Abstracts

Propagation of Interictal High-Frequency Oscillations in Focal Epilepsy

Abstract number : 1.042
Submission category : 1. Basic Mechanisms / 1C. Electrophysiology/High frequency oscillations
Year : 2018
Submission ID : 501485
Source : www.aesnet.org
Presentation date : 12/1/2018 6:00:00 PM
Published date : Nov 5, 2018, 18:00 PM

Authors :
Karina A. González Otárula, Montreal Neurological Institute and Hospital; Nicolás von Ellenrieder, Montreal Neurological Institute and Hospital; Carolina Cuello-Oderiz, Montreal Neurological Institute and Hospital; François Dubeau, Mon

Rationale: To investigate whether high-frequency oscillations show spatiotemporal propagation and assess the relevance of the earliest oscillations in relation to the seizure onset zone (SOZ) and resected brain tissue in a cohort of post-surgery seizure free patients.  Methods: We investigated the intracerebral EEG of patients with drug-resistant focal epilepsy who became seizure free after surgery. During one hour of interictal slow wave sleep we marked high-frequency oscillations, both ripples: 80-250 Hz and fast ripples: 250-500 Hz (i.e: events). We calculated the time delay between the onset of overlapping events in pairs of channels and constructed a timeline of the delay of each event with respect to the earliest event (first source channel). A network was defined if a statistically consistent temporal order could be established among the pairs of channels showing those overlapping events, with significant delays between channel pairs. The SOZ was defined as the first unequivocal ictal intracranial EEG change form background leading to a clear seizure discharge. Results: Fifteen patients were included. We found ripple networks in all patients, and fast ripple networks in 13. For ripples, first source channels were found in a higher proportion in the SOZ than the rest of the network channels (11/20, 52% vs. 109/399, 27%; p<0.05). First source fast ripples were also slightly more frequently found within the SOZ than the rest of the fast ripple network channels, but this difference was not statistically significant (11/16, 69% vs. 46/81, 57%; p=0.37). For both ripples and fast ripples, first source channels were resected more often than the rest of the network channels (Ripples: 10/20, 50% vs. 98/399, 25%; p<0.05. Fast ripples: 11/16, 69% vs. 18/81, 22%; p<0.005). Channels with the highest rates of ripples and fast ripples were resected in a similar proportion than the proportion of resected first source channels.        Conclusions: These results demonstrate that interictal high-frequency oscillations propagate from a source channel and are organized in networks, and indicate a possible need for the resection of earliest events.  However, these earliest events were resected in a similar proportion than the channels with highest rates of high-frequency oscillations; therefore their resection is not superior to the highest rate channels resection in terms of correlation with outcome. Funding: This study was supported by the Canadian Institutes of Health Research grant FDN 143208. K.G.O was also supported by the 2017 Epilepsy Canada Clinical and Basic Sciences Fellowship Grant.