Effectiveness and Safety of Non–Vitamin K Oral Anticoagulants in Comparison to Phenprocoumon: Data from 61,000 Patients with Atrial Fibrillation

Abstract All pivotal trials have evaluated non–vitamin K oral antagonists (NOACs) against warfarin. However, in some regions of the world, phenprocoumon is the most widely used vitamin K antagonist (VKA). There is little evidence documenting effectiveness and safety of NOACs compared with phenprocoumon in atrial fibrillation (AF). A retrospective cohort study using a German claims database was conducted to assess effectiveness (stroke, systemic embolism SE) and safety (bleeding leading to hospitalization) during therapy with NOACs and phenprocoumon in 61,205 AF patients. Hazard ratios (HRs) for effectiveness and safety outcomes were derived from Cox proportional hazard models, adjusting for baseline characteristics. Propensity score matching was performed as a sensitivity analysis. As a prespecified subgroup analysis, the effects of reduced NOAC dosing were compared with phenprocoumon. A total of 61,205 patients were identified in whom phenprocoumon (n = 23,823, 38.9%), apixaban (n = 10,117, 16.5%), dabigatran (n = 5,122, 8.4%), or rivaroxaban (n = 22,143, 36.2%) was initiated. After adjusting for baseline confounders, all three NOACs tested had significantly lower risks of stroke/SE compared with phenprocoumon (apixaban—HR: 0.77, 95% CI: 0.66–0.90; dabigatran—HR: 0.74, 95% CI: 0.60–0.91; rivaroxaban—HR: 0.86, 95% CI: 0.76–0.97). Apixaban (HR: 0.58, 95% CI: 0.49–0.69) and dabigatran (HR: 0.64, 95% CI: 0.50–0.80) were associated with lower bleeding risks than phenprocoumon, whereas the risk was similar for rivaroxaban and phenprocoumon. All three NOACs showed reduced risk of intracranial bleeding compared with phenprocoumon. Reduced doses of NOACs were predominantly used in patients with advanced age and comorbidities with generally similar effectiveness and safety benefits compared with phenprocumon as standard-dose NOACs.