Abstracts

Rationale: Corpus callosotomy (CC) is a palliative surgical procedure aimed to improve seizure control in patients with refractory epilepsy who are not candidates for resective surgery. While previously used to reduce frequency of tonic and atonic seizures, callosotomy has been recently attempted in patients with other seizure types with variable success. In Lennox-Gastaut syndrome, the improvement of seizures after CC was greater in patients with anterior-predominant spike-and-wave complexes compared to those with posterior-predominant discharges. Thus, localization of the epileptogenic zone within the rostral-caudal plane may be used to predict the success of CC. The goal of the present study is to examine the postoperative seizure outcomes in patients treated with CC and determine if invasive EEG (iEEG) findings could successfully predict seizure control. Methods: The retrospective chart review of adult and pediatric patients who underwent CC with variable focal resections between 2009-2016 at the University of Nebraska Medical Center revealed 26 patients. Twenty three patients underwent iEEG with subdural grids and strips; all patients had surface EEG (sEEG) recordings. All patients but two were previously treated with vagal nerve stimulator and received 4 anticonvulsants on average. The imaging revealed structural brain lesions, including abnormalities of cortical development, vascular malformations, and tuberous sclerosis in 10 patients. The outcomes were analyzed for tonic-clonic (TC), atonic (A) and complex partial (CP) seizure categories. The postoperative outcomes were graded according to the accepted seizure classification with grades 1 and 5 being consistent with seizure freedom without anticonvulsant use and worsening seizures, respectively. Results: The mean frequencies of TC, A and CP seizures prior to surgery were 189, 264 and 919 per year, respectively. Following the procedure, all seizure types significantly improved by 64, 70 and 38%, respectively (p < 0.05). There was no reduction in the number of anticonvulsants used after CC. The preoperative sEEG revealed bifrontal ictal onset in 16 patients; this was in agreement with the iEEG findings in 8 patients. Unilateral focal, multifocal or generalized ictal onset on sEEG were found in 5 patients; these results were discordant from those of iEEG. In recordings from subdural electrodes, the presence of ictal patterns characterized by voltage attenuation with superimposed fast activity as well as rhythmic spike wave discharges with frontal predominance appeared to predict greater postoperative seizure reduction compare to the other ictal patterns. The predictive value of the anterior-posterior gradient of ictal discharges in seizure control is being assessed. Conclusions: These findings suggest that CC could be an effective palliative surgical procedure for patients with intractable epilepsy, various seizure types and multiple seizure foci. When used in combination with surface EEG findings, the intracranial EEG data allows clarification of complex ictal patterns and may be used to predict surgical outcomes in CC patients. Funding: None