Abstract
Introduction
The odds ratio product (ORP) is a new highly-validated electroencephalogram biomarker of sleep depth. ORP has been validated as such by several studies investigating the effect of sleep disorders, responses to sleep deprivation and traffic noise. ORP during REM sleep varies considerably among individuals. Whether ORP reflects sleep depth also in REM sleep is unknown. We hypothesized that subjects with high REM ORP are more prone to REM sleep fragmentation.
Methods
Using data from the baseline (SHHS1; N=5,537) and follow-up (SHHS2; N=2,595) visits of the Sleep Heart Health Study, we calculated and summarized ORP in 30-second intervals corresponding to manually scored sleep stage epochs. We developed a heuristic to identify REM periods, defined as sequences of REM sleep epochs separated by no more than 10 minutes of other sleep stages or wake epochs. Using general linear models adjusted by age, sex, body mass index, race and ethnicity, we evaluated the relationship between REM ORP and total REM duration, number of awakening episodes per REM period and arousal index during REM sleep.
Results
Higher REM ORP was correlated with shorter total REM duration (ρ SHHS1=-0.12; p < 0.001, ρ SHHS2=-0.07; p < 0.001), more awakening episodes (ρ SHHS1=0.26; p<0.001, ρ SHHS2=0.30; p < 0.001) and higher arousal index (ρ SHHS1=0.18; p < 0.001, ρ SHHS2=0.16; p < < 0.001) during identified REM periods. In adjusted analyses, one-unit increase in REM ORP was associated, on average, with a 7 minute decrease in total REM duration (β=-7.10; p < 0.001), 1 more awakening episode per REM period (β=1.29; p < 0.001) and an increase of 6 arousals/hour (β=6.16; p < 0.001) during REM sleep periods.
Conclusion
We found that higher REM ORP was associated with shorter REM periods, higher proportion of awake during REM periods and higher REM arousal index. Although small, these differences suggest that ORP is consistent with the concept of sleep depth also during REM sleep.
Support
None