Obstructive Sleep Apnea in Pediatric Type 2 Diabetes—Prevalence, Demographics, and Screening Practices

Abstract

Background: Managing pediatric type 2 diabetes (T2D) requires provider proficiency in addressing co-morbidities. Adults with T2D are at high risk for obstructive sleep apnea (OSA) but data is limited in children. Study aims were to 1) develop a screening process to assess OSA symptom prevalence in children with T2D and 2) characterize patients with T2D and proven OSA (confirmed by polysomnography [PSG]). Methods: Michigan Sleep-Disordered Breathing questionnaire was implemented for patients age 8-18 years with T2D (2017). A score >0.33 defined positive screening for OSA symptoms, resulting in Sleep Medicine referral. Data collected included age, sex, race/ethnicity, duration of T2D, surgical and medication history, body mass index (BMI), blood pressure, lipid profile and hemoglobin A1c. Patients with positive screenings were followed prospectively to describe clinical course. The same data was collected on patients with T2D and proven OSA via retrospective chart review (2010-2016). Wilcoxon rank sum or Fisher’s exact test were used to compare 1) patients with and without OSA symptoms and 2) patients with and without proven OSA. Results: Over 10 months, 43 children with T2D completed OSA screening. Forty percent (17/43) screened positive for OSA symptoms leading to referral. Thus far, four children have completed PSG; three had OSA and 1 had primary snoring. No significant differences between youth with or without OSA symptoms were detected. Chart review of 225 children with T2D showed 42 had PSG and 29/42 (69%) had proven OSA. Only BMI and associated z-score were significantly higher when comparing children with proven OSA to those without OSA. Conclusion: Prevalence of OSA symptoms is high (40%) in children with T2D when routinely screened in a small cohort. Furthermore, the majority who complete PSG have proven OSA. There were no consistent factors that identified children with OSA symptoms; thus, screening all patients is crucial to prevent complications of delayed diagnosis. Disclosure K.A. Matlock: None. Y. Guo: None. J.C. Khoury: None. N. Gurbani: None. N.A. Crimmins: None.