0350 Characterizing Continuous Changes in Spectral Dynamics of Sleep EEG as a Function of Age. (27th May 2020)

• Record Type: Journal Article
• Authors: Kim, H; Prerau, M; Redline, S
• Is Part Of: Sleep. Volume 43(2020)Supplement 1
• Page Start: A133
• Page End: A133
• Publication Date: 2020-05-27
• Subjects: Sleep -- Physiological aspects -- Periodicals; Sleep disorders -- Periodicals; Sommeil -- Aspect physiologique -- Périodiques; Sommeil, Troubles du -- Périodiques; Sleep disorders; Sleep -- Physiological aspects; Sleep -- physiological aspects; Sleep Wake Disorders; Psychophysiology; Electronic journals; Periodicals; 616.8498
• DOI: 10.1093/sleep/zsaa056.347
• Languages: English
• ISSNs: 0161-8105

Abstract

Abstract: Introduction: Sleep is a continuous and dynamic physiological process. Current research practice, however, limits our ability to observe electroencephalography (EEG) oscillation dynamics by breaking sleep into discrete stages. In this study, we propose a novel quantitative framework that represents population-level changes in sleep EEG spectral dynamics as a function of age, preserving the information-rich spectral dynamics of sleep data. Rather than relying on sleep stages, our approach uses slow-oscillation power (SO-power) as an objective, continuous-valued correlate of sleep depth. Methods: We analyzed the EEG signal (Fz-Cz, 256 Hz sampling rate) from a subset of the Multi-Ethnic Study of Atherosclerosis (MESA) study participants (n = 2056, 53.6% female, age: mean 69.37 ± 9.12, range 54 - 94) who underwent polysomnography. For each subject, we computed the sleep EEG multitaper spectrogram and extracted the total baseline-normalized SO-power (0.1 - 1.5 Hz). We next computed mean EEG spectral power as a function of SO-power, which we then tracked across all subjects as a function of age in sliding windows. Results: The population analysis shows apparent, continuous changes in time-frequency domain features of the EEG as a function of a sleep depth along with age, that would be otherwise lost in traditional analyses. Moreover, by analyzing the directionality of the SO-power, we show that there is no apparent difference in neural activity during deepening sleep and lightening sleep; thus EEG sleep state is likely non-directional. Conclusion: Our results show that state-based sleep dynamics of the EEG power spectrum can comprehensively be represented using SO-power as a surrogate of sleep depth. This representation identifies state-based activity independent of the temporal evolution of sleep architecture. As such, it is a powerful tool for analysis and phenotyping of EEG activity in large cohorts. Support: The Biomedical Global Talent Nurturing Program through the Korea Health Industry Development Institute, funded by the Ministry of Health & Welfare, Republic of Korea (HI19C1065) to HK, National Institute of Neurological Disorders and Stroke (NINDS, R01 NS-096177) to MP.