Abstracts

EEG-fMRI combines the spatial resolution of MRI with the temporal resolution of EEG. It allows for visualization of BOLD signal changes in response to discrete EEG events e.g., spike and wave discharges (SWD) and, therefore, localization of SWD generators. Temporal clustering analysis (2dTCA) is a method to localize BOLD signal changes without the use of EEG. Both of these techniques may be useful in the investigation of SWD sources and non-invasive evaluation of the contributions of various brain regions to the process of SWD generation. The goal of this preliminary study was to compare these two techniques of SWD source localization in patients with IGE., Five patients with frequent generalized SWD on interictal EEG (2 with JME, 2 with adult onset IGE and 1 with JAE) were studied. 2-3 series of fMRI data (20 minutes each) were collected per patient at 4T. Concurrently, 64 channels of EEG were collected using MRI-compatible EEG system modified for this purpose and Scan 4.3.1 software (Compumedics Neuroscan, El Paso, TX). For the EEG-fMRI analysis, a reference time course file was generated based on the EEG with SWD coded as 1; normal EEG as 0; artifact as -1. EEG-fMRI data were analyzed using CCHIPS[reg]. BOLD signal changes were superimposed on a structural T1 image. The 2dTCA analysis was performed using all available fMRI datasets without EEG information. Reference time course files were generated by determining times at which a significantly large number of voxels have a large signal increase. Activation was determined using a fixed-effects analysis of all series., In 2 patients there was complete agreement in the SWD source localization between the two techniques. These patients had frequent SWD ([gt]9). The BOLD signal changes were mainly localized to paracentral frontal and parietal and lateral parietal head regions. The other 3 subjects had less frequent SWD (maximum 8 in combined runs) and there was minor movement artifact [ndash] the correlation between 2dTCA and EEG-fMRI was partial with localization of the BOLD signal changes to areas similar to the first 2 subjects. Thalamic activation was not observed., In this small study, 2dTCA and EEG-fMRI showed good concordance of BOLD signal distribution in patients with IGE. There was complete concordance in patients with large number of SWD and partial in patients with lower number of SWD. Further study with larger number of subjects using both techniques is needed to establish the degree of agreement between these data analysis tools and the factors that influence the concordance. Both techniques may be useful in evaluating the pathophysiology of SWD generators in patients with IGE., (Supported by TNI and DDF, UCCOM, Cincinnati, OH and NIH K23 NS052468 (JPS) and NIH EB00461 and The Epilepsy Foundation (VLM).)