Risk factors for development of postoperative cerebellar mutism syndrome in children after medulloblastoma surgery.

OBJECTIVE Postoperative cerebellar mutism syndrome (pCMS) occurs in 7%-50% of children after cerebellar tumor surgery. Typical features include a latent onset of 1-2 days after surgery, transient mutism, emotional lability, and a wide variety of motor and neurobehavioral abnormalities. Sequelae of this syndrome usually persist long term. The principal causal factor is bilateral surgical damage (regardless of tumor location) to any component of the proximal efferent cerebellar pathway, which leads to temporary dysfunction of cerebral cortical regions as a result of diaschisis. Tumor type, cerebellar midline location, and brainstem involvement are risk factors for pCMS that have been identified repeatedly, but they do not explain its latent onset. Ambiguous or negative results for other factors, such as hydrocephalus, postoperative meningitis, length of vermian incision, and tumor size, have been reached. The aim of this study was to identify perioperative clinical, radiological, and laboratory factors that also increase risk for the development of pCMS. The focus was on factors that might explain the delayed onset of pCMS and thus might provide a time window for taking precautionary measures to prevent pCMS or reduce its severity. The study was focused specifically on children who had undergone surgery for medulloblastoma. METHODS In this single-center retrospective cohort study, the authors included 71 children with medulloblastoma, 28 of whom developed pCMS after primary resection. Clinical and laboratory data were collected prospectively and analyzed systematically. Variables were included for univariate and multivariate analysis. RESULTS Univariate regression analysis revealed 7 variables that had a significant influence on pCMS onset, namely, tumor size, maximum tumor diameter > 5 cm, tumor infiltration or compression of the brainstem, significantly larger decreases in hemoglobin (p = 0.010) and hematocrit (p = 0.003) in the pCMS group after surgery than in the no-pCMS group, significantly more reported incidents of severe bleeding in the tumor bed during surgery in the pCMS group, preoperative hydrocephalus, and a mean body temperature rise of 0.5°C in the first 4 days after surgery in the pCMS group. Multiple regression analysis revealed that tumor size, tumor infiltration into or compression of the brainstem, and higher mean body temperature in the first 4 postoperative days were independent and highly significant predictors for pCMS. CONCLUSIONS The authors confirmed earlier findings that tumor-associated preoperative conditions, such as a maximum tumor diameter ≥ 5 cm and infiltration into or compression of the brainstem, are associated with a higher risk for the development of pCMS. Most importantly, the authors found that a 0.5°C higher mean body temperature in the first 4 postoperative days increased the odds ratio for the development of pCMS almost 5-fold. These data suggest that an important focus for the prevention of pCMS in children who have undergone medulloblastoma surgery might be rigorous maintenance of normothermia as standard care after surgery.