IDENTIFICATION OF PHYSICAL EXAM FINDINGS WITH HIGH PREDICTIVE VALUE FOR MODERATE TO SEVERE PEDIATRIC OBSTRUCTIVE SLEEP APNEA(OSA) IN OVERWEIGHT/OBESE CHILDREN

Abstract Introduction Obstructive sleep apnea(OSA) is highly prevalent and under-diagnosed in the overweight/obese pediatric population largely due to limitations of existing pediatric OSA screening instruments including lack of efficiency and practical implementation and lack of careful consideration of physical examination(PE) findings with high predictive value for OSA. We sought to identify PE finding(s) predictive of pediatric OSA in overweight/obese patients to inform development of an OSA screening tool. Methods Overweight/obese patients presenting to the Cleveland Clinic weight-management clinic between 2013-2018 with polysomnogram (PSG) data were included. The association of PE predictors: age, sex, race (white, black, other), neck (NC), waist circumference (WC), tonsil size (TS), height, systolic and diastolic blood pressure (BP) percentiles) in relation to OSA defined by apnea-hypopnea index (AHI)≥5,i.e. clinically significant pediatric OSA, were assessed using univariate and multivariate logistic regression models (OR,95%CI). Results Retrospective analysis of 180 overweight/obese patients (BMI percentile>85th for age and sex) and age 12.5±3.7 years were included. The multivariate model showed that only WC was significantly associated (1.03, 1.00 - 1.07, p=0.038) with OSA defined as AHI≥5. A statistically significant interaction of age and sex was observed such that the likelihood of OSA increased in males with older age and conversely decreased in females with older age. (1.26,1.04 -1.52, p=0.038) The reduced multivariate model, which included age, sex, WC, and age*sex interaction term, correctly discriminated AHI <5 vs. ≥ 5 66.5% of the time. Conclusion In this large clinic-based overweight/obese pediatric sample, males, older age and WC were significant predictors of OSA and TS was not. A significant interaction of age and sex was observed supporting increased OSA with increasing age in males. Data generated supports value of PE findings of age, sex and WC to incorporate in development of an OSA screening tool for overweight/obese children. Support