A differential effect of aspirin on prevention of stroke in atrial fibrillation.

Aspirin reduces the risk of stroke in patients with nonrheumatic atrial fibrillation, but its efficacy may not be uniform in patient subgroups. We sought patient characteristics predictive of aspirin efficacy. Exploratory multivariate analyses of a multicenter, double-blind, randomized clinical trial comparing aspirin (325 mg/ day) to placebo in 1,120 patients for prevention of thromboembolism were performed. Patients ≤75 years of age assigned to aspirin had a 62% (95% confidence interval, 27-81%; p = 0.002) reduction in thromboembolism, significantly different from older patients given aspirin (p = 0.03). The rate of thromboembolism in patients ≤75 years old treated with aspirin was 2.1% per year (95% confidence interval, 1.2-3.8). Patients with a history of hypertension were at high risk (9.0% per year) of thromboembolism during placebo treatment and had a 57% reduction (p = 0.005) when given aspirin; even so, the remaining absolute risk during aspirin therapy was 3.8% per year. These exploratory analyses suggest that aspirin is more effective in patients ≤75 years old than in those >75 years old. Although aspirin is effective in patients with a history of hypertension, aspirin may not lower the absolute risk adequately, particularly in patients >75 years old. These secondary results require confirmation before influencing the antithrombotic prophylaxis of patients with atrial fibrillation in clinical practice. The adequacy of aspirin prophylaxis according to patient age is being assessed in the second phase of the study.