ABSTRACT
BACKGROUND: The polysomnogram (PSG) is the "gold standard" for diagnosing obstructive sleep apnea (OSA). However, nocturnal oximetry is a practical screening tool for children with adenotonsillar hypertrophy (ATH). This study aimed to investigate the incidence of, and predictive factors for, OSA in children with ATH and normal / inconclusive overnight oximetry. METHODS: The prospective study enrolled children aged 3-15 years with ATH and normal / inconclusive overnight oximetry. All participants underwent full-night PSG. To evaluate the predictors of OSA, we used logistic regression analysis, including sex, history of allergic rhinitis, body mass index z-score, neck circumference-height ratio, and polysomnographic parameters (obstructive apnea-hypopnea index (OAHI), nadir oxygen saturation (SpO2), peak end-tidal CO2 , and arousal index). RESULTS: The participants were 189 children; 167 (88%) were diagnosed with OSA by PSG. A history of allergic rhinitis (P = 0.033), and the PSG findings for nadir SpO2 (P = 0.027) and arousal index (P = <0.001) predicted the diagnosis of OSA. We divided patients with OSA into two groups (mild versus moderate to severe OSA). Patients with OAHI ≥5/h were defined as having moderate-to-severe OSA. No clinical factors significantly predicted OAHI ≥5. Of the 189 participants, 58 children (31%) were diagnosed with severe OSA (OAHI ≥10). The only PSG factor that predicted severe OSA was the arousal index (P < 0.001). CONCLUSIONS: The observed incidence of OSA in children aged 3-15 years with ATH and normal/inconclusive overnight oximetry was very high. A history of allergic rhinitis may help to triage the patients. The arousal index was a predictor of pediatric OSA.
