PREDICTING SURVIVAL FOR 1 YEAR AMONG DIFFERENT SUBTYPES OF STROKE - RESULTS FROM THE PERTH-COMMUNITY-STROKE STUDY

Background oud Purpose Few studies have evaluated the factors influenc ing or predicting long-term survival after stroke in an unselected ser ies of patients in whom the underlying cerebrovascular pathology is cl early defined. Moreover, the relative importance of risk factors for s troke, including sociodemographic and premorbid variables, has not bee n described in detail. Methods The study cohort consisted of 492 patie nts with stroke who were registered with a population-based study of a cute cerebrovascular disease undertaken in Perth, Western Australia, d uring an 18-month period in 1989 and 1990. Objective evidence of the p athological basis of the stroke was obtained in 86% of cases, and all deaths among patients during a follow-up of 1 year were reviewed. Resu lts One hundred twenty patients (24%) died within 28 days of the onset of stroke. Among the different subtypes of stroke, the 1-year case fa tality (mean, 38%) varied from 6% and 16% for boundary zone infarction and lacunar infarction, respectively, to 42% and 46% for subarachnoid hemorrhage and primary intracerebral hemorrhage, respectively. Using Cox proportional-hazards analysis, a predictive model was developed on 321 patients with acute stroke (test sample). The best model containe d five baseline variables that were independent predictors of death wi thin 1 year: coma (relative risk [RR], 3.0; 95% confidence interval [C I], 1.1 to 8.4), urinary incontinence (RR, 3.9; 95% CI, 1.4 to 10.6), cardiac failure (RR, 6.5; 95% CI, 2.8 to 15.1), severe paresis (RR, 4. 9; 95% CI, 1.6 to 15.5), and atrial fibrillation (RR, 2.0; 95% CI, 1.1 to 3.5). The sensitivity, specificity, and negative predictive value of this model for predicting death were 90%, 83%, and 95%, respectivel y. When applied to a second randomly selected validation sample of 171 events, sensitivity was 94%, specificity 62%, and negative predictive value 92%, indicating stability of the model.Conclusions Although the case fatality, timing, and cause of death vary considerably among the different pathological subtypes of stroke, simple clinical measures t hat reflect the severity of the neurological deficit and associated ca rdiac disease at onset independently predict death by 1 year and may h elp to direct management.