fMRI correlates to contralateral interictal epileptiform discharges

RATIONALE: Interictal epileptiform discharges (IEDs) on scalp EEG are limited in their usefulness for identifying epileptogenic regions because of their propagation to regions distant from sources, generation within regions that do not produce seizures, and EEG s poor spatial resolution. fMRI performed with simultaneous EEG (fMRI/EEG) may depict focal regions of metabolic change correlated to IEDs, but the significance of the correlates remains to be determined. To assess significance, we compared the localizations of IEDs fMRI correlates to the localization of an epileptogenic region found using other techniques. METHODS: A subject with partial seizures underwent fMRI/EEG. Baseline EEG demonstrated left temporal slowing and bilateral, independent temporal spikes. Video-EEG monitoring indicated a left-sided onset through semiology and an electrographic ictal onset in the left anterior temporal region. Structural MRI depicted abnormality of the left anterior mesial temporal lobe. Interictal PET depicted hypometabolism of the left anterior temporal lobe. Ictal SPECT depicted hyperperfusion of the left mesial temporal lobe and insula. Overall, the evaluation supported an epileptogenic region localization within the left temporal lobe despite the right temporal IEDs. fMRI/EEG was performed with a 3 Tesla GE scanner, modified for Echo Planar Imaging (EPI) by Advanced NMR Systems, and a custom MRI-compatible EEG machine, designed in collaboration with Grass-Telefactor. Functional EPI Gradient-Echo scans (TR = 2500 ms, TE = 45 ms, 20 cm FOV, 64 x 64 matrix) were collected using 6 axial slices through the temporal lobes. Simultaneous EEG was recorded from 16 standard scalp sites. Maps of individual IEDs were made by convolving time of spike occurrence with an a priori hemodynamic response function, and correlating with the MR voxel intensities using scanSTAT software. Spatial smoothing was performed at 5 mm FWHM. RESULTS: Of the 7 IEDs occurring during the fMRI/EEG study, 6 produced fMRI correlates. Five of these 6 IEDs were right temporal and 1 was left temporal. The fMRI correlates for 4 of the 5 right temporal IEDs were within the left temporal lobe. This contralaterality is a novel observation. The fifth was within the right temporal lobe. The left temporal IED s correlate was within the left temporal lobe. Correlation coefficients all were 0.4 or greater. CONCLUSION: Although the fMRI correlates for 4 IEDs were contralateral to the IEDs scalp location, they co-localized with the presumed epileptogenic region. If fMRI signal change reflects metabolic change at an IED s anatomic source, then this subject s epileptogenic region appears to be a singular source for bilateral, independent IEDs. The presence of bilateral IEDs from a unilateral epileptogenic region is known to occur; however, fMRI previously has not been shown to have the potential to identify one source for multiple IED locations. This potential for fMRI/EEG requires replication with other subjects and support by additional analyses.