Surgery for Spinal Cord Ependymomas: Outcome and Prognostic Factors

BACKGROUND: Spinal cord tumors account for 5% to 10% of all primary central nervous system tumors. The most common intramedullary neoplasms are ependymomas, composing 50% to 60% of spinal neuroepithelial tumors in adults. OBJECTIVE: To evaluate the clinical and oncological outcomes of patients with spinal ependymoma primarily treated with microsurgery. METHODS: Patient charts and operative notes were analyzed to evaluate the clinical and oncological outcomes of 57 patients (33 men, 24 women) undergoing surgery for spinal ependymal tumors between 1987 and 2007. Mean follow-up was 67 months (range, 1-195 months; median, 56 months). Histopathological findings were 1 subependymoma World Health Organization (WHO) grade I, 16 myxopapillary ependymomas WHO grade I, 39 ependymomas WHO grade II, and 1 anaplastic ependymoma WHO grade III. Histopathological diagnoses were reviewed in 52 cases (91%) using the 2007 WHO classification. RESULTS: There were 47 complete resections (83%). Only 4 patients (7%) underwent (postoperative) radiotherapy. Forty-nine of 57 patients (86%) had stable or improved McCormick grades directly after surgery. A permanent decrease in the McCormick grade was seen in 4 (7%) patients. Multivariate logistic regression revealed only the preoperative neurological status of the patient as an independent predictor of functional outcome (P = .007). Recurrent tumors were diagnosed 12 to 72 months after surgery in 5 of 57 patients (9%) including 3 of 16 myxopapillary ependymomas (19%). In 4 of 5 patients, the primary tumor was incompletely resected. The progression-free survival rate was 89% and 84% for all patients at 5 and 10 years, respectively. An incomplete resection proved the only independent predictor of progression-free survival (P = .05). CONCLUSION: These results support early surgery aiming at complete resection as the primary treatment for presumed spinal ependymomas. The prognosis after surgery for some myxopapillary ependymomas seems worse than generally believed.