MEG-based identification of the epileptogenic zone in occult peri-insular epilepsy.

INTRODUCTION: Presurgical work-ups of patients with pharmacoresistant epileptic seizures can require multiple diagnostic methods if magnetic resonance imaging (MRI) combined with video-EEG monitoring fails to show an epileptogenic lesion. Yet, the added value of available methods is not clear. In particular, only a minority of epilepsy centres apply magnetoencephalography (MEG). This study explores the potential of MEG for patients whose previous sophisticated work-ups missed deep-seated, peri-insular epileptogenic lesions. PATIENTS AND METHODS: Three patients with well documented, frequent, stereotypical hypermotor seizures without clear focus hypotheses after repeated presurgical work-ups including video-EEG-monitoring, 3Tesla (3T) magnetic resonance imaging (MRI), morphometric MRI analysis, PET and SPECT were referred to MEG source localisation.
RESULTS: In two out of three patients, MEG source localisation identified very subtle morphological abnormalities formerly missed in MRI or classified as questionable pathology. In the third patient, MEG was not reliable due to insufficient detection of epileptic patterns. Here, a 1 mm × 1 mm × 1 mm 3T fluid-attenuated inversion recovery (FLAIR) MRI revealed a potential epileptogenic lesion. A minimal invasive work-up via lesion-focused depth electrodes confirmed the intralesional seizure onset in all patients, and histology revealed dysplastic lesions. Seizure outcomes were Engel 1a in two patients, and Engel 1d in the third. DISCUSSION: MEG can contribute to the identification of epileptogenic lesions even when multiple previous methods failed, and when the lesions are located in deep anatomical structures such as peri-insular cortex. For epilepsy centres without MEG capability, referral of patients with cryptogenic focal epilepsies to centres with MEG systems may be indicated.