Abstracts

Rationale: Subtle cognitive deficits have been reported in some patients with Benign Epilepsy with Centrotemporal spikes (BECTS), especially language difficulties. We aimed to investigate the effect of centrotemporal spike lateralization on language network reorganization using functional MRI.Methods: Twenty one patients who have either left (n=8, age range: 5-12 yrs, mean ± SD: 8.0±2.9 yrs, F/M=3/5) or right (n=13, 6-10, 8.3±1.8 yrs, 5/8) predominant spikes were included. The criterion of spike predominance was calculated as at least 60% left or right, based on the number of spikes during two hours of sleep, recorded by 24-hr ambulatory EEG. All patients participated in an fMRI story listening task, presented as 4 alternating 30-second blocks of auditorily presented stories and broadband noise in the frequency range of speech. After spatial normalization to the MNI template and motion correction, a general linear model approach was used to examine regional changes in BOLD signal for each spike -predominance group independently, and for the contrast between the groups, thresholded at z> 2.3, p< 0.05 corrected. Regions of significant activation for each group were explored in more detail by extracting the mean and maximum intensities of activation for each region. Language skills were assessed using Clinical Evaluation of Language Fundamentals (CELF) and Peabody Picture Vocabulary Test (PPVT).Results: Both groups of patients with either left- and right- predominant spikes showed bilateral superior temporal activations for stories>noise (Figure 1 a, b), similar to previous results in healthy children (Neuroimage. 2006;29(1):254-66, J Magn Reson Imaging. 2009;29(4):971-6). Analysis of significant regions of activation revealed that the left-predominant group had higher peak activation and a slightly higher mean intensity of activation on the right, whereas the patients with right-predominant spikes had a larger extent of activation and a slightly higher mean intensity of activation on the left. When the left- and right--predominant groups were contrasted, an increase in the left postcentral/supramarginal gyrus was observed for the left-predominant group compared to the right-predominant group (Figure 1c.) In addition, language skill assessments revealed that the patients with left-predominant spikes had lower language scores than the right-predominant group (Table 1.)Conclusions: This study offers preliminary evidence that language networks may reorganize based on the spike predominance in BECTS. Subtle shifts in lateralization of activation were observed contralateral to the predominant side of centrotemporal spikes. In addition, patients with left-predominant spikes showed an increase in activation in a region of the left hemisphere posterior to the region typically activated for the task, and also had lower scores on standardized language testing. Reorganization of language networks may underlie language difficulties in some BECTS patients.