0491 Night-to-Night Variability in Surrogates of Obstructive Sleep Apnea Severity: Main and Sex-Specific Effects on Next-Day Symptom Presentation

Introduction Obstructive sleep apnea (OSA) is a heterogeneous disease with complex pathophysiology and varying symptom presentation. Yet, most of what we know about how the disease expresses is based on the results of a single sleep study and symptom questionnaires that summarize individuals’ recent experience with OSA. Here, we leverage multi-night assessments of objective sleep parameters and daytime symptom presentation in individuals with OSA to assess associations between prior night’s sleep parameters on next morning symptoms. We also assess whether these relationships are moderated by sex. Methods Individuals with untreated OSA (apnea hypopnea index [AHI]≥5 events/h) were recruited from the community and invited to wear a cardiopulmonary coupling-based wearable ring for 15 consecutive nights, along with responding to ecological momentary assessments of daytime symptoms (sleepiness, fatigue, stress, mood and sleep quality). Associations between sleep parameters and next morning symptoms were assessed using linear mixed-effects models controlled for age, body mass index categories and baseline Epworth sleepiness scale score. Models incorporating an interaction term with sex were also evaluated. Results A total of 1,074 nights from 86 participants (52.3% women; mean [SD] age 54.5 [15.5] years; mean [SD] diagnostic AHI 19.0 [18.3] events/h) were included. We observed significant associations between increased prior night sleep efficiency and improved next day sleepiness, fatigue, mood and sleep quality; increased sleep duration and lower stress and improved sleep quality; increased wake after sleep onset (WASO) and worse fatigue and mood; and greater mean pulse rate and worse fatigue, mood, and sleep quality. Suggestive sex-specific effects on the association between sleep latency and stress were also observed (interaction p=0.08). No associations between OSA severity surrogates (AHI, time below 90% saturation, mean event duration, oxygen saturation index) and next day symptoms were observed. Conclusion Cardiopulmonary coupling estimated sleep parameters such as sleep efficiency, sleep duration, WASO, and mean pulse rate were associated with variations in next morning symptom presentations, with sex-specific effects observed on sleep latency. The effect of other surrogates of OSA severity on symptoms remain unclear. Support (if any) NHLBI (1R01HL170675-01); University of Pittsburgh School of Nursing Research Catalyst Award; Sleep and Circadian Science & Aging Research Hubs; SleepImage & SRS Foundation.