The introduction of non-vitamin K antagonist oral anticoagulants (NOACs) has been a major advance for stroke prevention in atrial fibrillation (AF). Patients and clinicians now have a choice between different NOACs, but have no direct comparative effectiveness evidence to guide decision-making. We aimed to compare the effectiveness and safety of dabigatran, rivaroxaban, and apixaban in clinical practice.
Using a large U.S. administrative claims database, we created three one-to-one propensity-score matched cohorts of patients with non-valvular AF who were users of dabigatran, rivaroxaban, or apixaban between 10/1/2010-2/28/2015 (Rivaroxaban versus dabigatran [N=31,574]; apixaban versus dabigatran [N=13,084] and apixaban versus rivaroxaban [N=13,130]). The primary outcomes were stroke or systemic embolism (effectiveness) and major bleeding (safety) that occurred on treatment. Cox proportional hazards models were used to compare outcomes in propensity-score matched cohorts.
We found no differences between the three NOACs in the risk of stroke or systemic embolism (hazard ratio [HR]: 1.00 [0.75, 1.32] for rivaroxaban versus dabigatran; 0.82 [0.51, 1.31] for apixaban versus dabigatran; and 1.05 [0.64, 1.72] for apixaban versus rivaroxaban). Apixaban was associated with lower major bleeding risk (HR 0.50 [0.36, 0.70], p<0.001 versus dabigatran; and 0.39 [0.28, 0.54], p<0.001 versus rivaroxaban). Rivaroxaban was associated with increased risk of major bleeding (HR 1.30 [1.10, 1.53], p<0.01) and intracranial bleeding (HR 1.79 [1.12, 2.86], p<0.05) compared to dabigatran.
Dabigatran, rivaroxaban, and apixaban appear to have similar effectiveness whereas apixaban may be associated with lower bleeding risk and rivaroxaban with elevated bleeding risk.