PEDIATRICS • ORIGINAL ARTICLE Association between REM sleep and obstructive sleep apnea in obese and overweight adolescents Orna Sever 1,2 & Eric J. Kezirian 2 & Emily Gillett 1,2 & Sally L Davidson Ward 1,2 & Michael Khoo 3 & Iris A. Perez 1,2 Received: 24 August 2018 /Revised: 27 November 2018 /Accepted: 4 December 2018 # Springer Nature Switzerland AG 2018 Abstract Purpose Overweight and obese children have demonstrated reduced rapid eye movement (REM) sleep, affecting energy balance regulation and predisposition to weight gain. Obstructive sleep apnea (OSA) is a known cause of decreased REM sleep. The purpose of this study is to examine the association between the percentage of REM sleep, BMI z-score, and OSA severity in overweight and obese adolescents. Methods We performed a cross-sectional study of 92 (43% female) overweight and obese adolescents (13–17 years old) who underwent overnight polysomnography (PSG) at Children’ s Hospital Los Angeles between 2010 and 2017. Results The average Body Mass Index (BMI) z-score was 2.27 ± 0.47, with 71% having BMI z-score ≥ 2. REM% during PSG was 15.6 ± 6.8, and obstructive apnea-hypopnea index was 17.1 ± 24.3. The distribution across categories of OSA severity was 27% none (≤ 1.5 events/h), 24% mild (> 1.5–5 events/h), 8% moderate (> 5–10 events/h), and 41% severe (> 10 events/h). REM% was not associated with BMI z-score, either on univariate or multivariate regression with adjustment for age, gender, and apnea-hypopnea index (AHI). When subdivided into OSA categories, a 1-unit increase in BMI z-score was associated with a 5.96 (p = 0.03) increase in REM% in mild OSA and an 8.86 (p = 0.02) decrease in REM% in severe OSA. There was no association between BMI z-score and REM% in none and moderate OSA. Conclusion Among overweight and obese adolescents, BMI z-score was associated with decreased REM% in severe OSA and unexpectedly increased REM% in mild OSA, but there was no association in none or moderate OSA. Keywords Adolescents . Obesity . Obstructive sleep apnea . Rapid eye movement Introduction Obstructive sleep apnea (OSA) and obesity, both known to have adverse cardiovascular, metabolic, and neuropsycholog- ical consequences [1], continue to increase in epidemic pro- portions in most industrialized countries [2–4]. Previous studies have shown that short sleep duration is linked to the development of obesity and this relationship may be mediated by endocrine and metabolic changes, partic- ularly the balance of ghrelin (associated with ratings of hunger) and leptin (associated with appetite) [4–6]. More re- cently, studies show that overweight and obesity risk in chil- dren and adolescents are not only associated with short sleep duration but also with changes in sleep architecture [6, 7]. Specifically, Liu et al. [6] attributed the increased overweight risk in children and adolescents to reduced REM sleep. In a study of healthy adults, Olson et al. also demonstrated that an increased percentage of REM sleep predicted a greater over- night change in leptin and a consequent decrease in morning leptin. Further, the investigators also showed that those with highest percentage of REM had lower morning leptin levels. In their cohort of patients, the authors did not find a relation- ship between sleep duration and leptin levels. Hence, an in- creased REM is needed to maintain lower leptin levels which would predict lower appetite and decreased risk of obesity [8]. REM sleep may affect energy balance via multiple path- ways that include regulation of feeding behaviors, food pref- erences, appetite, and metabolism [9]. Gonnissen et al. [10] found that reduced REM in healthy men resulted in weight * Iris A. Perez iaperez@chla.usc.edu 1 Children’ s Hospital Los Angeles, 4650 Sunset Blvd., #83, Los Angeles, CA, USA 2 Keck School of Medicine of USC, Los Angeles, CA, USA 3 Viterbi School of Engineering, University of Southern California, Los Angeles, CA, USA Sleep and Breathing https://doi.org/10.1007/s11325-018-1768-6