Pooled Prevalence of Diarrhea Among COVID-19 Patients

Dear Editor:

We read the article by D’Amico et al¹ discussing the pathogenesis, epidemiology, prevention, and management of diarrhea among patients with coronavirus disease 2019 (COVID-19), caused by the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). We agree on the importance of such reviews, primarily targeted to gastroenterologists. Nevertheless, we would like to discuss additional implications and analyses presented by D’Amico et al¹ regarding the prevalence of diarrhea derived from available published studies, and its comparison with SARS-CoV and the Middle East respiratory syndrome coronavirus (MERS-CoV) infections.

D’Amico et al¹ analyzed in the results from 20 studies in Table 1 in their article, with a pooled prevalence of 10.4%, although no information on the meta-analysis model was provided. We used that data, and additionally included 11 novel studies, published through May 3, 2020, also assessing the frequency of diarrhea among COVID-19 patients. With a total of 3335 patients from 31 studies, we used random-effects and fixed-effects models to determine the pooled prevalence. We used the Open Meta-Analyst software (Providence, RI). For the random-effects model, we found that 13.8% of patients presented with diarrhea (95% CI, 10.6%–17.0%), and for the fixed-effects model, 6.1% presented with diarrhea (95% CI, 5.4%–6.9%) (Figure 1 ). In the case of SARS-CoV and MERS-CoV, D’Amico et al¹ did not provide pooled prevalences for comparisons with SARS-CoV-2. Then, we used the data D’Amico et al¹ presented about these previous coronaviruses regarding diarrhea and we added 2 studies on SARS-CoV. Combining the data from 735 patients with SARS-CoV, the pooled prevalence yielded 12.4% (95 CI, 4.8%–20.0%) in the random-effects model, and 2.3% (95% CI, 1.3%–3.3%) in the fixed-effects model (Figure 1). In the case of MERS-CoV, this was 23.1% (95% CI, 10.7%–35.4%) and 17.5% (95% CI, 13.1%–21.9%), respectively (Figure 1).

Figure 1.

Random-effects and fixed-effects model for the pooled prevalence of diarrhea among COVID-19 patients.9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21

In addition to the cohort, cross-sectional, and case series studies, D’Amico et al¹ also included 3 case reports. We also looked at those and found 6 more case reports. Most of them are from China (5), followed by Japan (2), the United States (1), and Lebanon (1), with a median diarrhea duration of 3 days, with the number of evacuations ranging from 2 to 6 per day.2, 3, 4, 5, 6, 7

In a previous meta-analysis, we found that diarrhea was reported among 6.1% (95% CI, 2.4%–9.7%) of patients with COVID-19 (6 studies, 457 patients),⁸ which was the same prevalence obtained now with 31 studies and 3335 patients by the fixed-effects model in the current analysis (Figure 1). Although this would be considered a low prevalence, the COVID-19 pandemic has affected 10.49 million people worldwide (as of July 1, 2020), translating into between 566,897 and 724,368 diarrhea-associated cases (based on 95% CIs for pooled prevalence).

Because SARS-CoV-2 directly or indirectly may affect the enteric mucosa,¹ diarrhea and other gastrointestinal findings should raise clinical suspicion for COVID-19, with or without the presence of fever, cough, and other respiratory and nonrespiratory manifestations.⁸

Finally, as D’Amico et al¹ mentioned, fecal–oral transmission may be an additional source of transmission that deserves more research and subsequent preventive interventions.