Background Patients with atrial fibrillation who are vitamin K antagonist (VKA)-naive may have a higher risk of thrombosis and/or bleeding than VKA-experienced patients. Methods and results Using data from ARISTOTLE, we assessed baseline characteristics and the treatment effect of apixaban versus warfarin in the VKA-naive and VKA-experienced cohorts. We compared rates of study drug discontinuation and time-in-therapeutic range. Overall, 7,800 (43%) were VKA naive, and 10,401 were VKA experienced. At baseline, both groups were similar with respect to age and congestive heart failure, hypertension, age, diabetes, stroke score (CHADS(2)). Fewer VKA-naive patients had a history of prior stroke (18% vs 21%) or prior bleeding (10% vs 22%) and were more often female (39% vs 33%). The effect of apixaban on the primary efficacy and safety outcomes was similar in VKA-naive (stroke/systemic embolism: hazard ratio [ HR] 0.86, 95% CI 0.67-1.11 and major bleeding: HR 0.73, 95% CI 0.59-0.91) and VKA-experienced populations (stroke/systemic embolism: HR 0.73, 95% CI 0.57-0.95, P value for interaction = 0.39 and major bleeding: HR 0.66, 95% CI 0.55-0.80, P value for interaction = 0.50). Permanent study drug discontinuation was numerically less likely in patients receiving apixaban whether they were VKA naive (HR for discontinuation: 0.87, 95% CI 0.79-0.95) or VKA experienced (HR for discontinuation: 0.93, 95% CI 0.85-1.02). Among patients receiving warfarin, the mean/median times in therapeutic range were lower in the VKA-naive group (VKA-naive: 57.5/61.4, VKA-experienced: 66.0/69.1, P < .001). Conclusion The treatment effects of apixaban (vs warfarin) were not modified by VKA naivety. The rates of stroke/systemic embolism and major bleeding were numerically lower among the patients assigned to apixaban, irrespective of prior VKA use.
2013. Vol. 166, no 3, 549-558 p.