Background: Little is known about the net clinical benefit of warfarin therapy in elderly patients with atrial fibrillation.
Methods and results: We studied 2339 Chinese patients with nonvalvular atrial fibrillation aged≥80 years: 1805 with no antithrombotic therapy and 534 on warfarin therapy. Patients were stratified according to their CHA2DS2-VASc and HAS-BLED score. The primary end point was a composite of hospital admission with ischemic stroke or death. After 2.2-year follow-up (5199 patient-years), a primary end point had occurred for 1861 patients (79.6%): 66.9% in patients on warfarin (66.9%) compared with 80.8% in patients with no antithrombotic therapy (80.8%; hazard ratio, 0.53; 95% confidence interval, 0.48-0.58; P<0.001). This was related to substantially better mortality rate (hazard ratio, 0.40; 95% confidence interval, 0.37-0.45; P<0.0001) and ischemic stroke rate (hazard ratio, 0.64; 95% confidence interval, 0.54-0.77; P<0.0001) among patients on warfarin. For the net clinical benefit, 510 ischemic strokes and 42 intracranial hemorrhage were recorded. The annual incidence of ischemic stroke and intracranial hemorrhage was 11.3%/y and 0.6%/y, respectively, in patients prescribed no antithrombotic therapy and 7.1%/y and 1.1%/y, respectively, in those prescribed warfarin. The adjusted net clinical benefit favored warfarin for all elderly patients, and the best net benefit from warfarin was in those with high stroke and intracranial hemorrhage risk. In these high-risk patients, warfarin therapy was associated with 7.2 to 8.0 fewer events per 100 patient-years compared with no antithrombotic therapy.
Conclusions: In elderly patients with atrial fibrillation, warfarin therapy is associated with lower death and ischemic stroke and an overall net clinical benefit.
Keywords: aged; atrial fibrillation; warfarin.