Persistent Detection and Infectious Potential of SARS-CoV-2 Virus in Clinical Specimens from COVID-19 Patients

The Severe Acute Respiratory Syndrome Coronavirus (SARS-CoV-2) that emerged in December 2019 as the causative agent of Coronavirus 2019 (COVID-19) and was declared a pandemic by the World Health Organization in March 2020 has several distinctive features, including extensive multiorgan involvement with a robust systemic inflammatory response, significant associated morbidity and mortality, and prolonged persistence of viral RNA in the clinical specimens of infected individuals as detected by Reverse Transcription Polymerase Chain Reaction (RT-PCR) amplification. This review begins with an overview of SARS-CoV-2 morphology and replication and summarizes what is known to date about the detection of the virus in nasal, oropharyngeal, and fecal specimens of patients who have recovered from COVID-19, with a focus on the factors thought to contribute to prolonged detection. This review also provides a discussion on the infective potential of this material from asymptomatic, pre-symptomatic, and convalescing individuals, to include a discussion of the relative persistence and infectious potential of virus in clinical specimens recovered from pediatric COVID-19 patients.


SARS-CoV-2 Entry and Replication
SARS-CoV-2 is a positive-sense single-stranded RNA virus. Its genome is~30 kb which, like those of other coronaviruses, consists of genes for four structural proteins including surface (S), envelope (E), membrane (M), and nucleocapsid (N) proteins, as well as six accessory proteins, encoded by genes on open reading frames ORF3a, ORF6, ORF7a, ORF7b, and ORF8. Additional open reading frames encode a host of nonstructural proteins, including those that facilitate replication and transcription and others that enable the virus to evade the host immune response [24]. Like all viruses, SARS-CoV-2 relies on the replicative machinery of a vulnerable host cell to make copies of its genome, a process that begins with cell binding and entry. Attachment to and entry of SARS-CoV-2 into susceptible cells is mediated by the spike protein, which consists of two subunits: S1 and S2 [25]. The S1 subunit binds to angiotensin converting enzyme, ACE-2, a receptor on the host cell that is distinct from ACE-1 (the enzyme targeted by ACE inhibitors such as lisinopril, enalapril, and ramipril). The ACE-2 protein is widely distributed throughout the human body, and most abundantly expressed in the lung type II alveolar cells, enterocytes of the gastrointestinal tract, endothelial cells, smooth muscle cells, cortical neurons, and glial cells [26]. As S1 subunits bind to the membrane-bound ACE-2 protein molecules, the virus becomes enveloped in an endosome. Cell entry then continues by either of two processes. In the first, transmembrane protease, serine 2 (TMPRSS2) cleaves the S1 subunits from the S2 subunits and cleaves the ACE-2 proteins. The endosome is then endocytosed, and the virus is subsequently released into the cytoplasm after acidification or through the proteolytic action of cathepsins. Alternatively, TMPRSS2 effects an irreversible conformational change of the S2 subunits, and the virus fuses to the cell membrane.
Entry of the virus into the host cell and release from the endosome is followed by uncoating of the virus and release of viral RNA into the cytoplasm where it undergoes translation. The translation products include the replicase polyproteins pp1a and pp1ab that undergo further cleavage into smaller proteins including RNA-dependent RNA polymerase, helicase, and nonstructural proteins nsp3, nsp4, and nsp6. During translation, ribosomal frame shifting generates genomic and sub genomic moieties by discontinuous transcription. The coronavirus replication-transcription complex is then anchored to the intracellular membrane of the endoplasmic reticulum by nsp3, nsp4, and nsp6 to form double membrane vesicles. RNA-dependent RNA polymerase and helicase then drive the synthesis of sub genomic RNA from which structural and accessory proteins are produced, including the S, M, and E proteins, which are inserted into the endoplasmic reticulum and then transported to the endoplasmic reticulum-Golgi intermediate compartment (ERGIC). In contrast, the N protein binds the viral genomic RNA in the cytoplasm to form the nucleocapsid. The virions are then assembled in the ERGIC and released in vesicles from the cell by exocytosis (Figure 1) [27].

Duration of SARS-CoV-2 RT-PCR Positivity and Factors Associated with Prolonged Clinical Shedding
Prolonged clinical shedding has been described for several respiratory viruses, including SARS-CoV [28] and MERS-CoV [29], and the persistence of SARS-CoV-2 in respiratory secretions, as detected by RT-PCR amplification, was described early in the pandemic [30,31]. One of the earliest studies of infected patients reported the median and absolute duration of RT-PCR positivity to be 20 days and 37 days, respectively [32]. However, more recent reports describe the persistence of the virus for more than 60 days with a median duration of RT-PCR positivity of more than 30 days [33]. There are several factors that appear to correlate with prolonged clinical shedding, including severe illness. In one study, the median duration of the virus in respiratory specimens from patients with severe disease (21 days, 14-30 days) was significantly longer than in patients with mild disease (14 days, 10-21 days; p = 0.04) [34]. These findings are consistent with reports of longer clinical shedding times in ICU patients than in patients not in an ICU [35]. Moreover, the mean viral load in patients with severe illness appears to be significantly higher (around 60 times higher in one study) than that of patients with mild illness, suggesting that higher viral loads might be associated with a more severe clinical course [36]. However, at least one other study found no significant correlation between severity of illness and viral load [37]. Additionally, one study found an inverse correlation between disease severity and duration of RT-PCR positivity [38].

Duration of SARS-CoV-2 RT-PCR Positivity and Factors Associated with Prolonged Clinical Shedding
Prolonged clinical shedding has been described for several respiratory viruses, including SARS-CoV [28] and MERS-CoV [29], and the persistence of SARS-CoV-2 in respiratory secretions, as detected by RT-PCR amplification, was described early in the pandemic [30,31]. One of the earliest studies of infected patients reported the median and absolute duration of RT-PCR positivity to be 20 days and 37 days, respectively [32]. However, more recent reports describe the persistence of the virus for more than 60 days with a median duration of RT-PCR positivity of more than 30 days [33]. There are several factors that appear to correlate with prolonged clinical shedding, including severe illness. In one study, the median duration of the virus in respiratory specimens from patients with severe disease (21 days, 14-30 days) was significantly longer than in patients with mild disease (14 days, 10-21 days; p = 0.04) [34]. These findings are consistent with reports of longer clinical shedding times in ICU patients than in patients not in an ICU [35]. Moreover, the mean viral load in patients with severe illness appears to be significantly higher (around 60 times higher in one study) than that of patients with mild illness, suggesting that higher viral loads might be associated with a more severe clinical course [36]. However, at least one other study found no significant correlation between severity of illness and viral load [37]. Additionally, one study found an inverse correlation between disease severity and duration of RT-PCR positivity [38].
As with several other studies, after multivariate analysis disease severity was not an independent risk factor for viral persistence. A comparison of the median duration of SARS-CoV-2 RT-PCR positivity in respiratory specimens reported in these studies is depicted in Figure 2.

Persistent Detection of SARS-CoV-2 in Feces
Although SARS-CoV-2 is primarily transmitted via respiratory secretions and COVID-19 generally manifests as pneumonia, the widespread distribution of ACE-2 receptors makes COVID-19 a systemic infection. Enterocytes of the small bowel abundantly express the receptor on their brush borders and may become infected by the movement of virions from the airways into the gastrointestinal tract (e.g., by expectoration or mucociliary clearance) [42]. In a mechanism elucidated by Wrapp et al. [43], human proteases such as TMPRSS2 (transmembrane protease serine 2) and furin then cleave the polybasic bonds between the S1 and S2 subunits of the spike protein, resulting in a separation of the two into a pincer-like configuration. The S1 subunit then binds the peptidase domain of ACE-2 and the S-2 subunit effects fusion with the cell membrane, which is then followed by viral endocytosis [43]. This is likely clinically relevant, because necropsied mice infected with SARS-CoV-2 demonstrate enterocyte desquamation, edema, small vessel dilation, lymphocyte infiltration, as well as mesenteric lymph node hemorrhage and necrosis [42]. Moreover, COVID-19 patients may have gastrointestinal complaints including diarrhea, either preceding, along with pneumonia [44], or as a sole clinical manifestation of SARS-CoV-2 infection [45]. Whether this is due to direct cytopathic effects, a systemic inflammatory response, or some other mechanism (e.g., disruption of trans-membrane transporters or of the gut microbiome) has yet to be elucidated.
In addition to the clinical implications, gut involvement by SARS-CoV-2 raises the role of fecal shedding in transmitting the virus. In one systematic review of 55 studies (1348 patients), nearly half of collected stool samples had detectable virus. Moreover, the duration of fecal RT-PCR positivity (median 19 days) was significantly longer (p < 0.001) than that of respiratory RT-PCR positivity (median 14 days) [46]. In another meta-analysis, more than half of fecal samples had detectable virus for up to 70 days after the onset of symptoms and as long as 33 days (mean 12.5 days) after the virus was no longer detected in respiratory samples [47]. Despite the persistence of virus in the feces of infected individuals as detected by RT-PCR amplification, the infectious potential of fecal samples is uncertain. At least one study found quantitative titers to be below those of nasopharyngeal fluids and generally lower than those of enteric viruses such as norovirus and adenovirus [48]. Nonetheless, the presence of virus in the feces of COVID-19 patients raises concern for fecal-oral transmission, especially in situations in which adequate sanitation infrastructure is lacking, and the presence of SARS-CoV-2 in wastewater has been reported [49,50]. In this regard, young children may represent an important demographic, given their proclivity for unhygienic practices such as not washing their hands after stooling, sucking on their fingers, etc., and prolonged fecal RT-PCR positivity by pediatric COVID-19 patients has been described [51][52][53].

Persistent Detection of SARS-CoV-2 in Other Bodily Fluids
Extra-pulmonary tissue tropism has been documented for several coronaviruses including severe acute respiratory syndrome virus (SARS-CoV-1) and Middle East respiratory syndrome corona virus (MERS-CoV) [54,55] and it is increasingly apparent that SARS-CoV-2 is also organotropic, causing multi-system illness. Therefore, it is not surprising that acute kidney injury has been reported in more than a quarter of critically ill COVID-19 patients [56,57]. Although this may derive in part from hemodynamic instability and cytokine storm, several studies suggest a role for viral-mediated renal cytotoxicity; and in postmortem studies of COVID-19 patients, coronaviruses were identified in all kidney compartments examined, with apparent preferential targeting of glomerular cells [58]. In one meta-analysis of thirty studies in which the urine of COVID-19 patients was tested for virus using RT-PCR, the incidence of viruria was 8%, compared to 21.3% and 39.5% for blood and stool, respectively [59]. Although the infectious potential of SARS-CoV-2 in urine has yet to be elucidated, viable virus has been recovered from the urine of some COVID-19 patients, suggesting a possible role for genitourinary transmission [60], and at least one study has looked at the aerosolization of the virus from urinal flushing [61]. To date, however, viral RNA but not viable virus has been recovered from seminal fluid, and sexual transmission of SARS-CoV-2 has not been documented [62]. Similarly, viral RNA has been detected in tears, vomitus, and bile fluid, but the clinical significance of this has yet to be determined [63][64][65].

Infectious Potential of Clinically Shed SARS-CoV-2
The presence of SARS-CoV-2 RNA in respiratory secretions and other bodily fluids, as detected by RT-PCR, does not necessarily indicate viable virus; and their infectious potential is an area of active research with important public health implications. However, it is increasingly evident that the risk to others posed by post-convalescent COVID-19 patients may be negligible. For example, in one study of healthcare workers self-isolating due to persistent RT-PCR positivity up to 55 days after the onset of symptoms, no viable virus was recoverable in 29 of 29 nasopharyngeal/oropharyngeal samples tested [66]. In another similar study of 48 patients who had detectable viral RNA more than two weeks out from symptom onset, no virus could be recovered from any nasopharyngeal or salivary swab cultures [67]. The precise duration of infectivity likely varies, and the impact of several factors (such as viral load) is being studied. Nonetheless, data from a number of studies suggest that the risk posed by respiratory secretions is significantly reduced ten days after symptom onset [68], although viable virus may persist for as much as 15 days in saliva, urine and stool [69].
Based on these and similar studies, the Centers for Disease Control and Prevention (CDC) changed its guidance in August 2020 regarding the discontinuation of transmission-based precautions and disposition of patients with COVID-19 in healthcare settings from a test-based strategy to a symptom-based approach (Figure 3). According to the updated guidelines, transmission-based precautions for patients who are mildly to moderately ill and who are not severely immunocompromised may be discontinued when: (1) at least ten days have passed since symptoms first appeared; (2) at least twenty-four hours have passed since the last fever without the use of anti-pyretics; and (3) symptoms have improved. For patients with severe or critical illness or who are severely immunocompromised, the revised CDC guidelines recommend waiting at least ten and up to twenty days before discontinuing precautions. These guidelines acknowledge that prolonged clinical shedding of nonculturable virus occurs, and they replace previous guidelines that relied on negative RT-PCR results to clear a patient from transmission-based precautions [70]. documented [62]. Similarly, viral RNA has been detected in tears, vomitus, and bile fluid, but the clinical significance of this has yet to be determined [63][64][65].

Infectious Potential of Clinically Shed SARS-CoV-2
The presence of SARS-CoV-2 RNA in respiratory secretions and other bodily fluids, as detected by RT-PCR, does not necessarily indicate viable virus; and their infectious potential is an area of active research with important public health implications. However, it is increasingly evident that the risk to others posed by post-convalescent COVID-19 patients may be negligible. For example, in one study of healthcare workers self-isolating due to persistent RT-PCR positivity up to 55 days after the onset of symptoms, no viable virus was recoverable in 29 of 29 nasopharyngeal/oropharyngeal samples tested [66]. In another similar study of 48 patients who had detectable viral RNA more than two weeks out from symptom onset, no virus could be recovered from any nasopharyngeal or salivary swab cultures [67]. The precise duration of infectivity likely varies, and the impact of several factors (such as viral load) is being studied. Nonetheless, data from a number of studies suggest that the risk posed by respiratory secretions is significantly reduced ten days after symptom onset [68], although viable virus may persist for as much as 15 days in saliva, urine and stool [69].
Based on these and similar studies, the Centers for Disease Control and Prevention (CDC) changed its guidance in August 2020 regarding the discontinuation of transmission-based precautions and disposition of patients with COVID-19 in healthcare settings from a test-based strategy to a symptom-based approach (Figure 3). According to the updated guidelines, transmission-based precautions for patients who are mildly to moderately ill and who are not severely immunocompromised may be discontinued when: (1) at least ten days have passed since symptoms first appeared; (2) at least twenty-four hours have passed since the last fever without the use of anti-pyretics; and (3) symptoms have improved. For patients with severe or critical illness or who are severely immunocompromised, the revised CDC guidelines recommend waiting at least ten and up to twenty days before discontinuing precautions. These guidelines acknowledge that prolonged clinical shedding of nonculturable virus occurs, and they replace previous guidelines that relied on negative RT-PCR results to clear a patient from transmission-based precautions [70].  Recognizing that recovered COVID-19 patients may persistently shed inert non-infectious virus into clinical specimens, the CDC generally recommends a symptom-based rather than a test-based approach for symptomatic individuals [70]. (NA: Not applicable).

Infectivity of SARS-CoV-2 Shed by Pre-Symptomatic and Asymptomatic Infected Individuals
In a June 8 press briefing, the World Health Organization's (WHO) coronavirus technical lead stated that asymptomatic transmission of the SARS-CoV-2 virus was "very rare." The comment, which seemingly suggested that infected people without symptoms were not spreading the disease, generated confusion about the role of wearing masks, social distancing, and sheltering in place-something the public has been exhorted to do in order to curtail asymptomatic spread of the virus. This was followed by a clarifying comment from the WHO the next day that the use of the term "very rare" had been a "miscommunication" and had been based on a small number of studies done in "member states" that followed the contacts of infected but asymptomatic individuals [71].
A role for asymptomatic clinical shedding in new COVID-19 cases is in fact suggested by several studies. In one study of 94 infected patients and another 77 documented cases of transmission, an estimated 44% of transmission occurred during the pre-symptomatic period, with infectiousness starting from 2.3 days (95% CI, 0.8-3.0 days) before symptom onset and peaking at 0.7 days (95% CI, −0.2-2.0 days) before symptom onset [72]. In another study published in the New England Journal of Medicine that documented transmission of SARS-CoV-2 in a skilled nursing facility in King County, Washington, 56% of residents with positive test results were asymptomatic at the time of testing and "most likely contributed to transmission" [73]. In an accompanying editorial, the authors claim that the study shows that "asymptomatic persons are playing a major role in the transmission of SARS-CoV-2" [74]. Lastly, in a comprehensive meta-analysis, researchers at the Scripps Research Translational Institute reviewed the data from sixteen international studies in which a total of 45,394 individuals were screened for SARS-CoV-2 and found that of the 6738 individuals who tested positive, 40-45% remained asymptomatic [75].

SARS-CoV-2 Clinical Shedding by Children
The extent to which children infected with SARS-CoV-2 transmit the virus to others is not yet known. However, a number of studies have demonstrated prolonged clinical shedding by pediatric patients (Table 2), with an inverse correlation between the duration of RT-PCR positivity and age [52,76]. Moreover, detection of virus by RT-PCR tends to be longer in fecal than in respiratory specimens ( Figure 4) [51,52,77,78], and at least one study found the duration of clinical shedding to be longer among symptomatic than asymptomatic children [79]. Despite these findings, as well as published case reports documenting pediatric transmission of SARS-CoV-2 between children and from children to adults [80], there is limited evidence that children play a prominent role in propagating COVID-19. Several epidemiological studies of households, schools, and daycare settings suggest that children are rarely the index case and that secondary attack rates may be lower when the infector is a child [81]. In one analysis of thirty-one cases of household transmission of SARS-CoV-2 in southeast and southwest Asia, only three (9.7%) identified a child as the index case [82]. In a similar study of thirty-nine Swiss households, only three (8%) familial clusters began with a symptomatic child [83]. However, neither of these studies exclude the possibility of transmission from an asymptomatic child to others in the household.

Persistent Clinical Shedding, Relapse, and Reinfection
The detection of SARS-CoV-2 virus in respiratory specimens from recovered COVID-19 patients after one or more negative RT-PCR assays has been reported [89][90][91][92], raising the question as to whether this represents imperfect sampling, the limited sensitivity of the assay, intermittent shedding, relapse, or reinfection. Although data are limited, there are several small follow up studies of such individuals showing a lack of transmission to family members after the patients were discharged from the hospital, suggesting that they were clinically shedding inert virus [92]. Nonetheless, there are a small number of case reports of recovered COVID-19 patients clinically shedding virus that is genetically distinct from that which was originally isolated [93][94][95][96], a finding that is consistent with either reinfection or mutation of the original virus.
The extent to which detectable virus after a negative assay (i.e., re-positive) represents reinfection has not been clearly established. However, in one study of 87 patients in Guangdong, China, who retested positive, culturable virus or intact genomes consistent with possible reinfection were found in only 14% of cases, and the majority of patients were thought to be clinically shedding inert virus [97]. Moreover, in at least one animal study, nonhuman primates recovering from COVID-19 were protected from reinfection when challenged with the virus [98]. Collectively, these and other studies suggest that in most COVID-19 cases, persistent clinical shedding is not due to reinfection. An alternative explanation is that viruses might be sequestered somewhere in the body (e.g., in extracellular double-membrane vesicles or exosomes) and are released during a second round of cellular shedding [99]. This mechanism has been proposed for certain viruses, including Human Immunodeficiency Virus and Epstein Barr Virus [100]; and although such structures have been A limited role for children in the propagation of COVID-19 is also suggested by several school-based studies. In one study of SARS-CoV-2 transmission among children and staff in fifteen schools and ten early childhood education and care settings in the Australian state of New South Wales, there were just eighteen secondary cases identified among 1448 contacts [84]. Similarly, in Sweden, where primary schools and day care centers have remained open during the pandemic, the cumulative incidence for hospitalization with a non-incidental diagnosis of COVID-19 among children in Stockholm younger than 17 years was nine per 100,000, compared to 230/100,000 hospitalized adults during the same time period [85]. Lastly, in one meta-analysis of outbreaks in China, Hong Kong, and Singapore, mathematical modeling suggested that school closures would prevent only 2-4% of COVID-19-related deaths [86], a conclusion also drawn in a study of epidemic data from China, Italy, Japan, Singapore, Canada and South Korea [87]. Nonetheless, as of 18 March 2020, 107 countries had implemented national school closures in response to the pandemic, a number that had fallen to 23 by 25 November 2020 [88].

Persistent Clinical Shedding, Relapse, and Reinfection
The detection of SARS-CoV-2 virus in respiratory specimens from recovered COVID-19 patients after one or more negative RT-PCR assays has been reported [89][90][91][92], raising the question as to whether this represents imperfect sampling, the limited sensitivity of the assay, intermittent shedding, relapse, or reinfection. Although data are limited, there are several small follow up studies of such individuals showing a lack of transmission to family members after the patients were discharged from the hospital, suggesting that they were clinically shedding inert virus [92]. Nonetheless, there are a small number of case reports of recovered COVID-19 patients clinically shedding virus that is genetically distinct from that which was originally isolated [93][94][95][96], a finding that is consistent with either reinfection or mutation of the original virus.
The extent to which detectable virus after a negative assay (i.e., re-positive) represents reinfection has not been clearly established. However, in one study of 87 patients in Guangdong, China, who retested positive, culturable virus or intact genomes consistent with possible reinfection were found in only 14% of cases, and the majority of patients were thought to be clinically shedding inert virus [97]. Moreover, in at least one animal study, nonhuman primates recovering from COVID-19 were protected from reinfection when challenged with the virus [98]. Collectively, these and other studies suggest that in most COVID-19 cases, persistent clinical shedding is not due to reinfection. An alternative explanation is that viruses might be sequestered somewhere in the body (e.g., in extracellular double-membrane vesicles or exosomes) and are released during a second round of cellular shedding [99]. This mechanism has been proposed for certain viruses, including Human Immunodeficiency Virus and Epstein Barr Virus [100]; and although such structures have been observed in cultured SARS-CoV-2-infected cells, their role in spreading the virus remains speculative [99]. Lastly is the possibility of latent virus reactivation, as is seen with the herpes viruses, a phenomenon that was described in a COVID-19 patient who underwent treatment for B cell acute lymphoblastic leukemia [101].

Discussion
It has been one year since the novel coronavirus provisionally designated 2019-NCoV and subsequently renamed SARS-CoV-2 emerged from Wuhan, China. Since then, the virus has spread globally, and as of 23 November 2020, 59,127,000 COVID-19 cases with 1,396,017 deaths in 215 countries, territories, and conveyances have been reported [15]. COVID-19 has dominated the news and articles abound on transmission and case fatality rates, the utility of masking and social distancing, the efficacy (or lack thereof) of therapeutics, the pursuit of vaccines, as well as the timing of reopening schools and loosening of social restrictions. The virus has similarly become an intense focus of scientific research, with 70,731 results and 41,958 results in PubMed, using the search terms "COVID-19" and "SARS-CoV-2", respectively (accessed 4 November 2020). We have made significant strides in understanding both the virus and the disease it causes, and several fast-tracked vaccine candidates are currently in Phase 3 clinical trials [102]. However, many fundamental questions have yet to be answered, including the nature and duration of immunity, the long-term sequelae of infection, as well as the transmission risk posed by prolonged clinical shedding by convalescent persons. Regarding the latter, the persistent detection of SARS-CoV-2 in the bodily fluids and feces of convalescent patients introduces a measure of uncertainty with respect to their clinical management. At what point, for example, is it entirely safe to return an elderly recovered COVID-19 patient to a skilled nursing facility, or a cancer patient who has recovered from COVID-19 to the hematology/oncology clinic waiting room? Similarly, when is it appropriate to permit a child with persistently detectable virus in the stool to return to school or day care? Furthermore, what is the significance of a repeat positive SARS-CoV-2 RT-PCR assay on a respiratory sample from a recovered COVID-19 patient in the setting of a prior negative test? These are precisely the types of scenarios confronting physicians and public health officials and for which they are seeking evidence-based guidance.
Certainly, further studies are needed to determine both the mechanism of viral persistence and its implications for safeguarding public health. Nonetheless, based on the current available data, several conclusions can be inferred. Firstly, persistent detection of virus by RT-PCR in respiratory specimens collected from convalescent COVID-19 patients is a common phenomenon, as reported in the studies cited in this paper. In this respect, SARS-CoV-2 is like several other human coronaviruses including SARS-CoV and MERS-CoV [103]. However, among these viruses, MERS-CoV differs from SARS-CoV and SARS-CoV-2 in that prolonged detection of MERS-CoV viral RNA in the stool of infected individuals has not been widely reported [104]. Secondly, the persistent detection of SARS-CoV-2 in clinical specimens is unlikely to reflect either relapse or reinfection in most cases. This statement is supported by a number of studies showing that replication-competent virus is generally not recoverable after ten days following symptom onset in mild to moderate cases of COVID-19 [105] and after twenty days in severe or immunocompromised cases [106]. Similarly, in studies of individuals who have recovered from COVID-19 and subsequently redeveloped symptoms, replication-competent virus was not recoverable, even in the setting of a positive RT-PCR assay [97]. These data, combined with contact tracing studies of people exposed to convalescent COVID-19 patients, prompted the CDC to revise its guidelines regarding the discontinuation of isolation precautions from a test-based to a time-based approach [107].
For several reasons, children infected with SARS-CoV-2 represent a demographic of COVID-19 patients that has garnered much attention. These include observations that although virus remains persistently detectable by RT-PCR in the stool of children with COVID-19 [108], they tend to be less affected than adults, account for a relatively small percentage of diagnosed cases, and are rarely the index case in a household cluster [83]. Nonetheless, the extent to which children propagate COVID- 19 is an open question, and with every spike in COVID-19 cases, school districts ponder closures and the implementation of virtual classroom instruction [109]. Although there is no consensus on the best approach, proponents for and opponents of face-to-face instruction agree that there are significant advantages to having children return to the classroom. Some of these have been explicated by The CDC [110] and in a position statement the American Academy of Pediatrics "strongly advocates that all policy considerations for the coming school year should start with a goal of having students physically present in school" [111]. Nonetheless, one can anticipate that pending additional studies of the role of children in transmitting SARS-CoV-2, as well as the release of an effective COVID-19 vaccine, school districts will likely act reflexively to the local prevalence of COVID-19 in their communities.
Although it may seem as if the COVID-19 pandemic is interminable, it is only one year since the virus emerged, and the reality is that Nature does not readily give up her secrets. Hypotheses must be formulated and tested; results must be interpreted and reconciled; and conclusions must withstand the tincture of time. Consider, for example, that it was four years into the AIDS pandemic before the first HIV drug (zidovudine or AZT) was approved, and twelve years until the discovery of potent Highly Active Antiretroviral Therapy (i.e., HAART or "AIDS cocktails"). Similarly, reliably effective drugs against the hepatitis C virus, which was identified in 1989, were only available since 2013 [112]. Certainly, our understanding of SARS-CoV-2 will increase with time, but just assuredly, new questions will also arise. Perhaps there is some consolation to be found in the words of Jules Verne: "Science, my boy, is made up of mistakes, but they are mistakes which it is useful to make, because they lead little by little to the truth [113]".

Funding:
The publication cost of this manuscript was covered by the Martinsburg, West Virginia Veterans Affairs Medical Center.