Abstracts

Rationale: MEG has an inherently higher spatial resolution and may detect epileptiform discharges, which cannot be detected by scalp EEG. MEG is being increasingly used as a part of the non-invasive presurgical evaluation of patients with pharmacoresistant epilepsy. However, as is the case with scalp EEG, not all epileptiform discharges can be detected by MEG, and it is not clear which spikes are visible on MEG. Methods: We included three patients with pharmacoresistant epilepsy, who had stereotyped repetitive interictal spiking on intracranial stereo-EEG recordings. Patient 1 (18 year old, male): Seizures started at 15 months of age, and were characterized by right arm and leg tingling followed by generalized convulsion. Video-EEG monitoring (VEEG) showed interictal spikes in the bilateral paracentral regions and EEG seizures arising from the vertex region. MRI was normal and ictal SPECT suggested a left posterior insular onset. Patient 2 (30 year old, male): Seizures started at age 19 years, and were characterized by tunnel vision or psychic aura followed by staring. VEEG showed no interictal epileptiform discharges, but one EEG seizure arising from the right temporo-parietal region. MRI demonstrated a small area of abnormal cortical thickness and signal involving a deep sulcal aspect of the posterior segment of the right superior temporal sulcus. Patient 3 (30 year old, male): Seizures started at age 12 years, and were characterized by tingling sensation in the throat followed by drooling and garbled speech without loss of awareness. VEEG showed no interictal epileptiform discharges and non-localizable EEG seizures. MRI was normal. PET scan showed subtle hypometabolism in the right fronto-temporal operculum and ictal SPECT suggested a right posterior insular onset. MEG was recorded with a 204 planar gradiometer MEG system (Neuromag, Helsinki, Finland) with simultaneous stereo-EEG. To obtain stereo-EEG recordings intracranial depth electrodes were placed stereotactically following a comprehensive noninvasive evaluation and discussion in a multidisciplinary patient management conference. MEG source localization of interictal epileptiform discharges was obtained using standard equivalent current dipole (ECD) methods. Results: Despite technically good MEG recordings, in two of the three patients repetitive spiking was only detected by the stereo-EEG electrode traversing the restricted irritative zone in the parietal operculum and the depth of the superior temporal sulcus, respectively. In the third patient, MEG detected discharges corresponding to the repetitive spiking seen on stereo-EEG involving a wider area in the frontal operculum. A single tight cluster of ECDs was estimated in the right parietal operculum and posterior insula by MEG. Conclusions: MEG is a non-invasive technique capable of recording some deep spikes arising from a relatively large cortical region. Intracranial EEG (specifically stereo-EEG in these three patients) is more sensitive than MEG for the detection and localization of the epileptogenic cortex.