Fast ripples analyzed from whole-night intracranial EEG recordings accurately predict epilepsy surgery outcome
Abstract number :
249
Submission category :
3. Neurophysiology / 3A. Video EEG Epilepsy-Monitoring
Year :
2020
Submission ID :
2422595
Source :
www.aesnet.org
Presentation date :
12/6/2020 12:00:00 PM
Published date :
Nov 21, 2020, 02:24 AM
Authors :
Päivi Nevalainen, University of Helsinki and Helsinki University Hospital; Nicholás von Ellenrieder - Montreal Neurological Institute and Hospital, McGill University; Petr Klimeš - 1) Montreal Neurological Institute and Hospital, McGill University; Franço
Rationale:
Fast ripples (FRs) detected in intracranial EEG are a highly specific marker of the epileptogenic zone but are thought to suffer from low sensitivity. We examined whether FRs can become a better marker of the epileptogenic zone if analyzed from an overnight intracranial EEG recording, and if their absence reflects suboptimal spatial sampling rather than low sensitivity.
Method:
We analyzed FRs in 43 patients undergoing stereo-EEG. We detected FRs automatically from a whole-night recording edited for artefacts and visually from a 5-minute period of slow-wave sleep. We calculated the accuracy of removing ≥50% of total FR events or of channels with FRs to predict postsurgical seizure outcome based on the whole-night and 5-minute analysis approaches. We also examined any association of i) low overall FR rates or ii) absence or incomplete resection of one dominant FR area with poor outcome. We considered one dominant FR area to exist if the highest FR rate channels (in rank order, sufficient to reach ≥0.5 resection ratio) were confined to a continuous brain area.
Results:
The postsurgical seizure outcome was good (Engel class I) in 15, and poor (Engel II-IV) in 28 patients. The outcome prediction accuracy was highest (81%, 95% confidence interval 67-92%) when using the FR event resection ratio and whole-night recording (vs. 72%, 95% confidence interval 56-85%, for the traditional visual 5-minute approach). Absence of channels with high FR rate ( >6/min; p=0.001) or absence or incomplete resection of one dominant FR area (p< 0.001) were associated with poor outcome.
Conclusion:
FRs are an accurate predictor of epilepsy surgery outcome at individual level when using overnight recordings. Absence of high FR rate channels and absence of one dominant FR area are poor prognostic factors that may reflect suboptimal spatial sampling of the epileptogenic zone or multifocality, rather than an inherently low sensitivity of FRs.
Funding:
:The Canadian Institutes of Health Research, the Montreal Neurological Institute, the International Federation of Clinical Neurophysiology, and the Fonds de Recherche du Québec – Santé.
Neurophysiology