Pure motor stroke: a reappraisal.

We studied the correlations between the pattern of weakness, stroke type, topography, and etiology in 255 patients whose first stroke was manifested by isolated hemiparesis. They represented 14% of consecutively admitted stroke patients. The weakness distributions were as follows: face, upper limb, and lower limb (FUL) (50%); face and upper limb (FU) (29%); upper limb (U) (10%); and upper and lower limb (UL) (9%). Twenty-nine percent of the patients had dysarthria, which was of no localizing value. Less than one half of the patients had a deep infarct, and one third had a potential embolic source from the heart or large arteries. Logistic regression analysis showed that history of hypertension and type of weakness distribution were the main factors accounting for lesion localization: patients with FUL distribution and hypertension had a 90% probability of deep infarct; patients either with FUL distribution but no hypertension or with UL distribution and hypertension each had 70% probability of deep infarct. Pure motor monoparesis was almost never caused by a deep infarct. We suggest that the assumption of a lacunar etiology to a pure motor stroke should be applied only to patients with FUL involvement.