Predictors of seizure freedom after incomplete resection in children

Objective: Incomplete resection of the epileptogenic zone (EZ) is the most important predictor of poor outcome after resective surgery for intractable epilepsy. We analyzed the contribution of preoperative and perioperative variables including MRI and EEG data as predictors of seizure-free (SF) outcome after incomplete resection. Methods: We retrospectively reviewed patients <18 years of age with incomplete resection for epilepsy with 2 years of follow-up. Fourteen preoperative and perioperative variables were compared in SF and non-SF (NSF) patients. We compared lesional patients, categorized by reason for incompleteness, to lesional patients with complete resection. We analyzed for effect of complete EEG resection on SF outcome in patients with incompletely resected MRI lesions and vice versa. Results: Eighty-three patients with incomplete resection were included with 41% becoming SF. Forty-eight lesional patients with complete resection were included. Thirty-eight percent (57/151) of patients with incomplete resection and 34% (47/138) with complete resection were excluded secondary to lack of follow-up or incomplete records. Contiguous MRI lesions were predictive of seizure freedom after incomplete resection. Fifty-seven percent of patients incomplete by MRI alone, 52% incomplete by EEG alone, and 24% incomplete by both became SF compared to 77% of patients with complete resection (p = 0.0005). Conclusions: Complete resection of the MRI- and EEG-defined EZ is the best predictor of seizure freedom, though patients incomplete by EEG or MRI alone have better outcome compared to patients incomplete by both. More than one-third of patients with incomplete resection become SF, with contiguous MRI lesions a predictor of SF outcome. Neurology 2010;75:1448-1453