Incidence, Timing, and Predictors of Delayed Shunting for Hydrocephalus After Aneurysmal Subarachnoid Hemorrhage

Abstract

Although hydrocephalus is often treated with permanent cerebrospinal fluid (CSF) shunting during hospitalization for acute aneurysmal subarachnoid hemorrhage (SAH), little is known about the development of delayed hydrocephalus. Using administrative data on all visits to nonfederal emergency departments and acute care hospitals across California from 2005 to 2010, we identified patients with SAH and discharged without placement of a CSF shunt. Patients were followed for up to 7years to determine whether they subsequently developed delayed hydrocephalus, as indicated by hospitalization for a permanent CSF diversion procedure. In 8,889 patients discharged with SAH, 116 (1.3%) went on to develop delayed hydrocephalus. Most (>90%) diagnoses of delayed hydrocephalus occurred within the first year after discharge. Cox proportional hazards analysis identified microsurgical clipping (hazard ratio 2.0; 95% confidence interval 1.2-3.3), temporary ventriculostomy placement (2.5; 1.6-4.1), mechanical ventilation (1.7; 1.1-2.8), and discharge to a skilled nursing facility (2.9; 1.8-4.6) as being significantly associated with the development of delayed hydrocephalus. At 1year after discharge, the cumulative rate of delayed hydrocephalus was 0.9% (95% CI, 0.7-1.1%) for those without temporary ventriculostomy placement during the initial hospitalization, versus 5.7% (95% CI, 3.9-8.1%) in those who had received a temporary ventriculostomy. Delayed hydrocephalus after SAH occurs rarely overall, but in a substantial proportion of patients who required temporary ventriculostomy during the initial hospitalization. These results support vigilant surveillance of patients after removal of a temporary ventriculostomy, given the potential of delayed hydrocephalus to impair recovery or even result in clinical deterioration following SAH.