Abstracts

Rationale: Graph theory has gained a major role in characterizing the physiopathology of epilepsy, confirming changes in local connectivity and disruptions in long-distance connections, affecting important hubs of the resting-state networks^1,2. The aim of this work is to characterize these alterations using non-invasive High Density EEG (HD-EEG) in combination with source localization and directional connectivity measures³. Methods: Eleven patients were studied (6 females, mean age 35 years) with HD-EEG in resting state (eyes closed, no interictal discharges) before intracranial evaluation, which allowed the precise determination of the seizure onset zone (SOZ). Weighted Minimum Norm Estimate was used to estimate dipole source time courses in the brain based on scalp-recorded signals. Information inflow and outflow of atlas-based brain regions were estimated through partial directed connectivity from these dipole time courses. A set of graph measures was calculated for pairwise connections in standard EEG frequency bands^4,5 and compared using Wilcoxon tests. Results: From the patients studied with preoperative eyes-closed resting-state EEG 4 suffered left temporal epilepsy, 6 right temporal epilepsy and 1 of them right frontal epilepsy confirmed by intracranial EEG. Different patterns of information-flow were found depending on the frequency band. Inflow was significantly lower in the region containing the SOZ than in other regions in the theta (5-8 Hz, p=0.024) and alpha (9-12 Hz, p=0.023) frequency bands. In the theta frequency band, global efficiency, betweenness centrality, cost and degree were significantly higher in regions containing the SOZ (p=0.017, 0.002, 0.013 and 0.013, respectively). Conclusions: Neuronal connectivity abnormalities in epileptic patients were detected during resting-state periods. Inflow showed a higher value localizing the SOZ than outflow. Based on our results and previous works^5,6, we hypothesize that the SOZ has a dysfunction pattern complex and dynamic, so that during the interictal state there is a decreased inflow from the surrounding areas into the SOZ, causing disruptions in global connectivity which may increase the outflow from the SOZ and thereby trigger a seizure. We also found a higher global efficiency, betweenness centrality, cost and degree in the region containing the SOZ. These findings support a central role of the SOZ in the global brain network, enabling it to recruit and synchronize the activity of neurons throughout the cerebral cortex during the ictogenesis. Using graph-theoretic measures to analyze the epileptic network based on surface EEG recordings provides new possibilities for the diagnosis and treatment of patients with epilepsy.References:1. Stam CJ. Modern network science of neurological disorders. Nat Publ Gr. 2014;15(10):683-695.2. Englot DJ, Hinkley LB, Kort NS, et al. Global and regional functional connectivity maps of neural oscillations in focal epilepsy. 2015.3.  Coito A, Genetti M, Pittau F, et al. Altered directed functional connectivity in temporal lobe epilepsy in the absence of interictal spikes: A high density EEG study. Epilepsia. 2016.4.  Sporns O, Honey CJ, Ko R, Sciences B, Neurophysiology S. Identification and Classification of Hubs in Brain Networks. 2007;(10).5.  Nissen IA, Stam CJ, Reijneveld JC, et al. Identifying the epileptogenic zone in interictal resting-state MEG source-space networks. Epilepsia. 2016:1-12.6.  Varotto G, Tassi L, Franceschetti S, Sprea R, Panzica F. Epileptogenic networks of type II focal cortical dysplasia?: A stereo-EEG study. 2012;61:591-598. Funding: No funding was received in support of this abstract.