Association Between Anticoagulants and Interacting Drugs and Risk of Major Bleeding in Nonvalvular Atrial Fibrillation: Case-Control Study in SIDIAP, Catalonia, Spain.

PURPOSE: Interactions with anticoagulants causing an increase in their effect may potentially enhance the bleeding risk. We aimed to analyze the risk of major hemorrhage in anticoagulated patients simultaneously exposed to potentially interacting drugs which may enhance the bleeding risk.

METHODS

Case-control study nested in a cohort of anticoagulated patients in 2011-2020. Cases were all people hospitalized for a major hemorrhage (cerebral or gastrointestinal), matched to individuals without bleeding. The index date was the day of hospital admission for cases and the same date for the matched controls.

DATA SOURCES: SIDIAP database, containing information from primary health care electronic records, and the database of diagnoses at hospital discharge in Catalonia, Spain. We analyzed exposure to interacting drugs during 3 months prior to the index date. The association between hemorrhage and exposure to interacting drugs was calculated through multivariate logistic regression models.

FINDINGS: We included 2,811 cases (77.9% cerebral and 22.1% gastrointestinal hemorrhages), matched to 28,054 controls. We found association between hemorrhage in patients receiving vitamin K antagonists (OR 1.30, 95% CI 1.16-1.47). All types of interactions resulted in higher bleeding risk for all anticoagulants. Proton pump inhibitors were found protective for gastrointestinal (OR 0.55, 95% CI 0.46-0.65) but not for cerebral bleeding (OR 1.18, 95% CI 1.08-1.30).

IMPLICATIONS: We estimated the risk of major hemorrhage in anticoagulated patients simultaneously exposed to potentially interacting drugs which may enhance bleeding risk. Our study underscores the potential impact of interactions on cerebral and gastrointestinal bleeding risk in anticoagulated patients.