Abstract Rationale Asthma and obstructive sleep apnea (OSA) are two common chronic airway related disorders prevalent in children with obesity, each associated with significant morbidity. However, only few studies have focused on obese children with comorbid asthma and OSA. Here we aimed to evaluate whether OSA was associated with distinct characteristics in children with and without asthma. Methods As part of a prospective cohort study of the effects of obesity on asthma (GenOAAT), we recruited youth ages 6-20 years with and without asthma. Participants completed respiratory health questionnaires, spirometry, lung volumes, and biomarker assays (C-reactive protein CRP, total IgE, and peripheral blood eosinophil counts). OSA was defined as a physician/polysomnogram (PSG) diagnosis of OSA (ICD 10 code) in the electronic health record (EHR), or a Pediatric Sleep Questionnaire score ≥0.33. We evaluated characteristics associated with OSA using logistic or linear regression models, for categorical and continuous outcomes, respectively. Models were adjusted for sex and race (model 1), with the addition of obesity (model 2) or obesity and asthma (model 3). Results A total of 142 GenOAAT participants had complete data for analysis: 18 with OSA (mean age 15.0 years, 44% male) and 124 with no OSA (mean age 14.8 years, 53% male). Prevalence of asthma (78% vs 45%, p = 0.010) and obesity (89% vs 46%, p 0.001) was higher in children with OSA compared to no OSA. Adjusting for race and sex, children with OSA had 1.7 (95%CI=0.5-2.9, p = 0.006) higher percent body fat z-score and 2.5 (95%CI=1.6-3.4, p 0.001) higher basal metabolic rate z-score than children with no OSA (Figure 1, red-model 1). Adjusting for all covariates (model 3), children with OSA also had lower FEV1 (β=-1.5, 95%CI = -2.2, -0.85, p 0.001), FEV1/FVC (β=-0.99, 95%CI = -1.6, -0.4, p 0.001), and FEF2575 (β=-1.6, 95%CI = -2.3, -0.9, p 0.001) z-scores; as well as higher RV/TLC (β = 5.3%, 95%CI = 0.9-9.6%, p = 0.020), and RV %predicted (β = 30%, 95%CI = 8.7-51%, p = 0.007) than children without OSA (Figure 1, blue-model 3). Participants with OSA was associated with higher odds of elevated CRP (1 mg/dL) (OR = 7.2, 95%CI = 1.8-34.7, p = 0.008). Conclusions Children with OSA have a distinct phenotype which increases their risk for respiratory morbidity. In our analysis, they had higher prevalence of asthma and obesity and a higher inflammatory burden and caloric expenditure. Despite OSA being regarded as an upper airway disease, they also had worsened lower airway function. Future studies exploring the relationship between OSA severity in children with asthma and obesity can help identify those at higher risk for respiratory morbidity. This abstract is funded by: HL149693 (NIH)
Bandyopadhyay et al. (Fri,) studied this question.