Authors :
Presenting Author: Atsuro Daida, MD, PhD – UCLA Mattel Children’s Hospital, David Geffen School of Medicine, Los Angeles
Yuanyi Ding, MS – Department of Electrical and Computer Engineering – University of California, Los Angeles; Yipeng Zhang, MS – Department of Electrical and Computer Engineering – University of California, Los Angeles; Benjamin Edmons, MD – Division of Pediatric Neurology, Department of Pediatrics, UCLA Mattel Children’s Hospital – UCLA Mattel Children’s Hospital, David Geffen School of Medicine, Los Angeles; Shingo Oana, MD, PhD – Division of Pediatric Neurology, Department of Pediatrics, UCLA Mattel Children’s Hospital – UCLA Mattel Children’s Hospital, David Geffen School of Medicine, Los Angeles; Samuel Ahn, MD – Division of Pediatric Neurology, Department of Pediatrics, UCLA Mattel Children’s Hospital – UCLA Mattel Children’s Hospital, David Geffen School of Medicine, Los Angeles; Noriko Salamon, MD – Division of Neuroradiology, Department of Radiology – UCLA Medical Center, David Geffen School of Medicine, Los Angeles; Raman Sankar, MD, PhD – Division of Pediatric Neurology, Department of Pediatrics, UCLA Mattel Children’s Hospital – UCLA Mattel Children’s Hospital, David Geffen School of Medicine, Los Angeles; Aria Fallah, MD, MS – Department of Neurosurgery – UCLA Medical Center, David Geffen School of Medicine, Los Angeles; Richard Staba, PhD – Department of Neurology – UCLA Medical Center, David Geffen School of Medicine, Los Angeles; Jerome Engel, MD – Department of Neurology – UCLA Medical Center, David Geffen School of Medicine, Los Angeles; William Speier, PhD – Department of Bioengineering – University of California, Los Angeles; Vwani Roychowdhury, PhD – Department of Electrical and Computer Engineering – University of California, Los Angeles; Hiroki Nariai, MD, PhD, MS – Division of Pediatric Neurology, Department of Pediatrics, UCLA Mattel Children’s Hospital – UCLA Mattel Children’s Hospital, David Geffen School of Medicine, Los Angeles
Rationale: Sleep spindles are thought to be generated by the thalamocortical loop. High-frequency oscillations (HFOs) play an important role in hippocampal-cortical interaction through sleep spindles during NREM sleep for memory consolidation. However, the role of HFOs in the thalamus remains poorly studied. We hypothesize that HFOs play an integral role in the thalamus in the generation and maintenance of sleep spindles.
Methods: Stereotactic EEG (SEEG) data from nine patients with medication-resistant focal epilepsy who had thalamic coverage were analyzed. Interictal recordings from N2 sleep with a sampling frequency of 2000 Hz, up to 30 min per subject, were selected for further analysis. Sleep spindles (11-16 Hz) were detected by short-term energy methods with subsequent verification by human experts in all SEEG channels, scalp EEGs (Fz, Cz, Pz) channels, and thalamic channels covering the anterior nucleus (AN) and centromedian nucleus (CM) of the thalamus. HFOs co-occurring with the sleep spindles (slHFOs) were quantitatively analyzed in each channel.
Results: We detected a total of 3,135 spindles in the thalamic electrodes, 9,539 in the cortical electrodes, and 895 in the scalp electrodes. Among 351 channels that detected spindles, slHFOs were only observed in the thalamic channels (23 channels) (
Figure). slHFOs had a mean peak frequency of 331.18 (+/- 21.98) Hz and duration of 1.47 (+/- 0.38) seconds. The onset latency of slHFOs was earlier by 0.56 (+/- 0.51) seconds compared to that of sleep spindles (p < 0.01). The peak onset of slHFOs preceded that of the sleep spindles by 0.35 (+/- 0.43) seconds (p = 0.02). Sleep spindles associated with slHFOs had a higher density (mean 4.79/min vs. 1.36/min; p < 0.05) and longer duration (mean 1.21 vs. 0.91 seconds; p < 0.001) than those without slHFOs per channel. There was no difference in the peak frequency of the spindles between the spindles with slHFOs and without slHFOs (13.07 +/- 1.07 Hz vs. 14.04 +/- 0.77 Hz). slHFOs were detected in eight out of nine AN electrodes, contrasting with four out of seven CM electrodes. Sleep spindles in AN exhibited a lower mean peak frequency compared to CM (12.72 Hz vs. 14.08 Hz; p < 0.01), though no significant differences were noted regarding density and duration.