Prognosis of ischaemic stroke — a multivariate analysis

Publisher Website
Google Scholar
Abstract
Paucity of studies on the prognosis of ischaemic stroke prompted this study. Sixty-seven patients with CT-proven infarction were studied after a mean duration of 9 days to evaluate the role of a number of clinical, CT scan and evoked potential parameters in predicting the prognosis. Their mean age was 53 (range 22-82) years and 16 were females. Four of these patients died and 20 had poor recovery. Out of 24 clinical, CT scan and evoked potential parameters analysed by univariate analysis, 11 variables were found to be significant These were Canadian Neurological Scale, Glasgow Coma Scale (GCS), Motoricity Index, muscle tone, tendon reflex, incontinence, sensations, motor area involvement on CT scan, size of infarction, motor evoked potential and somatosensory evoked potential. Multivariate logistic regression analysis was employed to study the combined effect of different variables on the outcome. The best model included past history of stroke, GCS, pulse pressure and involvement of motor area on CT scan. The model proposed by us is simple and useful for predicting 3 month outcome of ischaemic stroke. A larger study is recommended to confirm these findings.-