Orbitofrontal epilepsy: distinct neuronal networks underlying electroclinical subtypes and surgical outcomes.

Objective: The aim of this study was to characterize the clinical and electrophysiological findings of epilepsy originating from the orbitofrontal cortex (OFC) as well as its surgical outcomes.

Methods: The authors retrospectively reviewed 27 consecutive cases of patients with drug-resistant orbitofrontal epilepsy (OFE) who underwent tailored resective surgery after a detailed presurgical workup. Demographic features, seizure semiology, imaging characteristics, resection site, pathological results, and surgical outcomes were analyzed. Patients were categorized according to semiology. The underlying neural network was further explored through quantitative FDG-PET and ictal stereo-electroencephalography (SEEG) analysis at the group level. FDG-PET studies between the semiology group and the control group were compared using a voxel-based independent t-test. Ictal SEEG was quantified by calculating the energy ratio (ER) of high- and low-frequency bands. An ER comparison between the anterior cingulate cortex (ACC) and the amygdala was performed to differentiate seizure spreading patterns in groups with different semiology.

Results: Scalp electroencephalography (EEG) and MRI were inconclusive to a large extent. Patients were categorized into the following 3 semiology groups: the frontal group (n = 14), which included patients with hyperactive automatisms with agitated movements; the temporal group (n = 11), which included patients with oroalimentary or manual automatisms; and the other group (n = 2), which included patients with none of the abovementioned or indistinguishable manifestations. Patients in the frontal and temporal groups (n = 23) or in the frontal group only (n = 14) demonstrated significant hypometabolism mainly across the ipsilateral OFC, ACC, and anterior insula (AI), while patients in the temporal group (n = 9) had hypometabolism only in the OFC and AI. The ER results (n = 15) suggested distinct propagation pathways that allowed us to differentiate between the frontal and temporal groups. Pathologies included focal cortical dysplasia, dysembryoplastic neuroepithelial tumor, cavernous malformation, glial scar, and nonspecific findings. At a minimum follow-up of 12 months, 19 patients (70.4%) were seizure free, and Engel class II, III, and IV outcomes were observed in 4 patients (14.8%), 3 patients (11.1%), and 1 patient (3.7%), respectively.

Conclusions: The diagnosis of OFE requires careful presurgical evaluation. Based on their electrophysiological and metabolic evidence, the authors propose that varied semiological patterns could be explained by the extent of involvement of a network that includes at least the OFC, ACC, AI, and temporal lobe. Tailored resections for OFE may lead to a good overall outcome.