Direct oral anticoagulant- or warfarin-related major bleeding: characteristics, reversal strategies and outcomes from a multi-center observational study

Background — Direct oral anticoagulants (DOACs) have expanded the armamentarium for antithrombotic therapy. While DOAC-related major bleeds were associated with favourable outcomes compared to warfarin in clinical trials, warfarin was reversed in <40% of cases, raising concerns about the generalizability of this finding.

Methods — Consecutive patients ≥66 years presenting to five tertiary care hospitals across three cities in Ontario, Canada with diagnoses that included hemorrhage from October 2010 to March 2015. Charts were screened for association with DOAC or warfarin use; eligible cases were abstracted and linked to administrative databases.

Results — Among 19,061 records screened, 2,002 (460 DOAC, 1542 warfarin) were eligible. Reversal agents were frequently used among warfarin bleeds (72.9% vitamin K, 40.7% prothrombin complex concentrates). Red blood cell transfusions occurred more often among DOAC bleeds than warfarin (52.0% vs. 39.5%, adjusted relative risk [aRR] 1.32 [95% CI 1.19 – 2.47]). However, units of blood products transfused were not different between the two groups. Thirty-four DOAC cases (7.4%) received activated prothrombin complex concentrates or recombinant factor VIIa. In-hospital mortality was lower following DOAC bleeding (9.8% vs. 15.2%, aRR 0.66 [95% CI 0.49 – 0.89], although differences in 30-day mortality did not reach statistical significance (12.6% vs. 16.3%, aRR 0.79 [95% CI 0.61 – 1.03]).

Conclusions — In this unselected cohort of patients with oral anticoagulant-related hemorrhage with high rates of warfarin reversal, in-hospital mortality was lower among DOAC-associated bleeds. These findings support the safety of DOACs in routine care and present useful baseline measures for evaluations of DOAC-specific reversal agents.