Response to Letter by van Werkum: Proton pump inhibitors and clopidogrel: a difficult dilemma

Abstract

We thank Dr van Werkum and colleagues for their interest regarding our recent article. Limitations regarding a nonrandomized study design have well been discussed in our article. The use of calcium-channel blockers in each group was b20%, and the differences between groups were not statistically significant. The goal of our study was neither to provide a solution regarding risk stratification between upper gastrointestinal hemorrhage and stent thrombosis by concomitant use of proton pump inhibitor (PPI) nor to investigate the effect of different PPIs on pharmacokinetics of clopidogrel. We appreciate the authors' effort to reemphasize the need for randomized clinical trials addressing these issues. Nevertheless, we felt the need to report nonrandomized, real-word data because data on class effects of PPIs were lacking. Several studies investigating PPI-clopidogrel interaction came to opposite conclusions, which could be explained by differences in study designs. It is important to distinguish between studies investigating the impact of concomitant PPIs treatment on clinical outcome and trials examining the effect of PPIs on pharmacokinetics/ pharmacodynamics of clopidogrel. Secondly, some trials investigated the effect of PPIs as a class; other studies examined particular PPIs. Two retrospective analyses have shown that patients receiving PPIs and clopidogrel had significantly more major cardiovascular events than patients taking clopidogrel alone. In contrast, the subgroup analysis of the Clopidogrel for the Reduction of Events During Observation (CREDO) trial showed no increase in adverse effect rates in patients on concomitant PPI treatment. A casecontrol study in patients after a myocardial infarction has indicated that PPI-clopidogrel interaction might not be a class effect—the concomitant treatment with PPIs other than pantoprazole increased the risk of reinfarction. Studies examining the effect of different PPIs on the pharmacodynamics of clopidogrel have shown that omeprazole decreased the effect of clopidogrel, whereas lansoprazole, pantoprazole, and esomeprazole had no influence on the antiplatelet effect of clopidogrel. In summary, although data are controversial whether PPIs alter the antiplatelet effect of clopidogrel, a trend indicates that clopidogrel-PPI interaction might not be a class effect.