to the editor: I found this review of proton pump inhibitors (PPIs) to be complete and well written. However, the reported risks of hip fractures associated with PPI use and of cardiac events when PPIs are used in combination with clopidogrel (Plavix) were overstated.

Although case-control studies based on administrative data showed an increased odds ratio for PPI use in persons who sustained hip fractures, this association did not hold for persons without other preexisting fracture risk factors.¹ Prospective cohort studies have not shown an increased hip fracture risk in PPI users. PPI use does not appear to be associated with reduced bone density, which calls into question the hypothetical mechanism of reduced calcium absorption. Any association between PPI use and hip fractures might be explained by other risk factors in patients who tend to require PPIs, such as those who smoke.

As for the suggested increased risk of cardiac events in persons taking clopidogrel and PPIs, more recent cohort studies, a recent randomized controlled trial, and meta-analyses of the highest quality studies have not shown any association.² Therefore, I would be reluctant to withhold PPIs from patients who might benefit from them based on undue concerns about unsubstantiated risks.

in reply: We appreciate the detailed analysis and comments by Dr. Konrad. The goal of our article was to make readers aware of potential adverse effects of long-term PPI use. Clinical data suggest that long-term use needs to be scrutinized, especially in older patients.

Dr. Konrad's reference from the U.S. Food and Drug Administration (FDA) states that over-the-counter (OTC) PPIs “are only intended for a 14 day course of treatment up to 3 times per year. … Healthcare professionals should be aware of the risk for fracture if they are recommending use of OTC PPIs at higher doses or for longer periods of . time than in the OTC PPI label.”¹

As we stated, the clinical significance of interactions between PPIs and clopidogrel is unknown. However, retrospective analysis has shown an increased risk of rehospitalization and adverse coronary events with concomitant use of PPIs and clopidogrel. Also, the FDA required a change to the prescribing information for clopidogrel to reflect the increased risk.² We agree that patients taking clopidogrel who would clearly benefit from PPI use should be prescribed a PPI. However, selecting a PPI other than omeprazole (Prilosec), administering the PPI at a different time than clopidogrel, or substituting a high-dose histamine H₂ antagonist may be preferred.