Methods: We searched three electronic databases (MEDLIN E [PubMed], Embase and the Cochrane Library) from inception to Aug. 28, 2009. Two evaluators independently extracted data. Because of heterogeneity, we used randomeffects meta-analysis to obtain pooled estimates of effect.

Results: We identified 31 studies: five case–control studies, three cohort studies and 23 randomized controlled trials. A meta-analysis of the eight observational studies showed that the overall risk of pneumonia was higher among people using proton pump inhibitors (adjusted odds ratio [OR] 1.27, 95% confidence interval [CI] 1.11–1.46, I² 90.5%) and histamine₂ receptor antagonists (adjusted OR 1.22, 95% CI 1.09–1.36, I² 0.0%). In the randomized controlled trials, use of histamine₂ receptor antagonists was associated with an elevated risk of hospital-acquired pneumonia (relative risk 1.22, 95% CI 1.01–1.48, I² 30.6%).

Interpretation: Use of a proton pump inhibitor or histamine2 receptor antagonist may be associated with an increased risk of both community- and hospital-acquired pneumonia. Given these potential adverse effects, clinicians should use caution in prescribing acid-suppressive drugs for patients at risk.