0329 Wake Intrusions in the EEG: A Novel Application of the Odds Ratio Product in Identifying Subthreshold Arousals

Abstract

Abstract Introduction Arousal scoring is controversial, due to its lack of objectivity and dependency on other intrinsic characteristics of sleep. Conventional arousal measurements have not consistently predicted insomnia severity. The present study used a novel application of the Odds Ratio Product (ORP), a measure of sleep depth ranging from 0 (deep sleep) to 2.5 (fully awake), to identify subtle wake intrusions in sleep (i.e., subthreshold arousals). We assessed whether these intrusions are better predictors of frequent insomnia symptoms when compared to conventional arousal measurements. Methods Using cross-sectional data from the Sleep Heart Health Study with high-quality ORP data (SHHS1; n = 5,771), accessed through the National Sleep Research Resource, we defined ‘wake intrusions’ as the number of times the ORP spiked above a wake threshold of 2.0 (99% agreement in wake scoring among sleep scorers) during sleep. The wake intrusion index (WII) was derived by calculating the quotient of the number of intrusions by total sleep time. Insomnia was defined from self-reported frequency of >15x/month of insomnia symptoms: delayed sleep onset, difficulty maintaining sleep, and/or early morning awakenings (n = 617). Linear models were used to examine the associations between insomnia and WII, as well as arousal index. Analyses were adjusted for age, sex, body mass index (BMI), and apnea-hypopnea index (AHI). Results There was a weak, but significant correlation between conventional arousal index and WII (r=0.25, 95%CI [0.22, 0.27], p< 0.001). After adjusting for age, sex, BMI and AHI, frequent insomnia symptoms were associated with WII (β=9.47, 95%CI [2.07, 16.87], p=0.012), but not with arousal index (β=0.16, 95%CI [-0.59,0.91], p=0.217). Conclusion Results suggest that the ORP WII may be more sensitive than the arousal index. Moreover, frequent insomnia symptoms were independently associated with WII in adjusted analysis, while not with the arousal index. The role of the ORP WII as a potential physiological marker of insomnia deserves further investigation. Support (if any) NIH (U01HL53916, U01HL53931, U01HL53934, U01HL53937, U01HL64360, U01HL53938, U01HL53940, U01HL53941, U01HL63463, R24HL114473, 75N92019R002).