Prevention of stroke is a crucial health care issue, as stroke is the third cause of death and the first cause of major disability in developed countries. The established role of platelet aggregation in TIA or minor and major ischaemic stroke has provided the rationale for many randomized trials of antiplatelet agents (aspirin, sulfinpyrazone, dipyridamole alone or in combination with aspirin, suloctidil, ticlopidine). The recent Antiplatelet Trialists' Collaboration (APT) meta-analysis (1994) based on 142 trials involving 100,000 vascular patients confirmed the data of the previous overview (1988). Aspirin, the only drug evaluated for primary prevention of ischaemic events, is not indicated for safety reasons in subjects at low risk of occlusive disease. Compared to control, antiplatelet therapy, notably aspirin which is by far the most widely used agent in trials, provides a 27% risk reduction of stroke, myocardial infarction or vascular death in patients suffering from ischaemic vascular events and a 22% risk reduction of these outcomes in patients having experienced a prior TIA/stroke. Aspirin (around 325 mg/day) and ticlopidine (500 mg/day) are currently the reference drugs for secondary prevention in cerebrovascular patients. The long term efficacy of ticlopidine, a specific antiaggregating agent, has been evaluated in two North American trials involving more than 4,000 patients. TASS showed ticlopidine to be significantly more effective in reducing the incidence of fatal or nonfatal stroke and death than aspirin in patients with TIA or minor stroke. The relative risk reductions over aspirin, the first year of greatest risk, were 41% for stroke and death and 46% for fatal or nonfatal stroke. CATS showed that ticlopidine compared with placebo induces a significant 30% relative risk reduction of stroke, myocardial infarction and vascular death over three years in patients who had suffered a recent thromboembolic stroke. The above results elicit two important issues: the optimal dose of aspirin and its tolerability compared to ticlopidine. The three controlled trials (UK-TIA, SALT, Dutch TIA) which have compared high (> or = 1 g/day) and low dose aspirin (< or = 300 mg/day) or various low doses of aspirin did not give a definite answer on the efficacy of low or very low (30 or 75 mg/day) doses of aspirin for reducing the risk of vascular outcomes in patients with stroke precursors. Even with low doses of aspirin there was still a risk of severe gastrointestinal bleeding, although minor side effects were less frequent.(ABSTRACT TRUNCATED AT 400 WORDS)