Low incidence of thromboembolism with Fiix-monitored warfarin compared to conventional warfarin and DOACs in patients with AF

Mixed population studies suggest that monitoring only coagulation factors II and X (Fiix) instead of conventional prothrombin time improves clinical outcomes in patients on warfarin. We hypothesized that Fiix-monitored warfarin (Fiix-warfarin) provides better real-world clinical outcomes than PT based international normalized ratio (PT-INR) monitored warfarin (PT-warfarin), apixaban, dabigatran, and rivaroxaban in non-valvular atrial fibrillation (AF) patients. We performed a retrospective population cohort study over a 5-year period including all long-term orally anticoagulated adult AF patients in the Greater Reykjavik area. Baseline characteristics differences were adjusted using inverse probability of treatment weighting. Principal outcomes were rates of total thromboembolism (TE), all-cause death, and major bleeding. Outcomes with Fiix-warfarin were used as reference. The study population consisted of 6417 patients anticoagulated long-term for 12 914 person-years (py), ie, Fiix-warfarin (n = 1257/py = 2514), PT-warfarin (n = 1904/py = 3998), apixaban (n = 1171/py = 1639), rivaroxaban (n = 1536/py = 3226) or dabigatran (n = 549/py = 1537). PT-warfarin (1.9% per py; hazard ratio (HR) 1.86 [P =.007]), apixaban (1.9% ppy; HR 1.94 [P = .02]), and dabigatran (2.2% ppy; HR 2.19 [P = .01]) had higher TE rates of than Fiix-warfarin (1.1% ppy). Similarly, rivaroxaban trended towards higher TE rates (1.6% ppy; HR 1.58; [P = .07]). Rivaroxaban had significantly higher all-cause mortality rate than Fiix-warfarin (3.0% vs 2.0% ppy; HR 1.48; [P =.04]). Major bleeding rates were similar. Warfarin anticoagulation variability was lower with Fiix-monitoring than with PT-monitoring. We conclude that Fiix-monitored warfarin could be the most effective long-term oral anticoagulant for patients with AF.