Gastrointestinal and hepatic manifestations of COVID-19: A systematic review and meta-analysis

Abstract

Background and aim: This review investigates the role of gastrointestinal and hepatic manifestations in COVID-19, particularly with regard to the prevalence of isolated gastrointestinal (GI) symptoms.

Methods: We searched PubMed, Embase, and Cochrane library for COVID-19 publications from 1 December 2019 to 18 May 2020. We included any study that reported the presence of GI symptoms in a sample of >5 COVID-19 patients. Data collection and risk of bias assessment were performed independently by two reviewers. Where ≥3 studies reported data sufficiently similar to allow calculation of a pooled prevalence, we performed random effects meta-analysis.

Results: This review included 17 776 COVID-19 patients from 108 studies. Isolated GI symptoms only occurred in 1% (95% confidence interval [CI] 0-6%) of patients. GI symptoms were reported in 20% (95% CI 15-24%) of patients. The most common were anorexia (21%, 95% CI 15-27%), diarrhea (13%, 95% CI 11-16%), nausea or vomiting (8%, 95% CI 6-11%), and abdominal pain (4%, 95% CI 2-6%). Transaminase elevations were present in 24% (95% CI 17-31%) of patients. Higher prevalence of GI symptoms were reported in studies published after 1st April, with prevalence of diarrhea 16% (95% CI 13-20), nausea or vomiting 12% (95% CI 8-16%), and any GI symptoms 24% (95% CI 18-34%). GI symptoms were associated with severe COVID-19 disease (odds ratio [OR] 2.1, 95% CI 1.3-3.2), but not mortality (OR 0.90, 95% CI 0.52-1.54).

Conclusions: Patients with isolated GI symptoms may represent a small but significant portion of COVID-19 cases. When testing resources are abundant, clinicians should still consider testing patients with isolated GI symptoms or unexplained transaminase elevations for COVID-19. More recent studies estimate higher overall GI involvement in COVID-19 than was previously recognized.

Keywords: COVID‐19; SARS‐CoV‐2; gastrointestinal; meta‐analysis; systematic review.

Figure 1

PRISMA flow diagram.

Figure 2

Proportion of coronavirus disease (COVID)‐19 patients with only gastrointestinal (GI) symptoms, without fever. Overall prevalence of COVID‐19 infection with a period of only GI symptoms without fever was 1% (95% confidence interval 0–6%). Whether anorexia was included in the definition of GI symptoms did not affect the proportion with GI symptoms in a statistically significant manner (P = 0.49).

Figure 3

Proportion of all coronavirus disease (COVID)‐19 patients with elevated transaminases, by studies restricted to patients with respiratory symptoms or not. Overall prevalence of elevated transaminases in COVID‐19 infection was 24% (95% confidence interval 17–31%). This proportion was 33% in studies restricted to patients with respiratory symptoms, and only 21% in studies including patients without respiratory symptoms (P = 0.20).

Figure 4

Odds of mortality in coronavirus disease (COVID‐19) infection with gastrointestinal (GI) symptoms, by continent. Odds of mortality were not significantly different (odds ratio [OR] 0.90, 95% confidence interval [CI] 0.52–1.54) in COVID‐infected patients with GI symptoms compared to those without. In the two US studies, however, the pooled odds ratio for mortality with GI symptoms was 0.51 (95% CI 0.35–0.73),versusOR 1.31 (95% CI 0.72–2.40) in studies from China (P = 0.04 for difference).