Epidemiologic predictors of 30-day survival in cerebellar hemorrhage.

ABSTRACT

BACKGROUND: Outcome after cerebellar hemorrhage is relatively poorly understood. OBJECTIVES: To describe cerebellar hemorrhage and to establish if there were epidemiologic risk factors associated with early mortality. METHODS: Computerized records were searched to identify intracerebral hemorrhage (ICD-9-CM code 431) from 1986 to 1996 at 2 hospitals. Charts were abstracted using a standardized protocol. The provincial vital statistics registry was used to confirm mortality data. A multivariable logistic regression model was developed to identify predictors of 30-day mortality. RESULTS: Of 629 identified cases of intracerebral hemorrhage, 79 (12.5%) were cerebellar. The 30-day mortality was 41.7%. Approximately one third (31.7%) had an identifiable underlying cause. The odds of survival at 30 days was significantly reduced (odds ratio = 0.24; 95% confidence interval, 0.08 to 0.74) if the hemorrhage was caused by an identifiable underlying cause. CONCLUSIONS: Cerebellar hemorrhage with an underlying cause is more likely to be fatal than idiopathic or hypertensive cerebellar hemorrhage.