fulltext
stringlengths 0
6.01M
|
|---|
Emerging infectious diseases are caused by pathogens that are newly recognized or whose incidence has either increased in the preceding 2 decades or threatens to increase. Viral diseases account for a large proportion of such infections. In the context of transplant recipients, important emerging viruses can be considered to be 1 of 3 types: (1) novel viruses; (2) known viruses increasing in incidence in the general population and, potentially, in transplant recipients; and (3) previously known viruses that cause disease of increased severity in the immunocompromised host. In this review, we begin by discussing viral diagnostics and the evolving field of viral discovery, which has increased the speed of virus identification but has created new challenges.
Our focus then shifts to specific emerging and reemerging viral pathogens in the transplant community. Viruses described in case series or multiple case reports are listed in Table 1 . Viruses described only in single case reports are listed in Table 2 . The potential risks of viral transmission as the result of xenotransplantation will not be addressed [1] .
Viral discovery has typically relied on the ability to detect new viruses in cell culture. Although clinical virology laboratories affiliated with transplant centers routinely perform viral culture, many pathogens do not grow well or do not grow at all, and viral detection using culture is further limited by the number of cell lines a laboratory can realistically maintain. Pathogen detection in the clinical laboratory is also limited by the available tests, which often target conserved sequences (polymerase chain reaction [PCR] or real-time [RT]-PCR) or specific antigens or antibodies to detect known viruses. Multiplex testing for clinical syndromes, particularly for respiratory-tract infections, allows for a less biased approach to viral diagnosis but still faces limitations in identifying emerging pathogens [2] . In rare situations, an unusual virus may be detected by testing for known pathogens, as in the case of a woman who presented with Usutu viremia, which gave a low-positive result by West Nile virus RT-PCR [3] .
A number of more rapid molecular methods are now being employed in viral discovery, categorized as sequence dependent (such as the pan-viral microarray) or sequence-independent techniques [4] . The pan-viral microarray is an array spotted with oligonucleotide sequences representing known viral pathogens. Novel viruses can be identified if sufficient similarity exists between sequences in the new virus and those on the array. Amplicons can then be recovered from the array, then cloned and sequenced [5] . This technology was used in the identification of SARS coronavirus from a cultured patient isolate [4] . PCR based on conserved sequences generally has limited applicability in viral diagnostics, as viruses do not contain highly conserved sequences analogous to 16S ribosomal RNA sequences utilized in bacterial identification [6] .
The sequence-independent amplification and sequencing of viral nucleic acids in biological samples has been termed viral metagenomics [4, 7] . Sequence-independent approaches include subtractive hybridization or representation difference analysis, sequence-independent single-primer amplification, and rolling circle amplification. These techniques have been used to identify agents such as human herpes virus 8, torque teno viruses (TTV), hepatitis E virus (HEV), Norwalk virus, parvovirus 4, and human bocavirus (HBoV) [4, 7] . Viral metagenomics has been aided by the development of a number of new sequencing platforms. Termed next-generation sequencing (NGS, or deepsequencing), such technologies allow for the rapid and parallel generation of 10 6 to over 10 9 sequences per run. Most current technologies rely on nonspecific amplification of viral DNA or RNA from samples treated to remove host nucleic acids. Amplification is followed by sequencing by synthesis using different technologies to detect base incorporation [6] [7] [8] . NGS has been utilized to identify novel viruses in patient samples and in studies of fevers of unknown origin [9, 10] . NGS has a great ability to detect both known and previously unknown (divergent) viruses, but mere detection does not demonstrate causation. For many of these viruses, classical Koch's postulates cannot be applied, and as demonstrated with TTV and HBoV, establishing a causative role for these agents can be difficult [6, 11, 12] . Mokili and colleagues [6] proposed "Metagenomic Koch's Postulates," but whether they are sufficient remains moot. At this time, NGS is largely a tool for research purposes. Sequencing reactions take a good deal of time to set up and perform [8] . These runs generate massive amounts of data that must be filtered prior to analysis using various alignment programs designed to handle the large numbers of short reads [8, 9] . Finally, results must be interpreted carefully. Contaminants from the laboratory and even from commercial reagents are often identified (eg, xenotropic murine leukemia virus-related virus), and confirming the presence of a virus identified with small numbers of reads may not be possible [9, 13] .
Human T-cell Leukemia Virus Type 1
Human T-cell leukemia virus type 1 (HTLV-1) seroprevalence rates range from 3% to 30% in endemic areas, to <1% in Western countries [14] . Chronic HTLV-1 infection is associated with adult T-cell leukemia (ATL) and HTLV-1-associated myelopathy (HAM) in 5% or fewer of those infected, but there is concern that immunosuppression in HTLV-1-positive transplant recipients may trigger progression to these complications [15] [16] [17] . Yoshizumi et al [17] identified 26 HTLV-1-positive, living donor liver transplant recipients. ATL developed in 4 patients at 181-1315 days post transplantation; all 4 patients died, including 3 from ATL. Overall survival rates did not differ between HTLV-1-positive recipients and 305 HTLV-1-negative liver transplant recipients from the same institution [17] . Case reports of ATL following renal transplantation in HTLV-1positive patients have been documented, though in case series of renal transplant recipients (totaling 46 patients with 5-17 years of follow-up), no cases of ATL or HAM developed [18] [19] [20] [21] . HAM has been reported in 1 HTLV-1 D+/R+ living-related liver transplant recipients [22] .
ATL responds poorly to conventional chemotherapy, with the highest median survival rates reported in clinical trials being approximately 13 months [23] . As a consequence, hematopoietic stem cell transplantation (HSCT) has been evaluated for the treatment of ATL in HTLV-1-positive patients. (HSCT will be used to describe the transplantation of multipotent stem cells from bone marrow, peripheral, or cord blood.) The largest study of HSCT for ATL involved the retrospective analysis of 386 patients with ATL who had undergone an allogeneic HSCT at 3 centers in Japan [23] . Their 3-year survival rate was 33%; ATL recurred in 41% of patients who survived to 30 days post transplant. Those who received transplants from a related HTLV-1 seropositive donor had a higher risk of disease-associated mortality relative to those whose related donor was HTLV-1 negative. HSCT recipients in complete remission at the time of transplantation had a higher rate of survival compared to patients not in complete remission (51% vs 26%) [23] . The transmission of HTLV-1 through transplantation or transfusion has been documented. In Spain, 3 HTLV-1-negative recipients of organs from a single HTLV-1-positive donor (1 liver and 2 kidney transplants) developed HAM within 2 years of transplantation [24] . Two case reports document the occurrence of HAM in a heart transplant recipient and an HSCT recipient who acquired HTLV-1 through blood transfusions [16, 25] . In low-prevalence areas, however, universal donor screening with Enzyme-linked immunosorbent assay followed by Western blotting resulted in many false positives, and the practice is no longer recommended by the United States Organ Procurement and Transplantation Network [14] .
HEV is a common cause of acute liver disease in the developing world, primarily from fecal-oral spread through contaminated drinking water. Infections in developed nations are well described, but typically result from the consumption of undercooked pork products. In immunocompetent hosts, HEV acute infection is self-limited with rare progression to fulminant liver failure. However, in the immunosuppressed host, chronic infection marked by persistent viremia and abnormal liver function with eventual progression to cirrhosis can occur [26] [27] [28] . Cases have been described in recipients of a variety of transplants, including kidney, liver, heart, and lung [27, 28] . In a multicenter review of 85 cases of acute HEV infection, 65.9% of the solid organ transplant (SOT) recipients developed chronic hepatitis of whom 14.3% developed cirrhosis. The use of tacrolimus compared with cyclosporine A was an independent predictor of chronic infection [28] . Rare cases of encephalitis and polyradiculopathy with HEV RNA detection in the cerebrospinal fluid (CSF) have also been described [26] .
The majority of HEV infections following SOT result from de novo infections and are unlikely to represent virus reactivation [26, 29] . Rare instances of transmission through blood transfusion or the donated graft have been reported [26] . Determining the overall incidence of HEV-related disease following transplantation is hampered by the available diagnostic tests, none of which are Food and Drug Administration (FDA) approved. Commercial serological assays have variable test characteristics, and tests for HEV RNA detection in serum or stool samples are not routinely available [26] .
Reports of HEV infection following HSCT are limited. While an individual case of HEV reactivation following HSCT has been reported, a review of 32 anti-HEV immunoglobin Gpositive patients prior to HSCT showed no evidence of disease reactivation [30] .
Treatment of prolonged HEV viremia often involves reducing immunosuppression. Pegylated interferon administration has been shown to induce a sustained virologic response in a limited group of patients [26] . Both approaches to viral control may increase the risk of graft rejection. Ribavirin monotherapy has induced sustained virologic responses without the risk of rejection and may represent the first-line agent for treatment [26] .
There are 5 clinically relevant non-SARS human coronaviruses (HCoV): OC43; 229E; HKU1 and NL63, both identified in the last decade; and the Middle East Respiratory Syndrome HCoV (MERS-CoV), identified in 2012. A prospective study found that 41% of HCoV infections were asymptomatic and none of 22 infected allogeneic HSCT recipients developed lower-respiratorytract infection, although prolonged viral excretion is frequent [30] . Nonetheless, there have been reports of fatal HCoV infection following HSCT. There is evidence among SOT recipients that HCoV can cause severe lower-respiratory-tract infections and increase the risk of graft rejection [31, 32] . HCoV-HKU1 and NL63 do not appear to be more virulent than the previously discovered HCoV-OC43 and 229E; however, the recently identified MERS-CoV has been associated with severe pneumonia and a high mortality rate. Cases have not yet involved immunocompromised hosts.
Efforts to identify novel respiratory pathogens have led to the discovery of HBoV and KI and WU polyomaviruses (KIPyV and WUPyV) [12, [33] [34] [35] . While these viruses have been detected in patients with respiratory symptoms, evidence to support a causative role for these agents in severe disease is lacking [12, 33] . Studies evaluating HBoV as a respiratory pathogen in immunocompromised adults have detected the virus infrequently and have not documented an effect on patient outcomes [2, 12] . The establishment of HBoV as a respiratory pathogen has also been complicated by high rates of copathogen detection and HBoV detection in asymptomatic patients [2, 12] . Viral dissemination in transplant recipients occurs, with HBoV detected in blood and stool. However, patients often had HBoV detected after weeks of hospitalization, and other pathogens were also detected during these episodes [36, 37] . Some commercial platforms for multiplex detection of respiratory pathogens include HBoV. No specific antiviral treatment is available [2] .
KIPyV and WUPyV have been detected in nasopharyngeal and bronchoalveolar lavage samples from SOT and HSCT recipients [33] [34] [35] . Detection of these viruses has been associated with sputum production and wheezing following HSCT [34] . However, similar to HBoV, these viruses are often codetected with other pathogens, and they have not been associated with severe respiratory tract disease or mortality [34, 35] .
Immunocompromised hosts are more susceptible to complications of influenza; however, it is not clear that emerging strains will necessarily cause more severe disease. One case series of 237 SOT patients with H1N1 influenza showed that 16% required ICU admission and 4% died [38] . A series among HSCT recipients showed similar findings [39] . In a comparison of outcomes in kidney transplants and immunocompetent patients with H1N1, there were no differences in morbidity or mortality [40] . To date, no cases of H5N1 or H7N9 influenza have been reported in transplant recipients.
The recent rise in measles incidence brings it into consideration here. The most significant manifestation may be subacute measles encephalitis (SME), though severe cases of pneumonia have been documented [2] . SME has developed in renal transplant recipients and a single HSCT recipient. Patients may present with a measlescompatible illness, which improves. They develop altered mental status and seizures 2-4 weeks later; fever is uncommon. The first imaging changes are seen by magnetic resonance imaging, and diagnosis is confirmed by immunoglobin M (IgM) seroconversion or RT-PCR. The clinical course is one of deteriorating mental status and treatment-refractory seizures [2] . Four of 6 transplant cases of SME have died. The 2 survivors both had significant neurological deficits [41] . The incidence of measles in transplant recipients, as well as the proportion with severe disease, is unclear. Two series identified 2 cases of interstitial pneumonia (1 fatal) among 24 HSCT recipients diagnosed with measles, though methodological limitations existed in both studies [2] .
Dengue virus (DENV) is the most common vector-borne viral disease worldwide and has been detected in an increasing number of countries over the last 40 years. In 2 case series involving 33 renal transplant recipients, only a single case of severe dengue developed, with no fatalities or loss of graft function [42, 43] . Severe cases of dengue, including 4 deaths, have been reported in renal transplant recipients along with fatal cases in a liver transplant recipient and an HSCT recipient [2] . In patients who died, disease typically developed within the first month post transplant. Human-to-human transmission of DENV as a result of SOT or HSCT has been postulated, and transfusion-related DENV infections have been reported [44] . FDA-approved diagnostics include tests for IgM detection and a Centers for Disease Control and Prevention-developed RT-PCR; management consists of supportive care.
Seventeen cases of rabies have been reported in transplant recipients, and to date, all have been transmitted through the transplanted tissue or organ [2, 45] . Nine cases followed corneal transplantation, including 8 deaths [2] . The sole survivor, reported in 1981, began postexposure prophylaxis (PEP) on postoperative day 1 [46] . Two clusters (Texas, 2004, and Germany, 2005) , totaling 7 rabies cases, have occurred following SOT [47, 48] . These cases followed the transplantation of liver, lung, kidney, kidney-pancreas, and iliac artery grafts. Patients typically developed encephalitis between 30 and 60 days post transplant, and all symptomatic patients died [48] . Patients in Germany received PEP and antiviral treatment, though not until postoperative day 45 [47] . The liver recipient in this cluster had been previously vaccinated and never developed disease [2] . Both donors were later determined to have rabies exposures (bat and dog bites, respectively) [47, 48] . A recent report (Maryland) documented a fatal case of rabies developing a year after kidney transplant. Transmission of raccoon-variant rabies through the donated graft was confirmed. Three other graft recipients from the same donor are alive, though full details are not available [45] .
The management of rabies focuses on prevention with vaccination in high-risk patients or PEP. Transplant recipients who receive PEP can mount adequate responses (antibody titers of 0.5 international units/mL), though titers are lower than in immunocompetent patients [2] . Based on the experience of the German liver transplant recipient, rabies vaccination may remain effective even after transplantation.
Cases of lymphocytic choriomeningitis virus (LCMV) transmitted through organ transplantation (4 clusters, including 14 cases and 11 deaths) document the ability of this pathogen to cause severe disease in the immunocompromised host [10, [49] [50] [51] . Another cluster involved the transmission of a related arenavirus in Australia, with similar outcomes (1 liver and 2 kidney recipients; 3 deaths) [10] . As with rabies infections post transplant, all cases resulted from transmission through organ transplantation [10, [49] [50] [51] . At this time, cases have not been described in the HSCT population.
The 4 case clusters of LCMV infection occurred in the United States and involved kidney, liver, and lung transplants [49] [50] [51] . Symptoms developed between 2 and 23 days post transplant and included fever, abdominal pain, nausea, diarrhea, and altered mental status. Patients often developed a peri-incisional rash and tenderness. CSF findings included elevated protein (often marked), normal to low glucose, and a mild pleocytosis [49] [50] [51] [52] . Three patients survived LCMV infection following SOT, 2 kidney recipients and a liver recipient. Ribavirin has been employed in some cases, though the benefit remains unclear [2] . Three corneal transplant recipients were potentially exposed to LCMV, though none of them developed symptoms or seroconverted [2] .
Contact investigation revealed exposure to rodents or positive testing for LCMV in 3 donors [49] [50] [51] . Investigation into the fourth donor revealed no exposure, and all testing performed on remaining tissues was negative [50] . It has been advised that immunocompromised patients avoid contact with rodents, including pets, though this was not the mode of LCMV acquisition in these outbreaks [53] .
For the majority of viral infections discussed here, data are insufficient to determine the true incidence of disease in transplant recipients. Measles, mumps, and yellow fever are vaccinepreventable illnesses, though these vaccines are live-attenuated and not recommended following transplantation. Also, antibody response to vaccines is less than in immunocompetent patients. Donor-transmitted rabies carries a dire prognosis, and though limited data exist, the use of PEP in transplant recipients appears safe.
Given their apparent rarity, screening for many of these diseases in organ donors is not recommended. The examples of HTLV-1 (discussed earlier) and LCMV are illustrative of some of the difficulties involved with donor screening. In the outbreak investigations for LCMV, only 1 of 4 donors had detectable antibodies. Indeed, RT-PCR from multiple samples failed to detect LCMV from 1 donor, and yielded a positive result in only a single lymph node in another [49] [50] [51] . It seems prudent to obtain a comprehensive history of potential organ donors, though it remains unclear how certain findings, such as rodent ownership, should affect one's donor status. Reporting rare infections in transplant recipients will help to identify agents for which more research is needed and screening may be warranted. However, it is likely that these infections are underdiagnosed as symptoms may be attributed to more common, and potentially coincident, posttransplant infections.
|
S evere acute respiratory syndrome (SARS) is an emerging infectious disease that was first reported in Guangdong Province in southern China in November 2002 and subsequently caused outbreaks in Singapore, Hong Kong, Southeast Asia, and Canada. 1, 2 The outbreaks not only resulted in many fatalities and hospitalizations but also had a huge economic impact on the nations involved, because of decreased revenue from tourism and industry. Although no other outbreaks have been reported since then, continued vigilance is essential in the event of future epidemics. To our knowledge, this is the first report of SARS in an occupational (naval) diver.
A 35-year-old Chinese man, working as a naval diver, presented to his primary care physician with fever 2 days after attending his mother's funeral. He first presented with symptoms on April 19, 2004, during the height of the SARS outbreak in Singapore. There was no travel history of note. His mother, who had had no previous medical problems, had collapsed suddenly and died at home after the onset of fever 3 days earlier, with no other symptoms. She had sought medical attention from her family doctor but was never formally diagnosed as having SARS. Although SARS was suspected as the cause of her fever, this was never proven; she had not been hospitalized because of the rapid progression of illness.
The patient was referred to the emergency department of a tertiary hospital and subsequently hospitalized for suspected SARS because of his complaint of fever and his positive contact history. Table I shows the case definitions defined in Singapore by the Ministry of Health during the SARS outbreak and the actions to be taken. In all, the patient's father, sister, three nieces, sister-in-law, and girlfriend were admitted as part of an infected family cluster (Fig. 1 ). The provisional diagnosis was SARS. The patient's father eventually died after 5 days in the hospital; however, other family members made uncomplicated recoveries.
The patient continued to have a fluctuating fever while he was hospitalized, and he complained of loss of appetite and loss of weight. However, he never developed shortness of breath or other respiratory symptoms. The patient was evaluated for other causes of febrile illness, including community-acquired pneumonia, malaria, dengue fever, rickettsial illness, and pulmonary tuberculosis. However, all tests were negative. The patient was treated with intravenously administered, broad-spectrum antibiotics (levofloxacin and then ceftriaxone and clarithromycin) and required oxygen supplementation for several days. However, he never required intensive care or mechanical ventilation.
The diagnosis of SARS was initially made clinically but later confirmed radiologically and serologically. On the fifth day of admission, the patient's chest X-ray showed infiltrates consistent with SARS. The SARS coronavirus serological test was negative on admission but became positive 1 week after admission. The patient showed improvement and was discharged after a 17-day stay in the hospital. After discharge, he was quarantined for 21 days according to protocol.
During review in our department after the quarantine period, the patient was deemed unfit to dive because of abnormal lung function test results (Table II) . The patient was otherwise well, and physical examination results were normal. However, the patient complained of dyspnea on exertion. Because he was a naval diver, the patient was referred for a respiratory consultation. A computed tomographic (CT) scan of the thorax 6 weeks after discharge showed no significant residual pathological condition. A high-resolution CT scan of the thorax 2 months after the initial CT scan also showed no abnormality. Results of a follow-up lung function test 5 months after discharge were normal. On the basis of all of these investigations, the patient was discharged from follow-up care by the respiratory physician. The patient gradually recovered and regained his normal fitness during the convalescent period. He passed chamber bounce dives at 10 m and 50 m and was subsequently certified fit to dive 6 months after quarantine. He is currently well and has not had problems diving since then.
Certain preventive measures to prevent the spread of SARS were taken after diagnosis of this patient. Principles of control against SARS focused on three levels of containment, i.e., (1) public health response and medical care, (2) organizational and administrative measures, and (3) social and personal adaptations. Therefore, our goals were twofold: (1) to detect SARS cases early and to break any chain of transmission and (2) to prevent the onset of infection. Public health measures included contact tracing and close monitoring of personnel who might have been exposed to the index case. Contact tracing was performed up to 1 week before admission for all individuals in proximity to the patient. Personnel who were in contact with the patient were required to monitor their temperature 3 times per day and were quarantined at home. Organizational measures included re-stricted access and contact between divers and other camp personnel, to prevent the potential spread of infection. The temperatures of all personnel entering the diving unit were checked before entry. Contact of the diving unit with other personnel in the camp was restricted. Diving unit accommodations were made out of bounds to other personnel except medical staff members. In addition, different meal times for divers and other personnel were enforced. Other measures to prevent the spread of disease included not allowing utensils in the cookhouse to be reused on the same day and having divers leave the camp at different times from other personnel. In the week following diagnosis of SARS in this patient, only essential personnel in the diving unit were required to report to work to prevent the possible spread of SARS. Personal measures included isolation of the index case after diagnosis and quarantine for 21 days after discharge from the hospital.
The etiology of SARS has been linked to a novel coronavirus, 3, 4 which was postulated to have crossed the species barrier from animal to human. This hypothesis is supported by reports that some of the early patients in Guangdong reported a history of occupational exposure to live caged animals, which were consumed as exotic "game food." The SARS coronavirus was also isolated from Himalayan wild civets. 5 Most of the cases resulted from "super-spreading events," in which some particularly infectious individuals were ultimately responsible for spreading the disease to tens of people.
The mean incubation period has been reported as 6 days, with a maximal incubation period of 14 days. 6 The primary mode of transmission is through direct or indirect contact with mucous membranes through infectious respiratory droplets and fomites. 7 Common presenting symptoms include fever, nonspecific symptoms (such as myalgia, malaise, and chest pain), anorexia, and respiratory symptoms (such as dyspnea). In early cases, the disease was misdiagnosed as a form of atypical pneumonia, with disastrous results.
Other findings reported include lymphopenia, thrombocytopenia, increased D-dimer levels, and increased activated partial thromboplastin time. In addition, deranged transaminase levels and elevated creatine kinase and lactate dehydrogenase levels were reported. 7 Chest X-ray findings, which provide clinical suspicion for SARS, include ground glass opacities, consolidation, and nodular and reticular opacities. 8 However, the standard for retrospective confirmation of infection is seroconversion in a whole-virus immunoassay (immunofluorescence assay or enzymelinked immunosorbent assay). Unfortunately, serological testing is positive only after the first week of illness. 7 Interestingly, a recent study demonstrated that polymerase chain reaction testing of tears could confirm infection in the first week of illness. 9 Age and coexisting illness have been reported as being prognostic factors for risk of death and the need for intensive care. 7 Treatment rendered has largely been supportive. Before serological diagnosis, empiric, broad-spectrum, antibiotic therapy aimed at both typical and atypical pneumonias is indicated. Although there is no definite cure for SARS, steroids and ribavirin have been used in some centers, with varying results. 7 The patient's pulmonary condition is monitored with serial chest X-rays, with ventilatory support as needed. However, prevention is paramount. Measures such as hand-washing, universal precautions, and public education are key to preventing the spread of SARS.
An infectious disease outbreak is a very real possibility in the armed forces because of the proximity of men and women living and working together for prolonged periods. In the context of a highly virulent and infectious disease such as SARS, an outbreak can lead to very high morbidity and mortality rates, as well as incapacitation of operational units. Therefore, prompt treatment and containment of spread are essential preventive measures. Other important measures include contact tracing, home quarantine, and widespread use of universal precautions. Lastly, a high index of suspicion must be maintained. For this patient, aggressive containment measures successfully prevented the spread of SARS to other unit personnel.
Because this patient was a naval diver, his convalescence and recertification for fitness to dive were particularly prolonged. Fortunately, he made a full uncomplicated recovery. Much of this could have been attributable to his young age and aboveaverage fitness as a diver, as well as the lack of other comorbidities. Because there are no reports defining the effect of SARS on fitness to dive, we postulate two categories of effect.
First, there is the early recovery period. Immediate problems during diving would include increased risk of hypoxia and hypercapnia because of inadequate lung ventilation in relation to exercise level. In addition, there would be increased breathing resistance attributable to weak respiratory muscles being unable to overcome the normal flow resistance of the breathing apparatus. This would lead to impaired ability to respond to nonrespiratory problems during diving, with a corresponding decreased ability to compensate because of poor residual lung function.
Second, there is the late recovery period. Long-term problems would include increased risk of diving complications such as pulmonary barotrauma, resulting from decreased lung compliance and gas trapping as a result of fibrosis and scarring within the lung parenchyma, which are known complications of SARS. On the basis of our experience, we suggest that CT scans of the thorax and lung function tests should play a vital role in the assessment of such patients during the convalescent period before certification of fitness to dive. Our experience has also shown that an individual is more susceptible to respiratory infections after contracting SARS. All of these potential problems would necessitate careful follow-up monitoring of service personnel after infection by SARS or future emerging respiratory pathogens before certification of fitness for diving.
The management of SARS in the armed forces is unique because of the increased potential for an outbreak. Preventive measures are essential to prevent the spread of the disease. Divers affected by SARS must be thoroughly investigated before they are allowed to resume their vocation, because of the potential problems posed after infection by SARS.
|
The rapid progression of the COVID-19 pandemic has created significant challenges for the public as well as healthcare professionals around the world. Knowledge regarding virus incubation, transmission and shedding is crucial for the reduction of new cases and protection of healthcare professionals. Guidance regarding isolation and protective equipment has changed as evidence has increased and developed.
The high incidence of cough and fever in COVID-19 is well established [1] . Gastrointestinal symptoms are also well documented suggesting a potential faeco-oral transmission route [2] . Discharge guidelines for hospitals for declaring a COVID-19 patient recovered in the UK are largely based on time from either symptom onset or positive test depending on the severity of illness and the discharge destination [3] .The European Centre for Disease Prevention and Control, on the other hand, has advocated the need for continued self-isolation and hand hygiene measures even 14 days post-discharge based on prolonged viral shedding in faeces and respiratory samples [4] . This evidence may influence the recommended duration of self-isolation, home sanitation practices during isolation and after discharge and the use of protective equipment for procedures involving the gastrointestinal tract. Evidence-based recommendations for specialities such as gastroenterology, gastrointestinal endoscopy and gastrointestinal surgery are required where there may be an exposure risk to virus shed in faeces. Despite viral RNA being detected in the air or other surface samples like toilets, it is still unclear whether it is viable to transmit infection through this route [5] .
The primary aim of this review is to assess the incidence and timing of positive faecal samples for SARS-CoV-2 in relation to the clinical course of patients with COVID-19.
Our secondary aims are to establish the incidence of patients with positive faecal samples after negative respiratory swabs and any evidence to suggest faecal virus transmitted infection.
Reports of cases or studies of COVID-19 patients with evidence of the virus in faecal samples were systematically identified and full text articles were reviewed for data extraction.
A comprehensive search was undertaken as per the search strategy outlined below for literature that included SARS-CoV-2 virus testing of faeces. MEDLINE was searched to find articles published until 3 April 2020. The defined search terms were created after collaboration between the authors experienced in gastroenterology, colorectal surgery and systematic review. Search terms reflected the aim to identify studies with evidence of faecal COVID-19 and included 'clinical', 'faeces', 'gastrointestinal secretions', 'stool', 'COVID-19', 'SARS-CoV-2' and '2019-nCoV'. Additional manual searches to identify the most recent evidence were performed in the American Journal of Gastroenterology, Gastroenterology, Gut, the Lancet Gastroenterology and Hepatology, the World Health Organization (WHO) Database, the Centre for Evidence-Based Medicine, the New England Journal of Medicine and the National Institute for Health and Care Excellence. COVID-19 preprints published until 10 April 2020 on medRxiv and bioRxiv and an independent search on social media (Twitter) by the authors (SS, SD) added more articles. The search strategy used for social media and a brief description of the WHO and other databases are provided in Appendix S1.
Articles describing COVID-19 patients who had faecal or stool specimens tested for the virus were included.
Considering the knowledge gaps existing for COVID-19 all articles were considered regardless of the number, age or gender of patients or the country of publication. Animal-based studies or articles without an available full text were excluded. Foreign language articles were considered but excluded unless the necessary language expertise was available within the research group.
Articles were sorted alphabetically by author name and divided between two reviewers (SG and JP). Abstracts were reviewed and classified by the same two authors through the Rayyan Web Application [6] to identify those for full text review. The same process was used for full text articles and these data were managed through EndNote (EndNote X9.3.1 license provided by Cardiff University). Articles were then discussed between the same reviewers to identify the final selection of full text articles. Any conflicts were solved by the supervising author if necessary. Reference lists and review articles were cross-referenced to identify any further original studies. All articles were categorized and described in a PRISMA flow chart.
The final data extraction was also carried by the two reviewers (JP and SG) and managed through Microsoft Excel files. The data parameters extracted from the studies are shown in Table 1 . The final data were verified by the two reviewers (JP and SG) with conflict resolution as described previously if necessary.
MEDLINE searches identified 565 articles and 194 were found through other databases. An overview of the selection process is shown in the PRISMA chart in Table 1 Data parameters for extraction.
Study reference 2.
Country of publication 3.
Number and type of patients in the study 4.
Type of sample taken (faecal sample, anal swab, RT-PCR, culture) 5.
Number of patients having faecal samples tested and number of positive samples 6.
Timing of positive faecal swab after symptom onset 7.
Duration of positive faecal specimen after negative nasopharyngeal swab 8.
Any Table 2 .
Most studies were from China (n = 20) with two from the USA and one each from Italy, Korea, Vietnam and France. The number of participants recruited in the studies ranged from 1 to 206 with ages ranging from 3 months to 87 years. Sample collection consisted of faecal samples or anal or rectal swabs. Quantitative reverse transcription polymerase chain reaction (RT-PCR) was the test performed on all samples to detect viral RNA.
The indication for faecal testing was not specified in most studies. In some the test was done in The predominant symptoms of presentation in the studies were persistent cough, fever and breathlessness with fewer patients reporting diarrhoea or vomiting. All studies had information regarding our primary aim of reporting faecal samples for the virus in those with COVID-19. Of these, 16 [7, 10, 11, [14] [15] [16] [17] [18] [19] 23, 24, [26] [27] [28] [29] [30] provided information on the duration of these tests after symptom onset and evidence of positive faecal samples after symptom recovery, discharge from the hospital or negative nasopharyngeal RT-PCR. The data extraction is summarized in Tables 3 and 4 which are divided based on the number of patients tested for faecal RT-PCR in the study (≤ 10 and > 10 respectively) and the detailed combined table is attached as supplementary results (Table S1 ). A total of 824 patients were included across the studies and 540 were tested for faecal viral RNA . Positive faecal RT-PCR tests occurred in 291 (53.9%). The timing of the first positive sample was available in 21 studies and varied from day 0 of symptom onset to day 17. Late positive tests do not necessarily equate to absence of the virus earlier in the illness but may reflect the heterogeneity in testing patterns amongst the studies. First stool samples were often reported late after hospital admission [11] or even after discharge [28] while some were analysed from day 1 of hospitalization or symptom onset [19, 20, 27, 29, 32] . There is a similar discrepancy in follow-up testing. Some tested until samples were found to be negative [17] while others did not [18, 29] .
Of 199 patients who tested positive for faecal viral RNA and who were followed up with stool testing, 125 (62.8%) showed persistent shedding of virus in the stool samples after a negative nasopharyngeal swab while in the individual studies it ranged from 23.3% to 100%. The duration for faecal shedding of viral RNA after clearance of respiratory samples ranged from 1 to 33 days and in one patient up to 47 days from symptom onset [26] .
None of the studies was designed to detect live virus in the faeces except for the study by Wang et al. [25] . Of 153 stool specimens tested in this study, 44 were PCR positive and, of four specimens cultured, live virus was detected in two [25] .
This rapid review demonstrates a high incidence and persistence of positive faecal RT-PCR tests for SARS-CoV-2 after negative nasopharyngeal swabs in patients with COVID-19. This may have important implications regarding measures to prevent the spread of disease, precautions recommended for the public and protective equipment for health professionals performing interventions involving the gastrointestinal tract.
A Chinese review performed by Tian et al. [33] summarized evidence on the importance of identifying gastrointestinal symptoms in addition to the respiratory symptoms of patients with COVID-19. Despite persistent shedding of SARS-CoV-2 virus in faeces there seems to be no correlation with the presence or severity of gastrointestinal symptoms based on the limited data available. Our review adds to this evidence from China and describes the plausibility of faeco-oral transmission.
Despite this review demonstrating a high incidence of positive tests for virus in the faeces, the absence of evidence to confirm infectivity from this must be emphasized. In order to adequately confirm this, good quality evidence is required to demonstrate infectious virus in faeces and its risk of transmitting disease between individuals. These data may then enable the development of reliable guidelines and recommendations. However, given the rapid development of the pandemic, this will take time and reviews such as this may help guide focused and valuable research questions for the future. The findings of our review provide a synopsis of the best available evidence regarding SARS-CoV-2 in the faeces at the current time.
Evidence regarding other coronaviruses may be helpful in this context. Similar patterns of virus isolation from stool and faeco-oral transmission were observed for other coronaviruses including SARS-CoV-1 [34] . Bio-aerosol generation of viral particles as a result of toilet flushing, the impact of disinfection on this [35, 36] and the persistence of coronaviruses on surfaces has been studied before [37] . Other indirect evidence of microbial exposure and contamination of the operator's face during endoscopy [38] and laboratory evidence of SARS-CoV-2 infection of the gastrointestinal tract and Table 4 Overview of data extracted from studies included in the review with > 10 patients tested for faecal virus [11, 12, 17, 20, [25] [26] [27] [30] [31] [32] .
Patients with positive faecal RT-PCR [39, 40] add to the evidence for plausibility of transmission.
The risk to healthcare professionals from patient exposure is well known, specifically in high aerosol generating procedures. Professional societies and investigator groups from countries with experience of managing COVID-19 in the context of gastrointestinal interventions [41, 42] highlight the risk to individuals in endoscopy departments and the need for necessary precautions including negative pressure rooms and personal protective equipment for both upper and lower gastrointestinal procedures. This review supports the importance of these measures given a high prevalence and persistence of SARS-CoV-2 virus in faeces. Isolation of live virus is confirmed only by one study [25] and the proportion of cases that might be transmitted by this route is unclear due to the heterogeneity in case selection and lack of standardization of study designs and protocols. Environments such as care homes may be particularly vulnerable to transmission of infection by this route and recommendations must take into account this evidence to ensure the protection of health and social care providers and the general public in the meantime. Application of these data to the population may be helpful in guiding the recommendations for isolation periods to reduce transmission rates.
Despite finding a high incidence of positive faecal samples for SARS-CoV-2 in the included studies, our review cannot confirm the true population prevalence of positive faecal samples or the rate of false negatives. This is due to the significant variability in study design which is an inherent problem with COVID-19 research at present. This heterogeneity was not formally assessed due to it being a rapid review but can be clearly identified on inspection of the study designs and outcomes. The variability in patient numbers and characteristics, sample timing, sample nature (faecal samples vs anal or faecal swabs) and follow-up testing should be considered when interpreting the reliability of the results. If other studies confirm viable virus in stool, then methods of culture also need to be described and standardized for comparison and replication in other populations. The majority of the included studies are small, heterogeneous, retrospective and often did not assess viral shedding in the faeces as their primary aim. At present, however, this is the only evidence available. There were two foreign language articles excluded due to lack of translation resources. The preprints are not peer reviewed and therefore should be treated with caution.
The duration of viral shedding in the faeces is mostly reported from 1 to 33 days after a negative nasopharyngeal swab but can continue for up to 47 days after onset of symptoms in patients with COVID-19. These positive samples can occur after negative nasopharyngeal swabs or resolution of patient symptoms. Isolation of live virus in stool specimens of two cases in a single study supports the possibility of faeco-oral transmission. Further research is needed to prove whether this viral shedding in stool results in a significant proportion of case transmissions in the community as well as within care institutions and secondary care. Until further evidence is generated appropriate precautions should be recommended for the protection of healthcare workers and patients.
Implications for the public 1 In addition to strict adherence to hand washing recommendations, home toilet sanitary and disinfection precautions should be taken in the case of isolation or contact with a symptomatic COVID-19 case with or without gastrointestinal symptoms. This statement is based on limited evidence of possible viable faecal virus excretion. 2 These precautions may need to continue for longer than the period of symptoms and the current recommendations for isolation after symptoms cease. This statement is based on limited evidence of the duration after the onset of symptoms that an RT-PCR stool test might still be positive.
1 Professional bodies' recommendations on protective equipment, endoscopic and surgical procedures for COVID-19 patients should be followed [43] [44] [45] [46] . Healthcare teams managing patients with gastrointestinal symptoms may need to consider the possibility of COVID-19 coexisting with or worsening symptoms of underlying conditions such as inflammatory bowel disease [47] .
1 Future studies on viral shedding and infectivity of SARS-CoV-2 should consider standardization of sampling methods in terms of the timing and the type of sample collection, with appropriate precautions for laboratory staff handling these samples until the situation is clearer. 2 Study designs may wish to consider repeat and parallel sampling with nasopharyngeal swabs at defined time points. This may be correlated with symptoms and serology to clarify the effect of neutralizing antibodies and viable virus excretion in the stool. 3 Study designs may benefit from testing stool samples from comparable groups. This could include symptomatic, asymptomatic or recovered individuals in and out of family clusters and with or without gastrointestinal symptoms. This may improve our understanding of clinical and public health implications and potential targets for intervention in these settings.
We would like to thank Dr SA Roberts for critical review and comments on the manuscript.
None declared.
SG, JP contributed to conception and design of the project, data collection, analysis and interpretation, drafting of the article, revisions and final approval. SS contributed to data collection, review of the manuscript and final approval of the article. JU contributed to data collection, interpretation, review of the manuscript and final approval of the article. SD contributed to the conception and design of the project, data collection, revisions and final approval of the article and was overall supervisor.
|
In this issue of Acta Haematologica Cheung et al. [1] review what they term a haematologist's perspective on the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) pandemic and on coronavirus disease 2019 (COVID-19). Their work is encyclopaedic with 128 references covering diverse haematological parameters including lymphocytes, T-and B-cell subsets, platelets, haemoglobin and coagulation parameters. They present data on hospital admission values and/or risk of death from COVID-19. They also discuss the impact of the SARS-CoV-2 pandemic on the blood product supply and on therapy of haematological cancers. Finally they discuss consensus guidelines from several societies.
Their review is timely. But is it a useful guide or a doorstop? Both. Readers will be able to quickly access relevant publications on these topics from the tables and references. On the downside there is little critical discussion or synthesis. For example, how common is lymphopenia and is it an accurate prognostic covariate? In Table 1 the authors list 26 studies including by my colleagues and me. Incidence rates of lymphocytopenia on hospital admission range from 26 to 83% (the text incorrectly says 26-80%). The problem is sample sizes range from 28 to 5,700 subjects. Should we give equal weight to all studies in the table? Probably not. Fifteen of the 26 studies have < 100 subjects each, whereas 2 studies each have > 1,000 subjects and 1 study 5,700 subjects, more subjects than the 20 oth-ers combined. Normally, authors provide a statistical analysis of this type of heterogeneous data, such as a weighted mean or median and confidence interval or interquartile range (IQR) followed by their conclusion. So, at the end of the day the reader is left wondering what the true incidence of lymphopenia on hospital admission is. My money is on results of the study of 5,700 subjects with a median of 63% (IQR 52, 75%), so about one-half of people. This lack of critical analyses of heterogeneous data is true of the other endpoints studied. Is this a killer? No, but critical statistical analyses would have been welcome. The issue of lymphocytopenia is especially important because it may correlate with survival. In a study of 1,571 subjects with COVID-19 we found a significant difference between 1,440 survivors (median 1.2 × 10E+9/L [IQR 0.9, 1.7 × 10E+9/L]) and 131 subjects who died (0.5 × 10E+9/L [0.4, 0.8 × 10E+9/L]; p < 0.001) [2] . Another example are data on admission prothrombin time. In the 8 studies cited in Table 1 , values range from 12.4 s in a study of 94 subjects to 13.4 s in a study of 1,099 subjects. Again, this is important. In another study my colleagues and I found that a prothrombin time > 12 s was also correlated with risk of death [3] . These and other data are summarized in a perspective in Acta Haematologica [4] .
Another gift of the authors are Tables 2-4 the International Society of Thrombosis and Haemostasis, American Society of Hematology (ASH), European Hematology Association (EHA) and International Myeloma. What the authors fail to warn us of is that although these guidelines and recommendations seem sensible, none are evidence based. Readers will recall many things in medicine which seemed to make sense but were later proved wrong or even harmful. For example, it made sense to think gastric ulcers are caused by stress, but unfortunately in stepped Helicobacter pylori and a Nobel prize for rejecting seeming common sense. People estimate one-half of what we teach in medical school is wrong. The problem is we don't know which half. Other famous medical reversals include radical mastectomy for breast cancer, intracerebral stents for transient ischaemic attacks, coronary artery stents for stable angina and vertebroplasty. In haematology readers will recall misguided opinions regarding autotransplants for advanced breast cancer. For those who think medical reversals are uncommon I recommend Ending Medical Reversal: Improving Outcomes, Saving Lives by Prasad and Cifu [5] . In one analysis of 363 medical procedures published as effective in The New England Journal of Medicine 2001-2010, 40% were later proved ineffective or harmful and it was impossible to confirm the benefit of a further 20% [5] . In another study of 3,000 medical interventions considered standard of care in The BMJ, 65% were determined to be unproved, ineffective or harmful [6] . So much for seeming common-sense recommendations that are not evidence based. Montaigne summed it up nicely: "Nothing is so firmly believed as that which man knoweth least."
Lest I be accused of publication bias, might I direct readers to an experiment where I fed large numbers of genetically identical mice shredded SARS-CoV-2-or CO-VID-19-related consensus guidelines or sheets of blank paper with their laboratory chow for 1 week. I then let them attack a round of Stilton cheese onto which I had written "kill COVID-19" [7] . There was no difference in cheese consumption between the cohorts. So much for consensus guidelines, at least for mice. As Abba Eban noted: "Consensus means that lots of people say collectively what nobody believes individually." Back to the article by Cheung et al. [1] . As I indicated, it is a useful reference document, but the reader will have to come to his/her conclusions about what the data mean. As for consensus recommendations: caveat emptor.
|
Background 3,181,642 cases and 224,301 deaths in 212 regions of the world-this is the status of COVID-19 (Coronavirus Disease 2019) pandemic as of May 1, 2020 [1] . This pandemic has managed to overwhelm the health care system of the most advanced countries in the world. As the whole of the medical fraternity stands robed as health care professionals to fight against COVID-19, specialty emergencies like trauma continue to pester the already overburdened health care community [2] . This situation calls for the need for a pandemic response protocol (PREP) in each specialty that helps the doctors to manage specialty emer-gencies without chaos and at the same time allowing them to play their part in pandemic management.
The PREP should allow a non-frontline pandemic specialist like an orthopaedician to be alert and trained to evolve as a frontline health care provider, as and when the situation demands, on the lines of development of a pandemic. We intend to formulate a response protocol based on the current guidelines from various national orthopaedic associations [3] [4] [5] and international orthopaedic organizations [6, 7] along with available relevant COVID literature that will help us to be well prepared for the upcoming pandemics or the second wave of COVID-19 if at all one comes.
Before any protocol that has to be designed for a pandemic situation, there are certain factors to be considered in their design. They include:
The aim of PREP is to tide over the pandemic crisis as an efficient health care workforce in the most effective way.
1. Attend to orthopaedic emergencies. 2. Patient and healthcare worker protection. 3 . Conserve and educate the orthopedic workforce on pandemic control. 4. Rise to the occasion as a frontline pandemic control team when needed.
The World Health Organization (WHO) divided the development of pandemic into six phases in 1999. This was further revised in 2005 and 2009 after the H1N1 flu pandemic [8] ( Table 1 ). The phases were devised in such a way that they apply to the whole world providing a global framework to help countries in pandemic preparedness and response planning [8] . A specialty PREP should closely evolve in stages with WHO phases of development of pandemic to achieve its aim as shown in Table 1 . The PREP activation should begin ideally when a country enters phase 2 of pandemic and continue until the beginning of the post-pandemic phase. The various stages of orthopaedic PREP are summarized in Fig. 1 .
This stage begins with phase 2 of the pandemic. Awareness about the pandemic is important and the national orthopedic forum should take up this responsibility and be in close lines with the WHO action plan. It should impart the information including signs and symptoms of possible pandemic and personal protective measures needed to combat the pandemic through scientific media to its fellow members. This prepares an orthopaedic surgeon to be socially more responsible and pick up the initial cases that he may come across. This also alerts him to be more responsive to the next stage of pandemic. Institutes should take adequate steps to ensure a continuous and adequate supply of personal protective equipment (PPE) for pandemic response. They should ensure the availability of adequate isolation beds and wards.
This stage begins with phase 3 of the pandemic. At this stage, essentially, the orthopaedic surgeons would have to become a part of the pandemic response task force which includes the entire health care workforce. Our primary aim must be the prevention of human-tohuman transmission thereby averting impending pandemic [9] while we balance orthopaedic care to the general public.
At this stage, the orthopaedic workload is not expected to decrease yet. However, as a containment measure, it is essential to split up the orthopaedic workforce into two dedicated teams including one inpatient and one outpatient team which alternate every week without a reduction in workforce. These teams do not come in contact with each other.
The doctors in the outpatient department are those that are about to pick up initial positive or suspect cases of the ongoing pandemic. So, they must hold a high vigil and at the same time be armed with adequate PPE. It is essential to have an institute specific history form to be filled in by all outpatients that warn the hospital of those with relevant travel history or symptoms. Symptomatic cases should be categorized as either suspect or non-suspect cases according to national or international guidelines [10, 11] and should be treated accordingly. The use of removable casts and splints should be maximized to reduce follow-up requirements.
Patients requiring emergency/early orthopaedic intervention must be attended to immediately as in normal scenarios. These include patients with orthopaedic oncology, trauma, and non-traumatic paediatric and spine conditions requiring immediate attention [12] . Day care procedures can still be carried out as they do not affect the inpatient beds available for pandemic preparedness. Day care procedures include arthroscopy surgery and simple procedures like implant removals which require < 23 h of hospital care [12] .
Elective procedures in high-risk patients such as joint replacement procedures and spinal decompression procedures in the elderly should be avoided completely. Adequate analgesics including intra-articular analgesic injections or nerve root blocks can be provided to prevent their frequent hospital visits. Documented evidence shows elderly patients with comorbid conditions and low physiologic reserves succumb to pandemic diseases [13] [14] [15] ; hence, they should be advised of the potential threat and their surgery should be deferred.
Orthopaedic trauma and oncological patients taken up for inpatient surgical care must be provided with prompt consultant driven surgical and anaesthetic care whenever possible which might aid in reducing the period of hospital stay. Principles of early total care in trauma patients whenever possible are to be followed. Rehabilitative services may be provided on a home-based approach through various online tools to aid in expedited discharges in reasonably quick time to have the manpower and beds available for the ongoing pandemic. We should avoid advising non-essential follow-ups. Local or telemedicine follow-up of post-operative patients, home visits, and local refilling of essential medications should be implemented where ever possible. Inpatient visitor's records should be maintained. Only one visitor for a patient should be allowed. Visitors should also be asked to fill the history form and must be screened for symptoms.
The institutes and departments must strictly follow the directives of the health ministry of the country and the WHO while handling the international patients.
The department must set standard protocols to handle interdepartmental referrals. Each referral must be categorized into a suspect and non-suspect referral and handled accordingly by a dedicated team.
The orthopaedic training for registrars and post-graduates in the department during this phase must essentially include daily updated knowledge on the nature of pandemic, the rationale behind PPE, and method of robing and disrobing PPE among others. This can be through video lectures or interdepartmental lectures if manpower is available.
This stage begins with phase 4 of the pandemic. Phase 4 of the pandemic with documented human-to-human transmission needs extreme containment measures [9] .
The department must be split into four groups. Specialty specific workload is expected to be reduced due to national/ regional lockdown.
1. Management of outpatient department 2. Management of theaters and inpatients 3. Training in intensive care/pandemic preparedness 4. Shifts in screening pandemics
The institute must cancel all the outpatient appointments with clear notice to the patient through their phone numbers. Only non-postponable paediatric outpatient procedures like casting for congenital deformity like clubfoot, vertical talus, and hip dysplasia can be addressed to avoid complex surgery with unsuccessful outcomes in the future. Other inevitable services like routine chemotherapy regimens can be administered at the nearest possible primary health centres under supervision. If such measures are not feasible, they can be given as a day care service in the hospital. Telemedicine must be encouraged whenever possible. Follow-up X-rays if needed can be advised to be taken near the locality of the patient and emailed to the institute. The institute must have a standard set of physiotherapy videos that can be advised to the patient through telemedicine.
Only emergency trauma surgeries and non-postponable spine and paediatric surgery like slipped capital femoral epiphyses (SCFE) fixation, growth rod lengthening, and growth modulation procedures should be taken up with as minimal hospital stay as possible. History forms should be maintained. Each patient should be categorized as suspect/non-suspect before being admitted.
Operating rooms with a negative-pressure environment, frequent air exchange, and separate access are needed. When the airborne spread of the pandemic is a concern during aerosolgenerating procedures such as drilling, intubation, or extubation, it is important to have proper PPE and protocols in place to limit the spread of infection in this setting [16] . Ante-rooms in which to put on and remove protective equipment should be available or even constructed adjacent to the operating room [17] . The operating room must work with minimal staff pattern. Adequate time between procedures is needed for decontamination. A senior-most surgeon on call should be advised to perform all the procedures. Surgical time and blood loss must be kept to minimal as far as possible. Regional and local anaesthesia is preferred to general anaesthesia when the pandemic is suspected to be airborne [11] . With the decrease in need of residents and fellows in operation theaters as the senior consultants perform most of the procedures, we should wisely engage the seemingly excess workforce. More number of them should be deployed in emergency rooms and casualties to relieve emergency physicians for a more responsible role in handling pandemic emergencies. Routine anaesthesia and intensive care postings of the residents should be preponed in the initial phase of the pandemic so that they get trained when the need arises and at the same complete their curriculum.
The psychological trauma to the doctors during the time of pandemic is multifactorial including the inability to fulfill their standards of patient care, separation from family, inability to fulfill the family needs, isolated working pattern, the spread of infection to colleagues, fear of the spread of infection, and long working hours [18] . The institute must put in place adequate measures to avert psychological trauma due to all of these issues.
This is the single most important part of PREP. A group of surgeons must take part in formal training in non-surgical skills like handling ICU equipment. They should have first-hand knowledge of screening and treatment of patients of the pandemic. They should be prepared to step up as the frontline pandemic response team when the situation demands [19] .
This stage begins with phase 5 of the pandemic. The surgical group should be split into two teams: one to handle the orthopaedic workload and the other for pandemic response. The fluency of specialty people to get transformed into the frontline pandemic workforce will have positive psychological and health impacts in pandemic mitigation [19] . This will satisfy the ultimate aim of our response protocol.
Pandemic orthopaedic case handling protocol [3-5, 20, 21] Emergencies 1. Polytrauma, pelvi-acetabular fractures with major haemorrhage, compartment syndrome, and exsanguinating injuries should be prioritized for early surgical management. 2. Dislocations should be attended immediately. 3. Septic arthritis and prosthetic joint infections should be attended immediately. 4. Localized abscesses without signs of sepsis should be drained in the emergency department.
should be managed by suppressive antibiotic therapy orally where ever possible.
1. Complex fractures fixations are planned in such a way to minimize the hospital stay and if a staged approach is used, the aim is to discharge and re-admit the patient whenever possible. Pre-pandemic and post-pandemic stages of PREP Stage 0: pre-pandemic stage
The twentienth century saw 3 influenza pandemics. Thus, the pre-pandemic stage typically lasts for years together. This phase should be used for research, development, and testing of pandemic response protocols which should also follow the lines of research and development of WHO pandemic response. Annual pandemic response mock drills should be a routine in every country.
This stage corresponds to the post-peak phase and possible new wave phase of a pandemic by the WHO [8] . In case of favourable regional scenario, stepping down of pandemic response to Stage 3 and then to Stage 2 should be carried out. Resting a part of the exhausted workforce is essential. Quasi emergency procedures that have been postponed because of pandemic response like a highly painful disc herniation on analgesics should be timed during this period. Every patient posted for surgery must be screened for the pandemic infection and necessary consent must be obtained for change in treatment options depending upon the results of the screening. Once the elective surgeries are resumed, the hospital must follow a priority list of waitlisted patients made based on the severity of the condition and age of the patient. We should, however, be prepared to step up the pandemic response if the second wave of pandemic occurs.
Unfolding its deadliness in more than 200 regions of the world, this COVID-19 pandemic has made it clear the importance of pandemic preparedness. One good example is the Disease Outbreak Response System Condition (DORSCON) of Singapore [12, 22] . This was developed after the 2003 Severe Acute Respiratory Syndrome (SARS) infection in Singapore. This system is a nation-wide color-coded guidance protocol to alert the health care system depending on the level of threat by the epidemic. Effective implementation of DORSON resulted in effective COVID-19 containment in Singapore [23] . Though WHO and most of the national health care agencies have their pandemic control strategies, they will not invoke the desired effects at the grass-root level unless they are supported coherently by departmental pandemic response systems. National associations and international organizations drawing new guidelines every fortnight in the middle of a pandemic will only create confusion among their members. Hence, a pre-defined PREP model prevents chaos, instills order, and provides moral support in the to-be stressed up health care system. We hope such a PREP model will evoke a better coherent response against the pandemic.
This pandemic response protocol is just a pilot model in this direction. The inclusion of pandemic response knowledge in the academic training of all specialties is a must. Additive knowledge of centuries is essential to deal with these oncein-lifetime catastrophes. So, literature must be analyzed, review articles must be published, and evidence-based changes should be made to the protocol in line with international pandemic preparedness protocols. A basic infrastructure that every institute or department should include must be standardized and improved.
This model applies to a full-fledged orthopaedic department in a multispecialty corporate or tertiary care teaching hospital. This does not apply to a dedicated orthopaedic specialty hospital or small-town multispecialty center where the scenario and resources are quite different. The efficiency of disaster management in any region is dependent on the availability of resources. It falls on the shoulders of the administration and national governing bodies to make sure adequate resources are available to meet the demands. Protocols like PREP makes difference in their ability to help us utilize the available limited resources to the maximum to mitigate the pandemic. Some customization of the protocol for every pandemic is going to be essential and unavoidable.
Integrated pragmatic approach under the WHO is essential in containing pandemics as they need international cooperation at various levels starting from knowledge sharing to monetary support [24] . PREP as the one described above in line with the WHO action plan will be an essential minimum response in a non-frontline pandemic response specialty like orthopaedics to combat and curtail the effects of a pandemic.
Compliance with ethical standards
|
Multisystem inflammatory syndrome in children (MIS-C) is a newly described condition associated with severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) exposure that is reminiscent of both Kawasaki disease and toxic shock syndrome. A recent surge in this disease has prompted health advisories by the US Centers for Disease Control and Prevention (CDC), 1 the Royal College of Paediatrics and Child Health, 2 and WHO. 3 As SARS-CoV-2 spreads and awareness of MIS-C grows, the number of reported cases continues to increase. As of June 30, 2020, 230 paediatric MIS-C cases and two deaths have been reported in New York state, USA.
Here we describe the case of an adult male who presented to NYU Langone Health in New York City, NY, USA, with a Kawasaki-like multisystem inflammatory syndrome in the setting of SARS-CoV-2 infection, similar to what has been reported in children. Although we caution readers from making broad conclusions from this single case, we report this presentation to heighten awareness of the possibility of a COVID-19associated Kawasaki-like multisystem inflammatory condition in adults.
The patient is an Hispanic man, aged 45 years, without any past medical history (body-mass index 26·6 kg/m²) who presented to the emergency department with 6 days of fever, sore throat, diarrhoea, bilateral lower extremity pain, conjunctivitis, and diffuse exanthem after having cared for his wife with SARS-CoV-2 infection 2 weeks earlier. The patient denied respiratory symptoms on presentation, although his respiratory rate was elevated (25-33 breaths per min), and he had not taken any medications before symptom onset. A SARS-CoV-2specific RT-PCR was positive, and chest x-ray showed diffuse interstitial haziness typical of COVID-19. Vital signs throughout admission were notable for persistent fever despite antipyretics (maximum temperature 39·4°C), hypotension (systolic blood pressure 80-90 mm Hg), tachycardia with episodes of atrial fibrillation with rapid ventricular response, and minimal oxygen requirements (1-2 L/min by nasal cannula). Physical examination revealed bilateral, nonexudative conjunctival injection, tender left neck swelling with palpable lymphadenopathy, periorbital oedema with overlying erythema, lip cheilitis, and targetoid erythematous papules and plaques with central duskiness involving the back, palms, neck, scalp, anterior trunk, and upper thighs. Images were obtained with patient consent and are shown in the appendix.
Complete blood counts showed leukocytosis (11 600-16 500 white blood cells per µL), with lymphopenia (0-700 lymphocytes per µL), neutrophilia (10 100-15 000 neutrophils per µL), atypical lymphocytosis (2% atypical lymphocytes), and increased band neutrophils (2-16% band cells), whereas comprehensive metabolic pan els showed hyponatraemia (serum sodium 124-135 mmol/L) and elevated hepatic enzymes (aspartate aminotransferase [AST] 96-198 U/L; alanine amino transferase 78-133 U/L). Notably, his platelet counts were normal. Inflammatory markers were ele vated, including an erythrocyte sedimentation rate of 120 mm/hr, ferritin of 21 196 ng/mL, C-reactive protein of 546·7 mg/L, D-dimer of 2977 ng/mL, procalcitonin of 31·79 ng/mL, and interleukin-6 (IL-6) of 117 pg/mL. Troponin was elevated (peak 8·05 g/mL), as was B-type natriuretic peptide (170 pg/mL). HIV-1 and HIV-2 antibodies and bacterial blood cultures were negative.
Contrast-enhanced CT of the neck revealed inflammation and oedema involving the bilateral lower eyelid and pre-septal space, as well as suboccipital reactive lymphadenopathy (largest lymph node measuring 1·8 cm). Electrocardiogram demonstrated ST elevations in the anterolateral leads, triggering left heart cardiac catheterisation, which showed angiographically normal arteries. A subsequent transthoracic echocardio gram displayed global hypokinesis of the left ventricular wall with a mild to moderately reduced ejection fraction of 40%. A slit lamp examination of both eyes confirmed diffuse conjunctivitis with chemosis, as well as the presence of inflammatory cells within the anterior chamber, indicative of uveitis. A 4-mm punch biopsy of the skin was performed on a papule on the back, with histology revealing rare intraepithelial collections of neutrophils with necrotic keratinocytes and a sparse interstitial, mixed-cell dermal infiltrate with vacuolar interface changes.
Given the patient's constellation of signs (fever for more than 5 days, erythema multiforme-like rash, bilateral non-exudative conjunctivitis, erythema or cracking of the lips, unilateral cervical lymphadenopathy measuring more than 1·5 cm in diameter), he met American Heart Association (AHA) criteria for Kawasaki disease, 4 and he was diagnosed with Kawasaki-like multisystem inflammatory syndrome associated with COVID-19. The patient underwent therapy with therapeutic dose low molecular weight heparin, intravenous immuno globulin (2 g/kg) over 2 days, and a single intravenous dose of the IL-6 inhibitor tocilizumab (400 mg). He was also enrolled in two randomised controlled trials (NCT04369742, NCT04364737) for COVID-19 treatment. He did not require vasopressor support or an intensive care unit level of care and was maintained on minimal oxygen requirements. Following intravenous to what degree our patient responded to intravenous immunoglobulin versus tocilizumab, and as treatment algorithms are designed for MIS-C, it will be important to monitor patients receiving tocilizumab for potential development of coronary artery aneurysms.
We highlight this case to draw attention to the presence of a Kawasakilike multisystem hyper inflam matory syndrome in an adult with SARS-CoV-2 infection and note clinical improvement following administra tion of anticoagulation, intravenous immunoglobulin, and tocilizumab. We emphasise the importance of multidisciplinary care and recognition of the possibility of this syndrome across specialties, as provision of care for our patient necessitated coordinated efforts between specialists in emergency medicine, internal medicine, infectious diseases, cardiology, rheumatology, dermatology, and ophthalmology. Although this patient's Kawasakilike presentation bears a strong resemblance to MIS-C, as recently described in paediatric cohorts, we acknowledge that this isolated case may represent a spurious finding rather than an instance of a larger disease pattern. Nevertheless, we present this case to raise awareness of a potential MIS-C-like condition in adults. Further investigation is warranted to better elucidate the possibility of an MIS-C analogue syndrome in adults as we continue to expand our understanding of SARS-CoV-2-related syndromes. SS and MG contributed equally. We declare no competing interests. seen in Kawasaki disease, and his diffuse conjunctivitis was not limbic-sparing. Biochemically, he demonstrated markedly elevated C-reactive protein, neutrophilia, and lymphopenia, which are more consistent with MIS-C than with classic Kawasaki disease. 7, 8 Although the cause of Kawasaki disease remains unknown, the most widely accepted theory is an aberrant immune response to an infectious trigger. Emerging reports depict the phenotype of MIS-C as a combination of Kawasaki disease, toxic shock syndrome, and macrophage activation syndrome (or haemophagocytic lympho histiocytosis), [5] [6] [7] [8] [9] all syndromes of dysregulated immune responses. Our patient's presentation also included features typical of these different multisystem inflammatory syndromes. Many of his laboratory findings were consistent with haemophagocytic lymphohistiocytosis (markedly elevated ferritin, elevated triglyceride and AST), although he did not have organomegaly and cytopenias typically observed with this syndrome. 11 His skin biopsy findings had features typical for Kawasaki disease (non-specific sparse inflammatory infiltrate) and features suggestive of toxic shock syndrome (few intraepidermal neutrophils with necrotic keratinocytes), providing histological support for a distinct inflammatory syndrome. MIS-C has several features resembling Kawasaki disease, but it is important to distinguish MIS-C from classic Kawasaki disease. Diagnostic distinction from classic Kawasaki disease might have meaningful implications: whereas treatments targeting IL-6 are currently being investigated among therapeutic options for COVID-19associated hyperinflammation, the IL-6 inhibitor tocilizumab might provoke the development of coronary artery aneurysms in patients with classic Kawasaki disease. 12 Given the recent recognition and evolving understanding of MIS-C, standardised treatment guidelines have yet to be established. It is unclear immunoglobulin and tocilizumab administration, he showed clinical improvement with defervescence, resolution of tachycardia and tachypnoea, improvement in rash, cheilitis, and conjunctivitis, and downtrending inflammatory markers. He was discharged 9 days after hospital admission. Upon outpatient followup, he had complete resolution of his diffuse cutaneous eruption and conjunctivitis as well as a normal repeat echocardiogram.
Our patient's clinical presentation and course share a striking resemblance to the newly characterised MIS-C. [5] [6] [7] [8] [9] With the exception of his age, our patient meets the current case definition for MIS-C according to both the CDC 1 and WHO. 3 Although it is postulated that children experience a Kawasaki-like MIS-C as a postinfectious phenomenon, it remains unclear whether our patient had an asymptomatic SARS-CoV-2 infection in the preceding weeks, with a persistent positive RT-PCR result, or whether his hyperinflammatory disorder occurred as a direct manifestation of acute infection. If the latter is true, it is notable that he did not experience the hypoxic respiratory failure most frequently associated with moderate to severe COVID-19, despite his abnormal chest x-ray findings.
Emerging reports have revealed a pattern of clinical differences distinguishing MIS-C from classic Kawasaki disease. [5] [6] [7] [8] [9] Notably, although our patient's presentation met AHA criteria for Kawasaki disease, he also exhibited many MIS-C-related features such as a predominance of gastrointestinal symptoms, generalised extremity pain, and prominent cardiac dysfunction, and his cardiac findings (elevated cardiac enzymes and left ventricular hypokinesis with a reduction in ejection fraction) resemble findings of myocarditis recently described in MIS-C. 10 Furthermore, our patient's palmar lesions are distinct from the acral erythema and swelling with subsequent desquamation typically
|
The recent coronavirus outbreak, better known at COVID-19 in the United States or novel coronavirus pneumonia in China, has reached pandemic proportions. The illness is caused by the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), also known as 2019-novel coronavirus (2019-nCoV). With the first adult cases reported in December 2019 in Wuhan, China, pediatric cases followed shortly thereafter with a strong tendency for familial clusters. 1 Interestingly, total numbers of symptomatic pediatric cases lag dramatically behind adult cases suggesting a protective effect of age. 2 As pediatric pulmonologists, we are used to common respiratory
viruses. Yet coronaviruses are unfamiliar. The first two human coronaviruses, HCoV-229E and HCoV-OC43 were identified in 1960s and are well-known causes of the common cold. 3 SARS-CoV-2 infection has unique features that offer clues to its underlying biology. The Chinese reports indicate that children are significantly less affected by COVID-19, and when children do develop symptoms, they include fever and dry cough that rarely advance to severe disease. The Chinese Center for Disease Control and Prevention identified 2143 children (aged 0-18 years) with laboratory-confirmed or suspected COVID-19; one death was reported in this group. 4 The most common symptoms are fever (65%) and cough (45%). 5 Up to one-third of reverse-transcription polymerase chain reaction confirmed-positive pediatric patients remain completely asymptomatic. 6 However, 6% of all pediatric patients with COVID-19 developed severe/critical illness, with upwards of 11% of infants under the age of 1 year developing severe/critical disease. 4 Interestingly, the radiographic changes that are characteristic of SARS-CoV-2 infection can also be seen in pediatric patients, even without progress to severe disease. Case series of pediatric patients in various regions of China reported that nearly 50% to 80% of the children had abnormal computed tomography (CT) findings, often displaying ground-glass opacities and nodules, mostly located in the lower lobe of both lungs near the pleural area. [6] [7] [8] Interestingly, one pediatric case series reported the chest CT findings of consolidation with a surrounding halo, suggesting parenchymal infection were seen in 50% of pediatric patients, a finding not previously Immunosenescence, the progressive decline in immune function with increasing age, is a possible explanation. Pediatric patients appear to maintain normal white blood cell counts and lymphocyte counts, 7 while adults often develop abnormalities in neutrophil counts and T-cell depletion. 10 However, the mechanism for this discrepancy is unclear.
Alternatively, lessons learned from prior coronavirus infections, namely SARS-CoV, the pathogen responsible for severe acute respiratory syndrome (SARS) outbreak in 2003, have shown than the virus binds to the angiotensin-converting enzyme type 2 (ACE2) receptor to allow viral entry into type II pneumocytes in the lung. As there is homology in the sequence of the receptor-binding domain for the two viruses, it is felt ACE2 receptor is also the site of entry for SARS-CoV-2. 11 As these ACE2 receptors are upregulated in individuals with COPD, hypertension, or smoking history, this may explain why these populations are at increased risk for more severe disease. 8, 12 As pediatric pulmonologists, we are on high alert for our most vulnerable patients with underlying lung disease. Although the overall pediatric numbers are low, rare lung disease in children may not fall into these categories. Cystic fibrosis (CF), for example, is rare in the Chinese population, so how both pediatric and adult CF patients fare with COVID-19 is yet to be seen. In the meantime, we should proceed with caution; optimizing treatments for underlying lung disease seems prudent. Furthermore, minimizing cough related to asthma could reduce the potential aerosolization of the virus in an otherwise asymptomatic carrier. In that same vein, choosing hydrofluoroalkane-inhaler administered medications over nebulizers will also help reduce the aerosolized virus.
Focusing on the pediatric population as a means of preventing disease spread is critical. Children are less likely to report symptoms
|
Respiratory viruses are a major cause of morbidity and mortality throughout the world and affect persons of all ages [1] [2] [3] [4] . In addition to >100 million office visits for upper respiratory infections each winter, hospitals fill to capacity with admissions due to community acquired pneumonia, acute exacerbations of chronic obstructive pulmonary disease, asthma and bronchitis and many of these illnesses are due to viral infection. "Pneumonia and Influenza" consistently ranks as the fourth most common discharge diagnosis, and each year, 270,000 to 540,000 hospitalizations and 7600 to 72,000 deaths in the United States are attributable to influenza [3, [5] [6] [7] [8] . Due to their epidemic nature influenza and RSV are widely recognized as pathogens in adults and children, respectively. However, the true burden of disease and the contributions of other viruses such parainfluenza viruses (PIV), human metapneumoviruses (hMPV), coronaviruses (CoV), and rhinoviruses (HRV) now being more fully recognized using modern molecular detection methods [9] [10] [11] [12] [13] . In addition to sensitive and rapid diagnostic testing, new molecular techniques allow an understanding of viral evolution, mechanisms and predictors of severe disease, interrogation of vaccine responses, improved bacterial and viral diagnostics and associations of viral infections with non-respiratory medical events. In this chapter the many ways molecular and precision medicine have impacted the field of respiratory viral disease will be reviewed.
In the past defining the epidemiology and impact of viral respiratory pathogens was significantly hampered by slow and/or insensitive diagnostic techniques such as cell culture and antigen detection [13, 14] . Polymerase chain reaction (PCR) has revolutionized the study of respiratory viruses and provides extremely sensitive, specific and rapid means for the detection of fastidious and non-cultivatable respiratory viruses [13] . PCR based epidemiologic studies now provide a more complete understanding of the clinical spectrum and age ranges of populations affected [15] [16] [17] [18] [19] [20] . In one study, conventional methods yielded a viral diagnosis in 14% of pneumonia cases, while use of PCR increased the yield to 56% [21] . Technology has rapidly evolved from single-plex PCR and gel electrophoresis to multiplex real time assays where products are detected by luminescent signals proportional to the target amplified [14] . There are currently a variety of commercially available assays that detect from 2 to 20 viral respiratory pathogens and maintain excellent sensitivity [14] . Many clinical microbiology laboratories are now moving to primarily molecular methods for viral detection and PCR formats are becoming increasingly simple so that nucleic acid extraction and PCR is fully automated with little operator input. Molecular point of care assays will soon be feasible [22] .
In addition to providing more sensitive means of detecting known viruses, molecular methods are extremely useful for viral discovery [23] [24] [25] . Over the past several decades a number of new respiratory viruses or variants have been identified including hMPV, novel strains of coronaviruses (HKU1, NL63, SARS-CoV, MERS-CoV), rhinovirus C, Human Boca virus, parechoviruses and new strains of avian influenza viruses. Molecular methods have been critical for the rapid identification of new viruses associated with dramatic lethal outbreaks but also for pathogen discovery for routine respiratory illnesses. Despite intensive investigation, in 12-62% of lower respiratory illnesses no pathogen can be identified suggesting additional agents may yet be discovered [26] [27] [28] . Several different genomic approaches for pathogen discovery have been used successfully and include random primer amplification, pan-viral DNA microarray and next generation sequencing ( Fig. 1 ) [24] .
If a viral class of an unknown pathogen or variant virus is suspected, consensus PCR using degenerate primers to detect sequences broadly conserved between members of a group can be used as was done to identify two new coronaviruses, HKU1 and MERS-CoV [29, 30] . Another technique for viral discovery is random primer amplification with conventional shotgun sequencing of PCR products [31] [32] [33] . Such was the case when Van den Hoogen discovered a new respiratory virus in 2001, in young children with bronchiolitis who tested negative for RSV [31] . After detecting paramyxovirus like particles in cell culture, RNA was subjected to random primer PCR and viral sequences were compared to all known pathogens. The new virus most closely aligned to avian pneumovirus but was determined to be a unique human pathogen and named human metapneumovirus (HMPV). Similarly, in 2003 Peiris identified a novel coronavirus as the cause of severe acute respiratory syndrome (SARS-CoV) using degenerate/random primers PCR amplification [32] . Using pan viral micro array investigators at the Center for Disease Control and Prevention independently identified the same SARS-CoV [34] . In this technique, after random primer amplification, PCR products are hybridized to microarrays consisting of 70mer oligonucleotides derived from every fully sequenced viral genome. Hybridized sequences are scraped from the microspot, amplified, cloned and finally sequenced [25] . Identification of completely novel infectious agents requires unbiased and sequence independent methods for universal amplification [23, 24, 35] . Conventional Sanger sequencing may have poor sensitivity for genomes at low quantity. Next Generation Sequencing (NGS) involves the analysis of millions of sequences and can detect small amounts of novel nucleic acid sequences in clinical samples. Continuous sequences are assembled, host sequences are subtracted and the residual sequences are analyzed for similarity to known microbial sequences. NGS has led to the discovery a numerous novel human and animal pathogens [24] . A recent study of nasopharyngeal aspirates of Thai children with respiratory illness using NGS identified a number of mammalian viral sequences belonging to newly described families of viruses such Anelloviridae as well as novel strains of HRV, enteroviruses and HBoV [35] . A critical step in viral discovery is the availability of bioinformatic tools to efficiently identify unique viral sequences in complex mixtures of host, bacterial and fungal sequences. New computational tools for analysis of the virome such as "VirusSeeker" are being developed [36] . Of note, detection does equate with causation and after discovery further studies are necessary to infer more than association.
The genetic and antigenic evolution of error prone RNA respiratory viruses, particularly influenza, has been of interest for several decades [37, 38] . Understanding the selective pressure exerted by pre-existing immunity on viral evolution may help design more effective influenza vaccines and surveillance of animal populations can be critical for early identification of emerging influenza viruses [39] . Advances in deep sequencing make it possible to measure low frequency within host viral diversity and factors such as antigenic diversity, antiviral resistance, and tissue specificity can now be studied to understand the complexities of viral evolution [40] . Influenza evolution at a population level has been studied years, yet, new antigenic variants are initially generated and selected at the level of the individual infected host. Within a host, influenza viruses exist as a "swarm" of genetically distinct viruses [41] . Sanger sequencing defines consensus sequences and cannot resolve minority variants below 20% of the viral population. Deep sequencing has been used in natural infection and human challenge studies to characterize between and within host genetic diversity [41, 42] . The identification of low frequency mutations in the hemagglutinin (HA) antigenic sites or near the receptor-binding domain in vaccinated and unvaccinated influenza infected persons highlight viral evolution within a host due to selective immune pressure [41] . Similarly, NGS can reveal the rapid evolution of drug resistant variants during therapy [43] . Using samples collected over time, the mutational spectrum of H3N2 influenza A virus in an immunocompromised child was delineated [44] . Individual resistance mutations appeared weeks before they became dominant, evolved independently on cocirculating lineages. The within host evolution of antiviral resistance reflected a combination of frequent mutation, natural selection, and a complex pattern of segment linkage and reassortment. Within host sequencing diversity has also been examined in an infant with severe combined immune deficiency with persistent RSV infection [45] . NGS was performed on 26 samples obtained before and after bone marrow transplantation. The viral population appeared to diversify after engraftment with most variation occurring in the attachment protein (G). In addition, minority viral populations with palivizumab resistance mutations emerged after its administration. Deep sequencing of HRV during human challenge studies has shown that HRV generates new variants rapidly during the course of infection with accumulation of changes in "hot spots" in the capsid, 2C, and 3C genes [46] .
A genome-wide association study (GWAS) involves rapidly scanning sets of DNA, or genomes, of many people to find genetic variations associated with a particular disease. Typically, the genomes of cases are compared to non-affected controls and search for single nucleotide polymorphisms (SNPS) or polygenic changes that are associated with risk or protection from susceptibility or severity of the condition. GWAS have been useful to find genetic variations and risk for asthma, cancer, diabetes, heart disease and autoimmune illnesses with relatively limited studies relating to infectious diseases [47, 48] . Recent studies examining host genetic factors conferring susceptibility to respiratory viruses such as pandemic H1N1 2009 influenza A, SARS-CoV and RSV now provide some insight into host genetic factors for respiratory viral infections [49] [50] [51] . Previously most influenza research focused on viral genetics of novel viruses, yet experience with H1N1 2009 and H5N1 clearly indicate host factors also influence disease severity [49, 50] . A number of candidate genes influencing respiratory virus susceptibility have been identified in animal and human studies and involve host virus interactions, innate immune signaling, interferon related pathways and cytokine responses (Table 1) [49] [50] [51] [69] [70] [71] [72] [73] [74] [75] . Over 20 studies have evaluated genetic polymorphisms associated with severe RSV disease and none demonstrates dramatic results [51] . Most focused on one or a few candidate genes resulting in only modestly increased odds ratios of severe illness. A relatively large study of almost 500 hospitalized children that examined 384 SNPS in 220 candidate genes demonstrated that susceptibility to RSV is complex with a several associations to a few innate immunity genes. These included a Vitamin D receptor gene associated with down regulating interleukin 12 (IL-12), gamma interferon (IFN-γ), nitrous oxide synthase (NOS2A), the JUN oncogene, an important transcriptional regulator for innate immune pathways, and IFN-α (IFNA5) an antiviral cytokine [68] .
The host transcriptional response can be analyzed to investigate disease pathogenesis using a variety of methods including in-vitro studies of bronchial epithelial cells (BEC), animal models and infection both natural and experimental challenge [76] [77] [78] [79] . In addition, two compartments, the respiratory epithelium and blood can be sampled in human studies and interrogated using different viruses or viral strains to develop gene signatures for prognosis, as indicators of severity and to identify potential therapeutic targets.
Most respiratory viral mechanistic studies have been performed using influenza viruses, RSV, HRV and coronaviruses [80] [81] [82] [83] . Using BEC, the common and (H5N1, H7N7, H3N2 and H7N9 ) and analyzed cellular responses using microarray [83] . Common proinflammatory cytokines and antigen presentation were identified although each viral response was unique and notably, H7N9 responses were most similar to H3N2. The response of different clinical isolates of RSV in A549 cells, and monocyte derived human macrophages demonstrated that the pattern of innate immune activation was both host cell and viral strain specific [85] . Using RNA seq, differences in IL-6 and CCL5 were noted among the responses to different clinical isolates suggesting different RSV strains may vary in inherent virulence. Human studies have shown significant differences in the blood transcriptional profiles which change over time and differ depending on the infecting respiratory virus. Mejias and colleagues were able to differentiate RSV, HRV and influenza in young children based on the blood gene profile (Fig. 3) . HRV infection exhibited the mildest innate and adaptive responses compared to RSV and influenza and neutrophil gene expression was greatest in RSV infection with marked suppression of B and T cell and lymphoid responses [79] . Notably, gene expression changes persisted up to 1 month after infection. Similarly, studies of H7N9 infected patients showed transcriptional profile changes persisting up to 1 month with a transition from innate to adaptive immunity [86] .
Because of the association of HRV and exacerbations of asthma, the host response to HRV has been of particular interest [87] [88] [89] [90] . Studies using BECs from asthmatic and healthy donors demonstrate different transcriptional profiles when infected with HRV [87] . HRV, similar to other picornaviruses induces gene expression down regulation by the 2A and 2C proteins. In both asthmatic and healthy control derived cells the majority of genes were down regulated after exposure to HRV. However, some significant expression differences in inflammatory, tumor suppressor, airway remodeling and metallopeptidase pathways have been noted in asthmatic derived cells. Asymptomatic HRV infection is quite common and its role in asthma pathogenesis has been questioned. Interestingly, Heinonen et al. did not find a difference in the blood transcriptome of asymptomatic HRV infected children compared to non-infected controls [91] . Whereas, Wesolowska-Anderson and colleagues demonstrated over 100 differentially expressed genes in the nasal epithelium of asymptomatic infected HRV patients [90] . Thus, the blood transcriptome may not be as informative as the nasal epithelial transcriptional response for asymptomatic HRV infection. Given the significant host response to asymptomatic infection, HRV may play a role in asthma exacerbations in the absence of clinically evident disease. Lastly, it may be possible to identify patients with asthma who are prone to frequent HRV related exacerbations by examining the gene expression response of their PBMCs stimulated with HRV [89] .
Gene expression studies focusing on illness severity may enhance our understanding of disease pathogenesis, can identify potential therapies to modulate harmful host responses and can be used to develop biomarkers for predicting life threatening disease [79, [92] [93] [94] . A number of studies have been undertaken to understand the pathogenesis of severe RSV in young children and have identified a variety of gene expression patterns in blood including under expression of T cell cytotoxicity/NK cells and plasma cell genes, as well as upregulation of JAK/STAT, prolactin, IL-9 signaling, cell to cell signaling, and immune activation pathways [79, 92] . Using nasal epithelial gene expression analysis, van den Kieboom identified 5 differentially expressed genes in 30 children with mild, moderate and severe RSV infection [81] . Ubiquitin D, tetraspanin 8, mucin 13, β microseminoprotein, chemokine ligand 7 were up regulated and differentiated mild from severe illness. Lastly, nasal gene expression is complicated by interactions of the nasal microbiota and host cell gene responses [95] . In nasal samples from children with RSV infection, H. influenzae and S. pneumoniae dominated microbiota, Toll like receptors and neutrophil/macrophage signaling were over expressed and the presence of H. influenzae and S. pneumoniae along with age and sex were predictive of risk of hospitalization due to RSV.
Transcriptional profiling related to severity has been analyzed in seasonal influenza as well as emerging avian pathogens with a recognition that disease is not only due to an infection with a novel virus in a non-immune host but may also be due to an exaggerated host immune response [78, 96] . In a study of primarily seasonal influenza (H1N1, H3N2), influenza infection was associated with a significantly stronger antiviral, cytokine, attenuation of T/NK cell response compared to patients with respiratory illnesses of unknown etiology regardless of severity [96] . Notably, IFN and ubiquitination was significantly down regulated in those with severe vs. mild to moderate disease. In a study of the lethality of 1918 H1N1 influenza and H5N1 Vietnam influenza virus in Macaques, upregulation of key components of the innate immune response and cell death pathways were noted were noted with 1918 H1N1 infection but were down regulated with H5N1 [78] . Early up regulation of the inflammasome likely resulted in some of the severe tissue damage noted with the 1918 H1N1 influenza infections.
In vitro, animal and human challenge studies have been used to identify new strategies control or prevent symptomatic or severe infection [82, 97] . In HRV challenge studies, virperin expression correlated with rhinorrhea and chilliness. Knockdown of expression resulted in increased viral replication in BECs suggesting virperin has antiviral actions and might have potential therapeutic use. Influenza challenge studies clearly show a definable transcriptomic profile in the blood prior to the onset of symptoms offering the possibility of earlier and more effective oseltamivir treatment [77, 98] .
Lastly, host gene expression studies may allow investigation into links between respiratory viral infections and specific non-respiratory events. There is ample epidemiologic evidence that influenza epidemics are linked with increased rates of strokes and myocardial infarction (MI) [99, 100] . Increased rates of falls and functional decline in nursing homes have also been associated with increased influenza activity [101, 102] . However, direct links of events to viral infection are scarce in part due the event of interest may follow the infection by several weeks when the virus is no longer detectable by traditional testing. Several gene profiling studies have identified viral infection signatures that may persist up to 1-month post infection [79, 86] . Thus, it might be possible to study patients with falls or cardiac events for evidence of recent viral infection using a host response viral signature. In addition, evaluating the host response can provide information on mechanisms of disease. A viral gene signature was used to evaluate patients undergoing cardiac catheterization [103] . Notably, 25% vs. 12%, P = 0.04 of those with a viral gene signature present vs. those without viral signatures, suffered an MI. Furthermore, H1N1 infected patients showed an increased gene platelet expression signature providing insight into how infection may induce a prothrombotic state.
Given the availability of rapid and accurate Multiplex PCR for viral detection, host-based diagnostics might seem unnecessary. However, current PCR assays use conserved known viral sequences but can miss novel or significantly mutated viruses. This issue was seen in 2009 with pandemic H1N1 when influenza PCR assays had to be adapted to optimally detect the new influenza strain [104] . The emergence of novel respiratory viruses are a persistent threat and methods to detect a "viral signature" in the setting of clusters of severe pneumonia cases could be very useful. Zaas and colleagues developed an acute respiratory viral gene signature using microarray analysis of the blood from volunteers experimentally infected with influenza A, HRV or RSV [105] . The signature was subsequently 89% sensitive and 94% specific in classifying as viral 25 influenza and 3 HRV infected patients presenting to an emergency room. Additionally, a distinct blood transcriptome signature was noted in patients with severe H1N1 pneumonia [106] . Upregulated genes included those related to cell cycle, DNA damage, apoptosis, protein degradation, and T helper cells. Down regulated genes were primarily in immune response pathways suggesting immunosuppression as a mechanism of severe influenza pneumonia. Investigators developed a 29 gene classifier which predicted H1N1 influenza A regardless of concomitant bacterial infection and such a predicator could guide antiviral therapy in the face of negative pathogen detection methods.
In most cases of respiratory infection, the precise microbial etiology is unknown and antibiotics are frequently administered empirically [27, 107] . Although sensitive molecular diagnostics (PCR) now allow rapid diagnosis of a wide variety of respiratory viruses, their impact on patient management and antibiotic prescription has been modest primarily due to concern about bacterial co-infection [108] [109] [110] . Approximately 40% of adults hospitalized with a documented viral respiratory infection have evidence of concomitant bacterial infection and thus clinician concerns are reasonable [109] . Importantly, sensitive and specific diagnostic tests for bacterial lung infection are currently lacking [111, 112] .
Although the site of infection is the respiratory tract, blood is a convenient sample comprised of components of the innate immune system (neutrophils, natural killer cells), as well as the adaptive immune system (B and T lymphocytes) [113] . Recent studies indicate that viral and bacterial infections trigger pathogen specific host transcriptional patterns in blood, yielding unique "bio-signatures" that may discriminate viral from bacterial causes of infection [114] [115] [116] [117] . In the largest study to date, Tsalik et al. used gene expression in blood to discriminate bacterial from viral infection or non-infectious illness in 273 subjects with respiratory illness [118] . These investigators defined 130 predictor genes in a model with an accuracy of 87% to discriminate clinically adjudicated bacterial, viral, and non-infectious illness. Most studies to date have used micro array but recently RNAseq has been used to differentiate viral and bacterial respiratory illness and in one study 141 genes were noted to be differentially expressed [119] . Three pathways (lymphocyte, α-linoleic acid metabolism, IGF regulation pathways) which included 11 genes as predictors for bacterial infection from non-bacterial infection (naïve AUC = 0.94; nested CV-AUC = 0.86). To date, a number of gene expression studies of adults and children have developed predictors with similar accuracy (AUC ranging from 78% to 94%), yet there has been little overlap in classifying genes identified [105, 106, [114] [115] [116] [118] [119] [120] [121] [122] . Diverse populations, types of infection, plus alternate analytic tools used, likely explain the different genes identified. More work needs to be done to refine predictive gene sets including patients with mixed viral-bacterial respiratory tract infection. Most studies to date have focused on blood; however, analysis of the nasal respiratory epithelium which is the site of infection might offer advantages. Although data are limited, several recent papers demonstrate that nasopharyngeal host response can also be used as a diagnostic for respiratory viruses [93, 123, 124] .
Immune response to influenza vaccine is variable and influenced by a variety of factors including prior vaccinations and infections, age, the presence of underlying conditions and the type of vaccine administered. Yet, even among a relatively homogeneous cohort of young healthy adults, antibody responses to vaccine can be variable [125] . Transcriptional profiling of whole blood provide insights into the mechanisms of variability, the effects of age, and vaccine types. The ability to predict vaccine response at baseline based on a transcriptomic signature would have significant clinical implications. To understand the biologic effects of live attenuated influenza vaccine (LAIV) compared to trivalent inactivated vaccine (TIV) blood transcription profiles from 85 young children were assessed by microarray at day 7 post vaccination [126] . Many more genes were differentially expressed in children receiving LAIV compared to TIV (245 vs. 49, respectively) and many modulated type 1 IFN. The efficacy of LAIV has been problematic in recent years and assessing stimulation of type 1 IFN genes could represent a potential biomarker for response to LAIV [126] . Bucasas and colleagues evaluated gene expression at multiple time points after vaccination of healthy young men with TIV [127] . They noted marked up regulation of gene expression of IFN signaling, IL-6 regulation, antigen processing and presentation genes within 24 h of vaccination and were able to define a 494 gene expression signature that correlated with the magnitude of antibody response.
In another study, a gene profile predictive of antibody response 28 days after influenza vaccination of young and older adults was developed [128] . Notably, the predictive genes were the same in young and old as well as a subgroup of subjects with diabetes suggesting similar pathways were involved despite differences in age and underlying medical conditions. Additionally, transcriptional profiling has been used to signatures in blood associated with B cell memory responses to vaccination. In a study of 150 older and middle aged adults vaccinated with TIV including an H1N12009 antigen, metabolic, cell migration/adhesion, MAP kinase and NF-ᵏB cell genes correlated with peak memory B cell ELISPOTs [129] . Finally, in a study of over 500 subjects vaccinated over several seasons, a predictive signature of nine genes and three gene modules were significantly associated with the magnitude of the serum antibody response (Fig. 4 ) [130] . Interestingly and in contrast to a previous study, the signature was distinct to the younger cohort. For example, inflammatory genes were associated with better response in the young but a worse response in the elderly. In summary, gene expression studies could be used to evaluate new vaccines and develop predictors of vaccine response in different subgroups of patients based on age and disease state allowing for individualized vaccine regimens.
Molecular analysis of respiratory viruses and the host response to both infection and vaccination have transformed our understanding of these ubiquitous pathogens. The ability to accurately diagnosis viral infections has not only impacted patient care but also changed our perceptions of the burden of disease and populations effected. Transcriptional profiling of blood and nasal epithelium may provide therapeutic targets to prevent and ameliorate illness as well as offer predictors of severe disease. (14) .)
|
The recent COVID-19 epidemics is posing formidable challenges both to the health and economic systems worldwide. In order to tackle the ongoing epidemic, the countries faced with the COVID-19 epidemic have used different strategies. In order to control the epidemic one has to lower the reproduction number (also called reproduction rate) ρ, eventually below the threshold, i.e ρ < 1. In a simple SIR Model [2] [3] [4] [5] [6] 1] , which we consider in this paper, ρ depends on the infection rate α, the removal rate β and the total population N . If ρ > 1 the epidemic starts with the number of infected individuals I growing exponentially, it reaches a maximum I P at a time t P , then decreases down to zero at time t E when the epidemic ends.
Although for severe epidemics like the COVID-19 it is very difficult to keep the reproduction number ρ from the beginning below the threshold, nevertheless containment measures soften the epidemic because they lead to a reduction of the epidemic peak I P .
Basically, there are therefore three types of containment measures one can use. One can act on α by lowering it, for instance * Corresponding author.
E-mail address: [email protected] by forcing social distancing or by prophylaxis measures. One can achieve the same result increasing β, e.g. by means of a prompt strict isolation of infected individuals. Last but not least one can reduce the total population N by separating it in strictly noncommunicating compartments. The policies of different countries for fighting the COVID-19 usually contemplate a mixture of the three types of measures mentioned above. Obviously, the choice of focusing on one kind of measure instead of another depends on a number of factors, which include not only its effectiveness but also its feasibility and its social and economic impact. For instance most of the European countries have focused their COVID-19 fighting strategies on measures aimed at reducing the infection rate α. Countries like Korea and Singapore favoured instead measures aimed at rising β.
A general feature of the epidemic dynamics is that the reduction of the epidemic peak I P can be only achieved at the expenses of increasing t P and t E , i.e of increasing the time-span of the epidemic. This is evident if one looks at the curves showing the behaviour of the number of infected I ( t ) as a function of time. Lowering α and/or increasing β will result in flattening the curves, i.e in a reduction of the epidemic peak together with an increasing of the width of the curve. The flattening of the epidemic curve may have both positive and negative effects, depending on a number of https://doi.org/10.1016/j.chaos.2020.109940 0960-0779/© 2020 Elsevier Ltd. All rights reserved. factors such as the capacity of the medical infrastructure to treat a big number of infected. On the one hand for epidemics like COVID-19 having a high rate of hospitalised infected with severe symptoms, reduction of I P and the slow down of the epidemic is necessary in order to allow the health systems to treat them properly. On the other hand, increasing t P means increasing the time-span in which the containment measures are effective, with potentially disruptive effects both on the economies and the live of the populations in the involved countries.
In this paper we investigate, in the framework of the SIR model, the impact of containment measures, which act on α and β, on I P , t P , t E and on the epidemic speed ( d ρ/ dt | P ). In order to be able to analyse, analytically, the dependence of the latter parameters from the former we need an exact form for the solution of the SIR dynamics in the time domain. Using a time reparametrization we linearize the SIR equations. This allows us to solve exactly the dynamics in the time domain. The form of the exact solution is then used to derive the scaling behaviour of I P , t P , t E and of the epidemic speed d ρ/ dt | P by reducing the infection rate α and by increasing the removal rate β by a factor of λ.
The main result of this paper is that containment measures which achieve the same reduction of the size of the epidemic ( I P , the total, I E and average ˆ I P number of infected) impact differently on its timescale. The occurrence time t P of the peak and the entire time-span t E of the epidemic can be reduced by a factor 1/ λ if the reduction of I is achieved by increasing the removal rate instead of reducing the infection rate. This means that, unless a drastic slow down of the epidemic is needed, epidemic containment measures based on tracing, early detection followed by prompt isolation of infected individuals are more efficient than those based on social distancing. In the final part of the paper we apply our results to the COVID-19 epidemic in Northern Italy. We show that the peak time t P and the entire time span t E could have been reduced by a factor 0.9 ≤ 1/ λ ≤ 0.34 with containment measures focused on increasing β instead of reducing α.
The structure of the paper is as follows. In Sect 2 , by means of a time reparametrization, we are able to linearize the equations for the SIR dynamics. This allows us to solve exactly the dynamics in the time domain and to derive the scaling behaviour of I P , t P , t E and d ρ/ dt | P , by reducing the infection rate α → α/ λ or by increasing the removal rate → λβ. We show that keeping I P fixed its occurrence time t P and t E can be be reduced by a factor 1/ λ by acting on the removal rate β instead on the infection rate α. This will be discussed in Sect. 3 . In Sect. 4 we discuss approximate solutions of the SIR model in which the reproduction number ρ < e . In Sect. 5 we apply our results to the COVID-19 epidemic in Northern Italy. We show that the peak time t P and the entire time-span of the epidemic could have been reduced by a factor 0.9 ≤ 1/ λ ≤ 0.34 with containment measures focused on increasing β instead of the reducing α. Finally, in Sect. 6 we state our conclusions.
The SIR model describes the deterministic dynamics of an infective epidemic, characterized by the fact that individuals, which have been infected and have recovered gain permanent immunity [2] [3] [4] [5] [6] 1] . Although the model is quite simple, it can be used to give at least rough estimates of epidemic dynamics, and in particular of the COVID-19 epidemic [7, [9] [10] [11] . A generalisation of the SIR model to take into account a large number of asymptomatic infectiveshence more apt to describe the COVID-19 epidemic-has been proposed in Ref. [9] [10] [11] .
The homogeneous and isolated population of N individuals exposed to the epidemic, is characterised at time t by the number of susceptible S ( t ), infected and infectives I ( t ) and removed (recovered, dead or isolated) R ( t ) individuals, with the conservation law N = S(t ) + I(t ) + R (t) . The timescale of the epidemics is assumed to be relatively short so that N can be assumed constant.
The dynamic describing the evolution of the epidemic is deterministic and described by the following, non linear, dynamical system:
The infective epidemic is characterised by two parameters: (1)
The infection rate (also called contact rate) α, which gives the transition rate between the class of susceptible and that of infected;
(2) the removal rate β, which gives the transition rate between the class of infected and that of removed (1/ β gives the characteristic time for the removal of infected from the dynamics). From Eq. (1b) it is immediately evident that number of infected individuals grows, i.e the epidemic spreads, only if
where γ is the epidemic threshold. Equivalently, one can introduce the reproduction number ρ( t ):
(see e.g. Ref. [10] ). This form of I ( S ) allows one to derive some qualitative and quantitative features of the epidemic dynamics but not its explicit time evolution. This latter can be only obtained by numerical integration of Eqs. (1a) , (1b),(1c) . For instance, one can easily find that, if initially we are above the threshold ρ 0 > 1, I ( t ) grows till it reaches a maximum I P , then it goes down to zero at a time t E when the epidemic ends.
The function I ( S ) allows to determine analytically the value of the peak I P but not the time t P of its occurrence, nor its entire time-span t E , nor the speed of the reproduction number V P := d ρ/ dt | P , nor the average value of the number of infected individuals at the peak ˆ I P . t P , t E , V P and ˆ I P have to be determined after solving numerically the dynamics. This is a quite unpleasant feature because it prevents a clear understanding of the dependence of t P , t E , V P and ˆ I P from the parameters α, β, which is a crucial information for fighting the epidemic.
In order to solve analytically the temporal dynamics let us reparametrize the time introducing a new time coordinate τ defined by d τ /d t = I(τ ) , i.e.:
where t 0 = t(τ 0 ) is the initial time. Using time-translations we can put without loss of generality, τ 0 = t 0 = 0 . The new time coordinate has a simple intuitive meaning, τ ( t )/ t gives the average value ˆ I (t) the number of infected at time t :
The time reparametrization (4) allows to linearise the system (1a),(1b),(1c) :
This can be easily integrate to give:
where S 0 , I 0 are initial data and R 0 = N − S 0 − I 0 . In the following
Exact solutions of the SIR model, which are equivalent to our Eqs. (7a) , (7b),(7c), (8) have been derived in Refs. [12, 13] using a completely different approach. Although, the integral (8)
The entire time-span of the epidemic t E can be computed setting I = 0 in Eq. (7b) . Because I 0 is usually small compared to S 0 it can be neglected, t E is obtained by first finding the (higher) root τ E of the transcendental equation:
and then using Eq. (8) to compute t E .
An other important quantity, which describes the intensity of the epidemics is the total number I E of individuals that are infected over the whole time-span of the epidemics. Taking into account that I 0 is rather small and that initially S 0 = N, I E can expressed in terms of the lower root S E of the transcendental equation obtained by setting I = 0 in Eq. (7b) (see Ref. [9] ) for details. We have
where S E is the lower root of the transcendental equation
In this section we investigate the scaling behavior of the peak quantities ( 9a ... (9d) , the total number of infected (11) , I E and t E by changing of the parameters α and β. It is already known that Eqs. (1a) ... (1c) are invariant under the scaling α → λα, β → λβ, t → λ −1 t [9] . This scaling transformation leaves invariant the epidemic threshold γ and tells us that we can increase (reduce) the timescale of the epidemic by simultaneously reducing (increasing) both α and β. However, this is not what we are interested in. Actually, we want to know what happens to the epidemic parameters listed above when we increase the threshold γ . Let us first observe that both the number of infected at the peak I P (see Eq. (9a) and the total number of infected I E (see Eq. (11) are decreasing functions of the parameter γ . In fact, we get from Eq. (9a) and Eq. (12) ,
We see that above the epidemic threshold ( S 0 / γ > 1), dI P / d γ is always negative, while being S E < N, dS E / d γ is always positive.
It follows that if we want to reduce the peak and the total number of infected we have to increase γ by a factor λ > 1. Because one can increase γ either by reducing α or by increasing β, we have to compare the effects on the peak parameters of these two different ways of increasing γ .
We are therefore lead to consider two different scaling transformations: transformation T (1) , which reduces the infection rate:
and transformation T (2) , which increases the removal rate:
The peak quantities in Eqs. ( 9a ... (9d) and the τ E of Eq. (10) do not transform in a simple way under T (1) and T (2) , however the ratios of T (1) and T (2) -transformed quantities follow simple scaling laws. In particular, they remain invariant whenever the quantity depends only on their ratio γ and not on α and β separately.
Using the following notation to denote rescaled quantities: I (1) P = I P (λ −1 α) , I (2) P = I P (λβ ) and similarly for the others quantities, we get,
The transformation law for t P and t E can be derived by first acting with the transformation T (1) on the integral (9d) , then acting with T (2) on the same integral and finally redefining the integration variable in the second integral τ → λ −1 τ . One obtains in this way:
Finally, using Eqs.
An important result follows from Eqs. (16a) , (16b),(17) and (18) : epidemic containment measures, which have the same effect for what concerns I P , R P , S P , ˆ I P and I E , have different im pact on the occurrence time t P of the peak, the whole time span of the epidemic t E and on the epidemic speed V P . Choosing measures increasing the removal rate β by a factor λ instead of reducing the infection rate α by a factor 1/ λ allows to drop t P and t E by a factor 1/ λ. For instance by implementing epidemic containment measures with λ = 2 we can reduce by a half both the time needed for the epidemic to reach the peak and the whole time-span of the epidemic. It should be noticed that this epidemic timescale reduction effect becomes more relevant for epidemics with high reproduction number ρ 0 > > 1. In fact the factor λ is limited by λ < ρ 0 , simply because for λ > ρ 0 the epidemic does not develop at all. Thus, if we have for instance ρ 0 = 5 we can reduce the peak time and the entire time-span of the epidemic until a factor of 1/5. Therefore, increasing β represents an efficient way to fight epidemics. If by increasing it we manage to bring ρ below the threshold we simply stop the epidemic, but even if we do not go so far, we can still reduce the size of an epidemic keeping under control its timescale.
The behaviour of the reproduction number speed V P in Eq. (16b) explains clearly what is going on. If one acts on β instead on α, V P increases, as expected, by a factor of λ. In short, increasing β instead of reducing α, allows one to speed up the epidemic dynamics keeping constant the number of infected at the peak, the average number of infected and the total number of infected. This is possible because the increasing of the removal rate allows prompt removal of infected individuals.
In the general case the integral (8) cannot be computed analytically. Therefore the function τ = τ (t) has to be computed numerically, by first performing numerical integration of the integral in (8) to find t = t(τ ) and then inverting it. There is, however, a situation in which the integral (8) can be computed analytically and the dynamics of the epidemic until the peak, can be expressed analytically in closed form in terms of the time t , albeit in approximate form.
For ατ < < 1 we can approximate the exponential in Eq. (8) by e −ατ ≈ 1 − ατ . This approximation allows to solve the integral and to invert the function t = t(τ ) . We find,
With this position we can easily write down the approximate form of the solutions (7a),(7b),(7c) in terms of the time t . We quote here only the form of I ( t ) and t P as a function of I P ,
Eqs. (19) and (20) are a good approximation only for ατ < 1. Because τ ( t ), is an increasing function of t , the approximation for the dynamics is good until the peak, if ατ P < 1, which implies from Eq. (9c) :
The recent development of the COVID-19 epidemic in Northern Italy represents an interesting case for applying the results described in the previous sections. The epidemic developed in the three main regions of Northern Italy (Lombardia, Veneto and Emilia-Romagna) we consider in this paper, starting from end of February 2020 (although it may be possible that the epidemic was circulating in the regions before that date).
Altogether these three regions have around 20 millions of inhabitants, we will therefore take N = 2 · 10 7 in our computations. We take as initial value I 0 = 100 , which approximately corresponds to the known cases of COVID-19 infected Northern Italy on February 23, 2020. The determination of the initial values of the other two parameters of the SIR model α 0 and β 0 is more involved.
These initial values are completely determined by the pathogen because they are not affected by the epidemic containment measures put into play. The value of α 0 can be determined from the exponential behaviour of the early dynamics, or equivalently from the initial doubling time [7, 8] . Using the raw data for the early dynamics of the epidemic in Northern Italy, Gaeta [7, 8] has given the estimate:
The determination of β 0 is even more problematic. This is because we expect it to be sensitive to the presence of large cohort of asymptomatic infectives. We can estimate β 0 from the basic reproduction number ρ 0 , using Eq. (3) . Rough evaluations of ρ 0 give a number between 2 and 2.5, however the indeterminacy related to the presence of large number of asymptomatic infectives may result in a much higher value for ρ 0 . To be rather conservative we assume here ρ 0 = 3 , so that Eqs. (3) and (22) give:
The COVID-19 containment strategies put in place in Northern Italy are a mixture of social distancing, social confinement, early detection and infection tracing. Although mainly focused of social distancing, these strategies contain all the previous ingredients, which modify in different ways the parameters α and β. Social distancing acts by reducing α, whereas systematic, prompt, and strict isolation of infected individuals as a result of early detection and tracing enhances β. Both effects rise γ and reduce in the same way the amplitude of the peak I P , the average number of infected ˆ I P and the total number of infected I E .
Because it is almost impossible to disentangle the effects of the various containment measures on α and β in the real situation, we will discuss and compare two hypothetical situations in which the rising of γ , γ → λγ 0 , with γ 0 = (2 / 3) · 10 7 , is obtained in two fully distinct and complementary ways:
• (1) We have exclusively social confinement containment measures: β is held fixed to its initial value β 0 , whereas α 0 is reduced by a factor 1/ λ.
• (2) We have exclusively containment measures consisting in prompt and strict isolation of infected individuals triggered by early detection and tracing of infected: α is held fixed to its initial value α 0 , whereas β 0 is increased by a factor λ.
Being 1 ≤ λ < ρ 0 we consider the following values: λ = 1 , 1 . 5 , 2 , 2 . 5 , 2 . 9 . Using Eqs. (9a) ... (9d) we compute for these values of the parameters α and β the peak quantities: peak amplitude I P , the time t P (in days) of occurrence of the peak (computed by numerical evaluation of the integral (9d) ), average number of infected ˆ I P and absolute value of the epidemic speed at the peak | V P | (in days −1 ). Moreover, using Eqs. (11), (12), (10) , together with Eq. (9d) , we compute numerically the total number I E of infected individuals during the epidemic and its whole time span t E (in days). The results are shown in Table 1 . Our results are in accordance with the scaling behaviour given by Eqs. (16a) , (16b), (17) and (18) .
We see from Table 1 that the raising of the epidemic threshold for γ from the initial value γ 0 first to 2 γ 0 then to 2.9 γ 0 let both the number of infected at the peak and their average number drastically sink from the order of magnitude 10 6 first to 10 5 and then to 10 3 − 10 4 . This reduction is the same independently of the fact that if it is achieved by reduction of α (way (1)) or by increase of β (way (2)). Similarly, the total number I E of infected individuals Table 1 Comparison of the effect of reduction of the infection rate α → (1/ λ) α versus increase of the removal rate β → λβ on epidemic parameter: peak amplitude I P , average value of infected ˆ I P , peak time t P , speed of reproduction number | V P | at the peak, total number of infected individuals I E and whole timespan of the epidemics t E . The total population is N = 2 · 10 7 , β 0 = 1 / 9 and α 0 = (1 / 6)10 −7 . The values of drops from the huge value 1 1.88 · 10 7 till 1.16 · 10 7 (for λ = 2 ) and then to 1.3 · 10 6 (for λ = 2 . 9 )
On the other hand, the two ways of reducing I P and ˆ I P and I E affect differently the occurrence time of the peak t P and the whole time span of the epidemic t E . By acting on β (way (2)) instead of on α (way (1)) we can shorten these times by 33% (for λ = 1 . 5 ), by 50% (for λ = 2 ) and even reduce it by almost 1/3 (for λ = 2 . 9 ).
Correspondingly, the speed of variation of the reproduction number | V P | will be enhanced by the same factors.
If the containment measures can manage to increase λ above 3, we go below the threshold for ρ and the epidemic does not start at all. Obviously, in a real situation reducing α or increasing β is not performed once for all at the beginning, but occurs in steps. Our main result is that in order to try to stop the epidemic it is much more convenient to rise β instead of lowering α because even if we do not manage to stop it, we are able to reduce its size and at the same time to shorten its timescale.
In this paper we have analysed, in the context of the standard SIR model for epidemic dynamics, the impact of different containment measures on size (the epidemic peak I P , the average number of infected ˆ I P and the total number of infected I E ), the timescale (the occurrence time of the peak t P and the whole time-span t E ) and the speed (time variation of the reproduction number | V P |) of epidemics. Using an exact solution for the epidemic dynamics we have been able to derive the scaling behaviour of these quantities under change of the two parameters (the infection rate α and the removal rate β) of the SIR model, which can be controlled by the containment measures. This allowed us to compare the impact on size, timescale and speed of the epidemic of containment measures acting either on α or on β.
The main result of our paper is that for a given reduction of I P , ˆ I P , I E , the timescale and the speed of the epidemic are to a great extend sensitive to the kind of measures we put into play. By increasing the removal rate β instead of reducing the infection rate by a factor λ one can reduce the timescale of the epidemic by a factor 1/ λ and increase the speed of the epidemics by a factor λ. Hence, flattening of epidemic curves I ( t ) achieved by reducing α or increasing β are not equivalent. Flattening the curve by acting on β instead on α allows to keep under control also the width of the curve.
This means that containment measures based on increasing β, e.g. based on tracing and removal of infectives, have a different effect than those that reduce α, e.g based on social distanc- 1 Notice that without containment measures at the end of the epidemic almost all individuals have been infected.
ing or lockdown. This his particularly true for what concerns the timescale of the epidemic. If by increasing β we manage to bring ρ below the threshold we simply stop the epidemic, but even if we do not go so far, we can still reduce the size of an epidemic keeping under control its timescale. This is surely an advantage when the sanitary system can deal with the epidemic because in this way containment measures do not have to be implemented for a long time span. The situation is of course different if and when the sanitary system is overwhelmed by the epidemic, as it happened in the first phase of the COVID-19 diffusion in many countries or regions, also as a result of its unexpectedly fast spreading.
An important point we have not addressed in this paper is the determination of the exact way in which the usual containment measures used to fight epidemic, impact on the values of the parameters α and β. Whereas it is quite clear that social distancing reduces the parameter α and does not change β, the effect of other measures like, early detection and contacts tracing is not a priori evident. Early detection and contact tracing increase β only if implemented on a large scale and followed by prompt and strict isolation of the detected infectives. If this is not the case, it is likely that these measures just bring a small reduction of α The recent analysis of Gaeta [10] of the different strategies used in Northern Italy to tackle the COVID-19 epidemic seems to confirm this result. He found that simple early detection and contact tracing, while having an impact on the epidemic peak, do not substantially affect the timescale of the epidemic. On the other hand he also showed that contact tracing if followed by prompt isolation is the only efficient way to reduce the size of the epidemic, without having to live with it a long time. The Veneto experience shows that this was one of the factors underlying the success of the containment strategy in that region. Thus the main lesson one can draw from our results is that, epidemic containment measures focused on tracing, early detection followed by prompt removal of infected individuals are more efficient to fight epidemics than those based on social distancing.
Let us conclude this paper with some comments about the range of validity of our results. The SIR model is an oversimplified model for epidemic dynamics. Generalisations of it are necessary in order to give a good descriptions of real epidemics. For instance, in the case of the COVID-19 epidemic a generalization of the SIR model seems to be necessary in order to take into account the presence of a large set of a asymptomatic infective [9] [10] [11] . On the other hand, the SIR model gives the bare bones of deterministic epidemic dynamics. For this reason we believe that, at least at qualitative level, the main result of this paper -the possibility to reduce the epidemic peak keeping under control its timescale by acting on removal rates-could remain true for generalized and improved SIR-like models.
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.
|
Thromboprophylaxis in COVID-19: Anti-FXa-the Missing Factor?
To the Editor:
Coronavirus disease (COVID-19) infection was declared a public health emergency of international concern in January 2020. The medical literature has since seen a succession of reports questioning a link between the disease in its severe form, a dynamic spectrum of coagulopathy, and a concerning incidence of thrombotic complications. As we accumulate observational data from around the globe and await well-designed prospective studies to inform best practice, clinical guidance on the management of thrombotic risk remains pragmatic.
We have read with interest initial reports from Wuhan, China, describing significant differences in D-dimer levels between survivors and nonsurvivors of COVID-19 and the overt presence of disseminated intravascular coagulation in over 70% of deaths (1) . In light of the histological features of thrombotic occlusion of the pulmonary vasculature at autopsy (2), the Shanghai Clinical Treatment Group advised the early application of anticoagulation therapy in severe COVID-19. This led to a retrospective comparison of patients who had not received any heparin before the guidance with those who had, and, unsurprisingly, heparin treatment was associated with a reduced mortality. A prophylactic dose of low-molecular-weight heparin (LMWH) was mostly used; however, the authors proposed that a higher dose may be more beneficial for non-Asian patients (3) .
With growing awareness of a distinct coagulopathy accompanying COVID-19 infection, the medical community has been keen to address the significant thrombotic risk for this patient group. Institutions have anecdotally reported what were perceived to be higher than expected rates of pulmonary embolus (PE), deep vein thrombosis, and occlusion of citrated circuits.
Klok and colleagues reported a 31% cumulative incidence of venous and arterial thrombosis, increasing to 49% after adjustment for competing risk of death, despite anticoagulant therapy in patients admitted to the intensive therapy unit (ITU). The authors suggested that ITU patients may warrant higher thromboprophylaxis dosing such as enoxaparin 40 mg twice daily versus 40 mg once daily (4).
Helms and colleagues compared a prospective cohort of ITU patients with COVID-19 with a historical cohort non-COVID-19 acute respiratory distress syndrome (ARDS). They also observed a high prevalence of clinically relevant thrombosis, most commonly PE (16.7%), despite prophylactic or therapeutic anticoagulation; this was usually diagnosed within a few days of ITU admission. The authors advised that anticoagulant treatment should be guided by anti-factor Xa (anti-FXa) activity and that higher targets would likely be required (5) .
Interestingly, Middeldorp and colleagues compared ITU patients with ward patients and commented on a much lower rate of venous thromboembolism (VTE) in the latter (6) .
The consensus from reports to date is that there appears to be a greater than expected VTE risk despite pharmacological thromboprophylaxis. Authors consistently acknowledge the current divergence in thromboprophylaxis dosing from what would usually be considered a standard dose (Table 1) .
We reviewed anti-FXa activity in patients admitted to hospital with COVID-19 infection, all receiving pharmacological thromboprophylaxis with enoxaparin 40 mg once daily, creatinine clearance of .30 ml/min, and platelet count of .30 3 10 9 /L (7).
We compared 4-hour after dose anti-FXa activity levels for 22 ward patients and 20 ITU patients ( Table 2) .
With a significantly lower mean anti-FXa activity of 0.1 IU/ml, 95% of ITU patients failed to achieve a target anti-FXa activity (0.2-0.4 IU/ml) compared with 27% ward patients. This difference appeared to relate to the degree of respiratory support required. Patients admitted with COVID-19 now receive weight-adjusted LMWH thromboprophylaxis with anti-FXa-guided dose escalation/reduction to achieve target anticoagulation levels.
These preliminary data suggest that patients admitted to ITU with COVID-19 may warrant a higher starting dose of pharmacological thromboprophylaxis, an approach that has already been adopted in some institutions. Furthermore, anti-FXa-guided LMWH dosing may have a role in ward patients because almost 30% of these patients demonstrate suboptimal anti-FXa target levels with standard dosing and merit early dose escalation.
LMWH is broadly used in hospital patients for the prevention and treatment of VTE owing to a relatively predictable pharmacokinetic profile and ease of monitoring. The anticoagulant effect is measured using the anti-FXa activity. Monitoring for prophylaxis is not routinely used; however, on the basis of studies published, a reasonable target range for prophylaxis has been suggested as between 0.2 and 0.4/0.5 IU/ml (8) . There are recognized situations, such as in pregnancy, renal impairment, and obesity, in which standard doses may not achieve optimal thromboprophylaxis. Uncertainty remains, however, regarding the value of anti-FXa monitoring in such groups. The consideration of patient-specific risk factors for thrombosis and hemorrhage, together with the relationship between anti-FXa activity and clinical outcome, may be more important (9) . This is of particular relevance in COVID-19, in which there is a recognized spectrum of thrombosis and bleeding risk in later stages of the infection (7) . Furthermore, the growing autopsy histology literature demonstrates a heterogeneity of thrombotic disease manifestations, including mutually exclusive deep vein thrombosis and PE, often despite anticoagulant therapy (10) . For pulmonary vascular occlusion that is more thrombotic than embolic, higher LMWH doses may not necessarily be more effective, and therefore the mechanism and relative contribution of the thrombotic burden to death and the best anticoagulation approach remain critical questions. Potential mechanisms for the development of what appears to be an acquired heparin resistance include reduced antithrombin levels, as seen in patients with sepsis requiring ITU care. The coagulopathy of COVID-19, however, appears to differ (1). As a result of a disrupted equilibrium and the well-recognized battle between inflammation and coagulation at the endothelial surface, more heparin is required to counteract excess thrombin generation in patients with severe disease (11) . We also know that significantly raised plasma concentrations of tissue factor and PAI-1 (plasminogen activator inhibitor-1) occur on approximately day 7 in patients with ARDS, which could in turn lead to increased alveolar fibrinolysis from an increase in local PAI-1 (12) . Taken in the clinical context, patients who require higher levels of ventilation or develop ARDS may warrant increased doses of LMWH thromboprophylaxis, resulting in discordant anti-FXa activity. Interestingly, ARDS has also been identified as a risk factor for VTE prophylaxis failure in critically ill patients with sepsis (13) .
Clinical thrombotic endpoints will undoubtedly form an important component of upcoming randomized controlled trials designed to define the relationship between optimal anticoagulation and thrombosis outcomes in COVID-19. Until then, the optimization of the anticoagulation strategy remains paramount. Amid consistent concerns for more effective prophylactic LMWH dosing, our data provide missing information regarding anti-FXa activity, confirming lower than expected activity, particularly in patients managed in the ITU. This informs the move by many institutions to start with higher thromboprophylaxis dosing pending the results of randomized controlled trials and provides additional clues as to the nature of the COVID-19-associated coagulopathy. n
|
Following the emergence of a new coronavirus named severe acute respiratory coronavirus 2 (SARS-CoV-2) in China in late December 2019 and subsequent global spread, a pandemic was declared by the World Health Organization on March 11, 2020. 1 As of July 26, 2020, more than fifteen million cases of the new respiratory disease named Accepted Article coronavirus disease 2019 (COVID-19) and 640,016 deaths have been notified globally. 1 In Greece the first cases were diagnosed on February 26, 2020. 2 During the first weeks of the epidemic, most COVID-19 cases were travel-associated, while as the epidemic progressed, community transmission was established. 2 The first evidence that SARS-CoV-2 can be transmitted from person-to-person was set when a member of a family in China who did not travel to Wuhan, became infected by the virus after several days of contact with family members who travelled to Wuhan; all patients had radiological ground-glass lung opacities, with adults presenting respiratory symptoms, while a 10-year child remained asymptomatic. 3 Soon after, it became evident that asymptomatic infection and mild clinical illness are more prevalent in children compared to adults. 4, 5 Herein, we studied the transmission dynamics of SARS-CoV-2 infection within families with children in Greece, focusing on the comparison of disease severity, outcome, and viral load between adults and children.
SARS-CoV-2 infection is notifiable disease in Greece. Surveillance of SARS-CoV-2 infection is performed by the National Public Health Organization on a case basis. Data are notified daily by all laboratories testing for SARS-CoV-2 using real-time reversetranscriptase polymerase chain reaction (RT-PCR). In addition, physicians notify all laboratory-confirmed COVID-19 cases using a standardized notification form. A passive comprehensive system for hospitalized cases is also in place, collecting data daily on admissions in intensive care unit (ICU), complications, and outcome. For every COVID-through telephone interview with the physician in charge.
Contacts of SARS-CoV-2 infected cases were traced. Close contacts were instructed to stay isolated for 14 days following the last contact with the COVID-19 case.
In case of onset of symptoms, contacts were advised to attend a COVID-19 referral hospital for testing.
The study period extended from February 26 (first COVID-19 case diagnosed in Greece) through May 3, 2020 (last date of lockdown in Greece). Family clusters were identified through the national registry of SARS-CoV-2 infections. We studied family clusters diagnosed in three reference laboratories for SARS-CoV-2 (two in Athens and one in Thessaloniki) where most cases were diagnosed. Families with at least one child were included in the study. Demographic, epidemiological and clinical data were collected. An adult family member (preferably the mother) was contacted through telephone in order to collect data about the possible source of infection of the first case, symptoms of household members and in-family contacts.
Patients' respiratory samples were tested by real time RT-PCR following commercial or in-house protocols. Based on the cycle threshold (Ct) value of the PCR, persons were categorized into three groups, those having high, moderate, or low viral load (Ct <25, 25-
Asymptomatic SARS-CoV-2 cases were defined as those with positive SARS-CoV-2 PCR in the absence of symptoms. COVID-19 cases were defined as those with positive SARS-CoV-2 PCR and compatible signs and symptoms. COVID-19 cases were classified as mild when patients were managed in the outpatient setting, moderate when patients were admitted to hospital and had a favorable outcome, while severe were classified those admitted to intensive care unit (ICU) or had a fatal outcome. Children were defined as persons<18 years of age. A family cluster was defined as the detection of at least two cases of SARS-CoV-2 infection within a family. Index case was defined as the first laboratory-diagnosed case in the family, which brought SARS-CoV-2 infection in the family under medical attention. In contrast, first case was defined as the first COVID-19 case in a family. Household contacts were defined as persons either living in the same residence or having close contacts with a family member for >4 hours daily in the family residence. Close contact was defined as a contact of >15 minutes within a distance of <2 meters with a COVID-19 case.
Categorical variables were compared by using the chi-squared test while for continuous variables t-test was used. P-values <0.05 were considered statistically significant. A logistic regression analysis was not performed due to the small sample size and the inadequate events per variable, given that small to moderate samples size such as less than 100 usually overestimate the effect measure. 6 The results are presented mainly in a This article is protected by copyright. All rights reserved.
descriptive form, including total numbers, frequencies, or percentages. Analysis was performed by using IBM-SPSS 26 (IBM Corp. Released 2016).
Written consent was not required, given that the data were collected within the frame of epidemiological surveillance. Data were managed in accordance with the national and European Union laws.
We studied 23 family clusters with a median number of 5 (range: 3-7) household members per family. In total there were 109 household members, including 66 adults and 43 children. An adult household member with COVID-19 was the first case in 21 (91.3%) family clusters and a child in 2 (8.7%). Among adults, fathers were identified as first cases in 9 clusters, mothers in 8, both parents in 2 and other relatives in 2. In terms of source of infection of the first case, 11 were community-acquired, 6 travel-associated, 3 healthcare-associated, while in 3 the source of infection could not be identified. In six (26.1%) family clusters children constituted the index cases, including five infants <3 months (clusters 13,14,15,20,22) and one adolescent girl (cluster 9). The median number of days between the onset of symptoms and the date of sample collection for SARS-CoV-2 test was 5 days, with significant difference between children (3.67±2.35 days) and adults (5.92±3.00 days) (p-value-0.019). Table 1 shows the characteristics of household members and the timeline of transmission of infection per family cluster. There was a median of 3 (range: 1-7)
infected persons per cluster. The median attack rate per family cluster was 60% (range: This article is protected by copyright. All rights reserved.
in 12 clusters transmission occurred from an adult to another adult. There was no evidence of child-to-adult or child-to-child transmission, although in 14 clusters there was close contact between infected children and non-infected adult household members. (Table 1 ).
We studied 23 family clusters of SARS-CoV-2 infection that occurred in Greece. We found a median attack rate of 60% (up to 100% in some clusters), which demonstrates the high transmission dynamics of SARS-CoV-2. Attack rates up to 75% have been also reported in other family clusters. 7, 8 In line with studies from Switzerland and China 7,9 , in This article is protected by copyright. All rights reserved.
our study adults accounted for almost all virus importations within families. Of note, five clusters were brought under attention because a young infant became ill and hospitalization was required. In terms of timing, the complete lockdown had an exceptional impact on the onset of family clusters, given that less than one-fifth of identified clusters occurred after that date.
We found no case of transmission of SARS-CoV-2 infection from an infected child to another child or an adult. In a cluster of COVID-19 that occurred in the French Alps, one infected, symptomatic child had many close contacts within three different schools, yet no case of transmission was identified despite an exhaustive epidemiologic and virologic investigation. 10 This may be attributed to the fact that children with SARS-CoV-2 infection more often have an asymptomatic infection or a mild course compared to adults. 5, [7] [8] [9] 11 In our study infected children were significantly more likely to have an asymptomatic infection or a mild disease and a favorable outcome, compared to adults.
The shorter time period that elapsed between the onset of symptoms and testing in children compared to adults, may be attributed to the increased awareness and high rate of healthcare seeking for ill children. It has been reported that patients with severe COVID-19 tend to have a higher viral load than those with mild disease. 12 Although the number of children in the present study was low, it was found that one third (4/12) of asymptomatic children presented high viral load. High viral load has been detected in children with no severe symptoms. 9 In our study, children were either asymptomatic or presented mild symptoms, and only infants presented a moderate form of the disease, and none presented a severe form of the disease. Most probably these findings are related with the immune response in the various age groups rather than the viral load. This article is protected by copyright. All rights reserved.
Limitations of the study were the relatively low number of clusters tested and the fact that the clinical samples were taken from different sites and on different days after symptoms' onset in each patient; thus any conclusion on association of the viral load and severity of the disease cannot be drawn.
In conclusion, the present study provides an insight into transmission dynamics of SARS-CoV-2 within families with children indicating that the prevalent direction of transmission is adult-to-child than child-to-adult. Contact tracing showed that in most cases the adults had contact with a confirmed COVID-19 case, thus, they were the primary source of the family infection. However, since the tracing was based on the dates of the PCR test and given that adults present symptoms in a higher proportion than children, it may happen that more adults have been identified first and the positive children were assessed as secondary cases. Therefore, a conclusion about the index case cannot be drawn with certainty and the role of children in virus transmission needs further investigation.
|
The quality of healthcare, and in particular gastrointestinal endoscopy, may be perceived differently both by healthcare professionals (HCPs) and by patients according to diverse perspectives or concepts [1] . Quality will vary according to different metrics, but there is usually room for improvement depending on the interventions available at different levels of the endoscopy process.
In 2018, the European Society of Gastrointestinal Endoscopy (ESGE) paid special attention to the quality of the environment in which a patient undergoes an endoscopy procedure. As a clinical community, we somehow need to recognize that, beyond endoscopy performance measures, which clearly depend on the endoscopist, the real or perceived quality of an endoscopy is dependent on other domains that start before and end after the procedure itself. In this regard, the service, and in particular the endoscopy team and the way it is managed, is of paramount importance. In fact, the importance of the leadership team and the defined roles within the team is supported by at least one trial showing a positive impact on endoscopy performance after educational interventions on leadership and endoscopy skills [1] .
Human factors play an integral role in teamworking. Teams have a distinct structure but with a "life" that includes goal setting with clear tasks and roles, establishment of working pat-terns, normalization of working styles, and autonomous performance [2] . Not surprisingly, in order to perform adequately, those who lead must embrace some of the key concepts. In endoscopy teams, the key elements relating to patient safety have been identified as human factors, leadership, and communication [3] . Moreover, as training occurs frequently in our endoscopy units, specific team members may change according to rotations. Team processes are interactive, allowing for mutual influence through a series of leading and following interactions [4] . This is opposed to the more traditional concept of leadership found in transformational and transactional leadership theories [5] , where teams have an assigned leader whose main responsibility is the achievement of tasks by motivating and rewarding members to overlook self-interest for the greater good of the team's success.
Despite such recommendations, the relevance of the concept of the "endoscopy team" as the foundation for the efficacy and safety of our activity has been somewhat undermined by the prominent role played by individual endoscopist's skills in key quality indicators, such as adenoma detection rate, withdrawal time, and adequacy of tissue sampling during colonoscopy.
Such (apparent) reluctance to adopt a team-based approach has now been recognized as one of the main reasons for the impact on endoscopy units of the global COVID-19 pandemic, Teams and endoscopy: another effect of the COVID-19 pandemic which started in 2020 [6] . It has been estimated that 5 % -20 % of HCPs in Western endoscopy units have been infected during the COVID-19 outbreak, in line with the estimated 10 % infection rate among HCPs generally. Such rates are at least 10-fold higher than similar rates reported in Asian countries, which may be related to different perspectives regarding the team or community versus individuals.
"What is interesting about this toolkit is that it provides a solid and consistent method for longterm implementation of teamworking in endoscopy. What is "new" is the consideration of the human factor as an organizational resource that must be maximized through structured interaction between different healthcare professionals."
The absolute need to execute the strictest strategies of infection prevention and control (IPC) against COVID-19 has generated the awareness of the importance of the team over the individual. Even a minimal breach of IPC by a single team member may lead to a cluster of COVID-19 infections. Adherence to policies of screening, triage, selection, separation, and monitoring should not be the responsibility of an individual HCP but requires the cooperation of the entire endoscopy team, including nurses, technicians, administrative personnel, and patients: patients must be carefully evaluated for triage and/or testing; COVID-minimization protocols, such as differentiation of endoscopy slots/ rooms according to the risk of infection, must be executed; and availability and proper use of personal protective equipment must be planned. All of this is made more complex by the lack of adequate infrastructure in a substantial proportion of our units. In addition, the profound suppression of any elective endoscopy activity during the 3-month lock-down period has presented the challenge of how to prioritize such burden of examination against an already stressed limited capacity [6, 7] .
In this issue of Endoscopy, Ravindran et al. provide a toolkit to support endoscopy teams during the current pandemic (which is not over yet!) [8] . The toolkit aims to help team processes and provides further evidence on its benefit. The adoption of these (or other) ways to consider nontechnical competences in healthcare and endoscopy teams may well be a benefit that we gain from this crisis. All six dimensions explored in the toolkitplanning and anticipating problems, optimizing communication, team cohesion, flattening hierarchy, sharing task burden, and providing supporthave clear implications and can be used beyond the current pandemic.
In fact, if we return to the current recommendations for endoscopy during the COVID-19 pandemic [6] , which provide direction on indications, how to perform procedures, and how to manage procedures, this guidance will benefit from (more than) daily exercise of leadership. For instance, clear communication, written as well as oral, must occur as in some cases teams are "mirrored;" all professionals must be aligned, and rescheduling of patients, sometimes at diverse timeslots compar-ed with before the pandemic, requires planning and team cohesion; flattening the hierarchy usually poses more burden to leadership but it may bring benefits and sustainable measures.
What is interesting about this toolkit is that it provides a solid and consistent method for long-term implementation of teamworking in endoscopy. What is "new" is the consideration of the human factor as an organizational resource that must be maximized through structured interaction between different HCPs. Such a transparent, proactive, and motivating approach to human interaction has been proven to be extremely effective in reducing the risk of error and improving efficiency within the medical field, such as preventing the administration of incorrect drugs or surgical mismanagements, as well as in other sectors such as the airline industry.
In conclusion, COVID-19 continues to place incredible stress on our endoscopy units. Unexpectedly, it has not impacted the endoscopy skills required to maintain quality assurance, as determined in recent years; however, it has exploited assumptions of our competence to act as a team and our ability to protect ourselves and our patients through shared strategies of IPC. But we have learned the lesson. As we embark on the post-lockdown period, if we can gradually increase our endoscopy capacity, it will be because we implemented a rigorous and structured team approach, with shared leadership and balanced approaches. Hopefully, as the impact of COVID-19 wanes, the strengthened teamworking culture will be a positive long-term effect of the pandemic.
|
Coronaviruses (CoVs), belonging to the family Coronaviridae, are positive-sense enveloped RNA viruses and cause infections in birds, mammals, and humans (1) (2) (3) . The family includes four genera, such as Alphacoronavirus, Betacoronavirus, Deltacoronavirus, and Gammacoronavirus (4) . Two infamous infectious coronaviruses in the genus Betacoronavirus are severe acute respiratory syndrome coronavirus (SARS-CoV) (5) and Middle East respiratory syndrome coronavirus (MERS-CoV) (6) , which have infected more than 10,000 people around the world in the past two decades. Unfortunately, the incidence was accompanied by high mortality rates (9.6% for SARS-CoV and 34.4% for MERS-CoV), indicating that there is an urgent need for effective treatment at the beginning of the outbreak to prevent the spread (7, 8) . However, this cannot be achieved with current drug development or an application system, taking several years for newly developed drugs to come to the market.
Unexpectedly, the world is facing the same situation as the previous outbreak due to a recent epidemic of atypical pneumonia caused by a novel coronavirus (2019-nCoV) in Wuhan, China (5, 9) . Thus, a rapid drug application strategy that can be immediately applied to the patient is necessary. Currently, the only way to address this matter is to repurpose commercially available drugs for the pathogen in so-called "drug-repurposing". However, in theory, artificial intelligence (AI)-based architectures must be taken into account in order to accurately predict drug-target interactions (DTIs). This is because of the enormous amount of complex information (e.g. hydrophobic interactions, ionic interactions, hydrogen bonding, and/or van der Waals forces) between molecules. To this end, we previously developed a deep learning-based drug-target interaction prediction model, called Molecule Transformer-Drug Target Interaction (MT-DTI) (10) .
In this study, we applied our pre-trained MT-DTI model to identify commercially available antiviral drugs that could potentially disrupt 2019-nCoV's viral components, such as proteinase, RNA-dependent RNA polymerase, and/or helicase. Since the model utilizes simplified molecular-input line-entry system (SMILES) strings and amino acid (AA) sequences, which are 1D string inputs, it is possible to quickly apply target proteins that do not have experimentally confirmed 3D crystal structures, such as viral proteins of 2019-nCoV. We share a list of top commercially available antiviral drugs that could potentially hinder the multiplication cycle of 2019-nCoV with the hope that effective drugs can be developed based on these AI-proposed drug candidates and act against 2019-nCoV.
Amino acid sequences of 3C-like proteinase (accession YP_009725301.1), RNA-dependent RNA polymerase (accession YP_009725307.1), helicase (accession YP_009725308.1), 3'-to-5' exonuclease (accession YP_009725309.1), endoRNAse (accession YP_009725310.1), and 2'-O-ribose methyltransferase (accession YP_009725311.1) of the 2019-nCoV replication complex were extracted from the 2019-nCoV whole genome sequence (accession NC_045512.2), from the National Center for Biotechnology Information (NCBI) database. The raw prediction results were screened for antiviral drugs that are FDA approved, target viral proteins, and have a Kd value less than 1,000 nM.
Molecule transformer-drug target interaction (MT-DTI) was used to predict binding affinity values between commercially available antiviral drugs and target proteins. Briefly, the natural language processing (NLP) based Bidirectional Encoder Representations from Transformers (BERT) framework is a core algorithm of the model with good performance and robust results in diverse drug-target interaction datasets through pretraining with 'chemical language' SMILES of approximately 1,000,000,000 compounds.
To train the model, the Drug Target Common (DTC) database (11) and BindingDB (12) database were manually curated and combined. Three types of efficacy value, Ki, Kd, and IC50 were integrated by a consistence-score-based averaging algorithm (13) to make the Pearson correlation score over 0.9 in terms of Ki, Kd, and IC50. Since the BindingDB database includes a wide variety of species and target proteins, the MT-DTI model has the potential power to predict interactions between antiviral drugs and 2019-nCoV proteins.
The 2019-nCoV 3C-like proteinase was predicted to bind with atazanavir (Kd 94.94 nM), followed by efavirenz, ritonavir, and other antiviral drugs that have a predicted affinity of Kd > 100 nM potency (Table 1 ). No other protease inhibitor antiviral drug was found in the Kd < 1,000 nM range. Although there is no real-world evidence about whether these drugs will act as predicted against 2019-nCoV yet, some case studies have been identified. For example, a docking study of lopinavir along with other HIV proteinase inhibitors of the CoV proteinase (PDBID 1UK3) suggests atazanavir and ritonavir, which are listed in the present prediction results, may inhibit the CoV proteinase in line with the inhibitory potency of lopinavir (14) . According to the prediction, viral proteinase-targeting drugs were predicted to act more favorably on the viral replication process than viral proteinase through the DTI model (Table 2 -6) . The results include antiviral drugs other than proteinase inhibitors, such as guanosine analogues (e.g., acyclovir, ganciclovir, and penciclovir), reverse transcriptase inhibitors, and integrase inhibitors.
Among the prediction results, atazanavir was predicted to have a potential binding affinity to bind to (Table 3) and are suggested as potential MERS therapeutics (15) . Recently, approximately $2 million worth of Kaletra doses were donated to China (16) , and a previous clinical study of SARS by Chu et al. (17) may support this decision (17) .
Another anti-HIV drug, Prezcobix of Johnson & Johnson, which consists of darunavir and cobicistat, was to be sent to China (16) , and darunavir is also predicted to have a Kd of 90.38 nM against 2019-nCoV's helicase (Table 3) . However, there was no current supporting literature found for darunavir to be used as a CoV therapeutic.
In many cases, DTI prediction models serve as a tool to repurpose drugs to develop novel usages of existing drugs. However, the application of DTI prediction in the present study may be useful to control Table 3 . Drug-target interaction (DTI) prediction results of antiviral drugs available on markets against a novel coronavirus (2019-nCoV, NCBI reference sequence NC_045512.2) helicase (accession YP_009725308.1). Ritonavir is expressed in canonical and isomeric form SMILES, and * indicates isomeric form SMILES of ritonavir.
|
illnesses, whereas others describe no influence on clinical presentation, [5] [6] [7] [8] [9] considering a non-homogeneous population of children aged less than 24 months.
In a previous study from our group, among 182 infants hospitalized for bronchiolitis, 14.4% had a viral co-infection, RSV + human bocavirus (hBoV). Infants with RSV + hBoV bronchiolitis had significantly higher clinical severity scores at admission and longer hospital stay, than those with human rhinovirus (hRV) and hBoV bronchiolitis. Infants with hRV bronchiolitis had higher eosinophil blood counts than infants with RSV and RSV + hBoV bronchiolitis. 2 Preliminary data showed that in infants with bronchiolitis, single and multiple viral infections manifest with similar clinical illnesses, however infants with a single virus had higher serum C-reactive protein (C-RP) than infants with multiple viruses, higher blood neutrophil numbers, and more frequently manifested fever. 10 Around 40-50% of infants hospitalized for bronchiolitis will have wheezing episodes in the first year of life. Although ample literature describes RSV, hRV, and the presence of higher blood eosinophil counts as factors predisposing to recurrent wheezing after bronchiolitis, 11 Our main aim in this study was to seek possible clinical or serological differences in a large series of infants with bronchiolitis from single and multiple viral infections. As the secondary outcome measure we compared the presence of wheezing at 12-24-36 months after bronchiolitis in infants with single and multiple viral detection.
We reviewed the clinical records for 486 full-term infants (263 boys, median age 2.03 months, range: 0.23-11.17) hospitalized for bronchiolitis in the Paediatric Emergency Department at "Sapienza" University Rome during 12 consecutive annual epidemic periods Bronchiolitis was clinically defined as the first episode of acute lower respiratory tract infection, characterized by the acute onset of cough, tachypnea, retraction, and diffuse crackles on chest auscultation in infants younger than 12 months. 13 Exclusion criteria were prematurity and underlying chronic diseases, such as cystic fibrosis, interstitial lung disease, congenital heart disease, and immunodeficiency.
To infants' parents, we administered a structured questionnaire seeking demographic information including age, gender, breastfeeding history, family smoking habits, family history for asthma, and atopy. On admission, we collected from the records the following clinical and serological data: total white blood cell count, blood neutrophil count, blood lymphocyte count, blood eosinophil count, C-reactive protein (C-RP), sodium serum level, chest radiological findings, and number of days hospitalization. On admission to hospital, each infant was assigned a clinical severity score ranging from 0 to 8 according to respiratory rate, arterial oxygen saturation on room air, presence of retractions, and ability to feed. 2 When clinically necessary, a chest x-ray was obtained at the Emergency Department, before hospitalization.
Categorical variables such as number and percentages, and continuous variables values were expressed as median and range. A χ 2 test was run
In 431 (88.7%) patients, RT-PCR detected only one virus and in 55 (11.3%) infants more than one virus. RSV was more frequently detected virus as a single than as a multiple viral infection (87.3%) ( Figures 1A and 1B) .
The most frequently isolated virus was RSV (365/486, 75.1%), followed by hRV detected in 89 infants (18.3%).
No differences were found in hospitalization stay, clinical severity score, O 2 supplementation and Paediatric intensive care unit (PICU) admission (Table 1) . Infants with co-infection presented less frequently fever (P = 0.05, by χ 2 test). Infants with a single virus infection had a higher serum C-RP level than infants with multiple virus infections (P = 0.012), higher blood neutrophil numbers (P = 0.005, by a non-parametric median test). The questionnaire answers indicated that children with multiple viral infections more frequently had a positive family history of asthma than those with a single virus infection (34.5% vs 22.2%, P = 0.043) ( Figure 1 ). When PCR detected RSV as a single virus it was associated with a higher C-RP level, than RSV associated with other viruses (P = 0.007) ( Table 2 ).
hRV was isolated in 60 infants as a single virus. hRV alone was associated with higher frequency of fever (P = n.s.), higher blood cell counts, higher neutrophils in the peripheral blood and higher C-RP levels than hRV coinfection (P = 0.029, P = 0.008, and P = 0.008). No differences were found for the median clinical severity score, nor for days hospitalization, nor for PICU admission, nor for O 2 supplementation.
In 29 (32.6%) infants hRV was isolated as a viral co-infection ( Figures 1A and 1B) . The most frequent co-infection was hRV + RSV in 20, followed by hRV + MPV in 5, hRV + PI in 1, hRV + AdV in 1, hRV + hBoV in 1, and hRV + RSV + MPV in 1. Infants who had an hRV coinfection more frequently had a positive family history of asthma than those with a single infection (41.4 vs 13.3%, P = 0.003). Infants with hRV alone more frequently had an eosinophil blood count higher than 400/mm 3 than those with a coinfection (18.3% vs 6.9%, P = ns) ( when comparing RSV versus RV versus RSV + hRV. Children with coinfection less frequently had fever than those with single infections (P = 0.012). The co-infection was associated with lower total white blood cell counts (P = n.s), lower neutrophil numbers in the peripheral blood, lower lymphocyte counts, and lower C-RP levels than the single infection (P = 0.008, P = ns, and P = 0.016). Infants with hRV detection alone more frequently had an eosinophil blood count higher than 400/mm 3 (P = 0.021) than those with RSV and RSV + hRV infections (Table 4 ).
In our study, 11.3% of the nasal swabs from the 486 full-term infants hospitalized for bronchiolitis prospectively and consecutively enrolled over 12 epidemic seasons contained a viral co-infection. Although we found no differences in clinical severity scores between infants with single or multiple viral infections, infants in whom RT-PCR detected multiple infections seemed to have a lower inflammatory response than those with single infections. No differences were found between the two groups for recurrent wheezing episodes.
The viral co-infection rate in this study is considerably lower than most reported rates. In children younger than 24 months with bronchiolitis, Chen et al 5 When we analyzed data for emerging and less frequently studied viruses, we detected hBoV more frequently as a co-infection than as infection. They also enrolled preterm infants or infants with underlying chronic disease, in whom they more frequently detected a multiple viral infection. In our previous paper in a smaller sample 2 we found a higher clinical severity score and longer hospital stay in 15 infants with RSV + hBoV, than in infants with RSV, hRV, hBoV infection alone.
Laboratory data analysis highlighted a higher neutrophil count and higher C-RP levels in infants with a single infection than in 21 but further studies are needed to clarify whether these two factors are involved in the clinical presentation of bronchiolitis with multiple viral detection. In accordance with our previous findings, 2 in our series infants hospitalized for hRV bronchiolitis alone more frequently had an eosinophil blood cell count higher than 400/mm 3 , and more frequently a higher prevalence of family history for atopy than in co-infection, but not statistically significant. These findings could suggest that severe hRV infections preferentially manifest in infants predisposed to atopy.
We also evaluated the possible correlation between co-infections and recurrent wheezing, hypothesizing that multiple viral infection might cause a more severe inflammation and this results in a higher recurrence of wheezing episodes. We found no association between co-infection and recurrent wheezing, despite the higher family history for asthma, in patients infected with a mixed than with a single virus.
This previously unreported finding suggests that recurrent wheezing after bronchiolitis depends on the type of virus the infants were found positive to, regardless of whether it was detected as a single or as a multiple virus. We previously reported a positive association between recurrent wheezing after bronchiolitis and hRV detection 11 and in a later study a higher RSV-RNA load. 12 This issue merits clarification in a study with a larger number of cases and a longer follow-up. In conclusion, our study shows that many infants with bronchiolitis (about 11%) in Rome, Italy are infected with multiple virus. No significant differences in clinical severity distinguish bronchiolitis in infants infected with single or multiple virus. Although co-infected infants seem to mount a lower inflammatory response than those without co-infections the immunological response to virus coinfection merits further in vitro studies. Co-infection seems not to influence the clinical presentation of bronchiolitis nor the recurrence of wheezing episodes after three years of follow-up.
We thanks Alice Crossman for her support with the English revision.
The authors have no conflicts of interest to declare.
What is known about this topic
Our study shows that multiple viruses detection (11.3%) in infants with bronchiolitis defined restrictively in Rome (Italy) is lower than compared to literature. No significant differences in clinical severity distinguish bronchiolitis in infants infected with single or multiple virus.
Co-infected infants seem to mount a lower inflammatory response than those without co-infections. Co-infection seems not to influence the clinical presentation of bronchiolitis nor the recurrence of wheezing episodes after three years of follow-up.
ORCID Fabio Midulla http://orcid.org/0000-0001-7476-5266
|
The 1953 discovery of the structure of DNA ushered a revolution in molecular biology, leading to an increased understanding of the central dogma and the subsequent development of invaluable molecular biology techniques, including the polymerase chain reaction (PCR), electrophoresis and automated sequencing, culminating in the completion of the human genome project (HGP) in 2003. The last decade has seen an avalanche of information gleaned in the post-HGP era, such as gene assignation, identification of disease related mRNA biomarkers, as well as the discovery of the importance of single nucleotide polymorphisms (SNPs) and methylated DNA. To date, the vast majority of genotyping techniques require a previous step of amplification, routinely carried out using the robust PCR thermal cycling methodology, and more recently quantitative real-time PCR (qPCR). However, these techniques inherently require the use of thermocycler and a reliable power supply, thus restricting their use to laboratories. To address requirements of amplification for use in low-resource settings, or at the point-of-need, isothermal DNA amplification methods have been developed, including nucleic acid sequence-based amplification (NASBA), strand displacement amplification (SDA), rolling circle amplification (RCA), the loop-mediated isothermal amplification (LAMP), helicase-dependent amplification (HDA), as well as the recombinase polymerase amplification (RPA). The characteristics of these isothermal approaches are summarised in Table 1 and the advantages and disadvantages of each technique has been extensively reviewed elsewhere [1e3] . RPA is remarkable due to its simplicity, high sensitivity, selectivity, compatibility with multiplexing, extremely rapid amplification, as well as its operation at a low and constant temperature, without the need for an initial denaturation step or the use of multiple primers. Overall, RPA positions itself very favourably for widespread exploitation in kits and assays for use at the pointof-care or point-of-need, as well as in affordable, sensitive, specific, user friendly, rapid, robust, equipment-free and delivered (ASSURED) devices, in low-resource settings.
In this review the reader will find the principles of RPA and a complete review of the majority of publications to date, detailing interesting aspects of RPA and diverse RPA approaches, covering different elements of the process, from sample pre-treatment, to amplification and detection strategies.
In 2006 Piepenburg et al. developed the RPA technology using proteins involved in cellular DNA synthesis, recombination and repair, which is currently commercialised by TwistDx (www.twistdx. co.uk) [4] . The RPA process starts when a recombinase protein uvsX from T4-like bacteriophages bind to primers in the presence of ATP and a crowding agent (a high molecular polyethyleneglycol), forming a recombinase-primer complex. The complex then interrogates double stranded DNA seeking a homologous sequence and promotes strand invasion by the primer at the cognate site. In order to prevent the ejection of the inserted primer by branch migration, the displaced DNA strand is stabilised by single-stranded binding proteins. Finally, the recombinase disassembles and a strand displacing DNA polymerase (e.g. large fragment of Bacillus subtilis Pol 1, Bsu) binds to the 3 0 end of the primer to elongate it in the presence of dNTPs. Cyclic repetition of this process results in the achievement of exponential amplification (Fig. 1 ).
Whilst it was initially believed that specifically designed primers of 30e35 bases in length were necessary for RPA, there are several reports demonstrating that normal PCR primers can be used and efficient amplification achieved [5, 6] . Longer primers (up to 45 nucleotides) can be used, but they could lead to secondary structures and potential primer artifacts. It is also recommended to avoid long tracks of guanines at the 5 0 ends while cytidines may be beneficial, whilst guanines and cytidines at the 3 0 tend to improve performance. A GC content below 30% or above 70% is not recommended and, as with PCR primers, sequences that promote primerprimer interactions, secondary structures or hairpins should not be used. RPA can amplify targets up to 1,5 kb but is better suited to amplicons between 100 and 200bp. The primer selection process is thus the same as that used for PCR and involves four steps: choice of target region, design of primer candidates, experimental screening, and, if necessary, secondary and tertiary candidate screening. To date, there is no software available to design primers for RPA. The use of self-avoiding molecular recognition (SAMRs) oligonucleotides can also be employed, where natural bases are replaced by A*, T*, G* and C*, where A* pairs with T, T* with A, G* with C, and C with G, but A* does not pair with T* and G* does not pair with C*, thus avoiding the formation of primer-dimers [7] .
The reaction can operate at temperatures ranging from 22 to 45 C and does not require a narrow temperature control [8e10]; however, most published reports are optimised for temperatures between 37 and 42 C. In order to control the reaction temperature different apparatus can be employed, including incubators, heating blocks, chemical heaters [9] and body heat [11] , and there are also examples of RPA working at ambient temperature in warm areas (above 30 C) [9] .
The crowding agent affects key biochemical process during the RPA reaction. Among them, it prevents the spontaneous recombinase-primer disassembly that occurs in the presence of the single stranded binding proteins needed for the amplification. However, the crowding agent has a negative impact on RPA performance at low target copy levels due to its viscosity, thus impeding the diffusion of reagents through the reaction mixture and inherently increasing amplification time. To minimise this effect, a mixing step is included 5 min after initiation of the RPA reaction, or, alternatively, mixing can be avoided by reducing the total volume of the reaction mixture to 5 mL [12] . An alternative strategy is to continuously mix the reaction solution, where an active matrix for electrowetting-on-dielectric facilitates continuous mixing of 270 nL or 750 nL of RPA cocktail, improving the limit of detection 100 times as compared to the benchtop assay [13] . The use of a phase-guided passive batch microfluidic chamber actuated by a syringe resulted in a reduction of the mixing time from hours to 1 min [14] .
The time required to amplify the DNA to detectable levels inherently depends on the number of starting DNA copies, but 20 min are usually adequate, although amplification times of as low as 3e4 min have been observed [15] . Long incubation times are unlikely to be beneficial in most applications, as for solution phase RPA the recombinase consumes all the available ATP within 25 min.
RPA can be used to amplify double stranded DNA, single stranded DNA, methylated DNA [16] , cDNA generated through reverse transcription of RNA or miRNA [17] (Tables 2e6). There are several reverse transcriptases that have been used with RPA, including Transcriptor R (Roche), Sensiscript R (Qiagen), or MuLV R (Applied Biosystems), with initial reports demonstrating that Transcriptor provides the best performance. cDNA can be produced prior to RPA or in the same reaction [18, 19] and RT-Freeze is also available from TwistDx.
RPA has successfully been used for different kinds of target organisms: bacteria, virus, protozoa, fungi, animals and plants, with diverse samples types, ranging from cultured microorganisms to body fluids (urine, sputum, respiratory washes, nasal, blood, plasma, saliva, vaginal and anal swabs), surgical biopsy specimens, organ tissues (skin, lymphatic nodes, liver, lungs, stomach, kidney), as well as animal and plant products (eggs, shrimps, rice, milk, fruit). Microfluidic devices incorporating a one-step digital plasma separation platform with autonomous parallel plasma separation and sample compartmentalisation for digital nucleic acid amplification have been developed for use with RPA [20] . A valveless microfluidic chip to pre-concentrate bacteria in urine using anion exchange magnetic beads prior to heat lysis has also been reported [21] , as well as an isotachophoresis chip for the extraction of DNA from Listeria monocytogenes in blood samples prior to RPA [22] .
Additionally, RPA has been also reported to indirectly detect non-nucleic acid targets, when aptamers are used as RPA template, and the first example of this was an aptamer based bio-barcode NASBA RNA 2 No 41 60e180 1 Yes Yes Yes SDA DNA 4 Yes 30e55 60e120 10 Yes No Yes RCA DNA/RNA 1 Yes 30e65 60e240 10 No No No LAMP DNA 4e6 Yes 60e65 60 z5 Yes No Yes HDA DNA 2 No 65 30e120 1 Yes No Yes RPA DNA/RNA 2 No 37e42 20e40 1 Yes Yes No assay [23] , which is based on the use of magnetic beads labelled with capturing antibodies and aptamers free in solution that are selective for different epitopes of the same target. In the presence of the target, a sandwich comprising magnetic beads, antibodies, target and aptamer is formed. The sandwich is then magnetocaptured, the solution removed and the bound aptamers are amplified using RPA and detected using fluorescence. Another example of the combination of RPA with aptamer detection was based on the immobilisation of b-conglutin on magnetic beads and following a competition assay, aptamers bound to the magnetic bead immobilised target are eluted, amplified by RPA and detected fluorescently [24] , or via lateral flow [16] .
Amplification can be executed in solution, with both primers in the solution phase, or, alternatively, on a solid phase, when one primer is immobilised on a surface and the other primer is in solution. In a more challenging approach, termed bridge amplification, both forward and reverse primers are immobilised on a surface. However, the vast majority of reports describing RPA exploit solution-phase amplification [25, 26] . In solution-phase, due to the unimpeded diffusion of primers and reaction reagents, amplification kinetics are favoured and the achieved limit of detection is subsequently usually better and amplification is achieved in a faster time than solid-phase. Nevertheless, solidphase and bridge amplification present some advantages, such as the potential for spatially resolved multiplexed amplification or the possibility to couple the amplification with diverse detection techniques including ring resonators [27e29], electrochemical [30e36] and colorimetric detection [5, 6, 30, 33, 37, 38] . Several methods have been developed with solid phase amplification with performances usually inferior to that achieved with solution phase amplification [5,27e30,39] as primer accessibility is more restricted impeding amplification efficiency, and future work will need to focus on strategies to decrease amplification time. Efforts to optimise the surface chemistry of the immobilised primers, exploiting vertical and horizontal spacers to enhance solid phase amplification has been reported [40] . To decrease the reaction time and improve the limit of detection, the surface-immobilised primer can also be introduced in the solution phase in an approach termed hemi-nested asymmetric solid-phase amplification [41e43]. Finally, when both primers are surface-tethered, bridge amplification can take place, but the required reaction time increases and the limit of detection can be compromised. , which scans DNA for homologous sequences (B). The primers are then inserted at the cognate site by the strand-displacement activity of the recombinase (C) and single stranded binding proteins stabilise the displaced DNA chain (D). The recombinase then disassembles leaving the 3 0 -end of the primers accessible to a strand displacing DNA polymerase (E), which elongates the primer (F). Exponential amplification is achieved by cyclic repetition of this process.
Nevertheless, bridge amplification allows multiplexing with a high number of different targets and novel labelling strategies could be exploited to improve the achievable detection limit [44] .
It has been demonstrated that RPA can be carried out directly in serum as well as in the presence of known PCR inhibitors, such as haemoglobin, ethanol and heparin [8] . However, RPA is inhibited by high genomic DNA concentrations in whole blood samples (20e100 ng/mL), but this problem has been reported to be partially solved via the use of a lateral flow-based enrichment of target DNA prior to amplification [45] . Another approach successfully implemented for the analysis of diluted crude DNA extracts from blood or swab samples consisted of heating the sample with AVL buffer and Trizol, followed by centrifugation [46] . RPA can also be carried out directly in urine [47] , pleural fluids [48] , seed powders [10] , milk [49] and stool samples [50] , only requiring heat lysis, direct lysis with nuclease free water or use of the EzWay™ Direct PCR buffer [49] . However, another study found that while 1,25% (v/v) of urine has no impact on amplification efficacy, 10% (v/v) did inhibit amplification when small amounts of target DNA were present in the sample (100 fg), but, this inhibition is not observed when the target DNA concentration is higher (10 pg), even at 10% (v/v) urine [51] . The robustness of RPA in the presence of traditional inhibitors facilitates amplification from crude extracts, which is not achievable using PCR. Whilst RPA pellets are more expensive than PCR reagents, the possibility to eliminate sample pre-treatment simplifies the assay and lowers costs [52] .
Multiplexing with RPA in the same solution is possible but is highly dependent on target sequences, amplicon size and primer design [39] . Primer, probe ratios and concentrations thus need to be carefully optimised for each multiplexing assay. Primers can compete for the recombinase proteins, with one of the reactions (continued on next page) consequently being suppressed [53] . Examples of successful multiplexing RPA in solution, include the detection of different MRSA alleles and an internal control [4] , a fluorescent duplex RPA assay for Staphylococcal Cassette Chromosome mec and an internal control [54] , and a real time fluorescent duplex RPA for C. coli and C. jejuni in chicken products [53] . A multiplex assay of three bacterial pathogens based on solid phase amplification and fluorescent detection using a reverse primer modified with a fluorescent tag has been described [39] , and a similar approach detailed the use of asymmetric solid phase multiplexing RPA for the detection of two human viruses and the bacterium E. faecalis using chemiluminescence detection [43] . Further examples include duplex RPA for cancer genotyping with label free Surface Enhanced Raman Spectroscopy (SERS) detection [55] and triplex RPA for three different plant pathogens using SERS nanotags and modified primers [56] . Finally, a triplex lateral flow assay for the detection of intestinal protozoa was developed, but still requires significant further optimisation to improve the detection limits [57] .
Other reports detail pseudomultiplexing platforms through parallelised single reactions, using foil based centrifugal microfluidic cartridges with stored reagents [49, 58] , digital versatile discs (DVD) [41, 42, 44] , vacuum degassed microfluidic cartridges [59] or polylactic acid/polycarbonate chips [6] .
The reagents necessary for RPA are sold in kits consisting of pellets, rehydration buffer and magnesium acetate, which is used as a reaction initiator and is thus not included in the rehydration buffer, and is provided separately. Pellets are stable for at least one year when stored in a freezer (<À15 C), fridge (2e8 C) and up to 6 months when stored at room temperature (22e28 C) [10] . The preparation of "homemade" pellets containing all the reagents necessary for RPA, including magnesium acetate, primers and the components present in the rehydration buffer has been reported. However, the resulting pellets should be stored at À20 C for optimum sensitivity, and reconstituted solutions can then be stored at 4 C but the achievable limit of detection was 10-fold less when compared with fresh solutions and it was not recommended to store these homemade RPA pellets at 37 C as they degrade and no amplification can be achieved [48] . 2.2.10. Specificity RPA has been described as highly specific, with 100% specificity for the target sequence in most cases. However, RPA has been reported to be dependent on the number and distribution of mismatches in the sequence of closely related DNA molecules, where 1 or more mismatches cannot be differentiated, depending on their distribution. However, more than 1 mismatch at the 3 0 end of primers has been observed to effectively prevent or reduce amplification, which has also been observed for 3 mismatches at both the 5 0 and 3 0 ends, or at the centre of the primer [60] . Whilst this may limit RPA's usefulness in using sequence specific primers, its tolerance to mismatches can be exploited to develop methods to determine the presence of emerging variant pathogens when it is not necessary to discriminate from the wild-type target as exemplified by a method developed to detect HIV-1 proviral DNA, where even 9 changes across the primer and probe binding sites are tolerated by RPA, allowing the detection of different virus strains [61] . However, this tolerance to mismatches can also lead to cross-reactivity as demonstrated by an RPA assay developed to detect the three different genotypes of Chikungunya virus that was [129] observed to have cross-reactivity with another related alphavirus, the O'nyong'nyong virus, based on 4 to 7 mismatches in the primers. A further example is an assay to determine Leishmania donovi that was observed to also amplify other Leishmania spp [62] . However, a method to detect EGFR mutations in lung cancer cells with specificity of just one base mismatch or single nucleotide polymorphism has been developed. Background amplification was reduced via the use of peptide nucleic acids, as PNA-DNA interactions are stronger than DNA-DNA, and one single mismatch is more destabilising than a normal DNA-DNA mismatch, thus improving specificity. However, an extra step is required to allow genomic DNA e PNA hybridization, heating to 99 C and then cooling down to 66 C, moving away from the attractive isothermal nature of RPA [38] . An alternative approach exploiting the use of shorter primers (19e21mer) to decrease the stability between primers and targets and increase specificity towards SNPs has also been reported, where a mismatch in the 3 0 -of the primer was included to increase the specificity. Furthermore, similar to the use of PNA, the use of natural dNTPs vs locked nucleic acids was compared. However, a loss of specificity was observed when multiplexing in the same reaction mixture was pursued, which was attributed to a competition between primers and amplicon [6] .
RPA can be monitored by end point detection (following amplification) or in real time (during amplification) and probes may be used depending on the detection strategy.
Several detection techniques can be used following amplification to determine the presence or absence of targeted nucleic acid [135] sequences. In general, end point detection requires less instrumentation than real-time detection, decreasing the overall cost of the test, and thus could be more appropriate for low resource settings.
The majority of reports detailing end-point detection of RPA products reported to date, rely on lateral flow assays, where results are obtained extremely rapidly in a visual read-out format. 3 different oligonucleotides (2 primers and 1 probe) and the Twist-Amp ® nfo kit are typically used for assay designs compatible with lateral flow strip detection [63] . The probe is recommended to be a 46e52 oligonucleotide modified at the 5 0 end with an antigenic label at the 3 0 end, with a polymerase extension blocking group and an internal abasic nucleotide analogue that substitutes one nucleotide found in the target sequence. The antigenic label is usually a carboxyfluorescein group (FAM), but others, including Alexa fluor488 or digoxigenin are also good candidates [57] . The abasic nucleotide (a tetrahydrofuran residue that replaces a conventional nucleotide, also called a dSpacer), is placed at least 30 nucleotides from the 5 0 end and 15 nucleotides from the 3 0 end. This dSpacer can be cleaved by an nfo nuclease, but only when the probe forms double stranded DNA. The cleavage produces a new 3 0 hydroxyl group in the probe, thus transforming the probe into a primer. In addition to the probe, an opposing amplification primer labelled at the 5 0 -end with another label (e.g. biotin) is required. The second primer used is a conventional primer equidirectional to the probe. The amplicon produced in the presence of the probe and the two primers will include the two labels on one DNA amplicon, ready to be detected in a sandwich assay format by antibodies or antibody/ streptavidin (Fig. 2) . Table 2 summarises reports detailing the combination of lateral flow and RPA. In all cases, the amplification and detection is performed in less than 1 h, achieving limits of detections as low as 1e10 DNA copies. There are also some reports detailing further innovations in lateral flow strip detection such as the use of inexpensive paper, glass fibre, as well as a plastic device in an origami format, which both stored lyophilised enzymes and facilitated mixing steps [64] , and was applied to the detection of Cryptosporidium, with a similar analytical performance to RPA in solution [65] . The same group reported another example of a paper and plastic microfluidic device that was self-sealing and self-contained once all reagents were loaded and only required a heat source, bringing the implementation of nucleic acid testing in a lowresource setting closer to reality [66] . An alternative RPA-lateral flow assay used tailed primers (primer containing a carbon stopper to generate double stranded DNA flanked by single stranded tails), to generate double tailed amplicons. Oligo-functionalised AuNPs were used as reporter probes and oligonucleotides as capture probes in the test and control line, instead of the conventional antigen label and antibody capture approach [67] , decreasing the cost of the strip.
Apart from lateral flow detection, other end point strategies can be exploited as summarised in Table 3 .
Agarose gel electrophoresis is a widely used technique for visualisation of amplification products, but post-amplification it is necessary to purify the amplicons to avoid smeared bands on the gel due to the presence of the proteins and the crowding agent present in the amplification mix.
Bridge flocculation assay is an equipment free assay that provides a binary naked eye visual read out, suitable for low-resource settings. The assay is based on the reversible flocculation of carboxyl-functionalised magnetic beads, which is dependent on the salt concentration, pH and length of DNA. A minimum DNA length of 100bp is needed for the crosslinking, amplicons can be facilely distinguished from primers. To execute the assay, a bead solution is added to the amplification products and following an ethanol wash, the beads are re-suspended in a low pH buffer and a positive answer is obtained if the beads remain flocculated [85e87].
DVDs and low reflectivity DVDs [41] are suitable substrates for the immobilisation of primers for solid phase or bridge amplification, facilitating multiplexing through parallelisation in individual reactors of the DVD. Once amplification is achieved a DVD reader can be used to read out the results in reflection [42] or transmission mode [44] . Additionally, the DVD drives provide centrifugal force to actuate microfluidics for aliquoting and mixing [41] .
Colorimetric detection can also be implemented with RPA. Primers modified with biotin, or biotin modified dNTPs can be used to produce labelled amplicons followed by addition of streptavidin-HRP and subsequently 3,3 0 ,5,5 0 -Tetramethylbenzidine (TMB) and H 2 O 2 , to produce a change in color, the intensity of which can be correlated to the concentration of the amplicons.
In some strategies RPA is carried out in solution and the product captured by magnetic beads [33] or on a microtitre plate following denaturation of the duplex RPA amplicon [37] . Other strategies involve immobilising one of the primers on a substrate and performing solid phase amplification [5, 30] followed by denaturation, hybridization with enzyme labelled reporter probe and optical/ electrochemical detection. In an alternative approach, chemiluminescence detection is achieved via the use of a biotinylated primer, and post-amplification incubation with streptavidinhorseradish peroxidase, luminol and H 2 O 2 [43] .
Fluorescence detection has also been employed in end-point detection approaches. Multiplexing can be achieved exploiting forward primers immobilised onto array spots, and fluorophore modified reverse primers. Following completion of RPA, the amplified product can be spatially resolved and visualised by laser scanner measurements [39] .
Quantum Dot (QD) barcodes are used as an alternative to traditional fluorophores for multiplexed fluorescence detection. One approach consists of polystyrene beads loaded with different types of QDs, which were functionalised with barcodes specifically designed for each of the targets, with one QD type used for each barcode. The beads are then distributed on microfabricated slides and the location of each QD detected using a Smartphone. Following RPA, single stranded DNA is generated and hybridised between the QD-barcode and an Alexa Fluor 647 labelled reporter probe, and the fluorescent signal again measured with the Smartphone. Correlation on the location of each QD-containing bead and the final fluorescent signal facilitated multiplexed detection [88] , and the strategy was validated using clinical samples [89] .
The TwistAmp Exo kit is normally used for real time-RPA with fluorescence detection but it has been used as an end point detection strategy in a multiplexed format, using a low cost, easyto-use, portable microfluidic cartridge system [59] .
Electrochemical transduction for the detection of RPA products via capture of single stranded DNA generated from the amplicon between a surface immobilised complementary probe, and an enzyme labelled reporter probe was described [30, 33, 34 ]. An alternative approach uses forward primers labelled with magnetic beads and reverse primers labelled with gold nanoparticles (AuNPs). The double tagged amplification product is captured by a magnet onto a working electrode and the AuNPs are detected directly through electrocatalytic hydrogen evolution [32] . The use of biotin-dUTPs to produce tagged amplicons was developed, where streptavidin e AuNPs bind to the amplicons on an electrode surface, and gold is oxidized to AuCl 4 À , which can be detected by differential pulse voltammetry [34] . An alternative approach is based on a solid phase RPA assay where one of the primers was tethered on a gold electrode surface and the other primer contained a biotin in the 5 0 , with post-amplification detection achieved using streptavidin-HRP in the presence of a precipitating TMB substrate [31] .
An electrochemical biosensor has also been reported for plant pathogen detection using modified primers to generate double tagged amplicons with biotin at one end and an oligonucleotide overhang at the other. Biotin was used to purify the amplicon using streptavidin magnetic beads, and the capture probe was used to bind to AuNP labelled with a complementary capture probe. Following purification, the amplicons were dropcast on screen printed carbon electrodes and the gold of the AuNP was measured using differential pulse voltammetry (DPV) [35] .
SERS has been exploited for the detection of RPA amplicons. A triplex assay to determine plant pathogens in vegetal tissues was developed using biotinylated reverse primers, tailed forward primers, and AuNPs functionalised with SERS nanotags and oligos complementary to the tails of the primers [56] . The same strategy was also used to develop a rapid multiplexed reverse transcription e RPA (RT-RPA) for the genotyping of prostate cancer tumor and urine samples, using SERS nanotags for a highly sensitive one-pot readout [90] . The same group furthered this work, describing multiplex RT-RPA, with label-free SERS detection, where purified amplicons are incubated with silver nanoparticles prior to SERS detection. The technology was applied to the analysis of 43 patient urinary samples, achieving very good sensitivity, specificity and accuracy [55] .
Schematic representations of different lateral flow assays, biosensors and POC devices developed using RPA are shown in Fig. 3 . The bridge flocculation assay [87] , and lateral flow approaches including a multiplexed lateral flow assay (57) and a disposable plastic and paper device (64) for RPA prior lateral flow assay are particularly suited to point of care devices due to the instrumentless naked eye read-out nature of the methods. Other approaches such as lab in a suitcase [62] , combine all the components needed to perform RPA in situ, using a portable fluorometer for the amplification read-out and portable solar panels and batteries as power sources. Other approaches such as electrochemical solid phase amplification [30] or solid phase amplification on DVDs [41] have potential for multiplexed detection of target at the point of need, but further research is required to reduce the number of steps or to automate the whole process.
RPA can be also monitored in real-time using fluorescent probes and a fluorimeter, facilitating quantification of DNA (Tables 4.1, 4.2 and 4.3). To make this approach accessible to low resource settings, portable and rechargeable fluorimeters have been developed, including the ESE Quant Tube scanner device (Qiagen), Genie III (OptiGene) and the Twista (TwistDx). These fluorimeters can be incorporated in a lab-in-a-suitcase or diagnostics-in-a-suitcase [62, 94] , where all instruments and disposables necessary to perform RPA in-field are packaged in a portable format. Nonspecific intercalating fluorophores such as SYBR Green [4] or Eva Green [23] can be employed for real time detection, but, as in the case of real-time PCR, these dyes cannot discriminate between amplicons and primer-dimer artefacts, thus giving rise to false positive results.
To obviate this problem, the use of specific probes, namely Exo probes and Fpg probes (Fig. 4) are recommended. Other PCR conventional probes such as Taq-Man probes are not compatible with RPA because the Taq-Man polymerases digest the displaced strand during the strand displacing process due to the 5 0 / 3 0 exonuclease activity, thus preventing the DNA amplification.
The Exo probe is an oligonucleotide with homology to the target amplicon that is blocked at the 3 0 to prevent probe elongation. The probe also has a dT-fluorophore and a dT-quencher flanking a tetrahydrofuran residue (dSpacer), which are separated by a maximum of 2e4 bases. The fluorophore signal is thus quenched when the single stranded DNA probe is in solution. However, when the Exo probe is annealed to a complementary DNA target, the DNA repair enzyme Exonuclease III, cleaves the probe at the dSpacer site, producing two probe fragments, separating the fluorophore from the quencher, and thus facilitating the generation of fluorescence [63] .
The Fpg probe, similar to the Exo probe, is an oligonucleotide with homology to the target amplicon that is blocked at the 3 0 to avoid probe elongation, and additionally contains a quencher and a fluorophore, separated by 4e5 nucleotides (7 at maximum). The quencher is placed at the 5 0 of the probe and the fluorophore is linked to an abasic nucleotide through a CeOeC linker, termed a dRgroup. In the absence of target, the fluorophore signal is quenched but when the Fpg probe is annealed to a complementary DNA target, the fpg enzyme cleaves the probe at the dR position, liberating the fluorophore, resulting in emission of fluorescence [63] .
It has been observed that the Exo probes provide higher sensitivity than nfo probes [61] , however, Exo probes can result in the exonuclease mediated degradation of DNA and therefore are not compatible with agarose gel electrophoresis [25] . qRPA can be achieved if reactions are protected from heat and light to avoid loss of enzyme activity and the photobleaching of probes, and magnesium acetate should be added immediately prior to fluorescence detection [95] .
Whilst real-time assays are routinely carried out in Eppendorf tubes, the use of a SlipChip platform for amplification has been described. The chip consists of plates clamped together and contains 3 lanes used to place sample, RPA master mix and magnesium acetate, separately. Once each lane is loaded, the plates can slip in order to mix all the components, and amplification is followed using a real time machine [96] . The use of a programmable digital microfluidic platform based on an active matrix electrowetting-on-dielectric (AM-EWOD) for real time detection has also been described. The automated platform incorporates 16,800 electrodes that can be controlled independently to simultaneously manipulate several droplets of around 45 nL. The system allows the continuous movement and heating of droplets achieving an improved detection limit (>2 orders of magnitude) as compared to benchtop assays [13] . In another report, a commercial 3D printer was modified and coupled with blue LEDs and a mobile phone camera to construct a robotic device for DNA/RNA extraction, amplification and real time detection in a multiplex format (up to 12 samples), and applied to ZIKA spiked urine samples [97] . An alternative approach combining ligation based assays with qRPA for the detection of fusion gene mRNAs was described. Right hand and left hand side ligation probes were designed to contain universal reverse and forward primer specific sequences incorporated at either side of the ligation site. Following ligation, the probes are amplified in separate reactions, and the signal due to intercalation of the SYTOQ fluorescent dye was measured, allowing simultaneous detection of three targets in 60 min [98] .
Real time detection is mainly restricted to fluorescence detection, however, there are some reports of alternative real-time strategies (Table 5) . Real-time, label-free and highly sensitive detection of RPA can be achieved using ring-resonator technology [29] , where primers are immobilised on a silicon ring resonator and the shift in the resonant wavelength is measured continuously during amplification. This approach has been demonstrated to have a sensitivity 100 times higher than benchtop RPA and conventional PCR methods and can be used to distinguish single point mutations [27] . Further examples of alternative real-time detection strategies include a label-free method that combines a dimethyl adipimidate supported on a thin film for the extraction and purification of DNA, and solid phase-RPA integrated with a Mach-Zehnder interferometer for combined amplification and detection [132] .
Fluorescence is the principle transduction technology that has been used to develop methods for absolute quantification in which the sample and reaction components are compartmentalised into several individual and parallelised reactions so that each reaction contains one or no copy of the target DNA ( Table 6 ). The compartmentalisation approaches developed include digital plasma separation [20] , centrifugal step emulsification [133] , SlipChip technology [134] and picoliter array based technology [135] . In digital plasma separation, the compartmentalisation and plasma separation is carried out passively using microfluidic chips with a microcliff structure that is actuated by passive degassed driven flow, inertia and sedimentation [20] . In centrifugal step emulsification, the compartmentalisation is achieved in droplets, produced by centrifugation using an inlet chamber. One channel is connected to a chamber by a nozzle and droplet production, and read-out of the amplification with a Smartphone-based device takes place in the same chamber [133] . As described previously, in SlipChip technology two plates are clamped together to create channels and wells for the creation of individual compartments [134] , and finally, picoliter array based chips on fabricated silicon and passivated with methoxy-PEG-silane agent facilitates the performance of up to 27,000 reactions in picoliter sized wells [135] .
RPA is a relatively new isothermal amplification technology that has experienced an exponential growth in terms of publications, popularity and applications since its first report in 2006. The majority of reports since then have focused on a wide range of different applications of RPA, but there are an increasing number of publications that detail methodologies to improve the performance of RPA and to further its capabilities. RPA is remarkable among isothermal amplification techniques due to its simplicity, high sensitivity, selectivity, compatibility with multiplexing, rapid amplification, as well as its operation at a low and constant temperature, without the need for an initial denaturation step or the use of multiple primers. RPA can amplify as low as 1e10 target copies in less than 20 min even in the presence of some known PCR inhibitors or in crude extracts. The technique has been successfully used to amplify both RNA and DNA targets in different kinds of organisms, in both the solution and solid phase. A wide variety of detection strategies are compatible with RPA, and some of these have been tested with real samples with performances similar or better than PCR. Table 1 outlines the properties of RPA as compared to other isothermal amplification techniques. Whilst most other isothermal amplification methods operate between 30 and 65 C, RPA takes advantage of enzymes and crowding agents to work at a low temperature ranging between 37 and 42 C and there is no requirement for a tight control of the temperature within this range, which is a particularly positive attribute not possessed by the other isothermal techniques. RPA, as well as NASBA and HDA do not require an initial denaturation step to generate ssDNA from the dsDNA target, in contrast to SDA, RCA and LAMP, highlighting its suitability for use in the field. In addition, RPA, in common with NASBA, RCA and SDA only requires 2 primers per target, which could position it to be more compatible with multiplexed amplification. RPA is also a very rapid method of amplification, markedly faster than other isothermal amplification methods and even though 15e25 min is recommended, efficient amplification can even be achieved in less than 5 min, depending on the target. RPA reagents are provided in a lyophilised format and are stable at ambient temperature for at least 6 months, whilst the reagents for all the other isothermal techniques require refrigeration, and this again positions RPA as being highly suited to implementation in point-of-need/care and ASSURED devices.
However, RPA does have some limitations, the principle one being that RPA kits are only sold by one company, which could have an impact on pricing, and the user also has limited flexibility in the kit formulation and whilst tailor-designed kits are available (e.g. without polymerase, without dNTPs), they are costly at low volumes. RPA normally requires purification/protein digestion following amplification, or will result in smearing or impaired flow in the cases of agarose gel electrophoresis and lateral flow, respectively. RPA, like PCR can be inhibited by high concentrations of genomic DNA, and as is the case with real time PCR, the use of SYBR Green [4] or Eva Green [23] cannot discriminate between amplicons and primer-dimer artefacts. Furthermore, real-time PCR conventional probes such as Taq-Man probes are not compatible with RPA because the Taq-Man polymerases digest the displaced strand during the strand displacing process due to the 5 0 /3 0 exonuclease activity, thus preventing amplification. In fact, realtime amplification using RPA is not straightforward as it is based a time threshold instead of a cycle threshold, which is dependent on RPA kinetics. This time threshold is dictated not only by the initial target concentration but also by the temperature and mixing step. It is advisable to slow down the RPA reaction rate in order to have a better control during real-time RPA and this can be achieved by decreasing the magnesium acetate concentration. However, as the time of adding the magnesium and the effectiveness of mixing will have a strong impact on RPA kinetics, and ideally real-time RPA should be completely automated. Whilst RPA seems particularly suitable for multiplexed amplification, this requires extensive optimisation of primer concentrations as primers compete for the recombinase proteins and ratios of each need to be tested experimentally as primers for one target can suppress the amplification of another target. Furthermore, to date there is no software available for the design of primers specific for RPA and this can result in lengthy optimisation of the primer sequences. Different DNA targets, even with the same GC content, primer melting temperature and amplicon length, can be amplified with extremely different efficiencies and the basis for this is still not well understood.
Given the tremendous advantages of RPA, as well as some of the current limitations of the technique, it can be expected that there will be exponential growth in the applications of RPA as well as improving and extending its performance. Recently RPA reagents have become available in a liquid format and it can be envisaged that increased flexibility in the kit formulation will allow an improved optimisation of assay conditions and facilitate a better understanding of the RPA mechanism. Currently "optimisation" depends on using a pellet, half pellet, quarter pellet etc., and as mentioned above, different targets are amplified with different efficiencies, and whilst RPA does appear to be particularly amenable to multiplexed detection, quite a laborious optimisation is currently required, but with more control of the amplification mix, this could become more simplified. Indeed, with the increasing interest in the simultaneous detection and sometimes also quantification of biomarkers, it is expected that there will be an exponential increase in the number of reports detailing parallelised amplification in solution-phase, in separate reservoirs in microfluidic systems, or on separate electrodes of an electrode array for solid-phase amplification, where multiplexing can be facilitated by spatial separation. Real-time RPA also requires extensive optimisation to truly control the amplification rate and to define properly the time threshold. To date real-time RPA has been achieved using fluorescent and ring-resonator detection, and other detection methodologies may further enhance the possibilities of real-time RPA, possibly even achieving highly multiplexed real-time quantitative RPA. The use and optimisation of RPA for differentiating single base differences (SNPs/mutation) or for the amplification of a family of species needs to be further explored as very few reports addressing this theme exists to date.
The focus of a large number of RPA related publications details the use of RPA in lateral flow formats, but to date there is no report of a completely integrated paper analytical diagnostic device, which only requires end-user addition of blood/saliva/urine/food/ environment sample. Innovative approaches for the application of temperature to facilitate efficient execution of RPA at the point-ofneed/care have been reported and cost-effective, efficient solutions are available. As yet RPA has not been approved by the FDA and is destined for research only applications and it can be expected that the technique will be validated and approved for medical diagnostics in the near future, facilitating the true implementation of RPA in lateral flow assays for companion diagnostics or as ASSURED devices in low resource settings.
In summary, RPA is a fascinating isothermal amplification technique that has already garnered a huge amount of attention due to its very attractive properties, having widespread application. Whilst to date the majority of interest has been the use of RPA in diverse areas, there is expanding interest in a deeper understanding of the underlying mechanisms of the technique, with the objective of a complete optimisation for real-time and multiplexed applications. RPA is exploited for laboratory-based analysis, portable analysis in laboratory-in-a-suitcase, analysis at the point-of-need/care with biosensors, lateral flow assays and microfluidic devices, and its exploitation in a range of commercial devices for molecular diagnostics, food quality control, environmental analysis and detection of biowarfare agents, amongst others, can clearly be anticipated in the near future.
|
The majority of biomolecules contain heterocyclic fragments in their structures. Synthetic heterocyclic compounds can imitate their natural analogs and interact with biological targets, exhibiting a broad spectrum of biological activity, therefore heterocyclic compounds have a broad range of applications as pharmaceutical agents. As many as 98% of synthetic drugs contain cyclic structural motifs in their molecules, while 87% of synthetic drugs represent heterocyclic derivatives. 1 Recently, major medicinal chemistry efforts have been motivated by the search for new antiviral drugs. Until this year, in all 93 antiviral drugs have been approved for clinical use. These drugs are used to treat only 9 types of viral infections -those that are caused by the human immunodeficiency virus (HIV), hepatitis В virus, hepatitis С virus, herpes virus, influenza virus, cytomegalovirus, varicella-zoster virus, respiratory syncytial virus, and human papillomavirus. 2 Among the antiviral drugs, 79 molecules contain a heterocyclic moiety. For the majority of viral diseases, there have been no drugs approved, thus an active search is continuing for drug lead compounds that could be developed into new pharmaceutical agents. 2b,3 For example, 105 antiviral compounds were at various stages of clinical trials in 2018, of which only 5 were approved for clinical use in the years 2018-2019. 2b-d The development of new antiviral drugs is complicated by the great genetic diversity of viruses and, consequently, the wide variety of molecular targets for antiviral therapy. The reproductive cycle of many viruses involves a group of proteins having the common function of increasing the membrane permeability. These proteins are known as viroporins. A typical characteristic of viroporins is their small size, usually up to 120 amino acid residues, the presence of one or several hydrophobic sequences and the ability of self-oligomerization with the formation of transmembrane channels that can transport small ions. 4 Viroporins have a key role in the process of viral replication: alteration or removal of the gene coding the respective viral protein results in a decrease or complete absence of virulence. 4e In recent years, viroporins have been gaining importance as drug targets, offering potentially new mechanisms for inhibiting the viral reproduction. The interest of researchers has been attracted by their small size, enabling the study of these proteins both by experimental methods of structural characterization and through the computational approaches available to medicinal chemists. Molecular dynamics calculations provide a relatively fast structural modeling of such small proteins, as well as their complexes with small molecule ligands. Virtual design is especially widely applied to the development of new inhibitors for the М2 ion channels of the influenza virus and р7 channels of hepatitis C virus (HCV p7), since their three-dimensional models have been determined, along with the mechanism of their action and details of interaction with small molecules (Fig. 1) . 5 Currently, there are only 29 viruses known that rely on ion channels in their reproductive cycle and simultaneously present risk to human health. The most dangerous of these viruses, requiring a focused search for therapeutic agents according to the current data from WHO (the World Health Organization), are the Ebola virus, the SARS and MERS coronaviruses, and the novel 2019-nCoV coronavirus. 5m Genomic analysis of the latter has revealed that it also contains genes encoding the viroporins characteristic for other coronaviruses. However, detailed studies have been performed only with regard to 8 viruses, for which ion channel inhibitors have been found ( Table 1) .
One of the first small molecule antiviral drugs with specific action was amantadine, which was approved as a medication for the prevention of influenza in 1966. 31 The efforts to characterize its mode of action led to the discovery of М2 ion channel in the influenza А virus. 32 Further studies led to the definition of a detailed mechanism for the operation of this channel and allowed to identify principles for its inhibition. It is currently believed that the М2 protein is essential to the influenza virus for the purpose of removing the capsid shell after entering the cell. The function of the channel is to transport protons into the virus particle, leading to a decrease in pH in the interior and a change in capsid protein conformation. This process results in a disruption of the capsid shell and release of viral RNA. 33 The structure of М2 protein consists of 97 amino acid residues and is divided into 3 segments: the extracellular domain (amino acid residues 1-23), intracellular domain (amino acid residues 47-97), and the transmembrane domain (amino acid residues . Four molecules of this protein are associated in a tetrameric channel. 34 The amino acid residues most important for the operation of the channel are histidine 37 and tryptophan 41, which form a "gate" controlling the channel. Upon protonation of the histidine 37 residue, the internal cavity of the channel expands due to electrostatic repulsion between neighboring amino acid residues, resulting in a proton flow through the channel (Fig. 2) . 35 The mechanism of blocking the channels formed of М2 protein subunits involves the entry of inhibitor molecule into the channel cavity and physical obstruction of the cavity. 36 According to the data of WHO, all influenza virus strains that have been identified by year 2011 showed resistance to amantadine, 37 therefore an active search for M2 ion channel inhibitors is ongoing, most often focused on heterocyclic compounds. The loss of effectiveness observed for small cage-like amine molecules has been explained by the absence of the wild type М2 ion channel (M2-WT) in the structure of the viral strains currently circulating in the population. The most common mutant type М2 ion channels are S31N, V27A, and L26F. 38 Nevertheless, the M2-WT ion channel is still used as a convenient laboratory model for preliminary assessment of antiviral activity. A series of compounds exhibiting strong inhibitory activity against the M2-WT ion channel are presented in Table 2 . Among nonaromatic structures, high activity was observed for five-and six-membered saturated heterocycles, which in most cases contained nitrogen atoms. Thus, half maximal inhibition was observed at microand submicromolar concentrations of pyrrolidines 1-5 and piperidines 6-8 fused to an adamantane moiety. The related pyrrolidin-2-ones 9-10 and piperidin-2-one 11 also exhibited a strong inhibitory activity. Compounds with strong inhibitory effects were also found among azapropellanes 12-21 containing a pyrrolidine ring. The spiro-fused pyrrolidine 22, piperidines 23, 24, piperazine 25, 1,3-dithiane 26, and thiazolidine 27 had half maximal inhibitory concentrations in the micromolar range, while in the case of thiazoline derivative 28 -in the nanomolar range. 3-Oxabicyclo[3.3.1]non-6-ene derivatives 29-31 showed a pronounced inhibitory activity against the М2 channel.
New research efforts aimed at finding ion channel inhibitors among aromatic heterocyclic compounds were started in year 2013. As a result, a range of structures were discovered, which contained an isoxazole ring (compounds 33, 34), imidazole ring (compound 35), or a pyrimidine ring (compound 37). These compounds showed inhibitory activity in the micromolar concentration range. Further structural optimization on the basis of studying the structure-activity relationships allowed to obtain a series of five-membered heterocyclic compounds 38-51 that showed inhibitory activity at submicromolar concentration range. Sufficiently active compounds were also identified among pyrazole derivatives (compound 52), thiophene derivatives (compound 53), selenophene derivatives (compound 54), and thiazole derivatives (compound 55). Two substituted tetrazoles 56, 57 exhibited antiviral activity in the micromolar concentration range. Among six-membered aromatic heterocycles, inhibitory activity against the М2 ion channel was observed in the case of pyridine derivatives (compounds 58, 64) and pyrimidine derivatives (compounds 59, 60) . Other examples of compounds with inhibitory activity against the М2 ion channel included New inhibitors of the М2 ionic channel have been historically developed on the basis of structural similarity with adamantane derivatives. It was assumed that effective binding to the channel requires the presence of a lipophilic molecular framework along with a basic functional group. Many of the compounds shown in Table 1 (compounds 1-26) were discovered according to this approach, however, new types of heterocyclic molecules with activity against the influenza virus have been discovered via screening of natural compounds and libraries of synthetic compounds (compounds 29-37), as well as molecular docking simulations (compounds 38-58) . Many of these discovered compounds interact not only with the М2 ion channel of the wild type influenza virus, but also with the channels of mutant strains. The highest activity was observed in the case of compounds 18 and 44-51, the half maximal inhibitory concentration of which was in the submicromolar range. The difference between the М2 ion channels in the mutant strains from those of the wild type virus was in the considerable weakening of the hydrophobic packing between the N-terminal ends of the transmembrane helices, providing a less rigid, more dynamic tetramer structure. The less constrained packing prevented binding of the inhibitor due to weakening of the hydrophobic contacts inside the pore. For this reason, the affinity of new ligands can be improved by adding new types of contacts (electrostatic interactions and hydrogen bonds, as well as π-π stacking and hydrophobic interactions) between the amino acid residues on the inside of the pore and the heterocyclic moieties of the inhibitors (Fig. 3) .
Few inhibitors are known for other virus channels, compared to the М2 ion channel of influenza virus. Among them, there is relatively more information available about the p7 ion channel of hepatitis C virus, for which computersimulated models have been constructed. Molecular docking studies have revealed a series of compounds (hexamethyleneamiloride 37, BIT-225 70, benzimidazolin-2-imine 71, indolin-2-one 72, 4,5-dihydropyrazin-2-one 73, piperidine 74, imidazole 75, pyrrolidine 76, and tetrazole 77, Table 3) , which have been shown to effectively block its functions, inhibiting viral replication. The ion channels of other viruses have been considerably less studied, therefore the search for their inhibitors relies on screening of compound libraries. The scarcity of literature published regarding these studies prevents us from evaluating the structure-activity relationships for ion channels of other viruses, taking into account their substantially different molecular architecture. 54 Hexamethyleneamiloride 37 has been characterized with strong activity against the Е ion channel of coronaviruses and moderate activity against the Vpu ion channel of HIV. Pyrazole derivative 70 (BIT-225) has shown strong activity against the Vpu ion channel of HIV. Some flavonoids exhibit noticeable activity against the ORF3a ion channel of coronaviruses, with the highest activity determined in the case of juglanin 78. Analogs 79 and 80 of the alkaloid lycorine showed strong inhibitory activity against the 2K ion channel of the Dengue virus. A range of alkylated iminosaccharides 81, 82 can effectively block the reproduction of human papillomavirus via binding to its ion channel Е5. Pyronin B (83) can quite effectively disrupt the function of SH ion channel of the respiratory syncytial virus (Table 3) .
Thus, heterocyclic fragments, known as the most common pharmacophores, can strongly bind to proteins, including viroporins, effectively blocking their functions. Even though this review article covers only compounds with proven activity against the ion channels of viruses, other compounds with pronounced antiviral activity but unknown mechanisms of action can probably act as ligands for these targets. At the same time, viroporins are small proteins with relatively simple molecular architecture. Their structures can be relatively easily determined both by experimental and computational methods, enabling effective virtual search for new lead compounds in drug development. The genome of a wide array of dangerous viruses encodes proteins that function as ion channels, for which effective inhibitors remain unknown. Obviously, the
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.