Aspirin for the primary prevention of stroke and other major vascular events - Meta-analysis and hypotheses

Background: Aspirin therapy reduces stroke by about 25% fbr persons with at herosclerotic vascular disease, but the effect in those without clinically apparent vascular disease is distinctly different. Objective: To define the effect of aspirin use on stroke and other major va scular events when given for primary prevention to persons without clinical ly recognized vascular disease. Data Sources and Extraction: Systematic review of randomized clinical trial s and large prospective observational cohort studies examining the relation between aspirin use and stroke in persons at low intrinsic risk. Studies w ere identified by a computerized search of the English-language literature. Data Synthesis: Five randomized trials of primary prevention included 52 25 1 participants randomized to aspirin doses ranging from 75 to 650 mg/d; the mean overall stroke rate was 0.3% per year during an average follow-up of 4.6 years. Meta-analysis revealed no significant effect on stroke (relative risk = 1.08, 95% confidence interval, 0.95-1.24) contrasting with a decrea se in myocardial infarction (relative risk = 0.74, 95% confidence interval, 0.68-0.82). The lack of reduction of stroke by aspirin for primary prevent ion was incompatible with its protective effect against stroke in patients with manifest vascular disease (P = .001). Intracranial hemorrhage was incr eased by the regular use of aspirin (relative risk = 1.35: P = .03), simila rly for both primary and secondary prevention. III 4 large observational st udies, self-selected use of aspirin was consistently associated with higher rates of stroke. Conclusions: The effect of aspirin therapy on stroke differs between indivi duals based on the presence or absence of overt vascular disease, in contra st with the consistent reduction in myocardial infarction by aspirin therap y observed in all populations. We hypothesize that the effect of aspirin th erapy on stroke for persons with major risk factors for vascular disease ma y be intermediate between a substantial decrease for those with manifest va scular disease and a possible small increase fur healthy persons due to acc entuated intracranial hemorrhage. When aspirin is given for primary prevent ion of vascular events, available data support using 75 to 81 mg/d.