Cerebellar mutism after posterior fossa tumor resection in children: a multicenter international retrospective study to determine possible modifiable factors

Purpose: A preliminary survey of pediatric neurosurgeons working at different centers around the world suggested differences in clinical practice resulting in variation in the risk of pediatric cerebellar mutism (CM) and cerebellar mutism syndrome (CMS) after posterior fossa (PF) tumor resection. The purposes of this study were (1) to determine the incidence and severity of CM and CMS after midline PF tumor resection in children treated at these centers and (2) to identify potentially modifiable factors related to surgical management (rather than tumor biology) that correlate with the incidence of CM/CMS. Methods: Attending pediatric neurosurgeons at British Columbia’s Children’s Hospital (BCCH) and neurosurgeons who completed a pediatric neurosurgery fellowship at BCCH were invited to provide data from the center where they currently practiced. Children aged from birth to less than 18 years who underwent initial midline PF tumor resection within a contemporary, center-selected 2-year period were included. Data was obtained by retrospective chart and imaging review. Modifiable surgical factors that were assessed included pre-resection surgical hydrocephalus treatment, surgical positioning, ultrasonic aspirator use, intraoperative external ventricular drain (EVD) use, surgical access route to the tumor, and extent of resection. CM was defined as decreased or absent speech output postoperatively and CMS as CM plus new or worsened irritability. Results: There were 263 patients from 11 centers in 6 countries (Canada, Germany, the Netherlands, India, Indonesia, and the USA). Median age at surgery was 6 years (range < 1 to 17 years). The overall incidence of postoperative CM was 23.5% (range 14.7–47.6% for centers with data on ≥ 20 patients). The overall incidence of CMS was 6.5% (range 0–10.3% for centers contributing data on ≥ 20 patients). A multivariate logistic regression on the full data set showed no significant association between pre-resection surgical hydrocephalus treatment, prone position, ultrasonic aspirator use, EVD use, telovelar approach, complete or near total resection, or treating center and either postoperative CM or CMS. Conclusions: While there was variation in surgical management of midline PF tumors among centers participating in this study, the factors in management that were examined did not predict postoperative CM or CMS.