0830 Novel OSA-driven metric predicts incident dementia in sleep apnea in two different community-based cohort studies

Abstract Introduction Obstructive sleep apnea (OSA) is associated with intermittent hypoxemia and sleep fragmentation. In OSA, sleep fragmentation is often measured by the number of electroencephalographic arousals per hour of sleep without considering their intensity. In this study, we sought to determine the association of “arousal intensity” with increased risk of dementia in two independent community-based cohorts: the Osteoporotic Fractures in Men Study (MrOS) and the Multiethnic Study of Atherosclerosis (MESA). Methods Arousal intensity was calculated from the baseline polysomnogram of the MrOS and MESA cohorts. We included participants with available data from baseline study to Visit 4 (median follow-up 10.7 [IQR: 10.3 – 11.1] years) in MrOS and from Mesa Sleep to 2018 in MESA (median follow-up 6.8 [IQR:6.2 – 7.3] years). Dementia at baseline or those who reported positive airway pressure therapy were excluded. The primary analysis assessed the association of arousal intensity with incident dementia (primary outcome). Dementia was defined as a significant change in the Teng Mini Mental Test plus self-report of medical diagnosis during the follow-up visit in MrOS and based on the ICD-9 codes from hospital discharge data in MESA. Multiple logistic regression analyses were used to obtain the adjusted odds ratio per 1 standard deviation increase in arousal intensity after adjusting for anthropometric/demographic variables, depression, alcohol, tobacco, and AHI. Results A total of 1,136 MrOS and 1,923 MESA participants with complete data were included, with a median age of 76 [IQR 72 – 80] years, (100% Male) and 67 [IQR: 61 - 75] years, (46.3% Male), respectively. In MrOS, 11.4% of men had events and every 1SD increase in arousal intensity was associated with a 23% [95%CI: 0 to 50; p< 0.05] increased risk of incident dementia. In MESA, 3.5% of men had events and every 1SD increase in arousal intensity was associated with a 29% [95%CI: 1 to 65; p< 0.05] increased risk of incident dementia. Conclusion In two independent cohorts, high arousal intensity was associated with an increased risk of incident dementia after adjusting for relevant confounders. Quantifying arousal intensity may provide added value for cognitive risk stratification in sleep apnea. Support (if any) AASM Foundation (grant number: 254-FP-21).