The association between proton pump inhibitors and the risk of gastrointestinal bleeding in oral anticoagulants users

Abstract

Current evidence of whether proton pump inhibitor (PPI) reduces the risk of gastrointestinal bleeding (GIB) associated with oral anticoagulants (OACs) is limited. Propensity score-weighted cohort and case-crossover studies were conducted separately in England and Hong Kong between 2011.01.01 and 2019.12.31. In the cohort design, we compared the hazards of hospitalised GIB in OAC + PPI users with OAC only users in people with atrial fibrillation and found higher hazard of GIB in OAC + PPI users in both settings. In the case-crossover design, elevated odds of exposure to PPI only, OAC only and OAC + PPI associated with GIB between 30-day hazard and referent periods were similarly found in both settings. Overall, the evidence of an elevated risk of OAC + PPI associated with GIB compared with OAC only was modest in the cohort study. Our case-crossover study suggested that residual confounding is likely to explain the association, suggesting that concomitant prescription of PPI with OAC did not modify GIB.