ObjectiveTo investigate the contributions of adenoid and tonsil size to childhood obstructive sleep apnea (OSA) and the interactions between adenotonsillar hypertrophy, age, and obesity in children with OSA.MethodsIn total, 495 symptomatic patients were recruited. The patients were assigned to four groups according to age:toddler (age 1-3, n=42), preschool (age 3-6, n=164), school (age 6-12, n=200), and adolescence (age 12-18, n=89). All subjects had tonsil size graded by otolaryngologists, adenoid size determined on lateral radiographs (Fujioka method), and a full-night polysomnography. The apnea-hypopnea index (AHI), adenoid size, and tonsil size were compared in obese and non-obese children in the four age groups. Adjusted odds ratios (ORs) and 95% confidence interval (CI) of adenotonsillar hypertrophy and OSA risk were estimated by multi-logistic regression.ResultsThe AHI was positively related to tonsil grade (r=0.33, p <0.001) and adenoid size (r=0.24, p <0.01) in all patients. Tonsil grade was positively related to AHI in all four age groups. Adenoid size was positively related to AHI in the toddler, preschool, school groups, but not in the adolescent group (r=0.11, p=0.37). Tonsil grade and adenoid size were both positively related to AHI in obese and non-obese children. In the regression model, obesity (OR=2.89; 95% CI 1.47-5.68), tonsillar hypertrophy (OR=3.15; 95% CI 2.04-4.88), and adenoidal hypertrophy (OR=1.89; 95% CI 1.19-3.00) significantly increased OSA risk.ConclusionsAdenotonsillar hypertrophy and obesity are the major determinants of OSA in children. However, the influence of adenoid size decreases in adolescence.