Abstracts

Rationale: Surgical treatment of partial epilepsy in children is challenging due to often unlocalizing clinical, neurophysiological, and radiological findings. MEG (magnetoencephalography)analyzed with a beamformer, synthetic aperture magnetometry (SAM) and SISCOM (Subtraction Ictal SPECT Co-registered to MRI) are two non-invasive tests that may contribute to localization of epileptogenic zone. While MEG detects interictal epiletiform discharges, SISCOM detects change in cerebral perfusion during the seizure. We examined whether these two tests concur with each other, and ultimately with ictal onset zone defined by intracranial EEG. Methods: We reviewed MEG SAM and SISCOM data on patients who underwent pre-surgical evaluation for intractable partial epilepsy at the Cincinnati Children’s Medical Center from 8/1/06 to 5/30/07. Among the patients who had both 275 channel MEG (VSM, Vancouver, CA) and ictal SPECT with 99mTc injection, we included only patients whose ictal SPECTs were injected within 30 sec following the onset of seizure. We used a statistical measure of increased kurtosis, SAMg2, to screen for increased spikiness at 323K virtual sensor locations in the brain. We visually inspected virtual sensor waveforms for spike morphology. SISCOM was acquired by subtracting ictal SPECT from interictal SPECT, followed by co-registering to the patient's MRI using imaging software ANALYZE (Mayo BIR, Rochester, MN). We compared SAM virtual sensor spikes locations with SISCOM hyperperfusion by lobar brain region and compared both to lobar localization by subdural grids and strips (IC-EEG). Results: Total of 14 patients met the inclusion criteria (M:F = 7:7; age, 3.20 ± 3.50; age of seizure-onset 10.2 ± 3.95). Nine patients had abnormal MRIs (3 tuberous sclerosis, 4 cortical dysplasia, 2 ischemic/traumatic injury). SAM and SISCOM agreed in 85.7% (12/14). Four patients (2 lesional and 2 non-lesional) underwent IC-EEG and all 4 patients had ictal onset zone co-localizing with SAM and SISCOM. Conclusions: Our data suggest that MEG (SAM) and ictal SPECT (SISCOM) has high concordance in localizing epileptogenic zone. Whether combination of MEG (SAM) and ictal SPECT (SISCOM) could improve the prediction of ictal onset onset zone and the surgical outcome compared to either method alone needs to be studied in the future.