Stroke and bleeding with non-Vitamin K antagonist oral anticoagulant or warfarin treatment in patients with non-valvular atrial fibrillation: A population-based cohort study

Abstract

Aims Oral anticoagulants (OACs) effectively reduce the risk of stroke in atrial fibrillation (AF). Three non-vitamin K antagonist OACs (NOACs) are introduced in regular care based on promising results compared with warfarin in randomized trials. This study compares outcomes with NOAC vs. warfarin treatment in OAC naïve AF patients in routine care, including primary care, in a large region with decentralized anticoagulant treatment. Methods and results Population-based cohort study. Individuals with non-valvular AF who initiated treatment with NOAC (n=9279) or warfarin (n=12 919) from 2012 to 2015 were identified in the Stockholm administrative health data register (population 2.2 million). Adjusted Cox regression analyses were performed to evaluate TIA/ischaemic or unspecified stroke/death, and severe bleeds (co-primary endpoints); and secondarily for components of the composites. NOAC patients were younger (72.9 vs. 74.1 years) and had lower CHA2DS2VASc scores (3.42 vs. 3.68) than warfarin patients. NOAC vs. warfarin treatment was associated with similar risks for TIA/ischaemic or unspecified stroke/death [hazard ratio (HR) 0.94; 0.85-1.05] and severe bleeds (HR 1.02; 0.88-1.19); lower risks of intracranial bleeds (HR 0.72; 0.53-0.97) or haemorrhagic stroke (HR 0.56; 0.34- 0.93), but a higher risk for gastrointestinal bleeds (HR 1.28; 1.04-1.59). The advantages with NOAC treatment were most pronounced with standard dose in patients below 80 years, and with dose reduction in patients aged 80 and above. Conclusion This population-based cohort study of routine care indicates similar or better effectiveness and safety with NOAC compared with warfarin treatment. NOACs were associated with fewer intracranial bleeds, but more gastrointestinal bleeds.