Abstracts

Rationale: Benign Epilepsy with Centrotemporal Spikes (BECTS) and Childhood Absence Epilepsy (CAE) are the two most common benign epilepsy syndromes, and represent focal and generalized epilepsy respectively. Previous studies suggest both are associated with neurocognitive comorbidities, however it is unclear whether these are the product of epilepsy in general or syndrome-specific effects. This study aims to directly compare the resting state connectivity of the language network between BECTS and CAE and determine to what extent spike burden in BECTS influences connectivity. Methods: Concurrent resting state fMRI and EEG data were collected on newly-diagnosed and drug naïve children with BECTS (n=16) and CAE (n=18). EEG was used to identify absence seizures and centrotemporal spike activity. All absence seizures were removed from fMRI analyses, in addition to 30 seconds pre and post seizure. fMRI resting state data was preprocessed in standard fashion via SPM12, and subsequent connectivity analyses were processed using CONN. Language network connectivity was measured by placing 20mm spherical seeds into bilateral inferior frontal gyri and posterior superior temporal gyri then compared between BECTS and CAE. Connectivity differences were analyzed both with and without the presence of centrotemporal spikes in the BECTS group. In the BECTS group, a linear regression was also performed using connectivity measures and quantity of centrotemporal spikes. Results: With centrotemporal spikes excluded from the study, the BECTS group showed significantly decreased connectivity between the left posterior superior temporal gyrus and other regions of interest compared to CAE (p-FDR<0.1, figure 1). With centrotemporal spikes included, the BECTS group showed additional pairs of significantly decreased connectivity among the language network seeds compared to CAE including bilateral inferior frontal cortex. However, within the BECTS group, linear regression analysis showed no relationship between connectivity and spike quantity. Conclusions: This study is the first to directly compare resting state language network connectivity between BECTS and CAE. Decreased connectivity in BECTS group likely represents a syndrome-specific effect which may contribute to language dysfunction. While adding scans acquired during centrotemporal spikes appeared to further decrease connectivity in the BECTS group relative to CAE, this finding does not appear related to spike burden within the BECTS group. Funding: No funding