Signals of bleeding among direct-acting oral anticoagulant users compared to those among warfarin users: analyses of the post-marketing FDA Adverse Event Reporting System (FAERS) database, 2010–2015

PURPOSE: To analyze and compare the signals of bleeding from the use of direct-acting oral anticoagulants (DOACs) in the US Food and Drug Administration Adverse Event Reporting System (FAERS) database over 5 years. METHODS: Reports of bleeding and of events with related terms submitted to the FAERS between October 2010 and September 2015 were retrieved and then analyzed using the reporting odds ratio (ROR). The signals of bleeding associated with DOAC use were compared with the signals of bleeding associated with warfarin use utilizing the FAERS databases. RESULTS: A total of 1,518 reports linked dabigatran to bleeding, accounting for 2.7% of all dabigatran-related reports, whereas 93 reports linked rivaroxaban to bleeding, which accounted for 4.4% of all rivaroxaban-related reports. The concurrent proportion of bleeding-related reports for warfarin was 3.6%, with a total of 654 reports. The association of bleeding and of related terms with the use of all three medications was significant, albeit with different degrees of association. The ROR was 12.30 (95% confidence interval [CI] 11.65–12.97) for dabigatran, 15.61 (95% CI 14.42–16.90) for warfarin, and 18.86 (95% CI 15.31–23.23) for rivaroxaban. CONCLUSIONS: The signals of bleeding varied among the DOACs, and the bleeding signal was higher for rivaroxaban and lower for dabigatran compared to that for warfarin.