827)*, LA-N-5 (RRID:CVCL_0389), SK-N-SH (RRID:CVCL_D044), IMR32 (RRID:CVCL_0346), KELLY (RRID:CVCL_2092), and LS (RRID:CVCL_2105) ; osteosarcoma: HOS (RRID:CVCL_0312)* and U2OS (RRID:CVCL_0042); pancreatic cancer: PaCa5072 (RRID:CVCL_C887)*, PaCa5061 (RRID:CVCL_C886)*, BxPC-3 (RRID:CVCL_0186)*, and PANC-1 (RRID:CVCL_0480)* ; prostate cancer: DU145 (RRID:CVCL_0105)*, LNCaP (RRID:CVCL_0395)*, PC-3 (RRID:CVCL_0035)*, and C5 ; ovarian cancer: OVCAR3 (RRID:CVCL_0465)* and SKOV3 (RRID:CVCL_0532)* ; head and neck squamous cancer: UT-SCC-2 (RRID:CVCL_7820)*, UT-SCC-16A (RRID:CVCL_7812)*, UT-SCC-24A (RRID:CVCL_7826)*, UT-SCC-24B (RRID:CVCL_7827)*, UT-SCC-60A (RRID:CVCL_A089)*, and Carey24 . Cell lines marked with an asterisk have been authenticated by the DSMZ-German Collection of Microorganisms and Cell Cultures GmbH, Braunschweig, Germany using short tandem repeat (STR) profiling.
For the production of TMAs, tissue cylinders with a diameter of 0.6 mm were punched from representative tumor or normal areas of each tissue block and transferred to recipient paraffin blocks. All tumor samples were obtained from the archives of the Institute of Pathology of the University Medical Center Hamburg Eppendorf. The use of archived diagnostic left-over tissues for the manufacturing of TMAs and their analysis for research purposes has been approved by local laws (HmbKHG, §12,1) and by the local ethics committee (Ethics commission Hamburg, WF-049/09). All work was carried out in compliance with the Helsinki Declaration. Freshly cut TMA sections were immunostained on one day and in one experiment (see below).
Preparation of cancer cells as formalin-fixed paraffin-embedded (FFPE) samples was carried out as previously described . Briefly, cancer cells were collected from culture flasks, fixed in formalin and next embedded into agar pellets. Agar pellets and formalin-fixed xenograft tumors were then subjected to standardized tissue infiltration using a Leica TP1020 tissue processor (Leica Biosystems, Nussloch, Germany). Subsequent paraffin embedding was performed using a Leica EG1160 Paraffin Embedding Center (Leica Biosystems, Nussloch, Germany). FFPE samples were sectioned with a thickness of 4 µm, mounted on HistoBond® glass slides (Paul Marienfeld, Lauda-Königshofen, Germany) and allowed to air-dry, followed by drying in an incubator at 37 °C overnight.
FFPE sections were de-paraffinized in two changes of xylene replacement (5 min each) and rehydrated in a series of graded ethanol (100, 96, 70 and 50% for 5 min each). Next, sections were rinsed for 5 min each with Tris-buffered saline/0.1% Tween 20 (TBS-T) and TBS (pH 7.6). The subsequent incubation steps were carried out in a moist chamber. Sections were blocked with goat serum (#X0907, Dako, Carpinteria, CA, USA; diluted 1:10) for 30 min at room temperature (RT). Directly afterwards, sections were incubated with an anti-YKL-40 primary monoclonal mouse antibody (MAb 201.F9, 2.1 mg/ml, IgG2b, epitope GAWRGTTGHHS, aa 210–220) diluted 1:100 in antibody diluent (#S0809, Dako, Carpinteria, CA, USA) for 60 min at RT. For isotype control, mouse IgG2b antibody (#16–4732-85, eBioscience, San Diego, CA, USA; diluted 1:50) was used. Staining specificity was validated by preincubation of antibody MAb 201.F9 with recombinant human YKL-40 protein (#2599-CH, R&D Systems, Minneapolis, MN, USA) for 1 h at RT to block YKL-40 binding sites and reveal possible nonspecific staining to the sections (Supplementary Fig. ). After incubation, slides were rinsed twice with TBS-T as well as with TBS for 5 min each. Subsequently, slides were incubated with secondary biotin-conjugated goat anti-mouse antibody (#E0433, Dako, Carpinteria, CA, USA) at a dilution of 1:200 in TBS for 30 min at RT, followed by rinsing twice with TBS-T and once with TBS for 5 min each. Next, sections were treated with Vectastain® ABC-AP Kit (#AK5000, Linaris, Drossenheim, Germany) according to the manufacturer’s recommendations for 30 min at RT and again washed in TBS-T and TBS as described above. Finally, alkaline phosphatase enzyme activity was visualized by incubating the sections with Permanent Red solution (#ZUC001-125, Zytomed Systems GmbH, Berlin, Germany) for 20 min and counterstained with hematoxylin for 4 s, with intermediate washes under running tap water (3 min) and in aqua dest (2 min). Slides were dehydrated in a series of graded ethanol (70% for 15 s, 96 and 100% for 5 min each) and three changes of xylene replacement (5 min each) and, finally, covered with Eukitt® Mounting Medium (#03989, Sigma-Aldrich, Taufkirchen, Germany) and coverslips.
IHC processed sections were first evaluated using a ZEISS Axiophot 2 microscope (Carl Zeiss, Jena, Germany). Digital images were obtained using a ZEISS Axio Scan Z1 slide scanner equipped with a ZEISS EC Plan-Neofluar 20x/0,50 Pol M27 objective (Carl Zeiss, Jena, Germany) and a Hitachi HV-F20SCL camera with 1600 × 1200 pixels (Hitachi Kokusai Electric America Ltd., NY, USA). For image acquisition, ZEISS ZEN 2.3 software was used (Carl Zeiss, Jena, Germany). Images were further processed using netScope Viewer software (Net-Base Software, Freiburg, Germany). YKL-40 staining was observed in the cytoplasm and as fine granular staining in the extracellular matrix. Staining intensity was assessed on a five-step scale (negative, weak, intermediate-low, intermediate-high, high). Further analysis included trichotomization of staining intensities (negative; weak and intermediate-low = ‘low’; intermediate-high and high = ‘high’).
YKL-40 protein expression in primary human tumor samples In our TMA analysis 2,310 of 3,079 tumor samples were interpretable. Non-informative cases (n = 769; 25%) were due to lack of tissue samples or absence of unequivocal cancer tissues in the respective TMA spots. In the total cohort, negative (‘0’) YKL-40 expression was found in 15.9% of all cancers, low (‘1’) in 20%, intermediate-low (‘2’) in 33.6%, intermediate-high (‘3’) in 22.2%, and high expression (‘4’) in 8.4% of the cancers (Table and Fig. ). YKL-40 protein expression was observed as a diffuse, often granular, cytoplasmatic staining with varying intensity. Cell membranes were consistently negative. In most tumors the extracellular matrix showed varying granular staining, suggesting that the cancer cells had released YKL-40. Tumor-infiltrating leukocytes residing in the adjacent tumor stroma frequently stained positive (Fig. ). The percentage of tumor samples with positive stroma cell staining and the proportion of tumor samples with staining intensity in stroma cells surpassing that in tumor cells are reported in Table as well. Representative IHC images of negative, intermediate-low, intermediate-high, and strong YKL-40 protein expression are shown in Fig. . Anti YKL-40 monoclonal antibody Mab 201.F9 was tested to be specific for human YKL-40. Cross-reactivity with murine YKL-40 was excluded (Supplementary Fig. ). YKL-40 is differently expressed in tumor subtypes Previously, YKL-40 was found to be elevated in sera of various tumor patients including those carrying solid tumors, leukemias and lymphomas . In our study, a striking difference in YKL-40 tissue protein expression was observed when comparing solid tumors with lymphomas. About 50% of all Hodgkin and Non-Hodgkin lymphomas did not show any YKL-40 expression using IHC (Fig. ). This observation suggests that in lymphomas a substantial proportion of serum YKL-40 may be derived from stroma and immune cells due to the tumor’s immunologic reaction, rather than from the tumors cells themselves. In solid tumors, on the other hand, tumor samples stained positive more frequently and YKL-40 was differently expressed depending on tumor subtype (Fig. ). For further analysis, we trichotomized our results into a YKL-40 negative group, a YKL-40 low (low and intermediate-low) group and a YKL-40 high (intermediate-high and high) group to more clearly highlight those tumors and subtypes with differentially regulated YKL-40 expression. Accordingly, in the total cohort, negative YKL-40 expression was found in 15.9% of all cancers, while low YKL-40 expression was found in 53.6%, and high expression in 30.5% of all cancers. In the past, elevated serum YKL-40 levels have been proposed as prognostic biomarkers in some cancer entities. This association can be recapitulated for some entities analyzed in our TMA, while for others prognostic associations are much less suggestive. In particular, we found that in thyroid cancer, the total proportion of YKL-40-positive tumors was relatively lower in the prognostically favorable papillary and medullary carcinomas (62.2% and 60%, respectively, Fig. ). Contrary, in prognostically less favorable follicular carcinomas, a considerably higher percentage of YKL-40-positive tumors (95.6%) was observed. Looking at the fraction of only YKL-40 strongly expressing tumors, it was found to be highest in undifferentiated anaplastic carcinomas (37%) having a median survival of only a few months . Similar, in colorectal cancer, positive YKL-40 tissue expression increased from low-grade over high-grade to adenocarcinoma (samples positive in 70.6, 78.9, and 96.9%, respectively; Fig. ). In other cancer entities, YKL-40 expression was found to be associated with certain tumor subtypes. In gastric cancer, YKL-40 expression was highly associated with the intestinal type (96.7% positive), while 65.5% of tumors of the diffuse type were negative for YKL-40 expression (Fig. ). In serous and endometrioid differentiated ovarian cancer, YKL-40 expression was present in 95 and 96% of the tumors, respectively. On the contrary, ovarian cancer samples of the mucinous type stained positive in only 56.5% of the cases, whereas benign Brenner tumors were positive in 20% of the cases (Fig. ). When comparing adenocarcinomas with squamous cell carcinomas, no general distinguishing pattern of YKL-40 expression was observed between these two entities (Fig. ). While in esophageal cancer, for example, YKL-40 expression was found to be linked to the squamous cell cancer subtype (high expression in 46.1% of samples versus 16.1% in adenocarcinomas, Fig. ). In cervical cancer specimens, however, YKL-40 was found to be linked to the adenocarcinoma subtype (high YKL-40 expression in 45.5% of samples versus 27% in squamous cell carcinomas, Fig. ). YKL-40 protein expression in xenograft tumors of human cancer cell lines Tumor models, especially cancer cell-derived mouse xenografts, are increasingly used in cancer research. However, protein expression levels may differ significantly from clinical patient tumor samples to those grown in cell culture or as xenograft primary tumors in mice. Therefore, we assessed YKL-40 protein expression in human cancer cell lines directly taken from 2D-culture and paired mouse xenograft tumors. Cell pellets were embedded in agar and routinely processed as FFPE blocks, so that from a methodological point of view they were treated in the same way as primary biopsy specimens. We found that YKL-40 expression was immunohistochemically undetectable in all cancer cells grown in vitro except in SW480, CaCo2 and OVCAR3 cells (data not shown). Interestingly, YKL-40 expression became frequently positive under in vivo conditions in mouse xenograft tumors, especially in melanoma, pancreatic and colorectal cancer models (Table , Fig. ). Moreover, our data suggest that human YKL-40 was released from tumor cells into murine extracellular matrix components (Fig. ). YKL-40 expression analysis in publicly available datasets First, we checked the Broad Institute Cancer Cell Line Encyclopedia (CCLE) online tool for YKL-40 mRNA expression . We found that in only in 19% of 1,019 cell lines with expression values for YKL-40, expression z-scores were confidently above zero. Furthermore, we accessed publicly available GEO dataset GSE48433 comprising mRNA expression profiles of 49 human cancer cell lines and xenograft tumor fragments at passages 1, 4 and 10 from the Developmental Therapeutics Program of the National Cancer Institute (NCI DTP) . In 33 out of 49 cell line/xenograft pairs, YKL-40 expression increased under in vivo conditions (Supplementary Fig. ). As the increased YKL-40 expression in vivo may result from interactions of cancer cells with their surrounding extracellular matrices, we searched the NCBI GEO database for expression data comparing cells mono-cultured and co-cultured with stromal cells. Only a limited number of experiments with YKL-40 expression data was available (GSE115052, GSE60035, GSE98154, GSE109577) [ – ]. A clear tendency of increased YKL-40 expression in co-culture could be observed in prostate and ovarian cancer cells (Supplementary Fig. ).
In our TMA analysis 2,310 of 3,079 tumor samples were interpretable. Non-informative cases (n = 769; 25%) were due to lack of tissue samples or absence of unequivocal cancer tissues in the respective TMA spots. In the total cohort, negative (‘0’) YKL-40 expression was found in 15.9% of all cancers, low (‘1’) in 20%, intermediate-low (‘2’) in 33.6%, intermediate-high (‘3’) in 22.2%, and high expression (‘4’) in 8.4% of the cancers (Table and Fig. ). YKL-40 protein expression was observed as a diffuse, often granular, cytoplasmatic staining with varying intensity. Cell membranes were consistently negative. In most tumors the extracellular matrix showed varying granular staining, suggesting that the cancer cells had released YKL-40. Tumor-infiltrating leukocytes residing in the adjacent tumor stroma frequently stained positive (Fig. ). The percentage of tumor samples with positive stroma cell staining and the proportion of tumor samples with staining intensity in stroma cells surpassing that in tumor cells are reported in Table as well. Representative IHC images of negative, intermediate-low, intermediate-high, and strong YKL-40 protein expression are shown in Fig. . Anti YKL-40 monoclonal antibody Mab 201.F9 was tested to be specific for human YKL-40. Cross-reactivity with murine YKL-40 was excluded (Supplementary Fig. ).
Previously, YKL-40 was found to be elevated in sera of various tumor patients including those carrying solid tumors, leukemias and lymphomas . In our study, a striking difference in YKL-40 tissue protein expression was observed when comparing solid tumors with lymphomas. About 50% of all Hodgkin and Non-Hodgkin lymphomas did not show any YKL-40 expression using IHC (Fig. ). This observation suggests that in lymphomas a substantial proportion of serum YKL-40 may be derived from stroma and immune cells due to the tumor’s immunologic reaction, rather than from the tumors cells themselves. In solid tumors, on the other hand, tumor samples stained positive more frequently and YKL-40 was differently expressed depending on tumor subtype (Fig. ). For further analysis, we trichotomized our results into a YKL-40 negative group, a YKL-40 low (low and intermediate-low) group and a YKL-40 high (intermediate-high and high) group to more clearly highlight those tumors and subtypes with differentially regulated YKL-40 expression. Accordingly, in the total cohort, negative YKL-40 expression was found in 15.9% of all cancers, while low YKL-40 expression was found in 53.6%, and high expression in 30.5% of all cancers. In the past, elevated serum YKL-40 levels have been proposed as prognostic biomarkers in some cancer entities. This association can be recapitulated for some entities analyzed in our TMA, while for others prognostic associations are much less suggestive. In particular, we found that in thyroid cancer, the total proportion of YKL-40-positive tumors was relatively lower in the prognostically favorable papillary and medullary carcinomas (62.2% and 60%, respectively, Fig. ). Contrary, in prognostically less favorable follicular carcinomas, a considerably higher percentage of YKL-40-positive tumors (95.6%) was observed. Looking at the fraction of only YKL-40 strongly expressing tumors, it was found to be highest in undifferentiated anaplastic carcinomas (37%) having a median survival of only a few months . Similar, in colorectal cancer, positive YKL-40 tissue expression increased from low-grade over high-grade to adenocarcinoma (samples positive in 70.6, 78.9, and 96.9%, respectively; Fig. ). In other cancer entities, YKL-40 expression was found to be associated with certain tumor subtypes. In gastric cancer, YKL-40 expression was highly associated with the intestinal type (96.7% positive), while 65.5% of tumors of the diffuse type were negative for YKL-40 expression (Fig. ). In serous and endometrioid differentiated ovarian cancer, YKL-40 expression was present in 95 and 96% of the tumors, respectively. On the contrary, ovarian cancer samples of the mucinous type stained positive in only 56.5% of the cases, whereas benign Brenner tumors were positive in 20% of the cases (Fig. ). When comparing adenocarcinomas with squamous cell carcinomas, no general distinguishing pattern of YKL-40 expression was observed between these two entities (Fig. ). While in esophageal cancer, for example, YKL-40 expression was found to be linked to the squamous cell cancer subtype (high expression in 46.1% of samples versus 16.1% in adenocarcinomas, Fig. ). In cervical cancer specimens, however, YKL-40 was found to be linked to the adenocarcinoma subtype (high YKL-40 expression in 45.5% of samples versus 27% in squamous cell carcinomas, Fig. ).
Tumor models, especially cancer cell-derived mouse xenografts, are increasingly used in cancer research. However, protein expression levels may differ significantly from clinical patient tumor samples to those grown in cell culture or as xenograft primary tumors in mice. Therefore, we assessed YKL-40 protein expression in human cancer cell lines directly taken from 2D-culture and paired mouse xenograft tumors. Cell pellets were embedded in agar and routinely processed as FFPE blocks, so that from a methodological point of view they were treated in the same way as primary biopsy specimens. We found that YKL-40 expression was immunohistochemically undetectable in all cancer cells grown in vitro except in SW480, CaCo2 and OVCAR3 cells (data not shown). Interestingly, YKL-40 expression became frequently positive under in vivo conditions in mouse xenograft tumors, especially in melanoma, pancreatic and colorectal cancer models (Table , Fig. ). Moreover, our data suggest that human YKL-40 was released from tumor cells into murine extracellular matrix components (Fig. ).
First, we checked the Broad Institute Cancer Cell Line Encyclopedia (CCLE) online tool for YKL-40 mRNA expression . We found that in only in 19% of 1,019 cell lines with expression values for YKL-40, expression z-scores were confidently above zero. Furthermore, we accessed publicly available GEO dataset GSE48433 comprising mRNA expression profiles of 49 human cancer cell lines and xenograft tumor fragments at passages 1, 4 and 10 from the Developmental Therapeutics Program of the National Cancer Institute (NCI DTP) . In 33 out of 49 cell line/xenograft pairs, YKL-40 expression increased under in vivo conditions (Supplementary Fig. ). As the increased YKL-40 expression in vivo may result from interactions of cancer cells with their surrounding extracellular matrices, we searched the NCBI GEO database for expression data comparing cells mono-cultured and co-cultured with stromal cells. Only a limited number of experiments with YKL-40 expression data was available (GSE115052, GSE60035, GSE98154, GSE109577) [ – ]. A clear tendency of increased YKL-40 expression in co-culture could be observed in prostate and ovarian cancer cells (Supplementary Fig. ).
Elevated YKL-40 serum levels have been described in a growing number of inflammatory and malignant diseases and for many of them, YKL-40 has been proposed as a prognostic biomarker . The biological role and function of YKL-40 in malignant progression, however, remains elusive for multiple reasons. First, cellular receptors for YKL-40 have not yet been identified. Secondly, there is opposing data for the role of YKL-40 on proliferation and tumor growth in xenograft models . Thirdly, the question of whether serum YKL-40 in tumor patients is derived from production and release by the tumor cells themselves or from immune and/or stromal cells due to the tumor's immunologic reaction is a matter of debate . This work complements a former immunohistochemical study on YKL-40 protein expression in normal adult human tissues by Ringsholt et al. . The authors found YKL-40 to be an omnipresent protein throughout the normal tissues analyzed, but the number of samples was relatively small. Protein expression data in tumor tissues, however, are only available for some selected entities and subtypes. So far, integrative analyses of YKL-40 expression in cancer, whether in primary human tumor tissues or human cancer cell lines, are only available on the mRNA level. The lack of actual protein expression data is critical for the interpretation of the functional role of YKL-40 in tissues. In a comprehensive analysis of YKL-40 expression in differentiating mesenchymal stem cells (MSCs), Hoover et al. found undifferentiated MSCs to transcribe significant levels of YKL-40 mRNA, whereas YKL-40 protein was absent both in cell lysates and media supernatants . Moreover, they found that YKL-40 mRNA was not immediately translated into protein, but with a delay of at least 24 h following the addition of osteoblast or chondrocyte differentiation media. The authors hypothesized that YKL-40 may be regulated by (one or more) microRNAs (miRNAs) binding to YKL-40 mRNA. This hypothesis was at least partly supported by association studies of miRNAs and mesenchymal markers (among them YKL-40) in glioblastoma , but needs further verification. These findings imply that mRNA quantification alone may not reflect the complex regulation of YKL-40 in cancer progression and provides a rationale for studying YKL-40 expression at the protein level. In accordance with immunohistochemical data on normal tissues , it is not surprising that YKL-40 protein could be detected in a high number of tumors from all entities in our study. On the other hand, in a significant proportion of tumors YKL-40 expression was specifically absent, whereas in many tumors the staining frequency and intensity of stroma cells surpassed those of the tumor cells. We detected intensely stained stroma cells that were surrounded by finely granular material. Morphologically, these cells most likely represent infiltrating neutrophilic granulocytes. While Ringsholt et al. found that this observation was exceptionally made in surrounding tissues of the appendix , we found that this observation was true for most of the tumors. To what extent YKL-40 serum levels are fueled by release from the tumor cells themselves or from specific granules of the neutrophilic granulocytes needs to be addressed in future studies. Further, YKL-40 expression in serum, tumor tissue and adjacent stroma should be studied coherently to answer the question of whether those patients with no YKL-40 expression in the tumor tissues still have elevated serum levels. The herein used antibody Mab 201.F9 has been validated for immunohistochemistry purposes by others in the past and was demonstrated to have specific, strong labeling properties with a lack of nonspecific background staining . Moreover, we found that Mab 201.F9 showed no cross-reactivity with murine YKL-40. This notion is of specific importance when interpreting our results in mouse xenograft tumor tissues derived from human cancer cell lines. Our results clearly suggest that human cancer cell lines, when exposed to in vivo conditions, upregulate YKL-40 expression and secrete YKL-40 into the surrounding mouse-derived stroma. It is known that three-dimensional cell-to-cell contact and cell-to-matrix contact may influence a cell's gene expression pattern. Our findings were further corroborated by in silico analysis of YKL-40 mRNA expression profiles in murine tumor xenograft tissues compared with its expression in the parental cell line taken from 2-D culture. In a glioblastoma xenograft model, Francescone et al. showed that YKL-40 binds to heparan sulfate of the ectodomains of syndecan-1 and initiates coupling with integrin α v β 3 . In endothelial cells this action was followed by activation of a signaling cascade engaging focal adhesion kinase (FAK) and the mitogen-activated protein kinase/extracellular signal-regulated kinase (MAPK/ERK) pathway leading to elevated levels of vascular endothelial growth factor (VEGF) and enhanced angiogenesis [ , , ]. Syndecans act as matrix co-receptors with integrins, thereby mediating cell–cell and cell–matrix adhesion and promoting proliferation, differentiation and migration . Upregulation of YKL-40 in vivo may be one essential component in the binding and activation of these receptors. In many of the cell lines tested in our study, YKL-40 mRNA expression increased markedly in vivo in the xenograft tumor tissues. These results may have implications for research on the role of YKL-40 in tumor models. In the past, much of the work on the effect of YKL-40 on malignant progression in tumor models has been based on glioblastoma cells. This restriction is most probably due to the fact that only a few cell lines produce YKL-40 protein in vitro and in vivo such as the glioblastoma cell lines U87, U1242MG, U343MG and U1231MG, and the osteosarcoma cell line MG63 . As in vitro analyses revealed lack of endogenous YKL-40 expression in other cancer cell lines, this notion provided reason to use them as the null background against which biological activities of YKL-40 have been examined, among them MDA-MB-231 and HCT-116 cells engineered to ectopically express YKL-40 . Our results suggest that this ectopic overexpression of YKL-40 may override the pathways that physiologically (up-)regulate its expression in vivo. Therefore, careful consideration should be taken when choosing a tumor model. Our study may pave the way for examining the effects of YKL-40 in tumor models of other entities as well. In conclusion, our data provide new insight into YKL-40 expression at the protein level in various tumor entities and its regulation in tumor models. Our results may provide a rationale for characterizing YKL-40 as a feasible distinguishing tissue marker in tumor subtypes. Our data also suggest that YKL-40 expression is a common feature of tumor progression in vivo, while the absence of YKL-40 expression in vitro may be an artifact of cell culture.
Below is the link to the electronic supplementary material. Supplementary file1 (PDF 1270 kb)
|
Perspectives on Using Pharmacogenomics to Guide Tobacco Cessation: Survey Results From an American Indian Community | bb3aa1bf-36bb-46f5-a8d8-32ca8e00ef7d | 11904310 | Pharmacology[mh] | Introduction Pharmacogenomics has the potential to tailor medication therapies based on an individual's genetic makeup, thereby optimizing treatment efficacy and minimizing adverse effects. The clinical utility of pharmacogenomics includes, but is not limited to, oncology, cardiology, psychiatry, pain, and infectious disease . By tailoring treatment based on individual variability, pharmacogenomics promises more effective and individualized healthcare. Integrating pharmacogenomics into tobacco cessation strategies presents an opportunity to better understand how genetics influence tobacco use behaviors and its effect on the chance of successful cessation . Widespread use of pharmacogenomics‐guided treatment, however, faces challenges due to our limited knowledge about genomic variation across diverse populations. Current research disproportionately includes populations of European ancestry and those living in large urban areas, leaving many other ancestry groups and rural populations underrepresented . This lack of diversity can lead to biased or incomplete understandings of how genetic factors influence drug response and disease susceptibility in different populations. American Indian and Alaska Native (AIAN) populations are notably underrepresented in pharmacogenomics research. The extent of pharmacogenomic variation among AIAN people remains largely unknown, with limited data suggesting potential variability among AIAN populations . Concerns expressed by AIAN communities related to genetic research are due to past experiences that cultivated historical distrust, often rooted in researchers' failure to engage with AIAN communities ethically and inclusively . AIAN communities advocate for community‐based participatory research (CBPR) approaches to enhance research practices, emphasizing mutual respect and shared decision‐making, as well as ensuring community needs and priorities are addressed . The emphasis on building trusting relationships and ethical engagement underscores the significance of incorporating community perspectives and values into research practices. Commercial tobacco and nicotine product use is a significant public health concern in Tribal communities ; however, identifying the most effective cessation method for individuals remains a challenge . The application of pharmacogenomics into tobacco cessation efforts (e.g., nicotine replacement therapy or pharmaceutical intervention) offers an opportunity to integrate genomic information into our understanding of tobacco use behavior and cessation success. Nicotine is primarily metabolized by the drug‐metabolizing enzyme cytochrome P450 2A6 (CYP2A6). Genetic variation in CYP2A6 leads to interindividual variability in the enzymatic activity of CYP2A6 and subsequently influences the usage of commercial tobacco products and the success of cessation interventions . Pharmacogenomics can aid in developing more personalized treatment plans that could improve the chances of successful cessation. By identifying relevant genetic variation and predicting CYP2A6 activity among AIAN patients, pharmacogenomics‐guided cessation strategies can also provide benefits to Tribal communities. Integrating pharmacogenomics into tobacco cessation relies on community support, highlighting the importance of addressing historical concerns within AIAN communities and alignment with CBPR principles. We established a community–academic research partnership between the Confederated Salish and Kootenai Tribes (CSKT) and the University of Montana (UM) to increase the participation of AIAN people in pharmacogenomics research . The CSKT community, residing on the Flathead Reservation in northwestern Montana, is comprised of the Salish, Pend d'Oreille, and Kootenai Tribes . A recent community health assessment reports there are approximately 8,000 enrolled CSKT members, with approximately 5,200 members residing within the Flathead Reservation . Our partnership—particularly with the CSKT Tribal Council, CSKT Tribal Health Department, and the community advisory board for the project—addresses health challenges with an awareness of the unique needs and concerns of the CSKT community. Our ongoing collaboration strives for practical, culturally sensitive healthcare interventions developed with and for the CSKT community. Our study aims to understand perceptions of utilizing pharmacogenomics to guide tobacco and nicotine product cessation in the CSKT community through the dissemination and analysis of a survey that was developed with a commitment to community‐centered research in mind. The survey assesses the community's interests in intervention, recognizes concerns with genetic research, and identifies cultural considerations. Through our partnership between CSKT and UM, as well as collaboration with researchers from the University of Colorado, we engaged the community as partners and advisors in pharmacogenomics research. Ultimately, we aim to utilize results from the survey to inform future implementation efforts using genetics to tailor tobacco cessation methods in the CSKT community, potentially expanding genetics research to an underserved and marginalized population.
Methods 2.1 Study Setting and Recruitment We distributed a survey with the help of CSKT partners to understand the perspectives of CSKT members and descendants on the utilization of pharmacogenomics to guide tobacco and nicotine product cessation. Flyers with information about the survey and a QR code were posted in locations across the Flathead Reservation, including Salish Kootenai College (SKC), CSKT Tribal Health Clinics, and CSKT governmental buildings; members of the study group also gave presentations about the survey at SKC. Information about the survey was also shared via email, on our website, and through social media. Due to these various survey distribution methods, it was not possible to determine how many individuals we reached with the survey; thus, we were unable to estimate a response rate. Data were collected from July 2022 to November 2023. Informed consent from participants was obtained through one‐on‐one interactions at in‐person events and through online consent prior to the start of the electronic survey. In‐person events served as an opportunity to foster our relationship with the CSKT community, building trust and offering transparency. All CSKT members (regardless of their use of commercial tobacco or nicotine products) were invited to participate in the survey. The study was approved by the UM, SKC, and University of Colorado Institutional Review Boards. 2.2 Study Design and Analysis The survey was developed collaboratively between members of the research group and was modeled after a prior research project . Members of the CSKT community advisory board completed a pilot test of the survey and provided input for revisions and cultural appropriateness. The survey included four key sections: participant demographics, perceptions on tobacco use in the CSKT community, personal tobacco and nicotine product use and cessation efforts (only current and former tobacco and nicotine product users were instructed to complete this section), and the use of pharmacogenomics as a tool to guide cessation efforts. The survey included either 40 questions (for current and former tobacco and nicotine product users) or 27 questions (for those that have never used these products); questions were formatted as Likert scale, multiselect, yes/no, and open ended. Participants ≥ 18 years old and a self‐identified CSKT member or descendant were eligible for the study. Results from participants who answered ‘yes’ to using nicotine‐only products were combined with those who answered ‘yes’ to using commercial tobacco; there were no questions specific to either group. The survey took 15–20 min to complete and was offered in a print and an online format via Qualtrics (Provo, UT, USA). Surveys completed in print were manually entered into Qualtrics by members of the study team. Participants received a $15 gift card upon completion of the survey. Survey responses were exported to an Excel spreadsheet from Qualtrics. Demographics were summarized, and we utilized Pearson's Chi‐square test of association to assess if there were significant differences in responses between age or gender. Members of the research group (M.L.W. and D.M.W.) independently analyzed the open‐ended questions in a deductive manner using thematic analysis and further collaborated to combine results and assess for potential bias. Responses were grouped into common themes and representative quotes were identified. We further used themes identified from the open‐ended questions to analyze all Likert scale, multiselect, and yes/no questions in an inductive manner. In response to a question raised by a CSKT community member during a presentation about the research results, we performed a post hoc analysis using Pearson's Chi‐square test of association between the use of tobacco for cultural reasons and participants' desire to quit tobacco products altogether. We also generated a word cloud from participants' written responses to the statement “Please share any questions, concerns, or comments you have regarding the use of pharmacogenomics in tobacco cessation” using Zygomatic (Amersfoort, Netherlands). The word cloud was used to identify common words expressed by participants to be further grouped with themes identified by the survey. Participants who did not provide comments ( n = 56) were not included. Common everyday words like “a,” “the,” and “and” were automatically excluded and these exclusions were double‐checked manually.
Study Setting and Recruitment We distributed a survey with the help of CSKT partners to understand the perspectives of CSKT members and descendants on the utilization of pharmacogenomics to guide tobacco and nicotine product cessation. Flyers with information about the survey and a QR code were posted in locations across the Flathead Reservation, including Salish Kootenai College (SKC), CSKT Tribal Health Clinics, and CSKT governmental buildings; members of the study group also gave presentations about the survey at SKC. Information about the survey was also shared via email, on our website, and through social media. Due to these various survey distribution methods, it was not possible to determine how many individuals we reached with the survey; thus, we were unable to estimate a response rate. Data were collected from July 2022 to November 2023. Informed consent from participants was obtained through one‐on‐one interactions at in‐person events and through online consent prior to the start of the electronic survey. In‐person events served as an opportunity to foster our relationship with the CSKT community, building trust and offering transparency. All CSKT members (regardless of their use of commercial tobacco or nicotine products) were invited to participate in the survey. The study was approved by the UM, SKC, and University of Colorado Institutional Review Boards.
Study Design and Analysis The survey was developed collaboratively between members of the research group and was modeled after a prior research project . Members of the CSKT community advisory board completed a pilot test of the survey and provided input for revisions and cultural appropriateness. The survey included four key sections: participant demographics, perceptions on tobacco use in the CSKT community, personal tobacco and nicotine product use and cessation efforts (only current and former tobacco and nicotine product users were instructed to complete this section), and the use of pharmacogenomics as a tool to guide cessation efforts. The survey included either 40 questions (for current and former tobacco and nicotine product users) or 27 questions (for those that have never used these products); questions were formatted as Likert scale, multiselect, yes/no, and open ended. Participants ≥ 18 years old and a self‐identified CSKT member or descendant were eligible for the study. Results from participants who answered ‘yes’ to using nicotine‐only products were combined with those who answered ‘yes’ to using commercial tobacco; there were no questions specific to either group. The survey took 15–20 min to complete and was offered in a print and an online format via Qualtrics (Provo, UT, USA). Surveys completed in print were manually entered into Qualtrics by members of the study team. Participants received a $15 gift card upon completion of the survey. Survey responses were exported to an Excel spreadsheet from Qualtrics. Demographics were summarized, and we utilized Pearson's Chi‐square test of association to assess if there were significant differences in responses between age or gender. Members of the research group (M.L.W. and D.M.W.) independently analyzed the open‐ended questions in a deductive manner using thematic analysis and further collaborated to combine results and assess for potential bias. Responses were grouped into common themes and representative quotes were identified. We further used themes identified from the open‐ended questions to analyze all Likert scale, multiselect, and yes/no questions in an inductive manner. In response to a question raised by a CSKT community member during a presentation about the research results, we performed a post hoc analysis using Pearson's Chi‐square test of association between the use of tobacco for cultural reasons and participants' desire to quit tobacco products altogether. We also generated a word cloud from participants' written responses to the statement “Please share any questions, concerns, or comments you have regarding the use of pharmacogenomics in tobacco cessation” using Zygomatic (Amersfoort, Netherlands). The word cloud was used to identify common words expressed by participants to be further grouped with themes identified by the survey. Participants who did not provide comments ( n = 56) were not included. Common everyday words like “a,” “the,” and “and” were automatically excluded and these exclusions were double‐checked manually.
Results 3.1 Participant Demographics We received 304 responses to our survey. Participants were excluded from analysis if self‐identified Tribal affiliations were not CSKT ( n = 21) and if survey completeness was less than 75% ( n = 10). After these exclusions, a total of 273 participants were included in the final analysis (Table ). Of the 273 participants, 222 identified as current or former users of tobacco or nicotine products and were asked additional questions regarding personal tobacco and nicotine use and cessation efforts (Table ). Participant demographics are comparable to those of respondents in a recent CSKT community health assessment report . We found no significant differences between responses based on age or gender. 3.2 Community Insight on Tobacco Use and Awareness We gained insights from participants regarding their awareness and opinions on tobacco and nicotine products. Our survey results revealed an overwhelming majority (92%) of participants had a high level of awareness concerning the health risks associated with tobacco products (Figure ). We also found that 76% of participants expressed that tobacco usage is a problem in the CSKT community (Figure ). One participant's comment demonstrated the community's focus on finding ways to decrease tobacco usage: “I think it would be a great help to the Native community if there was a way to increase the cessation of tobacco, as to my knowledge a large percent of us smoke.” Our study also highlighted a personal connection to the risks of tobacco usage, with 78% of participants reporting loved ones or close friends affected by tobacco consumption (Figure ). Some reasons expressed by participants to quit using tobacco products include: “Didn't want my daughter to think smoking was okay”; “My grandchildren”; and “My newborn baby.” These quotes show the influence of friends and family on the desire to quit tobacco products. These findings highlight the widespread recognition among participants of the community‐wide impact of tobacco use, emphasizing a collective sense of responsibility toward addressing this public health challenge with future generations in mind. An important finding from our survey was that 63% of participants use tobacco for traditional purposes, highlighting a unique cultural consideration for AIAN populations (Figure ). A CSKT community member raised an interesting question during a presentation of the research results: “were individuals who used tobacco for cultural reasons less likely to want to quit?” We found no statistical significance between those who use traditional tobacco and a desire to quit tobacco products altogether. Therefore, it is important to consider people's traditional tobacco use, but not to assume that this is always a barrier to cessation. This information can also be used to potentially recognize each individual's unique treatment goals as some participants had varying opinions on their cultural tobacco versus commercial tobacco use. 3.3 Participants' Desire for More Information About Pharmacogenomics Our results revealed that although some participants had experience with pharmacogenomics, an opportunity for further community engagement and education remains. We found that 11% of participants had previously provided a DNA sample for a research study (Figure ) and 6% had participated in pharmacogenomics research (Figure ). We found it encouraging that 29% of participants either “strongly agree” or “agree” that they are knowledgeable in pharmacogenomics; however, there is still work to be done to more effectively engage the community. It is crucial not to overlook perspectives of individuals without knowledge of pharmacogenomics and expand educational efforts alongside research endeavors, empowering community members to make informed decisions regarding research participation. The majority of participants (55%) agreed that they would participate in a pharmacogenomics research study if they had more knowledge and information (Figure ). One participant's comment was encouraging that community members see a community‐centric model of pharmacogenomics research as a useful tool to improve health, “I think [this approach] would be a better method for Native Americans to understand pharmacogenetics and epigenetics to motivate the healing process.” Several other participants expressed a desire for more information about pharmacogenomics: “More awareness in the community about what [pharmacogenomics] is. Our community has a huge tobacco issue.”; “How does [pharmacogenomics] work? And why hasn't it been used in more studies?”; and “I would like to know more about [pharmacogenomics].” 3.4 Importance of Community Involvement in Pharmacogenomics Research Our survey assessed the community's opinions on research methods in order to inform future research and implementation of pharmacogenomics. Our results demonstrated that the majority (64%) of participants agree that pharmacogenomics research should include the Tribal community in all aspects of the research process (e.g., planning and design, data collection, analysis, and distribution of results). Likewise, most participants (63%) would prefer that pharmacogenomics research be conducted in partnership with their local Tribal Health center or clinic. These preferences for Tribal partnerships underscore the importance of conducting studies within familiar and trusted community settings, aligning with principles of CBPR to promote culturally sensitive research practices. The community's desire for support was also exemplified in a word cloud highlighting the most commonly used word in describing participant's views on using pharmacogenomics for tobacco cessation (Figure ). The word cloud revealed that participants most often used words such as “community,” “help,” “know,” “people,” “research,” and “quit,” suggesting that the community is seeking support in various forms, whether it is through access to resources, knowledge about pharmacogenomics, or assistance with quitting tobacco use. The prominence of words like “community” and “people” underscores the importance of collective action and shared responsibility in addressing tobacco cessation within the CSKT community. Additionally, the frequent use of words like “help” and “know” indicates a desire for guidance and understanding, highlighting the need for education and outreach initiatives tailored to the community's specific needs and concerns. 3.5 Community Perspectives on Pharmacogenomics Research The CSKT community expressed support and interest in furthering research in pharmacogenomics for tobacco cessation and participants saw the research as beneficial for the community. The majority of participants (68%) said they believe using pharmacogenomics to guide tobacco cessation would be a good resource for their community (Figure ). Likewise, support was shown through short answer responses such as: “I believe that using [pharmacogenomics] research to guide tobacco cessation efforts would be a good resource for my community”; “I'm grateful it's being done, anything to help Indigenous people live healthy, full lives is so good in my eyes”; and “I think this would be very beneficial to our Tribal community”. The majority of participants (53%) indicated they would be willing to donate a sample of DNA for a future study (Figure ). This level of interest in the CSKT community is promising to further research efforts and to implement pharmacogenomics as a clinical tool. 3.6 Concerns With Handling Biospecimens and Results Dissemination Although the support many participants expressed toward pharmacogenomics research is promising, there is still a large proportion of participants (29%) with concerns regarding pharmacogenomics (Figure ). One common theme was concerns in the handling of biospecimens and genetic material. Some participant quotes are as follows: “The story of Henrietta [Lacks] always plays in my mind, when considering using genetic materials for anything in non‐white communities” and “The DNA results [are] what I would worry about. Privacy I'm sure would be a cause for hesitation.” Both of these quotes show the community's concern for data and sample stewardship. Participants also expressed the need to disseminate research results to the community with one participant saying, “My only concern is how we can put out [pharmacogenomics] information to the general public more accessibly.” This quote emphasizes why dissemination of research results is crucial to increasing awareness and developing a trusting and collaborative relationship. We are committed to prioritizing research dissemination, including publishing in the Tribal newspaper , creating informative videos and brochures, presenting at community events, and remaining accessible for questions.
Participant Demographics We received 304 responses to our survey. Participants were excluded from analysis if self‐identified Tribal affiliations were not CSKT ( n = 21) and if survey completeness was less than 75% ( n = 10). After these exclusions, a total of 273 participants were included in the final analysis (Table ). Of the 273 participants, 222 identified as current or former users of tobacco or nicotine products and were asked additional questions regarding personal tobacco and nicotine use and cessation efforts (Table ). Participant demographics are comparable to those of respondents in a recent CSKT community health assessment report . We found no significant differences between responses based on age or gender.
Community Insight on Tobacco Use and Awareness We gained insights from participants regarding their awareness and opinions on tobacco and nicotine products. Our survey results revealed an overwhelming majority (92%) of participants had a high level of awareness concerning the health risks associated with tobacco products (Figure ). We also found that 76% of participants expressed that tobacco usage is a problem in the CSKT community (Figure ). One participant's comment demonstrated the community's focus on finding ways to decrease tobacco usage: “I think it would be a great help to the Native community if there was a way to increase the cessation of tobacco, as to my knowledge a large percent of us smoke.” Our study also highlighted a personal connection to the risks of tobacco usage, with 78% of participants reporting loved ones or close friends affected by tobacco consumption (Figure ). Some reasons expressed by participants to quit using tobacco products include: “Didn't want my daughter to think smoking was okay”; “My grandchildren”; and “My newborn baby.” These quotes show the influence of friends and family on the desire to quit tobacco products. These findings highlight the widespread recognition among participants of the community‐wide impact of tobacco use, emphasizing a collective sense of responsibility toward addressing this public health challenge with future generations in mind. An important finding from our survey was that 63% of participants use tobacco for traditional purposes, highlighting a unique cultural consideration for AIAN populations (Figure ). A CSKT community member raised an interesting question during a presentation of the research results: “were individuals who used tobacco for cultural reasons less likely to want to quit?” We found no statistical significance between those who use traditional tobacco and a desire to quit tobacco products altogether. Therefore, it is important to consider people's traditional tobacco use, but not to assume that this is always a barrier to cessation. This information can also be used to potentially recognize each individual's unique treatment goals as some participants had varying opinions on their cultural tobacco versus commercial tobacco use.
Participants' Desire for More Information About Pharmacogenomics Our results revealed that although some participants had experience with pharmacogenomics, an opportunity for further community engagement and education remains. We found that 11% of participants had previously provided a DNA sample for a research study (Figure ) and 6% had participated in pharmacogenomics research (Figure ). We found it encouraging that 29% of participants either “strongly agree” or “agree” that they are knowledgeable in pharmacogenomics; however, there is still work to be done to more effectively engage the community. It is crucial not to overlook perspectives of individuals without knowledge of pharmacogenomics and expand educational efforts alongside research endeavors, empowering community members to make informed decisions regarding research participation. The majority of participants (55%) agreed that they would participate in a pharmacogenomics research study if they had more knowledge and information (Figure ). One participant's comment was encouraging that community members see a community‐centric model of pharmacogenomics research as a useful tool to improve health, “I think [this approach] would be a better method for Native Americans to understand pharmacogenetics and epigenetics to motivate the healing process.” Several other participants expressed a desire for more information about pharmacogenomics: “More awareness in the community about what [pharmacogenomics] is. Our community has a huge tobacco issue.”; “How does [pharmacogenomics] work? And why hasn't it been used in more studies?”; and “I would like to know more about [pharmacogenomics].”
Importance of Community Involvement in Pharmacogenomics Research Our survey assessed the community's opinions on research methods in order to inform future research and implementation of pharmacogenomics. Our results demonstrated that the majority (64%) of participants agree that pharmacogenomics research should include the Tribal community in all aspects of the research process (e.g., planning and design, data collection, analysis, and distribution of results). Likewise, most participants (63%) would prefer that pharmacogenomics research be conducted in partnership with their local Tribal Health center or clinic. These preferences for Tribal partnerships underscore the importance of conducting studies within familiar and trusted community settings, aligning with principles of CBPR to promote culturally sensitive research practices. The community's desire for support was also exemplified in a word cloud highlighting the most commonly used word in describing participant's views on using pharmacogenomics for tobacco cessation (Figure ). The word cloud revealed that participants most often used words such as “community,” “help,” “know,” “people,” “research,” and “quit,” suggesting that the community is seeking support in various forms, whether it is through access to resources, knowledge about pharmacogenomics, or assistance with quitting tobacco use. The prominence of words like “community” and “people” underscores the importance of collective action and shared responsibility in addressing tobacco cessation within the CSKT community. Additionally, the frequent use of words like “help” and “know” indicates a desire for guidance and understanding, highlighting the need for education and outreach initiatives tailored to the community's specific needs and concerns.
Community Perspectives on Pharmacogenomics Research The CSKT community expressed support and interest in furthering research in pharmacogenomics for tobacco cessation and participants saw the research as beneficial for the community. The majority of participants (68%) said they believe using pharmacogenomics to guide tobacco cessation would be a good resource for their community (Figure ). Likewise, support was shown through short answer responses such as: “I believe that using [pharmacogenomics] research to guide tobacco cessation efforts would be a good resource for my community”; “I'm grateful it's being done, anything to help Indigenous people live healthy, full lives is so good in my eyes”; and “I think this would be very beneficial to our Tribal community”. The majority of participants (53%) indicated they would be willing to donate a sample of DNA for a future study (Figure ). This level of interest in the CSKT community is promising to further research efforts and to implement pharmacogenomics as a clinical tool.
Concerns With Handling Biospecimens and Results Dissemination Although the support many participants expressed toward pharmacogenomics research is promising, there is still a large proportion of participants (29%) with concerns regarding pharmacogenomics (Figure ). One common theme was concerns in the handling of biospecimens and genetic material. Some participant quotes are as follows: “The story of Henrietta [Lacks] always plays in my mind, when considering using genetic materials for anything in non‐white communities” and “The DNA results [are] what I would worry about. Privacy I'm sure would be a cause for hesitation.” Both of these quotes show the community's concern for data and sample stewardship. Participants also expressed the need to disseminate research results to the community with one participant saying, “My only concern is how we can put out [pharmacogenomics] information to the general public more accessibly.” This quote emphasizes why dissemination of research results is crucial to increasing awareness and developing a trusting and collaborative relationship. We are committed to prioritizing research dissemination, including publishing in the Tribal newspaper , creating informative videos and brochures, presenting at community events, and remaining accessible for questions.
Discussion Our study highlights the attitudes, perspectives, and concerns regarding the utilization of pharmacogenomics to guide tobacco and nicotine cessation within AIAN communities. Through a comprehensive survey, we unveiled optimism among CSKT members about the potential benefits of pharmacogenomics initiatives for tobacco cessation. Despite historical issues and mistrust in genetic research, our findings suggest a willingness among CSKT community members to engage in transparent dialogue and participate in future pharmacogenomics research. Concerns regarding data stewardship, dissemination of results, and the need for further education highlight that work is still needed. Researchers should address community concerns by prioritizing communication and transparency, engaging community leaders in decision‐making processes, respecting Tribal sovereignty, ensuring that research benefits the community, and respecting community ownership of data and biospecimens. Inclusive research practices are essential in building trust and ensuring that pharmacogenomics research respects the autonomy and dignity of AIAN communities, leading to equitable health outcomes. We found that the majority of CSKT participants exhibited high awareness of tobacco‐related health risks and expressed concern about usage prevalence in their community. A significant number of participants had personal connections to tobacco use, with some sharing stories of loved ones affected by tobacco use and how their loved ones' usage inspired their cessation efforts. The majority of participants agreed that tobacco use is a problem in their community, highlighting that this research is a priority for the community. The CSKT community has prioritized tobacco cessation through a number of initiatives, including educational programs for community members on tobacco cessation, cessation treatment programs through the CSKT Tribal Health Department, and expanding research efforts. This highlights the critical need for alignment between community health priorities and research endeavors, emphasizing how research should directly address and support the needs of the community. Recognizing the role of cultural tobacco use is vital in addressing tobacco cessation among Tribal populations. Our study found that the majority of participants acknowledged using tobacco for ceremonial, traditional, or cultural reasons. It is important to recognize key differences between cultural and commercial tobacco usage, such as variations in formulation and contents of the tobacco, means of inhalation, and frequency of use. Individuals may think about the use of cultural tobacco and commercial tobacco differently, and therefore, the goals of care for CSKT patients may need to consider continued use for traditional purposes. Understanding this prior to the implementation of pharmacogenomics in tobacco cessation is critical to providing personalized care. While we were encouraged that a relatively high number of participants expressed prior knowledge of pharmacogenomics research, more outreach is needed, with many describing the need for further education alongside research and implementation efforts. As research leads to clinical implementation, providing adequate education is necessary to empower individuals to make informed decisions about their healthcare. Importantly, a significant portion expressed a preference for community involvement in all stages of pharmacogenomics research, emphasizing the role of education in fostering community participation and trust. Furthermore, the majority indicated a preference for conducting research in partnership with the CSKT Tribal Health Department and CSKT Tribal leadership, aligning with the CBPR principles of our CSKT‐UM partnership. Our findings underscore the significance of community engagement and education in shaping research agendas and promoting equitable health outcomes . The CSKT community described important considerations regarding how the research is conducted. Several participants explicitly expressed concerns regarding the handling of biospecimens belonging to community members and the accessibility of research results to the community. These concerns are grounded in a history of misuse of specimens, lack of confidentiality and transparency, and stigmatizing interpretations specifically for AIAN populations . Yet, barriers to participation in pharmacogenomics research may result in limited benefits for AIAN people in genetics‐guided treatment, potentially perpetuating health disparities. Researchers must prioritize understanding community needs and priorities through active engagement within the community. Our research has also shown the importance of ensuring accessibility to pharmacogenomics research and implementation, as well as addressing concerns around the handling of data and biospecimens. Our approach includes utilizing CBPR methods and building upon our longstanding relationships with the CSKT community, particularly with the CSKT Tribal Council, CSKT Tribal Health Department, and the community advisory board for the partnership. Maintaining open channels of communication and addressing community concerns will remain central to our research efforts and moving toward interventional strategies to incorporate pharmacogenomics into tobacco cessation efforts. As we navigate the outcomes of our research, it is important to acknowledge the limitations we encountered, which shed light on areas for improvement in our approach. The majority of participants enrolled in our study identified as women—partially explained by recruitment at a women's health and wellness event on the Flathead Reservation—which may mean the perspectives of men and other gender identities are less well characterized in our study. This is consistent with our previous research projects with the CSKT; however, higher participation of women is common . Another consideration is that our participant population may be biased toward the inclusion of individuals who already have a more positive view of research, possibly leaving out perspectives from people who may have less favorable views toward research, and thus were less interested in participating. Additionally, some terminology used in the survey is of a higher reading level due to the nature of the topic and the inability to change terms (e.g., DNA, gene, metabolism, pharmacogenetics, and pharmacogenomics), which could hinder comprehension. In order to address this concern, however, our survey began with a definition of these technical terms. Furthermore, our research presents perspectives from only one AIAN community, and AIAN communities are not homogenous. Given the dearth of perspectives from AIAN people in research, however, our findings represent opinions and considerations from communities that are often excluded from research. Although we acknowledge these limitations, they have served as invaluable lessons, guiding us toward a deeper understanding of our research process and underscoring the need for continued community collaboration. Our research serves as a foundational step in informing future pharmacogenomic studies with AIAN communities within the context of tobacco and nicotine cessation. Understanding the frequency of known CYP2A6 genetic variants and the identification of rare or novel variants that influence CYP2A6‐mediated nicotine metabolism in the CSKT population could inform tobacco cessation strategies specifically tailored to this community. Ultimately, our goal is to implement tobacco and nicotine cessation interventions informed by pharmacogenomics to improve patient care within the CSKT community through personalized treatment approaches. Beyond the CSKT community, our study sheds light on the importance of community‐centric approaches to further research when working with underserved and marginalized populations. Through continued collaboration, transparent communication, and ethical community engagement, our research paves the way for more equitable and effective healthcare practices for AIAN populations worldwide.
Wrote manuscript: M.L.W., D.M.W., S.R.K., K.G.C., and E.L.W. Designed research: L.I.M., B.N.C., K.G.C.; and E.L.W. Performed research: M.L.W., D.M.W., S.R.K., K.E.B., K.A., and J.S. Analyzed data: M.L.W., D.M.W., S.R.K., and J.S. Contributed new reagents/analytical tools: M.L.W., D.M.W., S.S.G.M., K.G.C., and E.L.W.
As an Associate Editor for Clinical and Translational Science, Erica Woodahl was not involved in the review or decision process for this paper.
|
Mental health-related communication in a virtual community: text mining analysis of a digital exchange platform during the Covid-19 pandemic | cbb36583-c0ad-4d2b-800f-c30feaa91b08 | 9233821 | Health Communication[mh] | The pandemic and measures imposed by the authorities to manage the spread of Covid-19 have led to an increase in loneliness, stress, anxiety and depression, a general decline in mental well-being and an increase in substance or alcohol abuse in the general population . People with pre-existing mental health problems were at increased risk of poorer mental health outcomes due to reduced access to mental health services or measures such as quarantine or social isolation . Lack of accessible mental health services might have led to people affected by the Covid-19 pandemic seeking alternative solutions for social inclusion. Especially during the first wave of the pandemic and the associated lockdown, individuals used digital solutions to maintain their connections with others by gathering in virtual communities (VC) . Virtual communities These VC differed regarding their aims, such as a VC for passive readers of daily Covid-19 infection reports or a VC with more active contributors through reciprocal emotional support and posting humorous content . However, all VC had one common theme: they were initiated and moderated by the users . VC make an important contribution to the promotion and maintenance of mental health in contemporary society and promoting the processes of recovery . As a result, they are firmly positioned in the range of available structures for health-related services. The advantages of VC are that they provide a forum for expressing and sharing service users’ opinions and experiences. Users with similar health-related problems and interests can exchange information in a protected environment or be in contact with each other regardless of time and location . In VC users who are less likely to be in direct contact are more willing to share personal information . In this context, exchanges with peers (i.e. persons with a comparable health-related problem) are particularly supportive in improving mental health , although only a few users are active in this way, with the majority only reading the shared experiences . However, one disadvantage of VC is that unstructured, unmoderated peer-supported Internet offerings can also lead to an increase in the feeling of stress among users . inCLOUsiv Low-threshold digital counselling and information services by health professionals during the Covid-19 pandemic have made a central contribution to the prevention of mental illness . In this regard, a group of Swiss health researchers, communication specialists and mental health service users developed inCLOUsiv, which is a digital communication platform that combines the advantages of VC with a peer-led low-threshold counselling and information service provided by professionals. inCLOUsiv is meant to be accessible to all groups of people interested in the topic of mental health, but particularly addresses people with mental health issues. The platform provides information on how to manage, maintain and promote mental health, as well as giving users an opportunity for interaction and networking. Users are able to post and comment on articles and posts, and in addition they are able to join live discussions in which a specific topic is discussed with a moderator in a chat. The platform is operated by a Non-Governmental Organization specialized in the field of mental health and involves a team of communication specialists, health professionals and peer support workers. inCLOUsiv combines the characteristics of a VC with structured inputs and moderation, therefore it cannot be assumed that users will interact in the same way as on user-led VC during the Covid-19 pandemic. Moreover, we could not find any other study describing the interaction among users on professional-led mental health-related VC during the pandemic. Therefore, this study aims to explore how users interact with each other on the inCLOUsiv platform during the first Covid-19-related lockdown. We further tried to answer the following questions: What Covid-19-related topics do the users exchange? To what extent is the exchange among the users benevolent? How and to what extent do users support each other?
These VC differed regarding their aims, such as a VC for passive readers of daily Covid-19 infection reports or a VC with more active contributors through reciprocal emotional support and posting humorous content . However, all VC had one common theme: they were initiated and moderated by the users . VC make an important contribution to the promotion and maintenance of mental health in contemporary society and promoting the processes of recovery . As a result, they are firmly positioned in the range of available structures for health-related services. The advantages of VC are that they provide a forum for expressing and sharing service users’ opinions and experiences. Users with similar health-related problems and interests can exchange information in a protected environment or be in contact with each other regardless of time and location . In VC users who are less likely to be in direct contact are more willing to share personal information . In this context, exchanges with peers (i.e. persons with a comparable health-related problem) are particularly supportive in improving mental health , although only a few users are active in this way, with the majority only reading the shared experiences . However, one disadvantage of VC is that unstructured, unmoderated peer-supported Internet offerings can also lead to an increase in the feeling of stress among users .
Low-threshold digital counselling and information services by health professionals during the Covid-19 pandemic have made a central contribution to the prevention of mental illness . In this regard, a group of Swiss health researchers, communication specialists and mental health service users developed inCLOUsiv, which is a digital communication platform that combines the advantages of VC with a peer-led low-threshold counselling and information service provided by professionals. inCLOUsiv is meant to be accessible to all groups of people interested in the topic of mental health, but particularly addresses people with mental health issues. The platform provides information on how to manage, maintain and promote mental health, as well as giving users an opportunity for interaction and networking. Users are able to post and comment on articles and posts, and in addition they are able to join live discussions in which a specific topic is discussed with a moderator in a chat. The platform is operated by a Non-Governmental Organization specialized in the field of mental health and involves a team of communication specialists, health professionals and peer support workers. inCLOUsiv combines the characteristics of a VC with structured inputs and moderation, therefore it cannot be assumed that users will interact in the same way as on user-led VC during the Covid-19 pandemic. Moreover, we could not find any other study describing the interaction among users on professional-led mental health-related VC during the pandemic. Therefore, this study aims to explore how users interact with each other on the inCLOUsiv platform during the first Covid-19-related lockdown. We further tried to answer the following questions: What Covid-19-related topics do the users exchange? To what extent is the exchange among the users benevolent? How and to what extent do users support each other?
To answer the research question, we conducted a descriptive study with the natural language processing method, text mining (e.g. frequencies of words used or sentiment analysis) . For unstructured datasets such as text from VC, text mining is suitable because data preparation and analysis is automated and facilitates the identification of new information and relationships within comprehensive unstructured datasets . Ethics approval and consent to participate This research was not covered by the Swiss Human Research Act, since no health-related personal data and user information was obtained (SR810.30, Article 2). Accordingly, no request to an ethics committee has been made. The study was conducted in accordance with the Declaration of Helsinki. Regarding informed consent, the users were informed about the study, voluntary nature of their participation and anonymization of extracted data in writing by e-mail, before completing registration on the platform and the Impressum of the platform. The users gave their informed consent by registering to the platform. Participants could withdraw from the study without giving a reason. Sample We invited people with experience of mental illness and mental health professionals to register on the platform. The platform was promoted to the target groups via newsletters, announcements on the partners’ websites and social media channels (LinkedIn, Facebook, etc.). The invitation was sent by e-mail to the researchers’ network and national established peer groups, such as Peerplus, Network Recovery, Experienced Involvement Switzerland, and others. The e-mail included information about the platform inCLOUsiv as well as information about the study. However, other interested people were also able to register on the platform. People were eligible to participate actively, if they had access to the internet and an own e-mail address. Registered users interacted with others in the forums, live discussions and posts, or they were able to access the contents (e.g. articles) without registration (read-only). Users were able to log onto the platform with an e-mail address, a name and a password of their own creation, to exchange ideas with other users and with the editorial team and moderators. Data collection Data collection took place over the period 24 March–17 May 2020. Data from the forums and live discussions were stored in the MySQL database of inCLOUsiv. The raw data were exported without profile information (name and e-mail address) after the test phase was completed. Data analysis Preparation and analysis of the data were carried out in the statistical program R 3.6.1 based on tidy data principles with the primary ‘tm’ packages and ‘tidytext’ . The dataset was prepared for analysis in several steps. (1) All the words were identified as individual elements in a corpus (this process is referred to as tokenization) and, subsequently, the distribution of the words in the corpus was calculated. (2) On this foundation, all stop words, numbers, punctuation marks and other words not relevant for the analysis (such as names, greetings) were deleted, so that the starting point for the analysis is an anonymized dataset consisting only of the core contents of the conversations. (3) The words were reduced to their dictionary root or base form (this process is known as lemmatization) using spaCyr packages with the German-language-specific package ‘de_core_new_lg’. This was important for the analysis, so that word forms with the same root, such as ‘makes’ [macht], ‘made’ [gemacht] and ‘make’ [mache], are aggregated in the basic form ‘to make’ [machen]. (4) On the basis of the pre-processed corpus, frequencies, correlations (phi coefficient), n-gram and sentiment analyses were carried out. An n-gram is a sequence of n elements from a given text: for example, the bigram (where n = 2) comprises a sequence of two words. For sentiment analysis, the German-language-specific resource SentimentWortschatz (SentiWS) was used. In its current version, SentiWS consists of 1650 positive basic word forms and 1800 negative basic word forms, with an overall assigned polarity from − 1 (strongly negative) to 1 (strongly positive). For example, the word ‘great’ [super] has a positive polarity with a value of 0.5012 and the word ‘bad’ [schlecht] has a negative polarity with a value of − 0.7706.
This research was not covered by the Swiss Human Research Act, since no health-related personal data and user information was obtained (SR810.30, Article 2). Accordingly, no request to an ethics committee has been made. The study was conducted in accordance with the Declaration of Helsinki. Regarding informed consent, the users were informed about the study, voluntary nature of their participation and anonymization of extracted data in writing by e-mail, before completing registration on the platform and the Impressum of the platform. The users gave their informed consent by registering to the platform. Participants could withdraw from the study without giving a reason.
We invited people with experience of mental illness and mental health professionals to register on the platform. The platform was promoted to the target groups via newsletters, announcements on the partners’ websites and social media channels (LinkedIn, Facebook, etc.). The invitation was sent by e-mail to the researchers’ network and national established peer groups, such as Peerplus, Network Recovery, Experienced Involvement Switzerland, and others. The e-mail included information about the platform inCLOUsiv as well as information about the study. However, other interested people were also able to register on the platform. People were eligible to participate actively, if they had access to the internet and an own e-mail address. Registered users interacted with others in the forums, live discussions and posts, or they were able to access the contents (e.g. articles) without registration (read-only). Users were able to log onto the platform with an e-mail address, a name and a password of their own creation, to exchange ideas with other users and with the editorial team and moderators.
Data collection took place over the period 24 March–17 May 2020. Data from the forums and live discussions were stored in the MySQL database of inCLOUsiv. The raw data were exported without profile information (name and e-mail address) after the test phase was completed.
Preparation and analysis of the data were carried out in the statistical program R 3.6.1 based on tidy data principles with the primary ‘tm’ packages and ‘tidytext’ . The dataset was prepared for analysis in several steps. (1) All the words were identified as individual elements in a corpus (this process is referred to as tokenization) and, subsequently, the distribution of the words in the corpus was calculated. (2) On this foundation, all stop words, numbers, punctuation marks and other words not relevant for the analysis (such as names, greetings) were deleted, so that the starting point for the analysis is an anonymized dataset consisting only of the core contents of the conversations. (3) The words were reduced to their dictionary root or base form (this process is known as lemmatization) using spaCyr packages with the German-language-specific package ‘de_core_new_lg’. This was important for the analysis, so that word forms with the same root, such as ‘makes’ [macht], ‘made’ [gemacht] and ‘make’ [mache], are aggregated in the basic form ‘to make’ [machen]. (4) On the basis of the pre-processed corpus, frequencies, correlations (phi coefficient), n-gram and sentiment analyses were carried out. An n-gram is a sequence of n elements from a given text: for example, the bigram (where n = 2) comprises a sequence of two words. For sentiment analysis, the German-language-specific resource SentimentWortschatz (SentiWS) was used. In its current version, SentiWS consists of 1650 positive basic word forms and 1800 negative basic word forms, with an overall assigned polarity from − 1 (strongly negative) to 1 (strongly positive). For example, the word ‘great’ [super] has a positive polarity with a value of 0.5012 and the word ‘bad’ [schlecht] has a negative polarity with a value of − 0.7706.
After the platform was publicly accessible, the number of registered users as well as daily unregistered visitors increased continuously. By 4 May 2020, almost 700 registered users were recorded. At the beginning of April there were up to 10,000 page views per day. However, the amount of website visits fell sharply during April. In the last week of the evaluation (28 April–4 May 2020), an average of 795 views per day were recorded (min. 62; max. 1159). An assessment by username and activity also revealed that for this portion of 2300 visits, the majority were passive consumers of the content and did not participate in the dialogue. As registration only required a valid e-mail address and a name, no further assertions can be made about the users. The analysis was based on an adjusted dataset of 31,764 words. Frequencies, n-grams and correlations Figure shows the most frequently used words ( n > 75). The most widely used reference word was ‘good’ ( n = 396), even though in the preparation, greetings (e.g. good morning) and sign-off words (e.g. goodbye) were removed. Topic-specific words such as ‘corona’ ( n = 144) and ‘virus’ ( n = 100) are apparent. The words ‘crisis’ ( n = 82) and ‘anxiety’ ( n = 98) give an initial indication of possible discussion points in the forums and live chats. Analysis of the bigrams (Fig. ), filtered according to a frequency of n > 5, showed several clusters, the majority of which consisted of two combined constructs, such as the topics prevailing at that time in the media: for example, ‘keep’ ‘distance’ or ‘home’ ‘office’ . The construct ‘corona’ was a nodal point that was associated, on the one hand, with stressing topics and, on the other hand, with a cluster consisting of supportive (i.e. empathetic) content. The participants exchanged views on the theme of ‘loneliness’ and the impact of the ‘crisis’ on ‘mental health ’. ‘Time’ also stands as a nodal point between stressing and supportive topics. Thus, ‘time’ was experienced during the test phase as a ‘difficult’ ‘situation’ that needs to be ‘taken’ ‘seriously’ . Also shared among the participants were tips for coping strategies, how to ‘take’ ‘time’ , to ‘do yourself some good’ or ‘go outside’ and ‘go for a walk’ . In addition to visualization of the bigrams, correlation analysis (phi coefficient) shows that some topics often come up together but do not directly follow each other in sequence, as is the prerequisite for analysis of the n-grams (Table ). Here, too, the challenges faced by participants during the first wave of Covid-19 show up as ‘anxiety’ due to the ‘pandemic’ or the ‘virus’ . On the other hand, there are also possible coping strategies, such as ‘relaxation’ and ‘sports’ or ‘listening’ and ‘hearing a voice’ , which in this context is not discussed as a symptom,but as a coping strategy. The analysis according to bigrams and trigrams, with a focus on advice on coping strategies with the words ‘idea’ or ‘tip’ , without narrowing down by frequency, revealed a variety of activities, such as sports (e.g. walking, jogging), games (e.g. board games, Lego), writing (e.g. letters, diary), reading and talking with relatives and affected people (e.g. forum, phone calls, WhatsApp) who were recommended to each other. Sentiment analyses Most of the words did not describe neutral sentiments. The results show a low tendency for positive sentiment between users, with a polarity ranging between negative (− 1) and positive (+ 1) with an average value of 0.04, in which words with a positive classification were represented with a preponderance of 72% (3737 vs. 1425). Table summarizes the 15 most commonly used positive and negative words. The word ‘good’ ( n = 396) with positive polarization was most frequently used, followed by ‘help’ ( n = 156). The most frequently used word with a negative polarization was ‘anxiety’ ( n = 98), followed by ‘crisis’ ( n = 82). An in-depth sentiment analysis focusing on thematic words such as ‘loneliness’ , ‘crisis’ , ‘corona’ and ‘time’ could show that every word, whether it appears to have a negative or positive connotation, carries both polarities (see Fig. ). For example, the word ‘loneliness’ is related to ‘isolation’ and also to ‘being creative’ . As the polarity of sentiments can range from − 1 to 1, it is apparent that relatively strong words with negative polarization were used in comparison with words with a positive polarization.
Figure shows the most frequently used words ( n > 75). The most widely used reference word was ‘good’ ( n = 396), even though in the preparation, greetings (e.g. good morning) and sign-off words (e.g. goodbye) were removed. Topic-specific words such as ‘corona’ ( n = 144) and ‘virus’ ( n = 100) are apparent. The words ‘crisis’ ( n = 82) and ‘anxiety’ ( n = 98) give an initial indication of possible discussion points in the forums and live chats. Analysis of the bigrams (Fig. ), filtered according to a frequency of n > 5, showed several clusters, the majority of which consisted of two combined constructs, such as the topics prevailing at that time in the media: for example, ‘keep’ ‘distance’ or ‘home’ ‘office’ . The construct ‘corona’ was a nodal point that was associated, on the one hand, with stressing topics and, on the other hand, with a cluster consisting of supportive (i.e. empathetic) content. The participants exchanged views on the theme of ‘loneliness’ and the impact of the ‘crisis’ on ‘mental health ’. ‘Time’ also stands as a nodal point between stressing and supportive topics. Thus, ‘time’ was experienced during the test phase as a ‘difficult’ ‘situation’ that needs to be ‘taken’ ‘seriously’ . Also shared among the participants were tips for coping strategies, how to ‘take’ ‘time’ , to ‘do yourself some good’ or ‘go outside’ and ‘go for a walk’ . In addition to visualization of the bigrams, correlation analysis (phi coefficient) shows that some topics often come up together but do not directly follow each other in sequence, as is the prerequisite for analysis of the n-grams (Table ). Here, too, the challenges faced by participants during the first wave of Covid-19 show up as ‘anxiety’ due to the ‘pandemic’ or the ‘virus’ . On the other hand, there are also possible coping strategies, such as ‘relaxation’ and ‘sports’ or ‘listening’ and ‘hearing a voice’ , which in this context is not discussed as a symptom,but as a coping strategy. The analysis according to bigrams and trigrams, with a focus on advice on coping strategies with the words ‘idea’ or ‘tip’ , without narrowing down by frequency, revealed a variety of activities, such as sports (e.g. walking, jogging), games (e.g. board games, Lego), writing (e.g. letters, diary), reading and talking with relatives and affected people (e.g. forum, phone calls, WhatsApp) who were recommended to each other.
Most of the words did not describe neutral sentiments. The results show a low tendency for positive sentiment between users, with a polarity ranging between negative (− 1) and positive (+ 1) with an average value of 0.04, in which words with a positive classification were represented with a preponderance of 72% (3737 vs. 1425). Table summarizes the 15 most commonly used positive and negative words. The word ‘good’ ( n = 396) with positive polarization was most frequently used, followed by ‘help’ ( n = 156). The most frequently used word with a negative polarization was ‘anxiety’ ( n = 98), followed by ‘crisis’ ( n = 82). An in-depth sentiment analysis focusing on thematic words such as ‘loneliness’ , ‘crisis’ , ‘corona’ and ‘time’ could show that every word, whether it appears to have a negative or positive connotation, carries both polarities (see Fig. ). For example, the word ‘loneliness’ is related to ‘isolation’ and also to ‘being creative’ . As the polarity of sentiments can range from − 1 to 1, it is apparent that relatively strong words with negative polarization were used in comparison with words with a positive polarization.
The aim of the study was to describe the users’ interaction on the platform inCLOUsiv during the first lockdown of the Covid-19 pandemic. The findings show that the users’ communication behaviour was well meaning and supportive. The positive and supportive interaction seems to go in line with other mental health-related communication in VC . Thus, the way in which users interact with each other does not differ between user-led VC and VC managed by organizations or professionals. This benevolent communication represents an important basis for a successful VC, as it motivates and thus stimulates further exchange . Furthermore, the benevolent exchange among people with mental health issues seems to be a necessary basis for the development of coping strategies , which may indicate that professional-led VC can be used as an adjuvant to existing treatment. The identified topics of ‘anxiety’ and ‘ loneliness’ in connection with the virus and isolation go hand in hand with the current literature: that the degree of anxiety has increased among the population . However, current literature on the effect of VC on users’ health status or mental health during the Covid-19 pandemic is sparse. Further research is needed to understand and describe the effect of such VC on the users’ health status as well as their well-being, but this is a challenge because, from the perspective of the users, the advantage of a VC and the support it provides is anonymity . Online platforms seem to be particularly relevant for younger people with mental health problems. The focus is on anonymity, sharing experiences or easy and unlimited accessibility . In this study, no information was collected on the demographic data of the participants, which makes it impossible to sub-analyse by group. Another important contribution to the sustainable success of such VCs seems to be the active participation of health professionals in the exchange, because the absence of these representatives could raise scepticism towards them . In this study, health professionals were encouraged to participate actively, but their involvement was limited to the content of the live discussions, in which users had the opportunity to engage with both peer support workers and professionals. How health professionals can be further motivated to participate actively in VC should be the subject of further research. For longer-term success, it appears to be important to promote the development of cohesion . The chosen participatory approach on the inCLOUsiv platform and the possibility of introducing relevant topics could indicate development towards a community, at least for the phase described. Cultural circumstances constitute an important foundation for this sustainable added value . Eichenberg and Brähler report a high level of acceptance of virtual support offerings among the population but point out that these are not yet used as much in standard health care offerings as is the case in everyday life. inCLOUsiv contributes to closing this gap and provides a virtual offering for the population that is used in everyday life and can provide an interface to the standard health care offerings. Users can access the platform with flexibility in terms of time and location. The low-threshold offer also seems to have attracted a great deal of interest, as access to inCLOUsiv was passive (i.e. used only for content consumption) in the majority of cases.
The innovative approach to analysis of the discussion content provides insights into the main topics during the Covid-19 pandemic and also the communication between users. The results were able to demonstrate the topics that people with mental illness are dealing with during the pandemic, which coping strategies they experience as helpful and also that they consequently exchange with each other on inCLOUsiv. One weakness of the study is the short survey period of 3 months, which generated a dataset that did not allow for sub-analysis of sentiments by forum or live discussion. In addition, the sampling procedure may have resulted in the sole voluntary participation of users who were already engaged and not overburdened. Furthermore, the proportion of active users was small in relation to the total number of visits. However, this is in line with another study about user activity in Covid-specific VC . While health professionals were also asked about active participation on the platform, this group seemed restrained and more present as moderators or participants in the announced live discussions. Due to a lack of personal and health-related information on users, no differentiating analyses could be made. However, this was planned to ensure anonymity and low-threshold barriers for registration on the platform, which is particularly important for people with mental health problems . In the first wave of Covid-19, mental health systems were not adequately prepared in terms of suitable services or reduced access to services , therefore Covid-specific VC might beneficially compensate for some aspects of mental health services for people with a mental illness. Thus, decision makers in the mental health system should consider expanding their services with VC as there was a demonstrable need for low-threshold exchange on mental health during the Covid-19 pandemic .
The results indicate that users have benevolent discussions and support each other in managing their mental health during the pandemic on professional-led VC. Users state that they benefit from the experience of others and exchange useful coping strategies among themselves. The benevolent communication among the users suggests that the users consciously communicate supportively and that there is no stigma or bullying among them. For health professionals, this means that such low-threshold support offerings can be used as an adjuvant to existing therapy, particularly in times of reduced access to local health services. For research, the analysis method can be used after developing the script for regular evaluation of content and sentiments on VC. Data from longitudinal studies can provide an in-depth insight into the course of the users’ sentiments towards mental health-related topics.
|
Burnout among paediatric residents during the COVID-19 outbreak in France | 44d5f624-6687-4ee2-9aa7-4f2db6478353 | 7788161 | Pediatrics[mh] | Physicians are at high risk of psychological disorders including burnout . According to the World Health Organization’s International Statistical Classification of Diseases and Related Health Problems, 11th Revision , burnout is “a syndrome conceptualized as resulting from chronic workplace stress that has not been successfully managed”. The condition is characterized by emotional exhaustion, depersonalization, and reduced personal accomplishment . Burnout is known to have very harmful effects on physician health, patient care, and health care systems; the effects respectively include (i) substance abuse, depression, and suicidal ideation, (ii) medical errors, longer recovery times, and lower patient satisfaction, and (iii) reduced physician productivity and increased physician turnover for health care system . The 2020 pandemic of coronavirus disease 2019 (COVID-19) disrupted health care systems and work organizations across the world and notably in France. The outbreak resulted in greater workloads and increased levels of stress and fatigue among physicians . Before the COVID-19 outbreak, the prevalence of burnout among paediatric residents was already high (31 to 40% in the USA; no data for European countries) . By disrupting the health care system, the COVID-19 outbreak may have increased this prevalence . Increases in (i) stress levels related to contact with COVID-19 patients, (ii) the workload associated with more frequent medical consultations in paediatric emergency departments, and (iii) physicians’ sick leave might also be responsible for greater burnout among paediatric residents. We reasoned that better epidemiological knowledge of (i) burnout among paediatric residents and (ii) the impact of the current COVID-19 outbreak would make it possible to implement preventive actions and adapt them if further outbreaks occur . Hence, the primary objective of the present study was to assess the national prevalence of burnout among paediatric residents in France during the COVID-19 breakout. The secondary objective was to identify factors associated with burnout in this population.
Study design and population We performed a nationwide, cross-sectional survey of paediatric residents affiliated with all 28 French university hospitals teaching paediatrics. We calculated that around 1300 residents were eligible for the study; this number was based on the number of new paediatric residents in France in 2019 ( n = 325) and the duration of the residency in paediatrics (4 years). An invitation to participate in the study (with a link to an online questionnaire) was e-mailed to each resident in the first week of May 2020 (i.e. during the national lockdown period). A reminder was sent 3 days after the first e-mail. Replies were accepted up until May 11, 2020 (the last day of the national lockdown period). We chose to stop collecting data at the end of the lockdown so that major societal change did not bias our interpretation of the results. The e-mail informed the residents about the study’s objectives and invited them to click on the link if they wished to participate in the study. The online questionnaire’s homepage provided information about the group conducting the study and how the participants’ data would be processed. Residents participating in the study were informed of their data protection rights, in line with French legislation and the European Union’s General Data Protection Regulation. The online questionnaire contained items on the residents’ sociodemographic characteristics, professional situation, and burnout. Respondents were also asked to state their level of anxiety before the epidemic and then during the epidemic on a numerical scale ranging from 0 (no anxiety) to 100 (highest level imaginable). The study protocol was approved by an independent ethics committee (Amiens University Hospital, Amiens, France; reference: PI2020_843_0037 (2020-04-10)). Burnout definition The main outcome was the prevalence of burnout, as assessed with the French version of the Maslach Burnout Inventory – Human Services Survey (MBI - HSS) . This scale evaluates three domains of burnout in caregiver populations and the French version has been validated . The questionnaire consists of 22 items, each of which is scored from 0 (never) to 7 (several times a day). Nine of the items assess emotional exhaustion, five assess depersonalization, and eight assess reduced personal accomplishment. The cut-off scores for the emotional exhaustion and depersonalization domains were respectively ≥ 30 and ≥ 12. The cut-off for reduced personal accomplishment was ≤ 33. Residents were considered to have experienced burnout when the level of emotional exhaustion or depersonalization exceeded the cut-off, regardless of the presence or absence of reduced personal accomplishment . Definition of the professional consequences of COVID-19 We assessed the association between the COVID-19 outbreak and burnout with regard to three different items. In order to study the impact of the epidemic on the resident’s workload, the various regions of France were categorized according to the incidence of admissions to intensive care units for COVID-19 in the week before the questionnaire was sent out. The two other items were declarative. The residents had to answer the two following questions: “Did your professional activity change as a result of the COVID-19 outbreak?” and “Did you participate directly in the management of patients with COVID-19?” Statistical analyses Descriptive statistics were calculated after assessment of the data distribution using the Shapiro-Wilk test. Given that some variables were not normally distributed, the data were expressed as the median [interquartile range] and percentages were presented with their exact 95% binomial confidence intervals (CIs). The associations between burnout and the residents’ sociodemographic and professional characteristics were assessed in bivariate analyses. Categorical variables were compared using a chi-square test, and median values were compared using a Wilcoxon test. Odds ratio were calculated with logistic regression coefficient. The threshold for statistical significance was set to p < 0.05. All analysis was performed using R software (version 3.5.0) .
We performed a nationwide, cross-sectional survey of paediatric residents affiliated with all 28 French university hospitals teaching paediatrics. We calculated that around 1300 residents were eligible for the study; this number was based on the number of new paediatric residents in France in 2019 ( n = 325) and the duration of the residency in paediatrics (4 years). An invitation to participate in the study (with a link to an online questionnaire) was e-mailed to each resident in the first week of May 2020 (i.e. during the national lockdown period). A reminder was sent 3 days after the first e-mail. Replies were accepted up until May 11, 2020 (the last day of the national lockdown period). We chose to stop collecting data at the end of the lockdown so that major societal change did not bias our interpretation of the results. The e-mail informed the residents about the study’s objectives and invited them to click on the link if they wished to participate in the study. The online questionnaire’s homepage provided information about the group conducting the study and how the participants’ data would be processed. Residents participating in the study were informed of their data protection rights, in line with French legislation and the European Union’s General Data Protection Regulation. The online questionnaire contained items on the residents’ sociodemographic characteristics, professional situation, and burnout. Respondents were also asked to state their level of anxiety before the epidemic and then during the epidemic on a numerical scale ranging from 0 (no anxiety) to 100 (highest level imaginable). The study protocol was approved by an independent ethics committee (Amiens University Hospital, Amiens, France; reference: PI2020_843_0037 (2020-04-10)).
The main outcome was the prevalence of burnout, as assessed with the French version of the Maslach Burnout Inventory – Human Services Survey (MBI - HSS) . This scale evaluates three domains of burnout in caregiver populations and the French version has been validated . The questionnaire consists of 22 items, each of which is scored from 0 (never) to 7 (several times a day). Nine of the items assess emotional exhaustion, five assess depersonalization, and eight assess reduced personal accomplishment. The cut-off scores for the emotional exhaustion and depersonalization domains were respectively ≥ 30 and ≥ 12. The cut-off for reduced personal accomplishment was ≤ 33. Residents were considered to have experienced burnout when the level of emotional exhaustion or depersonalization exceeded the cut-off, regardless of the presence or absence of reduced personal accomplishment .
We assessed the association between the COVID-19 outbreak and burnout with regard to three different items. In order to study the impact of the epidemic on the resident’s workload, the various regions of France were categorized according to the incidence of admissions to intensive care units for COVID-19 in the week before the questionnaire was sent out. The two other items were declarative. The residents had to answer the two following questions: “Did your professional activity change as a result of the COVID-19 outbreak?” and “Did you participate directly in the management of patients with COVID-19?”
Descriptive statistics were calculated after assessment of the data distribution using the Shapiro-Wilk test. Given that some variables were not normally distributed, the data were expressed as the median [interquartile range] and percentages were presented with their exact 95% binomial confidence intervals (CIs). The associations between burnout and the residents’ sociodemographic and professional characteristics were assessed in bivariate analyses. Categorical variables were compared using a chi-square test, and median values were compared using a Wilcoxon test. Odds ratio were calculated with logistic regression coefficient. The threshold for statistical significance was set to p < 0.05. All analysis was performed using R software (version 3.5.0) .
Three hundred and forty paediatric residents (26.1% of the ~ 1300 in France) completed the questionnaire. The median (interquartile range [IQR]) age was 27 (IQR 25–28), 285 (83.8%, 95%CI [79.5–87.6]) of the residents were women, and 242 (71.2%, 95%CI [66.0–75.9]) of the residents lived with a partner. The median number of night shifts per month and per resident was 4 (IQR 3–5) and 94 (27.6%, 95%CI [23.0–23.7]) of the residents worked more than 60 h a week. With regard to COVID-19, 271 (79.7%, 95%CI [75.0–83.9]) of residents reported a change in their professional activity as a result of the outbreak, and 136 (40.0%, 95%CI [34.8–45.4]) were directly involved in the management of patients with COVID-19 (Table ). The median anxiety score was 40 (IQR 20–60) before the outbreak and 50 (IQR 30–60) during the outbreak. The prevalence of burnout in the study population was 37.4%, 95%CI [32.2–42.7]. The proportion of residents reporting a high level of emotional exhaustion was 23.5%, 95%CI [19.1–28.4], and 28.2%, 95%CI [23.5–33.3] reported a high level of depersonalization (Table ). In bivariate analyses, we found that sex, the number of hours worked per week, and the anxiety scores were associated with burnout (Table ). There was no association between burnout and a change in activity due to COVID-19 outbreak, direct involvement in the management of patients with COVID-19, and the intensive care unit admission rate for COVID-19 in the resident’s geographic area (Table ). The emotional exhaustion (Appendix ) and depersonalization (Appendix ) domain scores were also associated with the working hours per week and the anxiety scores.
The prevalence [95%CI] of burnout among paediatric residents in France was 37.4% [32.2–42.7]. This value is high and similar to those reported by studies in the USA (ranging from 31 to 40% ), even though the countries’ respective health systems are organized differently. The level of burnout among paediatric residents is nevertheless lower than among other specialties. In a recent meta-analysis , the proportion was 42.5% for residents in general surgery, anaesthesiology, orthopaedics, and obstetrics-gynaecology. One possible explanation is that paediatric residents are less likely to be confronted with emergency situations than the other residents mentioned—even though the population of paediatrics residents is heterogeneous, and some residents are exposed to emergency-related stress in neonatal or paediatric intensive care units. With regard to the prevalence rates for the MBI - HSS subdimensions, the value for depersonalization (corresponding to a negative, cynical attitude toward patients) was higher in our study than in previous studies . A meta-analysis has reported that the prevalence rates depend on the specialty: the prevalence of depersonalization was higher among obstetrics-gynaecology residents, whereas the prevalence of emotional exhaustion was higher among cardiology residents . The self-reported working hours might explain (at least in part) the prevalence of burnout among paediatric residents in France . As described in the literature, we observed that the anxiety scores (both before and during the COVID-19 outbreak) were significantly associated with burnout . The prevalence of burnout was higher among male residents than among female residents, as also described in the literature . In contrast to previous studies, we did not observe a significant association between burnout and the resident’s level of training ( p < 0.16) . In the present study, the consequences of COVID-19 (the intensive care unit admission rate for COVID-19 in the resident’s region, direct participation in the management of patients with COVID-19, and changes in professional activity due to the COVID-19 outbreak) were not significantly associated with burnout. Earlier studies have shown various associations between physician burnout and natural disasters . After an earthquake in Italy in 2016, the prevalence of physician burnout was high (25.97%) . Similarly, the prevalence of physician burnout after Hurricane Irma in 2017 was 34.1% (95%CI [28.2–40.0]) . Yeo et al. focused on the responses of first-year trainee physicians to Hurricane Harvey in 2017 . The researchers observed that the prevalence of emotional exhaustion was about 10.2% in their cohort. The lack of an impact of the COVID-19 outbreak in the present study might be due to the low incidence of severe disease in children or the fact that the baseline prevalence of burnout in our study population was already high . Similar results were observed by Dimitriu et al.; the burnout rate in resident in first-line contact with patients with suspected or confirmed COVID-19 was not higher than in other residents . In Dimitriu et al.’s single-centre study, the prevalence of burnout among the residents was extremely high (76%), which meant that there was little room for a further increase. Our study had several limitations. Firstly, burnout can be defined in several ways. We only considered the three main generally accepted psychopathological subdimensions . However, we used the MBI-HSS (a validated scale for assessing all three domains of burnout) and applied a common definition of burnout (a level of emotional exhaustion or depersonalization exceeding the usual cut-off) . Secondly, our dataset came from a sample of 340 paediatric residents in France. In 2019, 325 new positions for paediatric residents became available, and the paediatric residency program lasts for 4 years. Although not all the available positions are filled and the deadline for replies was quite short, the response rate of 26.1% (340 out of ~ 1300 residents) was relatively high when compared with previous studies , and all the French university hospitals were represented. To the best of our knowledge, the present study was the first to have assessed burnout among paediatric residents in France (or in any European country, for that matter). Thirdly, our evaluation of the workload related to the COVID-19 outbreak was not very detailed. Nevertheless, we used three different items to try to increase the assessment’s sensitivity. The first two items were two “yes”/“no” questions about the resident’s professional activity. The third item was a quantitative question about the incidence of intensive care unit hospitalizations for COVID-19 in the week prior to the questionnaire being sent out. Fourthly, the study’s cross-sectional design prevented us from defining causal relationships between the sociodemographic data, the prevalence of burnout, and COVID-19. The findings must be confirmed in a longitudinal study with measurements before and after a natural disaster. Further longitudinal study of risk factors associated with burnout (such as chronic exposure to stress) is also required . Investigating the prevalence of burnout among the health care staff repeatedly (and not only after natural disasters) would help to focus attention on the health authorities’ decisions with regard to work organization, supervision, time management, and stress . Furthermore, support programs could usefully reduce work-related anxiety and could be monitored by regular assessing the prevalence of burnout—a major issue for both residents and the patients they care for . Similar prospective studies in other European countries would enable comparisons of various health systems.
The prevalence [95%CI] of burnout among paediatric residents in France was 37.4% [32.2–42.7]; 23.5% [19.1–28.4] of the respondents reported a high level of emotional exhaustion, and 28.2% [23.5–33.3] reported a high degree of depersonalization. Burnout was significantly associated with working hours per week and anxiety scores but not with exposure to the consequences of COVID-19; the latter observation might reflect the low incidence of symptomatic COVID-19 in children. Regular assessment of the prevalence of burnout within hospitals in France and in other countries might help to identify risk factors or protective factors for burnout; this would benefit both individuals and health systems/organizations.
ESM 1 (DOCX 20 kb). ESM 2 (DOCX 20 kb). ESM 3 (DOCX 21 kb).
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.