Assessment of Sleep Architecture in Obese Children With Obstructive Sleep Apnea

ABSTRACT Objectives To evaluate sleep in obese children with obstructive sleep apnea (OSA) and to determine the impact of OSA severity on their sleep architecture. Methods Records of children with OSA seen in the pediatric otolaryngology clinic were reviewed to collect information on body mass index categories, severity of OSA, and sleep architecture parameters including total sleep time (TST), sleep efficiency, sleep latency, proportions of non–rapid eye movement sleep stages (N1, N2, and N3), rapid eye movement sleep (Stage R), and wake after sleep onset (WASO). Pairwise comparisons were performed using multivariable linear regression, with statistical significance defined as p < 0.05. Results Out of 461 children with OSA (260 male, 201 female), 195 (42%) were obese. Compared to healthy weight children, TST and sleep efficiency decreased in Class II [( β = −28.34, 95% CI:[−49.36, −7.32], p = 0.008), ( β = −4.25, 95% CI: [−8.15, −0.34], p = 0.033)] and Class III obesity [( β = −38.12, 95% CI: [−58.83, −17.42], p < 0.001), ( β = −5.69, 95% CI:[−9.53, −1.84], p = 0.004)]. WASO increased in Class II ( β = 9.46, 95% CI: [0.16, 18.75], p = 0.046) and Class III obesity ( β = 10.94, 95% CI: [1.79, 20.09], p = 0.019). Stage R decreased in Class I obesity ( β = −1.94, 95% CI: [−3.65, −0.22], p = 0.027) and Class III obesity ( β = −3.08, 95% CI: [−5.06, −1.09], p = 0.002). N1% increased in Class III obesity ( β = 0.85, 95% CI: [0.03, 1.67], p = 0.043). N3% increased in Class I obesity ( β = 7.27, 95% CI: [2.81, 11.72], p = 0.001). Sleep latency and N2% were not different between obesity classes. BMI category was independently associated with TST ( p = 0.004), sleep efficiency ( p = 0.049), N3% ( p = 0.016), and REM% ( p = 0.019), while OSA severity was associated with N1% ( p < 0.001). Conclusions Sleep architecture parameters differ across obesity classes in children with OSA. The relationship between weight and sleep architecture warrants further investigation in children both with and without OSA. Level of Evidence 3.