Utility of Interictal Data in Guiding Pediatric Epilepsy Surgery

In pediatric patients with drug-resistant epilepsy, successful localization of the seizure onset zone (SOZ) is critical to surgical planning and outcome prognostication. SOZ localization requires the identification of seizures captured through intracranial electroencephalography (iEEG), necessitating prolonged hospital stays for invasive monitoring and two-stage surgical procedures at minimum. Localization of SOZ in automated fashion using only interictal data would enable a substantial reduction in the time required for pre-resection iEEG recording. Using iEEG to model brain functional connectivity (FC) is an approach that has shown significant potential.

We conduct a literature review on directed FC methods and their use in preoperative SOZ localization. Granger Causality, an approach originally used to analyze economic time series, has evolved into multiple similar methodologies for directed FC graph creation. Several modalities show strong correlations between electrodes with specific FC patterns and SOZs, but there is no current tool that can reliably predict SOZs from interictal iEEG data.

Multiple studies show a pattern of increased inward FC in electrodes located in the SOZ during interictal periods, with reversed information flow during seizures, suggesting the increased inward flow toward the SOZ may represent inhibitory pathways, which, when absent, lead to a more epileptogenic state. Further analysis of the changes in directed FC across longer periods may help elucidate how to select optimal segments for localization.