Abstracts

Rationale: Employment of bilateral intracranial EEG (BiEEG) survey may be an effective approach to explore focal epilepsy when non-invasive work-up turns inconclusive. During BiEEG monitoring, diffuse bisynchronous ictal discharges were observed in some patients. To address this situation, our previous study has demonstrated the viability of a multi-stage surgical procedure using BiEEG monitoring pre- and post- complete corpus callosotmy (CC) for further seizure lateralization or localization. However, the selection criteria for this procedure are still not clear. This study aims to find a predictor for possible focal seizure lateralization or localization during the first stage pre-CC BiEEG monitoring by analyzing the correlation among diffuse bisynchronous ictal discharges. Methods: 5 pediatric patients (age 4 mo - 8 yr, M 3 F 2) who underwent a multi-stage surgical procedure involving BiEEG survey pre- and post-CC were recruited in this study. The networking correlation among diffuse bisynchronous ictal discharges were quantified by measuring magnitude-squared coherence (0Hz to 500Hz) during seizure onset. Two iEEG electrodes were considered connected when the coherence coefficient was above a select threshold (0.9). Seizure lateralization was evaluated by calculating a laterality index using the normalized number of intra-hemispheric connection. Seizure localization was evaluated by the locations of the most connected electrodes. The results were then compared with final diagnosis based on the second stage post-CC BiEEG monitoring. Results: For all patients, the first stage pre-CC ictal BiEEG was inundated with diffuse bisynchronous epiletic discharges and could not provide adequate lateralization or localization. The second stage post-CC ictal BiEEG findings were successfully localized (n=3) or lateralized (n=2) , which led to resection of epiletogenic foci or functional hemispherectomy respectively. Surgical outcomes of 4 patients were ILAE class 1 and 1 patient was class 3 (mean follow-up duration 13.2 mos). The coherence analysis of the first stage pre-CC ictal BiEEG showed consistent results with final diagnosis based on the second stage post-CC ictal BiEEG findings. Figure 1 illustrates pre-CC and post-CC ictal BiEEG recordings of a patient. The corresponding pre-CC and post-CC ictal coherence analysis results are depicted in Figure 2. The proposed coherence analysis method could correctly lateralize 5 out of 5 cases and localized 2 out of 3 cases. Conclusions: Despite diffuse bisynchronous ictal discharges, the coherence analysis revealed helpful findings suggesting focal seizure onsets consisted with final post-CC diagnosis. The results indicated that it could be a good predictor for finding possible focal seizure onsets during the first stage BiEEG monitoring and provide a selection criterion for the multi-stage surgical procedure.