Abstracts

Rationale: Childhood absence seizures consist of brief 5-10 second episodes of unresponsiveness associated with 3Hz “spike-wave” discharges (SWD) on EEG. However, recent studies suggested that more subtle EEG and behavioral abnormalities may occur for many seconds both before and after these brief episodes. Functional magnetic resonance imaging (fMRI) is an effective non-invasive method which allows the investigation of longer-lasting events in brain networks before and after absence seizures. Methods: In this study, simultaneous EEG and fMRI measurement were conducted in 7 pediatric patients with childhood absence epilepsy, and in 7 age-matched normal controls. A dataset of 38 seizure episodes were tested using three different approaches: 1. the analysis of fMRI scans were performed using SPM2, but the onset of the hemodynamic response function (HRF) was systematically shifted from -20s to +20s relative to seizure onset. 2. A time course analysis (percent change of the fMRI signal) was performed for 13 cortical and sub-cortical regions using MARSBAR, in which all regions were segmented in terms of the SPM MRI template Colin27. 3. Correlation, timing, and PCA analyses were performed on the time courses for the identified regions. Results: Our results suggest that a complex sequence of fMRI increases and decreases occur before seizure onset and continue for at least 20 seconds after seizure end. SPM mappings with shifted HRFs showed early cortical fMRI increases in many regions, followed by large and long-lasting cortical decreases. The thalamus showed predominantly increases which occurred later than the cortical increases. Regional time course analyses confirmed that early cortical fMRI increases were present in the frontal lobes. Later peaks in mean activity were observed in rolandic and occipital cortex 7s and 7.6s respectively after seizure onset, and in the thalamus 11.7s after seizure onset. Persistent and delayed decreases were observed in most of the cortical regions. Based upon correlation analyses and PCA, regional fMRI changes were classified into three patterns: i. small early increases and late decreases (e.g. frontal cortex), ii. mainly late decreases (e.g. parietal cortex), or iii. mainly late increases (thalamus). In contrast, analysis of identical time epochs in normal control data displayed none of these changes. Conclusions: These results demonstrate the complex sequence of fMRI changes in absence seizures, which are not detectible using conventional HRF modeling. Cortical and subcortical network changes occur both before and after absence seizures. This may be important mechanistically for seizure initiation and termination, and may also contribute to changes in EEG and behavior which precede and follow absence seizures.