SARS-CoV-2 and the role of fomite transmission: a systematic review

Background: SARS-CoV-2 RNA has been detected in fomites which suggests the virus could be transmitted via inanimate objects. However, there is uncertainty about the mechanistic pathway for such transmissions. Our objective was to identify, appraise and summarise the evidence from primary studies and systematic reviews assessing the role of fomites in transmission. Methods: This review is part of an Open Evidence Review on Transmission Dynamics of SARS-CoV-2. We conduct ongoing searches using WHO Covid-19 Database, LitCovid, medRxiv, and Google Scholar; assess study quality based on five criteria and report important findings on an ongoing basis. Results: We found 64 studies: 63 primary studies and one systematic review (n=35). The settings for primary studies were predominantly in hospitals (69.8%) including general wards, ICU and SARS-CoV-2 isolation wards. There were variations in the study designs including timing of sample collection, hygiene procedures, ventilation settings and cycle threshold. The overall quality of reporting was low to moderate. The frequency of positive SARS-CoV-2 tests across 51 studies (using RT-PCR) ranged from 0.5% to 75%. Cycle threshold values ranged from 20.8 to 44.1. Viral concentrations were reported in 17 studies; however, discrepancies in the methods for estimation prevented comparison. Eleven studies (17.5%) attempted viral culture, but none found a cytopathic effect. Results of the systematic review showed that healthcare settings were most frequently tested (25/35, 71.4%), but laboratories reported the highest frequency of contaminated surfaces (20.5%, 17/83). Conclusions: The majority of studies report identification of SARS-CoV-2 RNA on inanimate surfaces; however, there is a lack of evidence demonstrating the recovery of viable virus. Lack of positive viral cultures suggests that the risk of transmission of SARS-CoV-2 through fomites is low. Heterogeneity in study designs and methodology prevents comparisons of findings across studies. Standardized guidelines for conducting and reporting research on fomite transmission is warranted.


Introduction
The SARS-CoV-2 (COVID-19) pandemic is a major public health concern. According to WHO statistics, there have been over 90 million confirmed cases and over two million deaths globally as of 18th January 2021 1 . Although many national governments have implemented control measures and vaccines are now being approved and administered, the rate of infection has not subsided as anticipated. Understanding the modes of transmission of SARS-CoV-2 is critical to developing effective public health and infection prevention measures to interrupt the chains of transmission 2 . Current evidence suggests SARS-CoV-2 is primarily transmitted via respiratory droplets and direct contact 3 , but other transmission routes have been suggestedaerosol and fomites.
While the respiratory, airborne, and direct contact modes of transmission have been investigated in detail, the role of fomites in the transmission of SARS-CoV-2 is less clear. Findings from previous systematic reviews have shown that viruses from the respiratory tract, such as coronaviridae, can persist on inanimate surfaces for some days 4 , and it has been suggested that SARS-CoV-2 can be transmitted indirectly through fomites or surfaces 5 . However, some authors have reported that there is a low risk of transmission of SARS-CoV-2 through fomites 6,7 and others have reported that the risk of such transmission is exaggerated 8 .
Several studies investigating the role of fomites in SARS-CoV-2 are now being published but the evidence from such studies has not been systematically evaluated. The objective of this review was to identify, appraise and summarize the evidence from primary studies and systematic reviews investigating the role of fomites in the transmission of SARS-CoV-2. Terminology for this article can be found in Box 1.

Methods
We are undertaking an open evidence review investigating factors and circumstances that impact on the transmission of SARS-CoV-2, based on our published protocol last updated on the 1 December 2020 (archived protocol: Extended data: Appendix 1 9 ; original protocol: https://www.cebm.net/evidencesynthesis/transmission-dynamics-of-covid-19/).
Briefly, this review aims to identify, appraise, and summarize the evidence (from studies peer-reviewed or awaiting peer review) relating to the role of fomites in the transmission of SARS-CoV-2 and the factors influencing transmissibility. We conducted an ongoing search in WHO Covid-19 Database, LitCovid, medRxiv, and Google Scholar for SARS-CoV-2 for keywords and associated synonyms. The searches for this update were conducted up to 20th December 2020. No language restrictions were imposed (see Extended data: Appendix 2 for the search strategies 9 ).
We included studies of any design that investigated fomite transmission. Predictive or modelling studies were excluded. Results were reviewed for relevance and for articles that looked particularly relevant, forward citation matching was undertaken and relevant results were identified. We assessed the risk of bias using five domains from the QUADAS-2 criteria 10 ; we adapted this tool because the included studies were not designed as diagnostic accuracy studies. The domains assessed were: (i) study description -was there sufficient description of methods to enable replication of the study? (ii) sample sources -was there a clear description of sample sources? (iii) description of results -was the reporting of study results and analysis appropriate? (iv) risk of bias -did the authors acknowledge any potential biases, if yes were any attempts made to address these biases? (v) applicability -is there any concern that the interpretation of test results differs from the study question? For each bias domain, the risk was judged as "low", "unclear" or "high". We extracted the following information from included studies: study characteristics, population, main methods, and associated outcomes including the number of swab samples taken with frequency and timing of samples, and cycle thresholds and samples concentrations where reported. We also extracted information on viral cultures including the methods. One reviewer (IJO) assessed the risk of bias and extracted data from the included studies, and these were independently checked by a second reviewer (EAS). We presented the results in tabular format, and bar charts used to present the frequency of positive tests. Because of substantial heterogeneity across the included studies, we did not perform a meta-analysis.

Results
We identified 709 non-duplicate citations of which 91 were considered eligible ( Figure 1). We excluded 27 full-text studies because they did not meet our inclusion criteria (see Extended data: Appendix 3 9 for the list of excluded studies and reasons for exclusion). Finally, we included 64 studies: 63 primary studies and one systematic review (see Extended data: Appendix 4; characteristics of studies in Table 1 and Table 2 9 ).

Quality of included studies
None of the included studies were linked to or mentioned a published protocol. The risk of bias of the included studies is

Amendments from Version 2
We have responded to the peer reviewers' comments and updated the competing interests section. shown in Table 4. Less than half of the studies (47.6%) adequately reported the methods used, and none used methods to minimise bias. The overall quality of the studies was rated low to moderate (see Figure 2).

Reviews
We found one "systematic review" investigating the role of fomites [Bedrosian 2020] ( Table 2). The authors searched two electronic sources -articles were last downloaded on July 10, 2020. There was no published protocol, and the authors did not assess the quality of included studies. A total of 35 relevant studies were included. Over half of the studies (25/35, 75%) were conducted in healthcare settings, and four compared environmental contamination before and after standard disinfection procedures. No study assessed viral infectivity or viability, but all tested the presence or absence of SARS-CoV-2 RNA.

Primary studies
We found 63 primary studies (Table 1). In general, the studies did not report any hypothesis but investigated epidemiological or mechanistic evidence for fomite transmission. Forty-one studies (65.1%) were conducted in Asia, 15 (23.8%) in Europe, five (7.9%) in North America, and one each in Africa and South America (1.6% each). A total of 44 studies were conducted exclusively in hospital settings, two in hospital and quarantine facilities, three in the laboratory, and the remaining in other non-healthcare settings (public places, community, banknotes, workplace, cruise ship, quarantine rooms and hospital outdoors). Four studies were conducted exclusively in ICU and another three in ICU plus hospital wards. Five studies used before and after study design.
In 59 studies (96.7%), fomite transmission was examined in high-frequency touch surfaces (Table 1); the remaining four studies examined circulating banknotes (1), disposable chopsticks (1) hospital staff PPE (1), and unspecified (1). The timing and frequency of sample collection and disinfection procedures were heterogeneous across studies (see Table 3). Fourteen studies (23%) performed sample collection before disinfection procedures, five studies collected samples before and after disinfection procedures, while 11 studies collected samples after disinfection. In 33 studies, the timing of sampling in relation to disinfection was not specified. In one study [Ryu 2020], disinfection procedures were not performed as required because of a lack of PPE and staff being afraid of contracting SARS-CoV-2. The number of samples per study ranged from five [Jin 2020] to 5400 [Marshall 2020].
Eleven studies (17.5%) set out to perform viral cultures; nine of these utilised the Vero E6 cell lines method while two did not specify the method used (see Table 1). Thirteen studies There is a knowledge gap on fomite contribution to SARS-COV-2 transmission and a need for testing method standardization to ensure data comparability.
There is a need for testing method standardization to ensure data comparability.

Frequency of SARS-CoV-2 positive test
All studies reported data on the frequency of positive tests (Table 5). (Figure 3 shows the graphical representation of these frequencies.) The frequency of positive SARS-CoV-2 tests across 51 studies (via RT-PCR) ranged from 0.5% to 75%; 12 studies (19%) reported no positive tests. The highest frequency of positive tests was found in residential isolation rooms. . In all four before and after studies, there was a substantial reduction in the frequency of positive tests after surface disinfection.

Viral load and concentration
A total of 17 studies reported data on viral concentration (Table 5); the units of measure used to report this data varied across the studies and included genomic copies/swab (4 studies), genomic copies/cm 2 (4 studies), genomic copies/mL (4 studies), and 1 study each for mean concentration, viruses/cm 2 , genomic units/m 2 , genomic copies/sample and RNA copies. We found it impossible to make any comparisons across the studies because of the heterogeneity in units of measurement.

Viral culture
Of the 11 studies that planned to perform viral culture, only two (18.2%) reported Ct values that could act as prompts to undertake viral isolation (Table 6). Only two studies provided information on the timing of sample collection for viral culture but were missing key details with respect to collection related to the timing of the onset of symptoms of the patients with respect to the collection and timing. One study of subjects in a cruise ship [Suzuki 2020] reported collecting samples for viral culture from 1-17 days after the cabin was vacated on a cruise ship and at least 17 days after the quarantining to cabins was ordered and 8 days after the first cabin cleaning, while another study of patients in residential isolation [Santarpia 2020] reported collecting the samples on "days 5-9" or "day 10" of occupancy at a medical centre or quarantine unit, all of whom were evacuated from the same cruise ship reported previously and would have been at least 2 weeks from the last day of quarantine [Suzuki 2020]. The incubation period ranged from 4-7 days and there were subtle differences in the culture media used across the studies ( Table 6). None of the studies reported success with viral culture despite positive RT-PCR detection tests. There were methodological issues with the techniques employed for viral culture across the studies (see Table 6).

Discussion
We found 63 primary studies investigating the role of fomites in SARS-CoV-2 transmission. The results of the majority of these studies show that SARS-CoV-2 RNA can be frequently detected on surfaces in both healthcare and nonhealthcare settings. However, there were no positive culture results for studies that attempted to culture for viable virus.
There is a wide variation in study setting and designs across studies, and the overall quality of published studies is low to moderate. The heterogeneity in study design and methodology makes it difficult to compare results across studies. The results of the systematic review (n=35) [Bedrosian 2020] showed that surface contamination was greatest in laboratories and least in households; however, none of the included studies addressed viral infectivity. The review authors did not assess the reporting quality of the primary studies and the search periods are now outdated.
The inability to culture the virus despite positive PCR detection tests suggests that SARS-CoV-2 RNA is more stable (and likely found in greater concentrations) on fomites than infective SARS-CoV-2 virus 11 . Factors known to affect the ability of fomites to serve as transmitters of respiratory viruses include the rate of decay of the virus on the surface and on the hands, the virus transfer rate (surface to hand, and hand to face), the frequency of touch between the hands and face, the dose-response curve of the virus, temperature and humidity, amongst others 12 .
The substantial reduction in positive detection rates before and after studies (and in some ICU settings) suggests that good hygiene procedures can minimise the risk of surface contamination. The inconsistency in describing a priori Ct values across the studies, coupled with the wide range in actual Ct values, suggests that the reported positive SARS-CoV-2 RNA detection rates are markers of previous viral presence from non-viable virus.
In a systematic review assessing the role of fomites in virus transmission in the Middle East Respiratory Syndrome (MERS) 13 , the authors reported possible evidence of fomite contamination but the evidence for fomite transmission was anecdotal. Our review findings are consistent with these observations. In an observational study of four hospitalised patients with MERS 14 , there was positive viral culture from fomites including bed sheets, bed rails, intravenous fluid hangers, and radiograph devices. In contrast to that study, published research on SARS-CoV-2 shows no evidence of positive viral culture We identified one non-peer-reviewed (pre-print) systematic review that assessed SARS-CoV-2 contamination in fomites 18 . The authors concluded that the quality of measurements was poor, and the reliability of the data is uncertain. Our findings are consistent with these. Compared to that review, we searched more databases, included more than twice the number of included studies, and accounted for the reporting quality of included studies.
Although there has been much research into fomite transmission of SARS-CoV-2, much uncertainty remains, and it is difficult to draw meaningful conclusions. Firstly, the variation in Ct across the studies suggests that there is no standardized threshold for detection of SARS-CoV-2 RNA. Some studies have shown that lower Ct correlates with higher genomic load 19 .
The studies included in this review used Ct of <35 to <43; these threshold values indicate that some of the positive tests reported in the studies may be misleading. Future research aimed at establishing internationally accepted Ct values should be considered a priority. The discrepancies in units of measurements for viral load and/or concentration also creates confusion. Therefore, standardized checklists for reporting of studies investigating SARS-CoV-2 transmission should be developed, including mandatory publishing of protocols, including the timing of the collection of any environmental specimens with respect to patient symptom onset. Looking for viable virus long after a patient has developed a significant innate and adaptive immunologic response will consistently yield negative results.
That all 11 culture studies failed to isolate the virus with significant fundamental methodological flaws indicates that the threshold for transmissibility from contaminated surfaces is unknown and more rigorous and carefully orchestrated studies are required before any conclusions may be drawn. One factor likely relates to the timing of sample collection after the onset of infection. Two studies reported the timeframe for sample collection but without precision while nine did not report any timelines. The mean incubation period of SARS-CoV-2 is 5-6 days 2 ; therefore, sample collection within the first few days of infection onset is likely to yield greater viral RNA load and result in better infectivity and culture results. Future studies should endeavour to collect surface samples of likely contaminated surfaces and medical equipment within useful timeframes and should also report this variable with their results.
As reported in the results, findings from one study [Lv 2020] showed that detection rates were different when qRT-PCR was compared with ddPCR. Interestingly, the authors of another included study [Bloise 2020] concluded that qRT-PCR is unable to differentiate between infectious and non-infectious viruses. Therefore, the use of RT-PCR as the gold standard for detection of SARS-CoV-2 requires further research. The positive findings from the before and after studies show that good hygiene procedures should continue to be a cornerstone for the management of SARS-CoV-2 and other communicable diseases.

Strengths and limitations
To our knowledge, this is the most comprehensive review to date that evaluates the role of fomites in SARS-CoV-2 transmission. We extensively searched the literature for published studies and included studies that are yet to undergo peer review. We also accounted for the quality of the studies and have presented summary data for some subgroups where possible. However, we recognize several limitations. We may not have identified all published studies investigating the role of fomites; indeed, several other studies may have been published after the last search date for this review. Heterogeneity due to variations in study designs and lack of uniformity in measurement metrics prevent us from statistically combining data across studies and limits the validity and applicability of the review results.

Conclusion
The evidence from published research suggests that SARS-CoV-2 RNA can be readily detected on surfaces and fomites. There is no evidence of viral infectivity or transmissibility via fomites to date but no studies to date have been found to be methodologically robust and of high enough quality to even adequately address the question. Good hygienic practices appear to reduce the incidence of surface contamination.
Published studies are heterogeneous in design, methodology and viral reporting metrics and there are flaws in the reporting quality. Standardized guidelines for the design and reporting of research on fomite transmission should now be a priority.

Data availability
Underlying data All data underlying the results are available as part of the article and no additional source data are required.

Open Peer Review
Department of Microbiology, Biochemistry and Molecular Genetics, New Jersey Medical School, Newark, NJ, USA I'm still not entirely convinced that the "bias" parameter included in the analysis is useful, since there is some element of subjectivity in making that assessment. Nevertheless, the authors have given a fuller accounting of what's involved in this parameter, so at least, a reader can draw their own conclusions regarding alleged "bias".
Overall, the authors have appropriately responded to my comments and suggestions, and I am satisfied that this manuscript is now acceptable as a peer-reviewed publication and a useful contribution to the literature.

Discussion
In the following sentence, "We identified one non-peer-reviewed (pre-print) systematic review that assessed fomite contamination in SARS-CoV-2." perhaps what the authors intended to say was, "We identified one non-peer-reviewed (pre-print) systematic review that assessed SARS-CoV-2 contamination in fomites."?

Figure 3
It would be more informative for the reader if the plot could distinguish between public spaces and rooms with COVID-19 patients. The authors could break the "other" category and include "isolation or quarantine room", since the most frequently contaminated spaces were quarantine rooms.
response curve of the virus, among other parameters. Even if the virus does not "grow" on the surface, there is still potential for transmission. Therefore, this sentence needs to be removed or clarified. Authors' response: We have revised the statement and added a reference. Now reads thus: "The inability to culture the virus despite positive PCR detection tests suggests that SARS-CoV-2 RNA is more stable (and likely found in greater concentrations) on fomites than infective SARS-CoV-2 virus [11]". Factors known to affect the ability of fomites to serve as transmitters of respiratory viruses include the rate of decay of the virus on the surface and on the hands, the virus transfer rate (surface to hand, and hand to face), the frequency of touch between the hands and face, the dose-response curve of the virus, temperature and humidity, amongst others [12]." Peer reviewer's comment: Second hypothesis: "The inability to culture the virus despite positive PCR detection tests may indicate … that the timing of collection was at a point of time where no viable virus would be likely to be found." -The authors suggest throughout the discussion that the 11 studies that cultured the virus showing not cytopathic effects have "significant fundamental methodological flaws" (which are not clearly pointed out in the manuscript), directing the reader to conclude that this is probably the reason why no attempt at culturing viruses from surfaces has been successful. It's helpful to point out that SARS-CoV-2 RNA is much more stable on fomites than infective SARS-CoV-2 virus, and it is found in much greater concentrations. For example, in an experimental study where SARS-CoV-2 was inoculated on surfaces, Paton et al. (2021) 1 found that the number of RNA recovered from the surfaces was 10 3 to 10 8 times higher than the number of viable virus in the samples. Given the relatively low concentration of RNA found on surfaces in the compiled articles, it is expected that the 11 studies that attempted to culture the virus could not find cytopathic effects. It is true, as the authors point out, that the transmission via fomites cannot be discarded. Nevertheless, the evidence that the authors compiled can already give an idea of how unlikely fomite-mediate transmission is in most settings. Authors' response: We have revised the statement.
revised as suggested.

Peer reviewer's comment: Discussion
In the following sentence, "We identified one non-peer-reviewed (pre-print) systematic review that assessed fomite contamination in SARS-CoV-2." perhaps what the authors intended to say was, "We identified one non-peer-reviewed (pre-print) systematic review that assessed SARS-CoV-2 contamination in fomites."? Authors' response: We have made the correction. Now reads thus: "We identified one nonpeer-reviewed (pre-print) systematic review that assessed SARS-CoV-2 contamination in fomites…" Peer reviewer's comment: Figure 3 It would be more informative for the reader if the plot could distinguish between public spaces and rooms with COVID-19 patients. The authors could break the "other" category and include "isolation or quarantine room", since the most frequently contaminated spaces were quarantine rooms. Authors' response: We understand the reviewer's point here. However, we reported the settings are described by the study authors -only 2 studies were described as "hospital and quarantine". Therefore we have not made any changes to the figure.
increasingly moved towards the view that fomite transmission is not a significant source of infection. This reviewer believes the authors are too timid in drawing that conclusion, only stating in the Abstract "Lack of positive viral cultures and variation in cycle thresholds create uncertainty about fomites as a mode of transmission." I believe a more forceful statement about the insignificance of fomite transmission is warranted.
In the second paragraph of the Introduction, the authors do not adequately characterize the content of some of the references cited.
The authors write, "it has been reported that SARS-CoV-2 can be transmitted indirectly through fomites or surfaces 5 ." This is inaccurate; reference 5 is a review of work published up to that time, and at most, suggests the possibility that the virus can be transmitted through fomites. None of the papers reviewed in reference 5 showed indirect transmission through fomites. The authors must revise this sentence. My suggestion is to replace "reported" with "suggested", i.e., "it has been suggested that SARS-CoV-2 can be transmitted indirectly through fomites or surfaces 5 ." The opposite situation occurs in the very next sentence. The authors write, "However, some authors have suggested that there is a low risk of transmission of SARS-CoV-2 through fomites 6,7 ." Reference 6 is more than a suggestion; the authors report the absence of infectious virus on surfaces in hospitals treating COVID-19 patients. My suggestion here is the reverse of my previous one, in this case, replace "suggested" with "reported" for reference 6, i.e., "However, some authors have reported 6 or suggested 7 that there is a low risk of transmission of SARS-CoV-2 through fomites." In the last paragraph of the methods section, the authors write: "One reviewer (IJO) assessed the risk of bias and extracted data from the included studies, and these were independently checked by a second reviewer (EAS)." However, the reader is not given a clue as to what the authors mean by "bias" in this context. The absence of a definition and/or criteria for alleged "bias" makes this parameter meaningless in the manuscript. If the authors are to include this parameter, they must revise the manuscript to make clear what it is they are assessing as "bias", and how that assessment was made.
Are the rationale for, and objectives of, the Systematic Review clearly stated? Yes

Is the statistical analysis and its interpretation appropriate? Yes
Are the conclusions drawn adequately supported by the results presented in the review? Yes Competing Interests: No competing interests were disclosed.

Reviewer Expertise: Microbiology, Virology
I confirm that I have read this submission and believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard, however I have significant reservations, as outlined above.

Competing Interests: None
The benefits of publishing with F1000Research: Your article is published within days, with no editorial bias • You can publish traditional articles, null/negative results, case reports, data notes and more • The peer review process is transparent and collaborative • Your article is indexed in PubMed after passing peer review • Dedicated customer support at every stage • For pre-submission enquiries, contact research@f1000.com