0499 Sleep and Neurocognition in Adolescents with ADHD: A Polysomnographic Study

Abstract

Abstract Introduction Sleep disturbances are common among adolescents with ADHD; however, few studies have characterized the nature of ADHD-related sleep problems using the gold-standard sleep measure, polysomnography (PSG), in adolescence. Additionally, although similar cognitive deficits are common across ADHD and sleep-disordered populations, the potential role of sleep in contributing to cognitive impairment in adolescent ADHD is unknown. This study investigates differences in PSG-measured sleep among adolescents with ADHD versus healthy controls without psychiatric disorders (HC) and associations with cognition. Methods Sixty-two adolescents aged 13-17 (31 ADHD, mean age=15.3, 50% female) completed a psychiatric evaluation and 3 nights of ambulatory PSG. Following the third night, participants completed the Cambridge Neuropsychological Test Automated Battery (CANTAB). Sleep variables were averaged over 3 nights. Linear regressions controlling for age and sex examined group differences in a range of traditional PSG and spectral EEG indices as well as relationships between PSG/spectral indices and cognition (two summary scores derived from CANTAB: response accuracy and response time) within the ADHD group. Results Adolescents with ADHD displayed reduced SWS% (F(3,51)=9.67, p=.003), increased N2% (F(3,51)=10.35, p=.002), increased relative sigma (F(3,47)=6.55, p=.01) and beta (F(3,47)=4.10, p<.05) power, and a trend toward reduced relative delta power (F(3,47)=2.95, p=.09) compared to HC. Within the ADHD group, greater REM% (r=.43), reduced N2% (r=-.55), greater relative delta power (r=.52), higher delta power peak (r=.56), steeper delta decline overnight (r=-.56), and reduced relative theta (r=-.53), beta (r=-.74), and gamma (r=-.67) power were associated with better response accuracy (p’s<.05). Greater relative delta (r=-.51), and reduced relative theta (r=.55) and beta (r=.63), power were associated with faster response times. Conclusion Although adolescents with ADHD did not differ from HC on traditional PSG measures (TST, WASO), they exhibited abnormalities in sleep stage distribution and non-REM EEG frequency spectral indices, including reduced SWS and low frequency power and increased stage 2 sleep and high frequency power overnight. Notably, similar parameters were associated with impaired cognition, suggesting sleep may contribute to cognitive deficits in ADHD. Future studies may clarify whether sleep plays a causal role in cognitive impairments in adolescent ADHD and if sleep treatments result in improved cognition in this population. Support (If Any) K23MH108704