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

Five-Year View

Currently, the world population is approximately 7 billion. Over the next 5 years the percentage of patients classified as geriatric is expected to increase as the 'baby boomer generation' reaches their sixties. Due to increasing numbers of comorbid states in the geriatric population, acute diseases such as pneumonia are more likely to require hospitalizations and incur significant healthcare costs. In the era of decreasing healthcare funding, it is imperative to evaluate all drug therapies for appropriate use. When evaluating appropriateness the following should be considered: adverse reactions, drug interactions, duration of therapy, correct dose (supratherapeutic/subtherapeutic), appropriate indication and route of administration. Our meta-analysis evaluated one adverse effect from PPIs, and found an association with PPI use and the development of CAP, particularly with increased dosage and shorter duration of PPI use.

Postmarketing experience has evaluated many adverse effects of PPIs not initially recognized in clinical trials, such as Clostridium difficile infections, osteoporosis, acute interstitial nephritis and pneumonia (both community and hospital acquired). These adverse effects are typically studied individually and authors make conclusions based only on the specific adverse effect that is studied. In clinical practice, it is understood to some degree that these adverse effects may happen, but it is hard to quantify results between different trials. For example, what would the number needed to harm (NNH) be when adding all of the adverse effects of PPIs together? Should different adverse effects be weighed differently when calculating a number-needed to harm? If the indication is appropriate what is the number-needed to treat (NNT)? Should the NNT based on outcome be weighed differently against a NNH based on adverse effect? What is the NNT and NNH difference when looking at H2RAs versus PPIs? All of these questions are important when trying to identify which patients will benefit the most from PPI therapy or when we should use alternate therapies such as H2RAs. Future studies should strive to answer these questions so a more systematic approach can be taken when deciding what the best choice is for a patient with an appropriate indication. It would also be important to look at all adverse effects in the same population, not compare incidences of adverse effects between different populations.

If PPIs continue to be overprescribed, the healthcare costs related to the adverse effects of PPIs will only increase with increasing geriatric populations. Several approaches can be taken to decrease these costs. First, PPI overuse needs to be curtailed. Educational programs, computerized reminders, emails, direct education, phone calls, pharmacist intervention and student intervention are all possible ways of discouraging overuse. Combining these methods would likely have the best results. Second, original studies that are conducted should evaluate all PPI adverse effects and efficacy so we can better quantify a benefit:risk ratio that is usable in clinical practice. By using these approaches, the unforeseen impact of PPIs on healthcare costs may be avoided at a time where decreasing healthcare funding is being carefully reviewed.