Abstracts

Rationale: While Benign Rolandic Epilepsy (BRE; Benign Epilepsy with Centrotemporal Spikes; BECTS) is associated with a favorable outcome regarding seizure remission, when active, BRE is associated with a broad number of behavioral and cognitive deficits that extend beyond the neural circuits associated with the interictal spikes. The widespread behavioral and cognitive deficits raise the questions as to whether BRE is a focal or a generalized seizure disorder. In this study we assessed whether centrotemporal spikes (CTS) are associated with widespread changes in brain connectivity that could explain the cognitive and behavioral consequences of the condition. Methods: To assess functional connectivity in the children with BRE we assessed spectral power and coherence during awake and sleep records in 38 children (20 girls and 17 boys with a median age of 8 years [range 2-18 yrs]) with CTS. Coherence represents the consistency of the phase difference between two EEG signals when compared over time and serves as a measure of synchronization between two EEG signals based mainly on phase consistency. Epochs of EEG with and without CTS were compared during both waking and sleep. Results: During the spike periods there was an increase in spectral power at all frequencies, although statistical significance was seen primarily in the delta, theta and alpha bandwidths (Fig. 1). This increase in absolute power was seen at all electrode sites and was similar in left and right-sided electrodes. When relative power was assessed there was a decrease in delta and gamma frequencies while increases were seen in theta and alpha bandwidths. During CTS there were significant changes in coherence compared to the EEG segments without spikes . Figure 2 shows heatmaps of p values for all 171 electrode pairs. In the delta bandwidth coherences were decreased during the CTS although with the corrected p values for false detection few electrodes pairs differed significantly. In the theta, alpha and beta bandwidths there were significant increases in coherence noted. The increases in coherences were widespread and bilateral and involved electrode pairs outside the central and temporal regions. To determine if there was a relationship between location of the spikes and coherence values, right-sided, left-sided and bilateral CTS were compared. There was no relationship between location of the CTS and coherence values. Conclusions: CTS were associated with marked increases in coherences and increases in theta, alpha and beta bandwidth power compared to EEG epochs without CTS. These findings suggest that while the EEG signature of BRE is focal spikes, CTS result are associated with widespread changes in spectral power and coherence. These findings indicate that BRE results in generalized changes in spectral power and connectivity and raise the argument that from a functional standpoint, BRE may be better classified as a generalized, rather than focal seizure disorder. While extrapolating EEG measures to behavior and cognition is challenging, there is increasing evidence that excessive coherence, or hyper-connectivity can result in less flexibility and be associated with abnormalities in behavior and cognition. Further studies are needed to determine if EEG coherence can serve as a biomarker for BRE-associated behavioral and cognitive dysfunction. Funding: No funding