Abstracts

RATIONALE: Magnetoencephalography (MEG) is increasingly utilized for pre-surgical localization of epileptic foci and eloquent cortex. The utility of MEG must be balanced against the high cost and technological investments required to perform these studies, particularly when less expensive localization methods are available. To help elucidate the value of MEG, we critically reviewed our experience with whole-head MEG in the management of neocortical epilepsy focusing on cases in which MEG provided unique localization data.
METHODS: MEG studies were performed in a 68 channel whole-head magnetometer (CTF Systems). Interictal spikes and somatosensory evoked field dipoles were calculated using a singe equivalent dipole model and registered to MRI images using standard methodology. Our epilepsy database was reviewed to identify all patients with suspected neocortical epilepsy that underwent MEG studies at our institution and subsequently had a surgical procedure. We reviewed the clincal course, electrocorticography (ECoG) findings, and surgical outcomes of these patients to determine if: 1) MEG data corresponded with extra-operative cortical stimulations or ECoG-defined seizure-onset zones; 2) MEG data provided critical localization data; 3) MEG-guided surgical resection correlated with surgical outcome.
RESULTS: Over a 16 month period 48 MEG studies were performed, including 23 patients with suspected neocortical epilepsy. Subdural grid or depth electrodes were subsequenlty implanted in 15 of these patients. MEG regionally correlated with ECoG and surgical outcome in 9 (60%) cases. In 4 cases MEG provided unique localization data not evident from other imaging modalities that strongly influenced the surgical management of the patient. This included 2 cases in which the MEG findings directed the placement of subdural electrodes to define a seizure onset zone, and 2 cases in which MEG defined regions of eloquent cortex in relationtion to the zone of seizure onset that was critcial to the safe treatment of the patient. In all cases the primary somatosensory region was correctly identified by MEG, as verified by intra-operative cortical stimulations. Median follow-up period is 17 months. One patient was deplanted without resection due to risk of injury to the primary sensory cortex. All patients that underwent a resection are seizure free with no neurological deficits.
CONCLUSIONS: MEG can provide localization information that is not available by other methods. MEG is especially valuable in cases of neocortical epilepsy. The utility of MEG in select cases of neocortical epilepsy justifies the expense and technological investment required by this developing technology.
Support: NIH NS20806, NCRR RR13176, Huntington Medical Research Institutes, Huntington Hospital, a gift from Robert S. and Denise Zeilstra.