0732 Quantitative Characterization Of Sleep Disordered Breathing Dynamics. (27th May 2020)

• Record Type: Journal Article
• Authors: Chen, S; Eden, U; Prerau, M
• Is Part Of: Sleep. Volume 43(2020)Supplement 1
• Page Start: A278
• Page End: A279
• 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/zsaa056.728
• Languages: English
• ISSNs: 0161-8105

Abstract

Abstract: Introduction: Sleep-disordered breathing (SDB) is a dynamic process in which the rate of respiratory events is highly influenced by numerous covariates, such as sleep stage, body position dominance, time of night, and overall instability of sleep architecture. Additionally, respiratory event rate likely has history dependence, such that the likelihood of a respiratory event is influenced by the timing of previous events. Despite its dynamic nature, clinical diagnosis collapses the complex process of SDB to a single number measuring the average rate of respiratory event occurrence— the apnea-hypopnea index (AHI). Thus, potentially valuable information is being lost by ignoring SDB temporal dynamics and history dependence. Methods: We apply a general point process framework to sleep apnea events to develop parametric models of a time-varying instantaneous apnea rate given clinical covariates (e.g. EEG power, sleep stage, body position). Develop models of apnea history dependence, describing the likelihood of events given past event times. In doing so, we are able to compute an "instantaneous AHI", which measures the moment-by-moment event rate, which evolves temporally as a function of other clinical observations as well as event history. We apply our model to data from the MESA cohort (include number of subjects, male/female here ). We then applied dimensionality-reduction methods to assess any population phenotypes. Results: For every subject, we computed a time-varying AHI for each time point in their polysomnogram and estimated the influence of each of the measured covariates on the instantaneous rate. Results showed that the greatest predictor of apnea events were related to history dependence. Clustering analysis showed no distinct clusters, but rather a constant gradient of changes in history dependence. Conclusion: These results suggest that the greatest predictor of an apnea event onset is the timing previous event. Moreover, the way in which previous events influence subsequent events can be used as a means of phenotyping, paving the way towards identifying optimal personalized treatment. Support: National Institute of Neurological Disorders and Stroke Grant R01 NS-096177