Correlation between pre‐ or postoperative MRI findings and cerebellar sequelae in patients with medulloblastomas

Le résumé fourni par la source

INTRODUCTION: Immediate and delayed cerebellar dysfunction may be expected after surgical resection of a medulloblastoma. We investigated whether pre-operative and delayed post-operative MRI may correlate with such sequelae. MATERIAL AND METHODS: The data of 31 patients in continuous complete remission after removal of medulloblastoma, irradiation and chemotherapy, were retrospectively reviewed. Magnetic Resonance Imaging (MRI) was analyzed for the following items: preoperative MRI (ratio of the surface of the tumor/posterior fossa, presence of ventricular dilatation or tonsilar hernia, involvement of the dentate nucleus) and delayed post-operative MRI (amount of cerebellar parenchyma removed, degree of cerebellar atrophy, presence of T1 hypointense regions in remaining cerebellar area and removal of region containing dentate nucleus). These data were correlated with immediate and long-term cerebellar syndrome and daily life repercussions. RESULTS: On preoperative MRI, the ratio of the surface of the tumor/posterior fossa and the presence of tonsilar hernia were significantly correlated with long-term sequelae on speech (respectively P = 0.027 and P = 0.05). Initial supratentorial ventricular dilatation was correlated with ability to sustain adequately daily tasks (P = 0.002). On delayed MRI, cerebellar atrophy was inversely correlated with ability to sustain daily tasks (P = 0.002). Hypointense T1 territory in remaining cerebellar parenchyma significantly correlated with immediate post-operative cerebellar syndrome (P = 0.01) and showed a tendency for post-operative mutism (P = 0.087) but was not correlated with any long-term sequelae. CONCLUSION: Increased cranial pressure on initial MRI and cerebellar atrophy detected on subsequent MRI studies correlated with immediate and long-term cerebellar sequelae.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.