Detection and modification of the risk factors of stroke may be the most effective strategy for preventing its often irreversible consequences. A longitudinal prospective study was implemented to evaluate the effect of several risk factors on the course of cerebrovascular disease. The study groups were composed of 3S8 normal volunteers, and 308 patients with transient ischemic attacks. The two groups were followed for 4.2 (. 3.3) and 2.S years, respectively. Endpoints were defined as the occurrence of the following: transient ischemic attacks, stroke, multii-infarct dementia, dementia of the Alzheimer's type, or death. .Stroke and death were 2.5 times more frequent in the second group. Hypertension was the single predictor of reaching the endpoints (P<0.014). By excluding the cases with dementia of Ahheimers type, no single predictor could be identified. This study suggests that ischemic events arc significantly more frequent inpatients with transient ischemic attacks. A dichowmous split was observed between neurogenic and cardiogenic endpoints.