PREDICTING DAYTIME SLEEPINESS FROM NIGHTTIME SLEEP IN VETERANS WITH AND WITHOUT PTSD

Disturbed sleep characterizes posttraumatic stress disorder (PTSD), and may cause daytime sleepiness and compromise readiness and performance. The extent to which nocturnal sleep predicts daytime sleepiness (DS) in military personnel with and without PTSD is unknown. Here, we investigated nighttime sleep predictors of objective and subjective DS in Veterans with and without PTSD. 37 post-9/11 Veterans with PTSD (PTSD) and 47 without PTSD (Control) completed a 48-hour lab stay. Subjective and objective DS were assessed five times daily with a mood questionnaire that measured global vigor (Vigor) and Multiple Sleep Latency Tests (MSLT) to assess sleep latency (SL). Nighttime polysomnography was conducted using high-density (64-channels) electroencephalography. Sleep parameters included natural log-transformed sleep efficiency (SE), percent time in non-REM and REM sleep, and the average global log-transformed absolute power for non-REM and REM in slow oscillation (0.5-1Hz), delta (1-4Hz), theta (4-8Hz), alpha (8-12Hz), sigma (12-16Hz) and beta (16-32Hz) activity bands. To limit adaptation effects, analyses were limited to day 2. Independent samples t-tests were used to identify group differences in DS and nighttime sleep. Linear regressions were performed to identify nighttime sleep predictors of DS. Significant predictors were mean centered and entered into regressions with group and corresponding interaction effects to explore whether nighttime sleep moderated relationships between group and DS. The PTSD group reported lower Vigor (mean±sd, 72.17±16.33) than the Control group (85.49±8.90; t=4.47, p < .001). Objective SL did not differ between groups. Only SE differed between groups (t=-2.18, p = .03) and was slightly lower in PTSD. REM beta activity predicted Vigor, independent of group (β = -0.36, p = .02). No significant interaction between SE and REM beta was observed for Vigor or SL. Subjective sleepiness was higher in PTSD than Control subjects, but this was not reflected in objective sleepiness. REM beta activity predicted subjective, but not objective sleepiness in both groups of Veterans. How objective and subjective DS and REM sleep measures relate to cognitive performance remains to be investigated. USAMRMC MOMRP PT-130572(PI: Reifman).