Introduction
Human coronaviruses have been shown to affect the human gastrointestinal (GI) tract: Middle East respiratory syndrome coronavirus infection (MERS-CoV) was demonstrated to infect the GI tract, with intestinal epithelial cells supporting replication-competent virus [Reference Zhou1], and SARS-CoV-1 was associated with GI symptoms and prolonged RNA shedding in faeces as demonstrated by the presence of replication-competent virus in cell cultures of faecal samples from affected patients [Reference Leung2].
It was shown early in 2020 that the SARS-CoV-2 uses the angiotensin-converting enzyme 2 (ACE2) receptor as a cell entry receptor to enter ACE2-expressing cells [Reference Zhou3]. Since the ACE2 receptor is abundant not only in lung epithelial cells but is also highly expressed on the luminal surface of intestinal epithelial cells [Reference Zhang, Penninger, Li, Zhong and Slutsky4], a potential route for infection via the GI system is considered biologically plausible. Studies have demonstrated that human GI tract cell lines with a brush border and colon-derived cell lines plus colonic organoids have robust viral growth after inoculation of SARS-CoV-2 and allow for persistent infection within these cell lines [Reference Guo, Tao, Flavell and Zhu5–Reference Yantiss7]. There have also been animal models, including non-human primates, which have been inoculated by intragastric intubation, bypassing the respiratory tract. These data demonstrate unequivocal evidence for direct invasion of the GI tract by SARS-CoV-2 with concomitant GI and lung pathology [Reference Jiao8–Reference Sun10]. Given the findings that SARS-CoV-2 is able to replicate within human GI tract cell lines and remain infectious on excretion and the establishment of invasive infection in multiple animal models [Reference Jiao8–Reference Sun10], it was reasonable to consider whether infectious SARS-CoV-2 could be found in faecal specimens or other sources representing potential oro-faecal sources, creating an oro-faecal transmission risk (also commonly referred to as faecal-oral) [Reference van Seventer and Hochberg11], which would be important for the application of appropriate infection control measures.
Involvement of the GI tract was one of the reasons for considering that the real rate of paediatric cases may have been higher than that officially reported, as children may present with only GI symptoms and signs. SARS-CoV-2 RNA has been identified in anal/rectal swabs and in stool specimens of COVID-19 patients even after upper respiratory tract virus clearance [Reference Puoti, Rybak, Kiparissi, Gaynor and Borrelli12]. In contrast to what has been observed in adult patients, higher proportions of fever, vomiting, and diarrhoea have been recorded on admission in paediatric cases [Reference Zhang13]. GI localization in children may represent an alternative site of viral shedding and transmission. For these reasons, sanitation measures in schools included interventions directed against oro-faecal transmission routes [Reference Dewan, Sharma, Panda and Banerjee14].
Previously, we systematically reviewed studies reporting on the possibility of transmission via the oro-faecal route and found several studies reporting that SARS-CoV-2 RNA can be shed from the GI tract [Reference Heneghan15]. However, the shedding of viral RNA does not necessarily equate to viral replication in the GI tract. Since that report, it has become more apparent that understanding the transmission of SARS-CoV-2 depends on studies using high-quality, replicable methods to assess the potential infectivity of samples, along with rigorous epidemiological data examining exposures and outcomes [Reference Jefferson16]. We therefore set out to identify, appraise, and summarize the evidence on the presence of replication-competent SARS-CoV-2 in human faecal and other source specimens that could represent faecal-oral transmission pathways, and if cases of human oro-faecal transmission of SARS-CoV-2 have been convincingly demonstrated.
Materials and methods
We performed this systematic review following our published protocol [Reference Heneghan17]. In brief, we searched for studies reporting data from participants with COVID-19 (with or without control groups) from whom biological samples were obtained. We verified whether there was a laboratory confirmation of SARS-CoV-2, and corresponding clinical data, including symptomatology, disease course, treatments, and comorbidities. We focused on studies reporting the results of viral culture from human faecal and other GI samples, coupled with reverse transcription polymerase chain reaction (RT-PCR) testing, to investigate what is known about the presence of replication-competent virus within faecal samples from individuals with COVID-19 and the demonstration of any onward oro-faecal transmission of SARS-CoV-2. To be included, studies must have complied with all the following criteria: included viral culture of faecal/GI samples from SARS-CoV-2-infected persons, and assessed the cytopathic effect and verification techniques of the isolated virus to ensure that the cultured virus was SARS-CoV-2. Studies with data only from RT-PCR testing of faecal samples from SARS-CoV-2-infected persons and/or reporting solely predictive modelling were excluded. In order to help reduce bias due to selective reporting, case studies of one or two cases were excluded as well.
We assessed the risk of bias within five domains, modified and extended from the Quality Assessment of Diagnostic Accuracy Studies criteria [Reference Whiting, Rutjes, Westwood and Mallett18]. Further details regarding the methods are reported in the Supplementary Materials (Appendix 1).
Results
Figure 1 shows the flow of studies through the screening and inclusion/exclusion processes. The literature search identified 1,496 studies, and 14 were found from other sources, including duplications. After screening the abstracts, 53 independent studies were found eligible. We excluded 40 studies for various reasons (Appendix 2. Studies excluded on full-text screening). Finally, we included 13 studies with a total of 1,625 overall participants, regardless of Ct values [Reference Akiyama19–Reference Pedersen31].
Figure 1. Flow diagram showing the process for inclusion/exclusion of studies.
Table 1 shows the main characteristics of the included studies and viral culture information on SARS-CoV-2-positive subjects. More detailed information is presented in Table S1. Eight of the studies were cohort studies, while the remaining five were case series. The 13 studies involved mainly hospitalized patients but also some persons from ambulatory or home settings. Seven studies were conducted in Europe, four in Asia, and two in South America. The study duration ranged from 2 weeks to 15 months.
Table 1. Descriptive features of included studies and results from samples for cell culture
† See Supplementary Materials for verification methods used.
* Faecal samples, rectal swabs, wastewater or other GI-related samples. GI = gastrointestinal. ¥ = rectal swabs. PY = publication year. NR = not reported. €: A previous protocol stated only done if <30; it was assumed the number of specimens equated to the number of patients.
§ For Albert et al., the number of samples used for viral culture was estimated from the manuscript, since the exact number was not clear and only undiluted specimens were counted.
A Japanese case series of 13 COVID-19 patients diagnosed between January and March 2020 reported data of viral load from faeces by days and symptoms [Reference Akiyama19]. One study reported on a series of eight patients with five faecal samples investigating patients in Spanish hospitals and local wastewater samples [Reference Albert, Ruíz, Pemán, Salavert and Domingo-Calap20]. One was a prospective multicentre Spanish cohort, which analysed 31 samples obtained from 27 patients within 2 weeks of admission with COVID-19 symptoms between March 2020 and February 2021 [Reference Cerrada-Romero24]. One was a case series from a hospital in France reporting on 46 patients admitted with COVID-19 between March and April 2020, from whom 106 stool samples were taken as part of standard care; viral culture was reported for just one patient with kidney transplantation under immunosuppressive therapy who was admitted for severe diarrhoea [Reference Dergham, Delerce, Bedotto, La Scola and Moal25]. One study from a hospital in Brazil presented a retrospective case series using stored samples collected from four patients between January and July 2020 for the ongoing surveillance of acute gastroenteric disease [Reference Fumian26]. A small Korean study presented data collected from five COVID-19 patients and performed a quantitative PCR (qPCR) to assess viral load [Reference Jeong27]. Specimens positive with qPCR were subjected to virus isolation in Vero cells and they reported log copy numbers for faecal samples from three patients. Joshi presented a multicentre case series from hospitals in India using stored samples collected between May 2020 and August 2021 [Reference Joshi28]. A further study from India described a multicentre cohort study involving an estimated 55 patients (55 samples) attending hospitals who tested positive for SARS-CoV-2 between May 2020 and August 2021 [Reference Lavania29]. A study from three clinical centres in Austria reported data on 206 hospitalized children diagnosed between December 2020 and June 2021 [Reference Nogueira30]. A study from Denmark reported data on 28 patients diagnosed between 23 October 2020, and 17 March 2021 [Reference Pedersen31]. Another cohort study [Reference Ribeiro21] described data from a slum in Rio de Janeiro, Brazil, collected from May to October 2020, regarding the SARS-CoV-2 test results of 333 rectal swabs from 33 persons, some of whom were symptomatic patients and others were their household contacts. A cluster of 4 epidemiologically linked cases (corresponding to 13 samples) managed in hospitals in Germany early in the pandemic was reported [Reference Wölfel22]. Finally, one Chinese study analysed data from stool samples of 3 of 3 patients from a cohort of 11 patients diagnosed between 22 January and 4 February 2020 [Reference Yao23].
Across these 13 studies, there were 308 total samples from an estimated 229 hospitalized patients, households, and other settings (exact numbers were not always clearly reported) analysed with viral culture. Among these, three studies reported finding evidence of an isolated virus (Table 1), which were confirmed positive by qPCR and/or genome sequencing of SARS-CoV-2 in the culture supernatant. One of these three studies was a cohort study, in which the positive samples identified via viral culture were obtained from an immunocompromised patient. The samples were obtained at variable time points during the course of illness, and only nine studies reported Ct values from the stool specimens at the time of initial attempts at culture. Of the studies that reported Ct values from the faecal and other samples, all the six samples found to have evidence of culturable virus had Ct values reported as ≤30, which is in the range where a high likelihood of culturable virus may be found [Reference Wölfel22, Reference Lin32, Reference Zhou33]. Considering only the cohort studies, there was a 1.2% (3/ 258 samples) positive frequency for those specimens.
Overall, we identified a low-moderate risk of bias across the included studies in four domains but a high risk of bias in one domain, with evident deficits in reporting viral cell culture methods.
Table S2 shows the risk of bias assessment results. The risk of bias was generally considered low for the criteria for the domains of diagnosing a case clearly and appropriately, reporting of patient/population characteristics adequacy, and reporting of methods used to obtain RT-PCR results being replicable and appropriate. Reporting of patient characteristics was absent or unclear in only one (8%) study, and methods for RT-PCR were unclear in four (31%) studies. However, the risk of bias was considered high for clear reporting of viral culture methods, where it was only well reported in five studies. Analysis and appropriate reporting of the results were considered to have low-moderate risk of bias with unclear reporting in two studies.
Discussion
In this systematic review, we included 13 cohort and case series studies that subjected faecal, rectal swabs, or other GI samples (from hospitalized patients and persons in the community with confirmed COVID-19) to viral culture. Among a total of 308 samples subjected to cell culture which were found to be RT-PCR+ for SARS-CoV-2, six (1.9%) were reported as positive from cell culture. The sampling methodology was poorly described in some studies and appeared to be more of a convenience sampling as opposed to a rigorous hypothesis driven type of research protocol, with some exceptions. Sampling stool at late stages of illness with no attention paid to cycle threshold values or quantitative viral load would lead to a large number of samples that would have a high likelihood of being negative for culture given our knowledge about the natural history of COVID-19 [Reference Wölfel22, Reference Lin32, Reference Zhou33].
Schedules of sample collection and information on symptoms across the disease course were incompletely reported. Coupled with the limited number of data points available, it was not possible to assess the relationship between the time of symptom onset, faecal sample collection, and results of viral culture. The timing of sample collection is likely to be important [Reference Bullard34], and most of the studies reported herein are too few, varied, and unclear in their descriptions to be able to determine whether the included participants were indeed within a potential infectious window. However, the high Cts of many of the faecal and respiratory samples reported suggest that most of the included subjects were unlikely to still be infectious at the time of faecal sample collection. It is also possible that SARS-CoV-2 typically has only a brief survival time with transit through the GI tract or after exiting it; this is as yet unknown.
Nonetheless, in those samples where attention was paid to the timing of faecal sample collection with respect to the course of the illness and/or with low Ct values, higher frequencies of virus isolation were reported. It is possible that these current studies underrepresent the presence of culturable virus in human stool specimens due to poor sampling strategies.
The likelihood of transmission via the oro-faecal route is definitely biologically plausible from evidence on other human and animal coronaviruses, including MERS-CoV and SARS-CoV-1 [Reference Zhou1, Reference Leung2, Reference Zhang35] along with studies early in the pandemic reporting findings that suggested entry of SARS-CoV-2 into GI cells and possible replication with the GI tract, along with many experimental animal model studies [Reference Jiao8–Reference Sun10].
In addition, Qian et al. [Reference Qian36] reported a study using electron microscopy to examine tissue obtained from the rectal mucosa of a patient undergoing rectal surgery who subsequently developed COVID-19 symptoms. The authors reported observing virions matching SARS-CoV-2 morphology in sections of the obtained tissue sample. As described above, however, viral morphology alone is not a sufficient indicator of identity [Reference Hopfer37], and in itself does not show that replication-competent virus is present; this observation could not be regarded as sufficient to indicate a transmission risk from faeces.
In February 2020, Zhang et al. [Reference Zhang38] reported testing stool samples by cell culture in Vero cells. The authors reported on the finding of successful cell culture in one sample taken from a laboratory-confirmed COVID-19 case with severe pneumonia 15 days after the onset of symptoms. They reported that viral particles with morphology typical of SARS-CoV-2 were observed under electron microscopy; no further details were reported and again the results of a verified replication-competent virus are uncertain.
To assess a chain of transmission for SARS-CoV-2, it is necessary to have reliable epidemiology, adequate reporting of symptoms and signs, sufficient follow-up, and evidence from human samples of the presence of a replication-competent virus [Reference Jefferson39]. We previously defined viral culture as encompassing several methods that can uniquely identify the replicating agent as SARS-CoV-2 [Reference Jefferson, Spencer, Brassey and Heneghan40]. Most commonly, this would be a plaque assay combined with an RT-PCR diagnosis or immunological staining, or gene sequencing of viral RNA.
Numerous studies have identified SARS-CoV-2 RNA in faecal samples and wastewater [Reference Guo, Tao, Flavell and Zhu5], suggesting the possibility of infectious virus remaining in faeces. Although useful for surveillance purposes, RT-PCR positivity in wastewater is of little significance in understanding transmission without viral load data and cultivability [Reference Jefferson16]. To the best of our knowledge, no review of the results of viral culture on faecal samples exists.
Strengths and limitations
We attempted to collate and synthesize all the relevant data from studies using viral culture of faecal samples to investigate the presence of replication-competent virus, which would suggest the potential for oro-faecal transmission of SARS-CoV-2. We used robust methods to search for the relevant studies, and we accounted for the reporting quality of included studies. However, studies investigating the presence of replication-competent virus within faecal samples have been few and have varied in design and methods. Methodologies have generally been lacking in scientific rigour. Comparison between studies is hindered by the lack of standard approaches for sampling schedules, RT-PCR cut-offs, and viral culture methods. In addition, there are acknowledged methodological difficulties in culturing virus from faecal samples due to the presence of heavy bacterial contamination and the presence of inhibitors in stool or direct cytotoxicity of stool specimens [Reference Arnold, Patton and McDonald41].
Viral culture has been reported infrequently in respiratory samples with cycle thresholds above 25 [Reference Lin32, Reference Bullard34], and it is likely to assume similar results within faecal samples with high Ct values. Selecting samples with an RT-PCR cycle count of higher than 25 greatly reduces the likelihood of being able to culture SARS-CoV-2 [Reference Lin32, Reference Zhang38, Reference Bruce42]. In the studies in this review, in general, samples with relatively high RT-PCR cycle counts were subjected to viral culture testing.
Standardized, appropriate methods for viral culture studies are essential to generate reliable evidence that can be compared between studies and allow the combining of results. A weakness found in our review was the reporting of rigorous virologic methods and use of controls. Only two studies [Reference Dergham, Delerce, Bedotto, La Scola and Moal25, Reference Lavania29] reported using negative/uninfected controls for the cell culture experiments; in no study were methods to reduce contamination described; and no study demonstrated an increasing viral load within cell culture supernatants related to a clear timeline and course of the disease.
Implications for research and policy
It is difficult to draw firm conclusions about the likelihood of infectiousness of faecal samples and the associated potential transmission via the oro-faecal route, given the limitations of the available data. Although our review suggests the presence of infectious virus appears to be of low frequency (1.9%) from these samples, it is important to recognize that not accounting for settings whereby Ct values are above a threshold allowing for a high likelihood of negative cultures and with testing in a non-standardized manner, the results are likely an underestimate. In addition, the study by Ribeiro et al. [Reference Ribeiro21] found the presence of replicating SARS-CoV-2 in 19.4% (6/31) of the samples tested, measured by the detection of sgN mRNA, suggesting the degree of infectious virus may be higher than that found by cell culture and this approach deserves further exploration.
Our findings are consistent with the findings in a recent systematic review whereby 8.3% of fomite samples which were RT-PCR+ were found to have infectious SARS-CoV-2 [Reference Onakpoya43]. They also reported that the highest frequency of detection was within 7 days of symptom onset and significantly associated with a Ct < 30. Placing our findings in context, when extrapolated to millions of COVID-19 cases globally, the real potential for an oro-faecal route of transmission in high-risk settings deserves more attention. However, it remains biologically plausible and this route of potential transmission needs further exploration. Given the stability and survivability of SARS-CoV-2 from environmental sources, taken together with our findings demonstrating the presence of infectious virus in 1.9% of faecal and/or rectal samples from humans, a network meta-analysis finding that meal-gathering by individuals had one of the highest risks associated with transmission of SARS-CoV-2 and the recent findings of infectious virus on several types of frozen foods, deli foods, meat, seafood fresh produce, and ice cream for prolonged periods of time from days to weeks, all add impetus to further explore this route of transmission [Reference Bryant, Wilks and Keevil44, Reference Zhao45].
It is important to address the possibility of oro-faecal transmission to develop effective infection control measures, particularly in high-risk settings, such as hospitals, residential care settings, and nursing homes, where facilities may be shared and where healthcare workers have contact with a number of at-risk individuals. Evidence from studies using rigorously performed viral culture is needed to investigate the presence of replication-competent virus and relate this to cycle thresholds or other surrogates of infectivity to assess the likelihood of transmission of SARS-CoV-2 via the oro-faecal route.
Conclusions
This review concludes that there is evidence of infectious SARS-CoV-2 in faeces, rectal swabs, and other GI specimens in the settings described.
Our review highlights the need for further high-quality research, ideally from prospective cohort studies, using appropriate timing of specimen capture and using standardized high-quality methods for viral culture with appropriate negative controls, coupled with robust evidence on possible transmission events to ascertain the likelihood of transmission of SARS-CoV-2 via the oro-faecal route. Animal models and human challenge studies are both viable options to help address this research gap.
Supplementary material
The supplementary material for this article can be found at http://doi.org/10.1017/S0950268824001481.
Data availability statement
All data used in this study are available as fully published data as referenced.
Author contribution
Data curation: E.C.R., E.S., J.C., J.B., S.G.; Validation: E.C.R., D.E., S.M., T.J.; Writing – review & editing: E.C.R., A.P., I.O., J.C., J.B., S.M., T.J., S.G.; Supervision: A.P., C.H., D.E., I.O., J.C., S.M., T.J.; Conceptualization: C.H., E.S., J.C.; Methodology: C.H., T.J.; Investigation: E.S., J.C., S.G.; Writing – original draft: E.S., S.G.; Resources: J.B.; Formal analysis: S.G.
Funding statement
This work is part funded by the NIHR School for Primary Care Research (project 569). The World Health Organization funded the first iteration of this review: Living rapid review on the modes of transmission of SARs-CoV-2 reference WHO registration No. 2020/1077093.
SG acknowledges the support of a grant funded by the European Union’s Horizon Europe Research and Innovation Programme under Grant Agreement No. 101046016 (EuCARE project: EUROPEAN COHORTS OF PATIENTS AND SCHOOLS TO ADVANCE RESPONSE TO EPIDEMICS).
Competing interest
TJ disclosure is available here: https://restoringtrials.org/competing-interests-tom-jefferson/.
CJH holds grant funding from the NIHR, the NIHR School for Primary Care Research, the NIHR BRC Oxford, and the World Health Organization for a series of Living rapid reviews on the modes of transmission of SARs-CoV-2, reference WHO registration No2020/1077093, and to carry out a scoping review of systematic reviews of interventions to improve vaccination uptake, reference WHO Registration 2021/1138353-0. He has received financial remuneration from an asbestos case and given legal advice on mesh and hormone pregnancy test cases. He has received expenses and fees for his media work, including occasional payments from BBC Radio 4 Inside Health and The Spectator. He receives expenses for teaching EBM and is also paid for his GP work in the NHS out of hours (contract Oxford Health NHS Foundation Trust). He has also received income from the publication of a series of toolkit books and appraising treatment recommendations in non-NHS settings. He is the Director of CEBM, an NIHR Senior Investigator, and an advisor to Collateral Global.
DHE holds grant funding from the Canadian Institutes for Health Research and Li Ka Shing Institute of Virology relating to the development of Covid-19 vaccines and the Canadian Natural Science and Engineering Research Council concerning Covid-19 aerosol transmission. He is a recipient of World Health Organization and Province of Alberta funding which supports the provision of BSL3-based SARS-CoV2 culture services to regional investigators. He also holds public and private sector contract funding relating to the development of poxvirus-based Covid-19 vaccines, SARS-CoV2 inactivation technologies, and serum neutralization testing.
JMC has held grants from the Canadian Institutes for Health Research on acute and primary care preparedness for COVID-19 in Alberta, Canada and was the primary local Investigator for a Staphylococcus aureus vaccine study funded by Pfizer, for which all funding was provided only to the University of Calgary. He is co-investigator on a WHO-funded study using integrated human factors and ethnography approaches to identify and scale innovative IPC guidance implementation supports in primary care with a focus on low-resource settings and using drone aerial systems to deliver medical supplies and PPE to remote First Nations communities during the COVID-19 pandemic. He also holds grants from the Synder Institute and a Catalyst Grant from the VPR Office at the University of Calgary for studies on the transmission of SARS-CoV-2 in K18 mice and received funding from BioMérieux Canada for accommodations and travel expenses to attend a meeting on AMR and accommodations and travel expenses from the ICPIC meeting to attend an IPC Think Tank meeting outside the submitted work. He is a member and Chair of the WHO Infection Prevention and Control Research and Development Expert Group for COVID-19 and a member of the WHO Health Emergencies Programme (WHE) Ad hoc COVID-19 IPC Guidance Development Group, both of which provide multidisciplinary advice to the WHO and for which no funding is received and from which no funding recommendations are made for any WHO contracts or grants. He is also a member of the Cochrane Acute Respiratory Infections Working Group.
JB is a major shareholder in the Trip Database search engine (www.tripdatabase.com) as well as being an employee. In relation to this work, Trip has worked with a large number of organizations over the years; none have any links with this work. The main current projects are with AXA and Collateral Global.
SM is a pharmacist working for the Italian National Health System since 2002. She had been a member of one of the three Institutional Review Boards of Emilia-Romagna Region (Comitato Etico Area Vasta Emilia Centro) from 2018 to 2023. She has no conflict of interest to disclose.
AP holds grant funding from the NIHR School for Primary Care Research.
IJO, EAS, and ECR have no interests to disclose.
SG is an epidemiologist/biostatistician group leader at the European Institute of oncology in Milan (Italy) and the European Institute of Oncology, Milan, Italy is partially supported by the Italian Ministry of Health with Ricerca Corrente and 5 × 1,000 funds. She has no conflict of interest to disclose.
Institutional review board statement
No ethics committee approval was necessary for this study because all data used were previously publicly available.
Open access
