• endoscopy
• infectious disease
• health service research
• clinical decision making
• infectious diarrhoea

Over 3000 healthcare workers (HCW) in China are suspected of having coronavirus disease 2019 (COVID-19) and over 1700 tested positive.1 These statistics underline the need for robust preventative measures against the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). Endoscopy departments are fertile grounds for viral spread because aerosolisation of bodily secretions occurs during procedures. A single viral-shedding patient with a high viral load can contaminate an entire endoscopy room with the virus that remains viable for up to 3 days, putting uninfected patients and HCWs at risk.2 3

Singapore previously had the largest cohort of COVID-19 patients outside China in the early phases of the outbreak. Given its novelty, the effectiveness of new preventative measures implemented within our endoscopy services was unknown. To determine best practice, we conducted systematic searches of literature and official websites for gastroenterology and endoscopy societies (n=28) in the 15 most-affected countries to scrutinise recommendations and associated evidence. Methodology is available on request.

In summary, we found careful patient selection was commonly advised but protocols for screening and triaging differed (table 1). The two most important differences observed were: (1) type of personal protective equipment (PPE) recommended and (2) postprocedure risk management (table 2). Only 32% (9/28) of all gastrointestinal (GI) related societies reviewed had provided guidance as of 16 March 2020. A universal gold standard was lacking. One article reported the effect of preventative measures on the incidence of new COVID-19 cases but the sample size was small and period of observation abrupt.4

Patient screening undoubtedly is the foremost step at preventing nosocomial transmission; timely detection allows postponement of non-urgent procedures until the infection has resolved, significantly reducing transmission risk to patients and staff. However, the median incubation time of the virus is 5.1 days but can extend to 14 days (99th percentile), meanwhile patients remain asymptomatic or have subclinical symptoms and may be infectious.5 6 This limits screening protocols reliant on symptomatology. GI symptoms of COVID-19 are also non-specific. Travel history becomes limited when COVID-19 becomes more rampant in local communities so contact screening for exposure to individuals who have symptoms of COVID-19 may be more useful. Nonetheless, data on the accuracy of question-based screening tools were not identified.

Current limitations of screening place greater importance on risk management strategies postprocedure. Detecting ‘false negatives’ that slip through processes allows for the identification of HCWs and patients with infection risk after exposure to asymptomatic or subclinical carriers in the viral incubation period at the time of endoscopy. A robust contact screening programme is then necessary to contain the spread of COVID-19 among exposed staff and patient contacts. Only one guideline identified in our review has advised on postprocedure patient follow-up on day 7 and day 14 by telephone.7

No evidence of SARS-CoV or SARS-CoV-2 transmission through endoscopy was identified. SARS-CoV-2 has been isolated in gastric, duodenal and rectal biopsies, and faecal viral RNA is detectable in half of all COVID-19 patients although there is a poor correlation to GI symptoms.8 9 Nonetheless, reports may surface in the future and suspicion for faecal-oral transmission should remain high. US and UK guidelines regarded lower endoscopy as low risk and therefore were less stringent with PPEs compared with China or Singapore (table 2). We have erred on the side of caution because the microbial contamination of surroundings after lower endoscopy has been reported.10 11 Differences in recommendations may also have been influenced by resource availability and health policies.

In our experience, resource allocation for staff education, decontamination and management of the physical and mental well-being of HCWs were also crucial. In conclusion, better evidence is needed to inform current practice. A postprocedure risk management programme can help prevent the nosocomial and community spread of SARS-CoV-2 and should not be neglected.