P01. Transient spike-related connectivity changes in temporal lobe epilepsy: A high density EEG study

Purpose We aimed to analyze the dynamic behavior of epileptic networks through the study of the effective connectivity at a whole-brain scale during interictal spikes in temporal lobe epilepsy patients (TLE) using high-resolution EEG signals. We aimed to understand the connectivity pattern differences in right versus left TLE (RTLE vs LTLE). Method Sixteen patients, 8 with RTLE and 8 with LTLE, were selected for the study. We assessed the connectivity changes of cortical networks during interictal spikes compared to baseline periods at high-temporal resolution, using high-density EEG recordings. The source activity was obtained for 82 regions of interest using an individual head model and a distributed linear inverse solution. A multivariate, time-varying and frequency-resolved Granger causality analysis was applied to the source signal of all ROIs. A non-parametric statistical test was carried out to assess the difference in outflow between interictal spikes vs baseline in each ROI. Results In both groups, the key driving structures were located in the anterior temporal pole, and their driving towards other regions was higher at the time of the spike. In LTLE the keys drivers were only ipsilateral while in RTLE the key drivers were both ipsilateral and contralateral. Moreover, in RTLE we observed a driving pattern from the ipsilateral to the contralateral regions that was not seen in LTLE. Both groups showed driving from the anterior temporal structures to the frontal lobe. In the RTLE and LTLE groups, the pattern of connectivity changes was concordant with the cognitive impairment. Conclusion The used approach was able to identify the major contributors to interictal epileptic activity in both RTLE and LTLE, concordant with invasive electro-clinical findings. Furthermore, a different connectivity pattern was observed in RTLE and LTLE, suggesting that they are not simply symmetrical entities. This enhanced characterization of the epileptic networks increases our understanding of these conditions and could have clinical implications for epilepsy surgery.