Slow-wave Activity Decay in Nights Preceding Epileptic Seizures
Abstract number :
3.495
Submission category :
3. Neurophysiology / 3A. Video EEG Epilepsy-Monitoring
Year :
2024
Submission ID :
1558
Source :
www.aesnet.org
Presentation date :
12/9/2024 12:00:00 AM
Published date :
Authors :
Valérie Costa, MSc – Research center of the Centre Hospitalier de l’Université de Montréal (CRCHUM)
Tian Yue Ding, MD candidate – Research center of the Centre Hospitalier de l’Université de Montréal (CRCHUM)
Amirhossein Jahani, PhD – Research center of the Centre Hospitalier de l’Université de Montréal (CRCHUM)
Presenting Author: Laura Gagliano, PhD – Research center of the Centre Hospitalier de l'Université de Montréal (CRCHUM)
Manon Robert, MSc – Research center of the Centre Hospitalier de l’Université de Montréal (CRCHUM)
Pietro Luca Ratti, MD, PhD – University of Oxford
Dang Nguyen, MD, PhD, FRCPC – CRCHUM, Department of Neuroscience of the Université de Montréal
Valérie Mongrain, PhD – Department of Neuroscience, Université de Montréal
Elie Bou Assi, PhD – Department of Neuroscience, Université de Montréal
Rationale: The bidirectional relationship between sleep and epilepsy is well-known: disturbed sleep can trigger seizures, and seizures can impair sleep quality. In earlier work, we demonstrated a significant link between reduced sleep efficiency, measured by a smart shirt, and subsequent seizures1. Sleep ensures the brain's energy stability through synaptic reorganization during slow-wave sleep (N3 stage of NREM sleep), known as sleep homeostasis. We hypothesize that impaired sleep homeostasis may increase the likelihood of seizure occurrence. This study aims to investigate whether changes in the exponential decay of slow-wave activity (SWA) on scalp electroencephalography (EEG), a key marker of sleep homeostasis, are associated with the occurrence of seizures the following day.
Methods: Patients with confirmed epilepsy were recruited from the Centre Hospitalier de l'Université de Montréal (CHUM) epilepsy monitoring unit between April 2019 and July 2022. Nights were classified as "pre-seizure" if followed by a seizure within 24 hours upon awakening and "non-pre-seizure" otherwise. Nights with sleep deprivation, less than 6 hours of sleep, or nocturnal seizures were excluded. Continuous EEG recordings were analyzed to extract slow-wave activity (SWA, 1-5 Hz) from 20-second sleep epochs. Sleep cycles were identified by intersecting SWA signals with local minima generated using a minimum filter. An exponential decay function (a·e-b·t+ c) was fitted to cycle peaks2 as well as to the area under the curve of each cycle. Statistical comparisons between pre-seizure and non-pre-seizure nights were conducted, testing various configurations: 1) fitting parameters a, b, and c with and without positivity constraints; 2) normalizing (or not) SWA to the average nightly SWA.
Results: A total of 246 nights from 52 patients were analyzed, including 60 pre-seizure and 186 non-pre-seizure nights. Significant differences in SWA exponential decay parameters were found between pre-seizure and non-pre-seizure nights (p< 0.05). Specifically, the decay rate parameter
Neurophysiology