Children with posterior fossa tumors (PFTs) may present with hydrocephalus. Persistent (or new) hydrocephalus is common after PFT resection. Endoscopic third ventriculostomy (ETV) is sometimes performed prior to resection to 1) temporize hydrocephalus prior to resection and 2) prophylactically treat post-resection hydrocephalus. The objective of this study was to establish, in a historical cohort study of pediatric patients who underwent primary craniotomy for PFT resection, whether or not pre-resection ETV prevents the need for post-resection CSF diversion to manage hydrocephalus.
The authors interrogated their prospectively maintained surgical neuro-oncology database to find all primary PFT resections from a single tertiary pediatric neurosurgery unit. These data were reviewed and supplemented with data from case notes and radiological review. The modified Canadian Preoperative Prediction Rule for Hydrocephalus (mCPPRH) score was retrospectively calculated for all patients. The primary outcome was the need for any form of postoperative CSF diversion within 6 months of PFT resection (including ventriculoperitoneal shunting, ETV, external ventricular drainage [EVD], and lumbar drainage [LD]). This was considered an ETV failure in the ETV group. The secondary outcomes were time to CSF diversion, shunt dependence at 6 months, and complications of ETV. Statistical analysis was done in RStudio, with significance defined as p < 0.05.
A total of 95 patients were included in the study. There were 28 patients in the ETV group and 67 in the non-ETV group. Patients in the ETV group were younger (median age 5 vs 7 years, p = 0.04) and had more severe preoperative hydrocephalus (mean frontal-occipital horn ratio 0.45 vs 0.41 in the non-ETV group, p = 0.003) and higher mCPPRH scores (mean 4.42 vs 2.66, p < 0.001). The groups were similar in terms of sex and tumor histology. The overall rate of post-resection CSF diversion of any kind (shunt, repeat ETV, LD, or EVD) in the entire cohort was 25.26%. Post-resection CSF diversion was needed in 32% of patients in the ETV group and in 22% of the patients in the non-ETV group (p > 0.05). Shunt dependence at 6 months was seen in 21% of the ETV group and 16% of the non-ETV group (p > 0.05). The median time to ETV failure was 9 days. ETV failure correlated with patients with ependymoma (p = 0.02). Children who had ETV failure had higher mCPPRH scores than the ETV success group (5.67 vs 3.84, p = 0.04).
Pre-resection ETV did not reliably prevent the need for post-resection CSF diversion. ETV was more likely to fail in children with ependymoma and those with higher mCPPRH scores. Based on the findings of this study, the authors will change the practice at their institution; pre-resection ETV will now be performed based on a newly defined protocol.