[Epidemiology of cerebrovascular accidents]

Recent studies concerning secular trends in stroke incidence and mortality and identification of independent risk factors for stroke are reviewed. Stroke mortality has declined in many industrialized countries in recent decades. In France, it has been declining by more than 30% between 1968 and 1982 in all age groups and in both sexes except for women under 40 years. The decline in stroke mortality seems to be partly real and partly apparent. In the community-based study of Rochester, Minnesota, stroke incidence decreased by 54% between 1945-49 and 1975-79. Recent data from Rochester, however, suggest that the incidence of stroke may no longer be declining. Survival after stroke has also apparently been improving but several sources of potential bias may also have influenced the decrease in reported survival rates. Hypertension is a major risk factor for stroke. Prolonged differences in "usual" diastolic blood pressure of 5 to 10 mmHg are associated with about 40% difference in stroke incidence. Recent analysis suggests that stroke incidence reduction could arise rapidly after hypertension control and that a lower blood pressure should confer a lower risk of vascular disease, even in individuals conventionally considered as "normotensive". There is evidence that cigarette smoking is an important risk factor for stroke with an overall relative risk of 1.5 and that the risk of stroke declines rapidly after the cessation of smoking. A cardiac condition may be a marker for another risk factor or the primary substrate for cerebral embolism. In patients with atrial fibrillation, the risk of stroke is increased through both of these mechanisms. Diabetes mellitus, chronic alcohol consumption (> 3 drinks/day), and high fibrinogen levels are other independent risk factors for stroke. While high levels of cholesterol may be associated with ischemic stroke, an inverse association of the serum cholesterol with the occurrence of intracerebral hemorrhage in men has been reported. In patients with asymptomatic internal carotid stenosis, higher degrees of stenosis convey a higher risk of stroke. However, far from all these strokes are due to thromboembolism from an atheromatous plaque in the ipsilateral internal carotid artery. The relative risk of stroke during the first 5 years following a transient ischemic attack is 7 times that in persons without transient ischemic attack. More than a third of the subsequent strokes occur in a vascular territory different from that of the incident TIA. While the use of oral contraceptives may increase the relative risk of stroke, postmenopausal estrogen treatment may have a protective effect on the risk of vascular diseases.