Direct oral anticoagulation versus warfarin in atrial fibrillation: a nationwide cohort study of effectiveness and safety among frail patients

Abstract Background Frail patients with atrial fibrillation carry a high risk of both stroke and treatment-related bleeding complications, but evidence for the safety and effectiveness of anticoagulation regimens in frail patients remain sparse. Objectives To determine the comparative effectiveness and safety of DOAC versus warfarin in frail patients with atrial fibrillation. Methods A nationwide registry-based cohort study applying a new-user design including all frail patients with atrial fibrillation who redeemed a prescription for DOAC or warfarin between January 2012 and December 2020. We applied inverse probability of treatment weighting (IPTW) to account for baseline confounding and applied a marginal structural modelling approach with weighted pooled regression to compute hazard ratios (HR) and risk differences for thromboembolic events and major bleeding comparing specific DOAC doses with warfarin. Results We identified 32048 anticoagulation naïve frail patients (median age 80 years, 53% female) with atrial fibrillation who started anticoagulation treatment: 6747 (21.1%) patients initiated warfarin therapy, 17076 (50.3%) initiated standard dose DOAC, and 9179 (28.6%) patients with indication for dose reduction initiated reduced dose DOAC. The median CHA2DS2-VASc score was 5. The comparative effectiveness analyses showed a similar risk for thromboembolism with weighted HRs of 0.95 (95% CI 0.80-1.13) for standard dose DOAC versus warfarin, and 0.97 (95% CI 0.77-1.23) for reduced dose DOAC versus warfarin (see Table). The thromboembolic event free survival difference at one year was -0.2% for DOAC regardless of dosing when compared with warfarin. The comparative safety analyses revealed significantly lower risk of major bleeding among patients initiating standard dose (weighted HR 0.69, 95% CI 0.59-0.87) or reduced dose DOAC (weighted HR 0.67, 95% 0.55-0.81) when compared with warfarin. The risk difference of bleeding at one year for DOAC users varied between -1.3% and -3.0%, with the greatest difference for patients with high frailty level initiating reduced dose DOAC. Conclusion We observed similar risk of thromboembolism and significantly lower risk of major bleeding associated with both standard and reduced DOAC treatment regimens compared with warfarin in frail patients with atrial fibrillation in routine clinical practice.