No consistent evidence of differential cardiovascular risk amongst proton-pump inhibitors when used with clopidogrel: meta-analysis

Int J Cardiol. 2013 Aug 10;167(3):965-74. doi: 10.1016/j.ijcard.2012.03.085. Epub 2012 Mar 30.


Background: Data from pharmacokinetic and pharmacodynamic studies indicate that the adverse clopidogrel-proton pump inhibitor (PPI) interaction may vary between PPIs, with pantoprazole considered relatively less problematic. We aimed to evaluate systematically whether individual PPIs differ in their risk for cardiovascular events when concomitantly administered with clopidogrel.

Methods: We searched MEDLINE, EMBASE and Cochrane Trials Register up to December 2011 for randomized and non-randomized studies that reported adverse cardiovascular events with exposure to specific PPIs in patients receiving clopidogrel. We performed random effects meta-analysis, and assessed heterogeneity using the I(2) statistic.

Results: A total of 23 studies with 222,311 participants were included. Meta-analysis of major adverse cardiovascular events was mostly limited by moderate-substantial heterogeneity. Pooled estimates of cardiovascular risk were significantly elevated for individual PPIs such as omeprazole, esomeprazole, lansoprazole, and pantoprazole when used with clopidogrel. However, meta-analysis of adverse cardiovascular risk in seven observational studies reporting on PPI therapy alone (without concomitant clopidogrel) also found an elevated odds ratio of 1.28 (95% CI 1.14-1.44) compared with no clopidogrel/no PPI exposure. Meta-analysis of two randomized controlled trials did not show significant adverse cardiovascular effect from omeprazole or esomeprazole.

Conclusions: The absence of consistent evidence on differential cardiovascular risk amongst PPIs (particularly regarding safety of pantoprazole) is in direct opposition to the platelet function and pharmacokinetic data. Our findings of increased cardiovascular risk with PPIs in the absence of clopidogrel suggest that confounding and bias are strong possibilities. The clinical validity or relevance of the hypothesized PPI-clopidogrel interaction remains questionable.

Keywords: Clopidogrel; Drug interaction; Meta-analysis; Proton pump inhibitors.

Publication types

  • Meta-Analysis

MeSH terms

  • Cardiovascular Diseases / drug therapy*
  • Cardiovascular Diseases / epidemiology
  • Clinical Trials as Topic / methods
  • Clopidogrel
  • Drug Interactions / physiology
  • Drug Therapy, Combination
  • Humans
  • Observational Studies as Topic / methods
  • Platelet Aggregation Inhibitors / administration & dosage*
  • Platelet Aggregation Inhibitors / adverse effects
  • Proton Pump Inhibitors / administration & dosage*
  • Proton Pump Inhibitors / adverse effects
  • Randomized Controlled Trials as Topic / methods
  • Registries
  • Risk Factors
  • Ticlopidine / administration & dosage
  • Ticlopidine / adverse effects
  • Ticlopidine / analogs & derivatives*


  • Platelet Aggregation Inhibitors
  • Proton Pump Inhibitors
  • Clopidogrel
  • Ticlopidine