Outcomes following cerebrospinal fluid shunting in high-grade glioma patients.

OBJECTIVE The clinical course of high-grade central nervous system gliomas is occasionally complicated by hydrocephalus. The risks of shunt placement and clinical outcome following CSF diversion in this population are not well defined. METHODS The authors retrospectively reviewed the outcomes of patients with pathologically confirmed WHO grade III or IV gliomas with shunt-treated hydrocephalus at their institution. Outcomes of patients in this cohort were compared with those of patients who underwent shunt treatment for normal pressure hydrocephalus (NPH). Hospital-reported outcomes in a national database for malignant primary brain tumor patients undergoing a ventricular shunt procedure were also reviewed. RESULTS Forty-one patients undergoing CSF shunting between 2001 and 2016 at the authors' institution were identified. Noncommunicating and communicating hydrocephalus occurred at similar rates (51.2% vs 48.8%). Symptomatic improvement after shunting was observed in 75.0% of patients. A major complication occurred in 17.1% of cases, with 2 patients suffering an intracranial hemorrhage. Prior administration of bevacizumab was significantly associated with the incidence of hemorrhage (p = 0.026). Three patients (7.3%) died during admission, and 8 (19.5%) died within 30 days of shunt placement. The presence of ependymal or leptomeningeal enhancement was more common in patients who died within 30 days (75.0% vs 11.1%, p = 0.001). Six patients (18.1%) required readmission to the hospital within 30 days of discharge. Revision surgery was necessary in 7 patients (17.1%). The median time from shunt placement to death was 150.5 days. In comparison with patients with NPH, shunt-treated high-grade glioma patients had increased in-hospital (7.3% vs 0.5%, p = 0.008) and 30-day (19.5% vs 0.8%, p < 0.001) mortality but no difference in the incidence of revision surgery (17.1% vs 17.5%, p = 0.947). Similarly, in the national Vizient Clinical Database Resource Manager, shunt-treated patients with malignant primary brain tumors had an increased length of stay (6.9 vs 3.5 days, p < 0.001), direct cost of admission ($15,755.80 vs $9871.50, p < 0.001), and 30-day readmission rates (20.6% vs 2.4%, p < 0.001) compared with patients without brain tumors who received a shunt for NPH. CONCLUSIONS Shunting can be an effective treatment for the symptoms of hydrocephalus in patients with high-grade gliomas. However, the authors' results suggest that this procedure carries a significant risk of complications in this patient population.