Mixing Antiplatelet Agents and Proton-Pump Inhibitors

David A. Johnson, MD


April 15, 2010

Outcomes With Concurrent Use of Clopidogrel With Proton Pump Inhibitors

Ray WA, Murray KT, Griffin MR, et al.
Ann Intern Med 2010;152:337-355

The Antiplatelet Effect of Aspirin Is Reduced by Proton Pump Inhibitors in Patients With Coronary Artery Disease

Würtz M, Grove EL, Kristensen SD, Hvas AM
Heart 2010;96:368-371


A tremendous amount of controversy has arisen with respect to the concomitant use of clopidogrel and proton-pump inhibitors (PPIs). Despite strong evidence from prospective trials that this potential interaction has minimal or no significant cardiovascular adverse outcomes, government authorities such as the US Food and Drug Administration (FDA), as well as the European Medicines Agency (EMEA), have issued statements that these agents should not be used concurrently.[1,2] The FDA cautions healthcare professionals about prescribing omeprazole and esomeprazole, in particular, along with clopidogrel.[1]

Most recently, the EMEA's Committee for Medicinal Products for Human Use, in a document dated March 17, 2010, recommended an amendment to the existing warning about the concomitant use of clopidogrel and PPIs. The class warning for all PPIs has been replaced with a warning stating that only the concomitant use of clopidogrel and omeprazole or esomeprazole should be discouraged.[3]

Study Summaries

Ray and colleagues' retrospective cohort study used automated access to data to evaluate patients who had received clopidogrel during a focused analysis from 1999 to 2005 in a Tennessee Medicaid program. Of 20,596 patients who were hospitalized for myocardial infarction, coronary revascularization, or unstable angina, 7593 were taking concomitant PPIs. Primary outcomes were events of gastrointestinal bleeding or cardiovascular disease complications, including any cardiovascular-related event or death.

The adjusted risk for gastrointestinal bleeding in patients taking concomitant clopidogrel and PPIs was reduced by 50% (hazard ratio, 0.50 [95% confidence interval (CI), 0.39-0.65]) but no significant association between this exposure and more adverse cardiovascular disease complications was found. Of note, for patients defined as being at the highest risk for gastrointestinal bleeding, the PPI exposure was associated with an absolute risk reduction of 28.5 (95% CI, 11.7-36.9) in the number of hospitalizations for gastrointestinal bleeding per 1000 patient-years.

Würtz and colleagues evaluated a possible negative influence of PPIs on the antiplatelet effect for concomitant aspirin use. Investigators from Denmark evaluated 54 patients receiving PPIs and aspirin (75 mg/d) from a total group of 418 patients with stable coronary artery disease who were taking aspirin. Patients receiving concomitant PPIs had higher platelet aggregation (P = .003), soluble serum P-selectin (a cell adhesion molecule stored in the alpha granules of platelets) levels (P = .005), and higher serum thromboxane B2 levels (P = .01), all of which suggested that these patients may have a reduced antiplatelet cardiovascular protective effect. The proposed method of action for this altered effect was an increase in intragastric pH caused by PPIs that resulted in reduced lipophilicity of aspirin. Under physiologic acidic conditions, according to the pH partition hypothesis, aspirin is absorbed in its lipid state by passive diffusion across the gastric mucosal membrane. The investigators suggested that if the pH potentially rises above the pKa (3.5) of aspirin, there may be a pronounced reduction in the lipophilicity of aspirin. According to previous studies, such chemical changes might compromise the bioavailability and therapeutic efficacy of aspirin.


Clearly a barrage of new data continues to challenge the cardiovascular safety of concomitant use of PPIs and antiplatelet therapy. Although somewhat limited by the retrospective study design, the study by Ray and associates provides further evidence that the combined use of PPIs and clopidogrel for patients deemed at high risk for gastrointestinal complications of antiplatelet therapies is safe and effective for the intended effect (prevention of gastrointestinal bleeding). No trade-off cost (eg, precipitation of potentially more formidable cardiovascular complications) was associated with concomitant use of these 2 drugs.

Although all of the PPIs were analyzed for potential risk-associated events, most PPI exposure was with pantoprazole (62%). Given the small event rates for the other PPIs, it would be inappropriate to base a weighted recommendation for a specific PPI on these findings.

Würtz and colleagues did not present any data on differences in adverse cardiovascular clinical outcomes that might be associated with the observed differences in platelet function. These data seem to have a distinct parallel with the clopidogrel data, where intermediate endpoints of physiologic differences in platelet function have not been clearly associated with clinical outcomes. Accordingly, before clinical adverse cardiovascular consequences can be inferred, further study is warranted to investigate potential causality of these adverse effects. Furthermore, studies will need to assess both gastrointestinal and cardiovascular harm or benefits to assess the full spectrum of clinical outcomes for combined aspirin and PPI therapy. At present, however, it would be an error in judgment to use this study to alter well-established clinical recommendations for the use of PPI prophylaxis in patients receiving aspirin who are deemed at high risk for upper gastrointestinal ulcer-related complications.