Introduction: Extent of resection is a known prognostic factor for progression free survival (PFS) of patients with intramedullary ependymomas.

Methods: From 2002-2010, 74 patients were operated for intramedullary ependymoma; 25 were excluded for <12 months follow-up/incomplete records; 49 remaining patients were reviewed, 48 had complete imaging. Kaplan-Meier survival analysis and log-rank tests assessed PFS. Pearson chi-square test assessed categorical and nominal variables. Univariate and multivariate logistic regression assessed imaging, surgical resection, and disease progression.

Results: Of the 49 patients, 41 (56% male; age 42±12 years/range 14-67; mean follow-up 49±37 months/range 12-125) underwent gross total resection (GTR) and 8 (88% male; mean age 43±15 years/range 18-59; mean follow-up 80 ± 64 months/range 18-205) underwent subtotal resection (STR). PFS was significant between groups (Fig 1, p=0.003). Of the 41 GTRs, 3 underwent en bloc resection without recurrence, and 39 patients underwent intralesional GTR with 3 recurrences. Of the 8 STRs, 7 patients recurred including 2 dead of disease. Surgical resection type, evidence of disease progression, and postoperative complications were independent of spinal location, tumor enhancement, and presence of capsular cyst (p>0.05). Between presentation and final follow-up, the likelihood of bowel/bladder dysfunction was significantly greater with STR vs GTR (50% vs 15%;p=0.02); but not for motor function (50% vs 32%;p>0.05), sensory (50% vs 27%;p>0.05) or dysesthestic pain (12.5% vs 17.5%;p>0.05). 46 patients had Grade II ependymomas with 8 recurrences and 3 had Grade III with 1 recurrence, although statistically histologic grade was independent of surgical resection achieved and disease progression (p>0.05).

Conclusions: Extent of resection significantly impacts PFS in patients with intramedullary ependymoma. Progression occurred as early as 11 months and as late as 205 months from surgery supporting continued long-term patient surveillance especially after subtotal resection. Preoperative imaging characteristics did not predict degree of resectability or disease progression.

Patient Care: Encourage surgeons to pursue a GTR to improve overall prognosis and assist in counseling patients prior to operative intervention

Learning Objectives: 1. Discuss the potential implications of degree of resection on PFS. 2. Discuss the revelance of preoperative imaging to surgical resection

References: 1. Cervoni et al., Recurrence of spinal ependymoma: risk factors and long-term survival. Spine 19(24):2838-2841, 1994 2. Kucia et al., Surgical treatment and outcomes in the treatment of spinal cord ependymomas, Part 1: Intramedullary Ependymomas. Neurosurg 68(ONS Suppl 1): ons57-ons63, 2011 3. Nakamura et al., Long-term surgical outcomes for myxopapillary ependymomas of the cauda equina. Spine 34(21):E756-E760, 2009 4. Whitaker et al., Postoperative radiotherapy in the management of spinal cord ependymomas. J Neurosurg 74:720-728, 1991 5. Chang et al., Surgical outcome and prognostic factors of spinal intramedullary ependymomas in adults. J of Neuro-Onc 57: 133-139. 2002. 6. Quigley et al., Outcome predictors in the magement of spinal cord ependymoma. Eur Spine J 16: 399-404, 2006. 7. Oh, et al. Prognosis by tumor location in adults with spinal ependymomas. J Neurosurg Spine 18:226-35, 2013.