Post-operative paediatric cerebellar mutism syndrome: time to move beyond structural MRI.

PURPOSE: To determine the value of structural magnetic resonance imaging (MRI) in predicting post-operative paediatric cerebellar mutism syndrome (pCMS) in children undergoing surgical treatment for medulloblastoma. METHODS: Retrospective cohort study design. Electronic/paper case note review of all children with medulloblastoma presenting to Great Ormond Street Hospital between 2003 and 2013. The diagnosis of pCMS was established through a scoring system incorporating mutism, ataxia, behavioural disturbance and cranial nerve deficits. MRI scans performed at three time points were assessed by neuroradiologists blinded to the diagnosis of pCMS. RESULTS: Of 56 children included, 12 (21.4%) developed pCMS as judged by a core symptom of mutism. pCMS was more common in those aged 5 or younger. There was no statistically significant difference in pre-operative distortion or signal change of the dentate or red nuclei or superior cerebellar peduncles (SCPs) between those who did and did not develop pCMS. In both early (median 5 days) and late (median 31 months) post-operative scans, T2-weighted signal change in SCPs was more common in the pCMS group (p = 0.040 and 0.046 respectively). Late scans also showed statistically significant signal change in the dentate nuclei (p = 0.024). CONCLUSIONS: The development of pCMS could not be linked to any observable changes on pre-operative structural MRI scans. Post-operative T2-weighted signal change in the SCPs and dentate nuclei underlines the role of cerebellar efferent injury in pCMS. Further research using advanced quantitative MRI sequences is warranted given the inability of conventional pre-surgical MRI to predict pCMS.