Stefan Fuertinger1, Kristina Simonyan1,2, Michael R Sperling3, Ashwini D Sharan4, Farid Hamzei-Sichani4,5. 1. Department of Neurology, Icahn School of Medicine at Mount Sinai, New York, New York, U.S.A. 2. Department of Otolaryngology, Icahn School of Medicine at Mount Sinai, New York, New York, U.S.A. 3. Department of Neurology, Sidney Kimmel College of Medicine, Thomas Jefferson University, Philadelphia, Pennsylvania, U.S.A. 4. Department of Neurosurgery, Sidney Kimmel College of Medicine, Thomas Jefferson University, Philadelphia, Pennsylvania, U.S.A. 5. Department of Neurosurgery, Icahn School of Medicine at Mount Sinai, New York, New York, U.S.A.
Abstract
OBJECTIVE: Cortical high-frequency oscillations (HFOs; 100-500 Hz) play a critical role in the pathogenesis of epilepsy; however, whether they represent a true epileptogenic process remains largely unknown. HFOs have been recorded in the human cortex but their network dynamics during the transitional period from interictal to ictal phase remain largely unknown. We sought to determine the high-frequency network dynamics of these oscillations in patients with epilepsy who were undergoing intracranial electroencephalographic recording for seizure localization. METHODS: We applied a graph theoretical analysis framework to high-resolution intracranial electroencephalographic recordings of 24 interictal and 24 seizure periods to identify the spatiotemporal evolution of community structure of high-frequency cortical networks at rest and during multiple seizure episodes in patients with intractable epilepsy. RESULTS: Cortical networks at all examined frequencies showed temporally stable community architecture in all 24 interictal periods. During seizure periods, high-frequency networks showed a significant breakdown of their community structure, which was characterized by the emergence of numerous small nodal communities, not limited to seizure foci and encompassing the entire recorded network. Such network disorganization was observed on average 225 s before the electrographic seizure onset and extended on average 190 s after termination of the seizure. Gamma networks were characterized by stable community dynamics during resting and seizure periods. SIGNIFICANCE: Our findings suggest that the modular breakdown of high-frequency cortical networks represents a distinct functional pathology that underlies epileptogenesis and corresponds to a cortical state of highest propensity to generate seizures. Wiley Periodicals, Inc.
OBJECTIVE: Cortical high-frequency oscillations (HFOs; 100-500 Hz) play a critical role in the pathogenesis of epilepsy; however, whether they represent a true epileptogenic process remains largely unknown. HFOs have been recorded in the human cortex but their network dynamics during the transitional period from interictal to ictal phase remain largely unknown. We sought to determine the high-frequency network dynamics of these oscillations in patients with epilepsy who were undergoing intracranial electroencephalographic recording for seizure localization. METHODS: We applied a graph theoretical analysis framework to high-resolution intracranial electroencephalographic recordings of 24 interictal and 24 seizure periods to identify the spatiotemporal evolution of community structure of high-frequency cortical networks at rest and during multiple seizure episodes in patients with intractable epilepsy. RESULTS: Cortical networks at all examined frequencies showed temporally stable community architecture in all 24 interictal periods. During seizure periods, high-frequency networks showed a significant breakdown of their community structure, which was characterized by the emergence of numerous small nodal communities, not limited to seizure foci and encompassing the entire recorded network. Such network disorganization was observed on average 225 s before the electrographic seizure onset and extended on average 190 s after termination of the seizure. Gamma networks were characterized by stable community dynamics during resting and seizure periods. SIGNIFICANCE: Our findings suggest that the modular breakdown of high-frequency cortical networks represents a distinct functional pathology that underlies epileptogenesis and corresponds to a cortical state of highest propensity to generate seizures. Wiley Periodicals, Inc.
Authors: W J E M Zweiphenning; M A van 't Klooster; E van Diessen; N E C van Klink; G J M Huiskamp; T A Gebbink; F S S Leijten; P H Gosselaar; W M Otte; C J Stam; K P J Braun; G J M Zijlmans Journal: Neuroimage Clin Date: 2016-11-04 Impact factor: 4.881