Abstracts

Rationale: The extent of cortical resection in epilepsy surgery was initially guided by peroperative electrocorticography (ECoG), until it became clear that interictal ECoG abnormalities were insufficient to precisely delineate the ictal onset zone, because nonspecific. Therefore, ECoG was replaced  by invasive presurgical investigations, such as SEEG, and more recently by an number of increasingly more accurate non-invasive presurgical investigations. These improvements are particularly relevant for type II focal cortical dysplasias (FCD) in which complete resection is predictive of a good outcome.Recently, a specific electrophysiological signature of type II FCD has been described, giving a reason to revisit ECoG. Using this technique peroperatively remains difficult because recording some cortical zones - the deeper parts of the cortex and the vertical part of the gyri - requires heavy source reconstruction algorithms. To pushback this limit we developed a 3D deep electrocorticography (3D ECoG). The goal of this preliminary study is to investigate the feasibility of the technique and its influence on the resection limits during surgery. Methods: From 2015 to 2016, all patients undergoing type II FCD resection for drug-resistant focal epilepsy were included and benefited from 3D ECoG during surgery. Phase I and phase II (if necessary) preoperative investigations followed the current clinical practice. 3D ECoG was performed under general anesthesia using Target Controlled Infusion protocol with PROPOFOL using MicroDeep 5® recording contact electrodes (DIXI-Medical®, Besançon France). Functional mapping using direct electric stimulation was performed if necessary. Results: Five patients were included, four females and one male (ages 15 to 49). Pathological analysis confirmed type II FCD in all patients. Four patients benefited from SEEG prior to the surgical resection. Three patients benefited from peroperative functional mapping, none of which limited the resection. No complication occurred.We found that type II FCD specific EEG spiking activity could be detected in all patients. 3D ECoG-guided resection and presurgical planning were identical in one patient; in four patients, 3D ECoG lead to extend the limits of the resection as planned presurgically, so that at the end of the surgical procedures, cortical zone showing spiking activity had been completely removed in all patients. Conclusions: peroperative 3D ECoG is feasible, and can lead to modify the limits of a surgical resection in type II FCD, even after exhaustive presurgical investigations. Such finding prompts to reconsider the presurgical strategy. With such a technique, preoperative investigations could be aimed only to detecting type II FCD and its gross localization. Accurate delineation could be performed using peroperative 3D ECoG.These preliminary results, coupled with promising simplifying perspectives, need to be confirmed on a larger population, with long-term follow-up on seizure outcome. Funding: None