Abstracts

Rationale: Patients with medically refractory epilepsy often undergo invasive evaluation guided surgery, with a suboptimal 50% success rate. The unsuccess results are possibly explained by poor/incomplete identification of the epileptic networks responsable for the generation and early propagations of the seizures. Consequently, new techniques for mapping and treating epileptic networks are needed. This work introduces a new method using fMRI with simultaneous direct electrical stimulation followed by resection or focal laser ablation of the epileptic nodes within the epileptic networks as defined by invasive recordings.Methods: Fifteen patients with extra-temporal medically intractable epilepsy who underwent SEEG evaluations were studied. After the extra-operative SEEG recording periods, patients underwent fMRI procedures under direct electrical stimulation (DES) (25 Hz, 1-3mA, 300ms PW) of the ictal onset areas as previously defined by the SEEG method, using SEEG depth electrodes. In addition, stimulation of cortical areas not involved with the epileptic activity was also performed (control). The areas in the brain showing fMRI responses were compared with the previous epileptic zone localization hypotheses by SEEG recordings and cortico-cotical evoked potentials. In restricted EZs, laser ablations of the seizure onset areas were performed and DES-fMRI studies were conducted after the ablation.Results: Robust fMRI maps of activation networks were easily produced, which also showed a significant positive correlation between quantitative measures of BOLD activity and measures of electrical connectivity in response to direct electrical stimulation (mean correlation coefficient of 0.38). In 14 patients, successful surgical outcome was consistent with the resection/laser ablation of brain areas containing high local fMRI activity, which anatomically corresponded to the previous SEEG localization hypotheses. In three patients, laser ablations of ictal onsets were performed and DES-fMRI images after the ablations resulted in extinguishment of the BOLD signals, when compared with pre-ablation images. In the last group, all patients had favorable outcome with a minimal follow up period of 6 months, No complications or adverse events were observed.Conclusions: The described method demonstrates the feasibility of simultaneous direct electrical stimulation and fMRI in humans, allowing the study of brain connectivity with high resolution and full spatial coverage. It also demonstrates the capability of acute and chronic network modulation by small but strategically placed lesions guided by SEEG data, acutely verified during surgery. These results suggest that DES-fMRI could be used for full-brain volume evaluation of brain connectivity, refining laser ablative and other neuro-modulation procedures for epilepsy.