Abstracts

Rationale: Over the past several years, there has been increased use of magnetoencephalography (MEG) as a diagnostic tool in the presurgical evaluation of epilepsy patients. MEG is based on the ability of super-conducting quantum interference devices (SQUID) to study magnetic fields in the brain. Previous case series have shown that MEG is an accurate tool in the presurgical evaluation of pediatric patients with localization related epilepsy. However, MEG data analysis in published studies involves only the single equivalent current dipole method. We performed MEG studies on pediatric epilepsy surgical candidates using both dipole fit and synthetic aperture magnetometry (SAM) analysis. Methods: Between 2004-2008, 18 pediatric patients were referred to the National Institutes of Health for MEG. Five of these patients underwent surgical resection. All patients had a presurgical workup, including brain MRI, video EEG, MEG, and intracranial EEG or intraoperative electrocorticography (ECoG). Neuromagnetic data were recorded using a CTF 275 MEG system composed of a whole-head array of 275 radial 1st order gradiometer/SQUID channels housed in a magnetically shielded room. Simultaneous EEG was recorded using the international 10-20 system. MEG data was analyzed using both equivalent current dipole fit and synthetic aperture magnetometry (SAM). Data were coregistered to MRI. Findings were compared to the intracranial EEG/ECoG results and surgical outcomes. Dipole fit and SAM data were compared for concordance. Results: Five patients, ranging from 5 to 14 years old, underwent presurgical evaluation (including MEG) and subsequent epilepsy surgery. All patients had at least two seizures per week and had failed multiple antiepileptic drugs. All patients had lesions on MRI. The lesions included a left MCA infarct, a right occipital cortical dysplasia, a right parietal meningioangiomatosis, left temporal atrophy secondary to perinatal stroke, and right frontotemporal atrophy secondary to encephalitis. In all patients, the seizure focus found on intracranial EEG/ECoG corresponded to the MRI lesion. MEG dipole fit analysis corresponded to intracranial EEG/ECoG findings in four out of five patients. SAM analysis corresponded to the intracranial EEG or ECoG findings in four out of five patients. Dipole fit and SAM had good concordance in three out of five patients. In the two patients with discordant dipole fit and SAM findings, the seizure focus was found using one of the two methods. All patients underwent seizure focus resection. Four patients had an Engels outcome of I or II with a mean follow-up time of 33 months. One patient had an Engels outcome of III. Conclusions: SAM and dipole fit analysis both provide accurate information on the location of seizure foci in pediatric patients with discrete lesions. Though SAM and dipole fit data were not always concordant, one of the two methods accurately identified the seizure focus in all patients studied. Therefore, using SAM and dipole fit analysis together in pediatric epilepsy is a reliable method for seizure focus localization.