MEDB-29. Application of Rotterdam Post-Operative Cerebellar Mutism Syndrome Prediction Model to Patients Operated for Medulloblastoma in a Single Institution

BACKGROUND: Post-operative cerebellar mutism syndrome (CMS) develops in up to 30% of children. The Rotterdam model (RM) predicts a 66% risk of CMS in patients with a score ≥100. However, our findings suggested that surgical experience contributes to CMS risk. The aim of this study was to retrospectively apply the RM and report incidence of CMS in high-risk patients from our institution. METHODS: Participants had to have first tumor resection at our institution and be enrolled on SJMB12 protocol (NCT01878617). All participants got structured serial neurologic evaluations. CMS, when present, was categorized into type 1 (complete mutism) and type 2 (paucity of speech with an inability to string 3-word sentence). Rotterdam score is calculated based on pre-operative imaging parameters and study neurologist (RBK) obtained it while blinded to CMS status. RESULTS: Of the 40 (14 female, 26 male) study participants, 4 (10%) had CMS (3 CMS1, 1 CMS2). Median age at tumor resection was 11.7 years (range 3.5-17.8). Tumor location was midline in 30 (75%), right lateral 6 (15%) and left lateral 4 (10%). Median Evans index was 0.3 (0.2-0.4) and 34 (85%) were ≥0.3 (indicative of hydrocephalus); 5 participants needed ventricular shunt. Median tumor volume was 50 cm3 (2-180.6). Gross total resection was achieved in 35 (87.5%), near total in 4 (10%) and subtotal in 1. Twelve tumors were SHH, 7 WNT, and 29 NWNS. Median RM score was 90 (25 – 145). Eighteen participants had a score of ≥100 and 16.7% of these (n=3) had CMS. Scores for the 4 with CMS were 85, 125, 145 and 145. CONCLUSION: At our institution, the incidence of CMS in those that had RM of ≥100 was much lower than reported risk of 66%. This data supports our hypothesis that neurosurgical experience remains a significant risk factor in the development of CMS.