Predictors of in-hospital shunt-dependent hydrocephalus following rupture of cerebral aneurysms

Abstract

The development of shunt-dependent hydrocephalus is a well-recognised complication after aneurysmal subarachnoid haemorrhage, and negatively impacts on outcomes among survivors. This study aimed to identify early predictors of shunt dependency in a large administrative dataset of aneurysmal subarachnoid haemorrhage patients. We reviewed the National Hospital Morbidity Database in Australia for the years 1998 to 2008 and investigated the incidence of ventricular shunt placement following aneurysmal subarachnoid haemorrhage admissions. Putative risk factors were evaluated with univariate and multivariate logistic regression analysis to identify independent predictors of outcome. The following variables were considered: poor admission neurological grade; aneurysm location; intracerebral haemorrhage; intraventricular haemorrhage; acute hydrocephalus requiring the insertion of an external ventricular drain; surgical clipping; endovascular coiling; meningitis; and prolonged period of external ventricular drainage. A total of 10807 patients hospitalised for aneurysmal subarachnoid haemorrhage were identified. Among them, 701 (6.5%) required a permanent cerebrospinal fluid diversion procedure during the same admission as the aneurysmal subarachnoid haemorrhage. On multivariate analysis, poor admission neurological grade, acute hydrocephalus, the presence of intraventricular haemorrhage, ruptured vertebral artery aneurysm, surgical clipping, endovascular coiling, meningitis, and a prolonged period of external ventricular drainage were significant predictors of shunt dependency. A patient with a ruptured middle cerebral artery aneurysm was unlikely to develop shunt dependency (odds ratio 0.58; 95% confidence interval 0.46–0.73; p<0.001).