Authors Senoo and Lip investigated the relationship between bleeding risk and age in patients taking warfarin for nonvalvular atrial fibrillation. They used the warfarin arm of the AMADEUS trial to conduct a post-hoc analysis. The patient population consisted of patients with atrial fibrillation taking warfarin, and they were divided into tertiles based on age: age <67, age 67-74, and age ≥ 75. The principal safety outcome was any clinically relevant bleeding, and the primary clinical outcome was the composite of cardiovascular death and stroke/systemic embolism. 
Analysis was conducted on 722 patients in the youngest tertile, 747 patients in the middle tertile, and 824 patients in the oldest tertile. The rates of cardiovascular death and stroke/systemic embolism were highest in the oldest age category. The relative risk of any clinically relevant bleeding did not differ significantly across age categories, although the oldest strata did have a higher risk of major bleeding. Additionally, the average amount of time spent in the therapeutic range of warfarin was inversely related to bleeding risk, cardiovascular death and stroke risk in this population.
The authors concluded that elderly patients with atrial fibrillation have a high risk of cardiovascular death and stroke/systemic embolism. These highest risk patients may derive the greatest net clinical benefit from being on warfarin, while their bleeding risk remains similar to other age categories. However, the authors also pointed out that this was a post-hoc analysis, and the results should be interpreted as hypothesis-generating. Nevertheless, this analysis confirms prior studies showing that the net benefit of oral anticoagulation increases as patients get older.
For future research, it will be interesting to see similar analyses using the new oral anticoagulants.
