Risk of community-acquired pneumonia with outpatient proton-pump inhibitor therapy: a systematic review and meta-analysis

Abstract

Background: Proton-pump inhibitors (PPIs) are among the most frequently prescribed medications. Community-acquired pneumonia (CAP) is a common cause of morbidity, mortality and healthcare spending. Some studies suggest an increased risk of CAP among PPI users. We conducted a systematic review and meta-analysis to determine the association between outpatient PPI therapy and risk of CAP in adults.

Methods: We conducted systematic searches of MEDLINE, EMBASE, CINAHL, Cochrane Central Register of Controlled Trials, Scopus and Web of Science on February 3, 2014. Case-control studies, case-crossover, cohort studies and randomized controlled trials reporting outpatient PPI exposure and CAP diagnosis for patients ≥18 years old were eligible. Our primary outcome was the association between CAP and PPI therapy. A secondary outcome examined the risk of hospitalization for CAP and subgroup analyses evaluated the association between PPI use and CAP among patients of different age groups, by different PPI doses, and by different durations of PPI therapy.

Results: Systematic review of 33 studies was performed, of which 26 studies were included in the meta-analysis. These 26 studies included 226,769 cases of CAP among 6,351,656 participants. We observed a pooled risk of CAP with ambulatory PPI therapy of 1.49 (95% CI 1.16, 1.92; I2 99.2%). This risk was increased during the first month of therapy (OR 2.10; 95% CI 1.39, 3.16), regardless of PPI dose or patient age. PPI therapy also increased risk for hospitalization for CAP (OR 1.61; 95% CI: 1.12, 2.31).

Discussion: Outpatient PPI use is associated with a 1.5-fold increased risk of CAP, with the highest risk within the first 30 days after initiation of therapy. Providers should be aware of this risk when considering PPI use, especially in cases where alternative regimens may be available or the benefits of PPI use are uncertain.

Fig 1. Flow Diagram of Study Selection, adapted from PRISMA Group 2009 Flow Diagram.

Studies could be excluded for more than one reason, therefore the sum of exclusion reasons exceeds total studies. Examples of studies lacking extractable data included abstracts without full results (such as abstracts reporting an odds ratio without exposure or outcome definitions) or papers summarizing outcomes qualitatively in the text (such as manuscripts reporting adverse event categories that grouped pneumonia among other events).

Fig 2. Summary Forest Plot of Overall Risk of Community-Acquired Pneumonia with Outpatient Proton Pump Inhibitor Use, subdivided by study design and effect estimate.

Solid diamond represents effect estimate. Shaded box size is proportional to the weight of the study in the meta-analysis. Confidence intervals are denoted by horizontal lines, with arrows where confidence interval extends beyond figure. Vertical dashed line represents the null effect. The open diamond is centered at the summary effect estimate and proportional to the confidence interval.