Abstract P388: Childhood Obstructive Sleep Apnea May Track Into Adolescence and Role of Weight Dynamics

Abstract

Introduction: Childhood and adolescent OSA share certain pathophysiological mechanisms. Thus, adolescent OSA may be in a continuum of childhood OSA. However, this hypothesis remains controversial. Hypotheses: Childhood OSA is associated with higher odds of adolescent OSA. Weight changes in this transitional period are associated with the development of OSA. Methods: We, firstly, conducted a meta-analyses to associate childhood and adolescent OSA based on literatures in which apnea hypopnea index (AHI) and/or OSA during both childhood and adolescence were reported. Major search engines were used to identify relevant reports. An AHI ≥ 1 was used as homologized definition of OSA at childhood and adolescence periods across all studies. From eligible studies, numbers of participants with none, remitted, incident, and persistent OSA from childhood to adolescence were abstracted. Crude relative risks (RRs) between childhood and adolescence OSA were computed by using Cochran-Mantel-Haenszel tests. A random-effects model, based on the DerSimonian and Laird method, was used to summarize the findings across studies. We, then, used data from the Penn State Child Cohort (PSCC) to analyze the association between weight changes from childhood to adolescent and incident/persistent OSA. Based on the changes of participants’ weight status (normal: BMI percentile < 85 vs. overweight/obese: BMI percentile ≥ 85) during childhood and adolescence, 4 mutually exclusive groups were formed and compared. Logistic regressions were used to associate weight gain (normal to overweight/obese) and weight loss (overweight/obese to normal) during this development stage and incident/persistent OSA. Sensitivity analyses were performed by using an AHI ≥ 5 to define OSA. Results: Four cohort studies, which included a total of 1473 participants, were included in the meta-analyses. The mean age during childhood and adolescence examinations were 8.5-9.8 years and 13.7-20.2 years. The pooled remission rate of childhood OSA was 52.8% (31.6%, 74.1%), while the incidence of adolescent OSA was 31.6% (15.8%, 47.4%). After weighing and summarizing RRs from individual studies, the overall RR was 1.54 (1.14, 2.08). The association was stronger [RR: 2.51 (1.65, 3.82)] in the sensitivity analysis. Based on PSCC data, we found a significant association between weight gain from childhood to adolescence and higher odds of incident OSA [OR: 2.61 (1.01, 6.81)] and a non-significant trend between weight loss and lower odds of incident OSA [OR: 0.23 (0.05, 1.06)]. Conclusion: While a substantial proportion of childhood OSA remits in the transition to adolescence, children with OSA are more likely to have adolescent OSA than those without OSA in childhood. Weight gain and loss may contribute to the natural course of OSA during transition from childhood to adolescent, especially the onset of incident OSA.