Abstract
STUDY OBJECTIVES: Adenotonsillar hypertrophy is the major cause of obstructive sleep apnea (OSA) in prepubertal children, but children without enlarged lymphoid tissues may still suffer from OSA. This study aimed to identify other potential anatomic features associated with childhood OSA. METHODS: This prospective study took place between January 2010 and April 2014. Prepubertal children suspected to have OSA, aged 6 to 11 years, were recruited. They underwent anthropometric measurements, nocturnal polysomnography, tonsil size evaluation, x-ray cephalometry, and sonographic measurement of lateral parapharyngeal wall (LPW) thickness. Linear regression analyses were used to test for the association between anatomic measurements and OSA severity. Logistic regression analyses were used to identify potential anatomic markers for different cutoffs (obstructive apneahypopnea index (OAHI) ≥ 1 and ≥ 5 events/h) for OSA. RESULTS: Forty-seven children with OSA (20 with moderate to severe disease) and 43 children for the control group were recruited. Sonographic measurement of LPW thickness and position of hyoid bone taken from x-ray cephalometry were risk factors associated with OSA. Linear regression analyses found that these two phenotypes were associated with OAHI. Multivariate models adjusted for age, sex, body mass index, z score, and tonsil size revealed that lower position of hyoid bone was independently associated with higher risk for OSA, whereas both lower position of hyoid bone and greater LPW thickness were associated with higher OAHI and also a higher risk for moderate to severe OSA. CONCLUSIONS: Position of hyoid bone and LPW thickness are anatomical markers of childhood OSA independent of obesity and tonsil size. Screening tools may include cephalometry and sonographic measurement of LPW to allow better delineation of OSA risk.