The Risk of Community-Acquired Enteric Infection in Proton Pump Inhibitor Therapy: Systematic Review and Meta-analysis

Background: Previous meta-analyses suggest that users of proton pump inhibitors (PPIs) have a higher risk of developing enteric infections compared with nonusers. These previous meta-analyses have considerable heterogeneity, and it is not clear whether the effect of PPIs is different for different types of microorganisms. Objective: The aim of this study is to update previous meta-analyses, concentrating on enteric infection in community settings and exploring potential sources of heterogeneity. Methods: A systematic search was conducted on electronic databases (all available years until November 2017). PubMed, EMBASE, Cochrane, and Web of Science were searched using specific keywords related to PPI therapy and community-acquired enteric infection. Eligible studies were selected based on prespecified criteria. Results: A total of 9 observational studies evaluating community-acquired enteric infection were eligible, including 12 separate analyses. The meta-analysis showed that PPI users have an increased risk of developing community-acquired enteric infection (pooled odds ratio [OR] = 4.28; 95% CI = 3.01-6.08). There was significant heterogeneity between the studies (I-2 = 85%; P < 0.001), which was partly explained by type of microorganism. The strength of the association was similar for Salmonella (pooled OR = 4.84; 95% CI = 2.75-8.54; I-2 = 58.7%; P = 0.064) and Campylobacter (pooled OR = 5.09; 95% CI = 3-8.64; I-2 = 81%; P < 0.001) but lower for studies that combined all bacteria (pooled OR = 2.42; 95% CI = 0.96-6.14; I-2 = 94.3%; P < 0.001). Conclusion: PPI users have an increased risk of developing community-acquired enteric infections compared with nonusers. The heterogeneity was partially explained by type of microorganism; the association is stronger for Salmonella and Campylobacter than for all bacteria combined.