Are Proton Pump Inhibitors Associated With the Development of Community-Acquired Pneumonia?

A Meta-analysis

Christopher Giuliano; Sheila M Wilhelm; Pramodini B Kale-Pradhan


Expert Rev Clin Pharmacol. 2012;5(3):337-344. 

In This Article

Expert Commentary

Our meta-analysis shows that the use of PPIs is associated with an increased risk of developing CAP. CAP is hypothesized to develop during PPI therapy due to an increased gastric pH, leading to subsequent bacterial colonization and aspiration of bacteria.[19–21] This meta-analysis also showed an association with higher dose (defined as higher dose than the usual dose) and shorter duration of PPI use with the development of CAP. No increased risk was seen with long-term therapy. The higher dose of PPI leading to an increased risk of CAP could be hypothesized to be related to a further increase in pH over standard dosing leading to an increased risk of bacterial colonization. Long-term PPI use showing no increased risk for CAP has been hypothesized to be related to the following reasons: decreased immune response regulation from PPIs; patients more likely to develop pneumonia developing it within the first 30 days of PPI use; and decreased compliance with long-term PPI use leading to lower suppression of gastric acid.[7,8]

Two other meta-analyses have evaluated the risk of CAP with PPI use.[7,8] Our findings are in agreement with these meta-analyses; however, there are some differences between our meta-analysis and the others. We have included a greater number of studies than previous meta-analyses, and we conducted a prespecified sensitivity analysis based on the quality of studies determined by the NOQAS and the patient population. This allowed us to determine possible reasons for heterogeneity, although heterogeneity still remained high except in some of the subgroup analyses. We also excluded one study from our primary analysis as PPI and H2RA data were inseparable,[24] although a previous analysis included this study.[8] Similar to other meta-analyses, we found a significant relationship between the development of CAP with high PPI dose and shorter duration of use.

There are several limitations to this meta-analysis. The studies included in our analysis had a high heterogeneity. This may be due to a number of factors. The quality of studies varied in our analysis based on NOQAS scores, although when lower quality studies were excluded heterogeneity remained high. Additionally, the NOQAS tool exhibits inter-rater variability.[36] Certain studies only evaluated specific patient groups, although when only population-based studies were included, heterogeneity still remained high. Among the studies, varying definitions of current PPI use and the differences in adjustment for confounding variables in each of the studies may have contributed to the high heterogeneity in our analysis. There are several patient factors that have been shown to increase the risk of CAP and are independent of acid suppression.[37] Several of these confounding variables were accounted for in the included studies. These variables included sex, age, tobacco use, alcohol use, total hospitalizations and/or physician office visits, past CAP, chronic obstructive pulmonary disease, asthma, congestive heart failure, chronic renal failure, cirrhosis, diabetes mellitus, stroke, cancer, myocardial infarction, dementia and specific medications. However, none of the included studies controlled for gastroesophageal reflux disease, and many studies did not control for alcohol or tobacco use, which are also risk factors for development of CAP.[37] Table 1 lists CAP risk factors which were not accounted for in each of the studies. In addition to the primary outcome, subgroup analyses were carried out to analyze the PPI dose and duration. However, there were few studies that could be included. There were inconsistencies in the classification of the duration of PPI use in the included studies. Therefore, we could not evaluate the association of PPIs and CAP between 30 and 180 days of PPI use.

AST is commonly utilized in the institutional practice setting for prophylaxis of stress-related mucosal damage. Up to 71% of hospitalized patients receive AST and 73% of these patients lack appropriate indications.[38,39] In addition, up to 69% of patients who receive inappropriate AST while hospitalized continue therapy after discharge.[38–41] The widespread use of PPIs may have been proliferated due to the perception that PPIs have a favorable adverse effect profile. The current meta-analysis shows that short-term PPI use is associated with development of CAP, a serious adverse effect. Patients currently receiving PPIs, particularly for less than 30 days or high dose, showed an association with CAP. PPIs should be initiated in patients that have appropriate indications for use, and doses should be optimized for indications. Practitioners need to be vigilant about the adverse effects of PPIs and consider alternative therapies.