(CCSP049) MAJOR BLEEDING WITH OACS IN PATIENTS WITH ATRIAL FIBRILLATION: EFFECTS OF CO-PRESCRIBED BLEEDING RISK MODIFYING DRUGS

Background: Patients with atrial fibrillation (AF) prescribed oral anticoagulants (OACs) are at elevated risk of major bleeding. Concurrent non-steroidal anti-inflammatory (NSAID) and antiplatelet drugs are known to increase this risk, while proton pump inhibitors (PPIs) are often prescribed to reduce major bleeding risk in OAC recipients. Because few real-world assessments exist, this study aimed to evaluate serious bleeding outcomes associated with bleeding risk modifying drugs in Canadians with AF taking OACs.

METHODS AND RESULTS: We used linked population-based administrative data containing physician billing, hospitalization, and prescription records of 5 million British Columbians during 1996-2020 in a retrospective observational design. We developed an incident cohort of patients with AF who were new users of OACs since AF diagnosis. Exposures of interest were prescriptions for NSAIDs or antiplatelets and PPIs within 180 days of hospitalization for major or gastrointestinal (GI) bleeding. Multivariable Cox proportional hazard models were used to evaluate time to first hospitalization for major bleeding and GI bleeding.

The study cohort included 33,765 patients (mean age 70.1, 45% female, mean HAS-BLED score 2.1). During median follow-up of 6.2 years, 3839 major bleeding events occurred (51.4% GI, 7.8% intracerebral hemorrhage). Median time to first event was 3.8 years, and bleeding was fatal in 7.9% of cases. NSAIDs or antiplatelets were prescribed to 13% of patients within 180 days of major bleeding, and ≥1 prescription (vs. none) was associated with 24.0% (95%CI 13.4-35.4) and 48.9% (95%CI 32.3-67.6) increased hazard of major bleeding and GI bleeding, respectively. This effect was significantly greater in warfarin recipients than DOAC recipients: Major bleeding: HR 3.276 vs. 1.289; GI bleeding: HR 3.319 vs. 1.310; test for contrast p< 0.001 for both.

PPI was prescribed in 27.0% of patients. Three or more PPI prescriptions (vs. none) within 180 days of major bleeding was associated with a 28% (95%CI 20.3-34.9) and 16.9% (95%CI 5.1-27.2) hazard reduction for major bleeding and GI bleeding, respectively. Fewer prescriptions did not reduce the hazard. Number of PPI prescriptions per patient was weakly but positively correlated with HAS-BLED score [r=0.135 (95%CI 0.125, 0.145)].

Conclusion: In AF patients receiving OACs, prescription of NSAID or antiplatelet drugs was common and sharply increased the hazard of major bleeding. PPIs modestly reduced major and GI bleeding hazard, but three or more prescriptions (i.e., continuous therapy) were needed to achieve this. Prescribing appeared rational in that PPI prescribing correlated weakly but positively with HAS-BLED score.