Purpose: Research into epileptic networks has recently allowed deeper insights into the epileptic process. Here we investigated the importance of individual network nodes for seizure dynamics.
Methods: We analysed intracranial electroencephalographic recordings of 86 focal seizures with different anatomical onset locations. With time-resolved correlation analyses, we derived a sequence of weighted epileptic networks spanning the pre-ictal, ictal, and post-ictal period, and each recording site represents a network node. We assessed node importance with commonly used centrality indices that take into account different network properties.
Results: A high variability of temporal evolution of node importance was observed, both intra- and interindividually. Nevertheless, nodes near and far off the seizure onset zone (SOZ) were rated as most important for seizure dynamics more often (65% of cases) than nodes from within the SOZ (35% of cases).
Conclusion: Our findings underline the high relevance of brain outside of the SOZ but within the large-scale epileptic network for seizure dynamics. Knowledge about these network constituents may elucidate targets for individualised therapeutic interventions that aim at preventing seizure generation and spread.
Keywords: Centrality; Epileptic networks; Graph analysis; Node importance; Seizure onset zone.
Copyright © 2014 British Epilepsy Association. Published by Elsevier Ltd. All rights reserved.