Effectiveness and safety of non-vitamin K antagonist oral anticoagulants versus warfarin among nonvalvular atrial fibrillation patients with prior bleeding events

Abstract Background Patients with non-valvular atrial fibrillation (NVAF) use oral anticoagulants such as warfarin or non-vitamin K antagonist oral anticoagulants (NOACs) for the prevention of stroke. However, the effectiveness and safety of warfarin and NOACs can be influenced by pre-existing patient comorbidities, such as a history of bleeding, and limited evidence are available to inform the choice of the most appropriate anticoagulant treatment for NVAF patients with bleeding history. Purpose This study used five United States insurance claims databases to evaluate the risk of stroke/systemic embolism (S/SE) and major bleeding (MB) among NVAF patients with prior bleeding events who were prescribed NOACs versus warfarin. Methods This retrospective observational study used data from 5 databases (CMS Medicare and four commercial databases, covering >180 million beneficiaries) to select adult NVAF patients who were treated with apixaban, dabigatran, rivaroxaban, or warfarin (01JAN2013–30JUN2019). Patients were required to have a prior bleeding event, defined as a hospitalization with a diagnosis for intracranial hemorrhage (ICH), gastrointestinal (GI) bleeding or bleeding at other key sites prior to or during the index treatment episode. In each database, three 1:1 NOAC-warfarin propensity-score-matched (PSM) cohorts were created before pooling the results. Outcome measures were time to first stroke/SE, (ischemic stroke, hemorrhagic stroke, and SE), and time to first MB (gastrointestinal bleeding, intracranial hemorrhage, and MB at other key sites), and were measured from the index treatment episode to treatment discontinuation or switch, death, health plan disenrollment, or end of study period. Hazard ratios of S/SE and MB were calculated using Cox proportional hazards models. Results Among the eligible NVAF population, 8.2% of patients had a prior bleeding event (ICH: 12.3%; GI: 60.7%; Other: 27.0%). After PSM, a total of 43,092 apixaban-warfarin, 11,295 dabigatran-warfarin, and 32,723 rivaroxaban-warfarin patient pairs with prior bleeding were selected with a mean follow-up of 8–9 months. Apixaban and rivaroxaban were associated with a lower risk of S/SE, and dabigatran was associated with a similar risk of S/SE when compared to warfarin. Apixaban and dabigatran were associated with a lower risk of MB, and rivaroxaban was associated with a similar risk of MB, compared to warfarin (Figure). Conclusion Among NVAF patients with prior bleeding events, NOACs