Risk factors for stroke in different levels of cerebral arterial disease

Despite recent studies on risk factors of stroke, it is still unclear whether certain risk factors differently affect different levels of cerebral arteries in stroke patients. We attempted to study risk factors and lifestyle factors in patients with cerebrovascular diseases affecting different levels of the cerebral arteries.

We studied 343 Korean patients (222 men and 121 women) with acute stroke. Strokes were subdivided into large vessel infarction (LVI, n = 156), small vessel infarction (SVI, n = 105), intracerebral hemorrhage (ICH, n = 82) according to modified TOAST criteria. Patients with cardiogenic embolic infarction, undetermined causes, and miscellaneous causes were excluded. Among the patients with LVI and SVI, 104 and 49 underwent angiographic studies, respectively, who were designated as having definite LVI and definite SVI. Definite LVI was further subdivided into proximal LVI when the internal carotid artery (n = 25), vertebral artery (n = 13) or basilar artery (n = 14) were primarily involved, and distal LVI when the middle cerebral artery (n = 35), anterior cerebral artery (n = 7) or posterior cerebral artery (n = 10) were involved. Using a structured interview, we assessed the risk factors and lifestyles: hypertension, diabetes mellitus (DM), cigarette smoking (CS), alcohol consumption, physical activity and exercise, and the amount of consumption of salt, vegetables, fat, fish and fruits. Serum cholesterol, body mass index and total body fat were also measured.

Multivariate analysis showed that in men, DM was a discriminant factor favoring brain infarction (BI) over ICH, definite LVI or ICH and proximal LVI over ICH. Decreased fruit intake differentiated distal LVI from definite SVI, and distal LVI from ICH. In women, DM and abdominal obesity differentiated BI from ICH, while DM favored definite LVI over ICH, proximal LVI over ICH and definite SVI over ICH. Multivariate analysis including sex as a separate variate showed that DM was a factor favoring BI over ICH, definite LVI over ICH, proximal LVI over ICH, distal LVI over ICH and definite SVI over ICH. CS was a factor favoring BI over ICH, while heavy CS was a factor discriminating distal LVI from definite SVI. Male sex was a factor favoring definite LVI over ICH and proximal LVI over ICH. A lesser amount of fruit consumption favored definite SVI over distal LVI.

Our results demonstrate that DM in all sexes and central obesity in women are factors discriminating BI from ICH. However, these factors as well as others were not found to discriminate the sizes of the vessels involved in the patients with ischemic stroke, except that being male may be a factor favoring proximal large vessel disease over ICH. Genetic factors or other not yet identified risk factors may be responsible for the differences in the subtypes of ischemic stroke among different ethnic populations.