A predictive model for obstructive sleep apnea and Down syndrome

Abstract

Obstructive sleep apnea (OSA) occurs frequently in people with Down syndrome (DS) with reported prevalences ranging between 55% and 97%, compared to 1-4% in the neurotypical pediatric population. Sleep studies are often uncomfortable, costly, and poorly tolerated by individuals with DS. The objective of this study was to construct a tool to identify individuals with DS unlikely to have moderate or severe sleep OSA and in whom sleep studies might offer little benefit. An observational, prospective cohort study was performed in an outpatient clinic and overnight sleep study center with 130 DS patients, ages 3-24 years. Exclusion criteria included previous adenoid and/or tonsil removal, a sleep study within the past 6 months, or being treated for apnea with continuous positive airway pressure. This study involved a physical examination/medical history, lateral cephalogram, 3D photograph, validated sleep questionnaires, an overnight polysomnogram, and urine samples. The main outcome measure was the apnea-hypopnea index. Using a Logic Learning Machine, the best model had a cross-validated negative predictive value of 73% for mild obstructive sleep apnea and 90% for moderate or severe obstructive sleep apnea; positive predictive values were 55% and 25%, respectively. The model included variables from survey questions, medication history, anthropometric measurements, vital signs, patient's age, and physical examination findings. With simple procedures that can be collected at minimal cost, the proposed model could predict which patients with DS were unlikely to have moderate to severe obstructive sleep apnea and thus may not need a diagnostic sleep study.

Keywords: Down syndrome; obstructive sleep apnea; trisomy 21.

Conflict of interest statement

CONFLICT OF INTEREST

The authors have no financial conflict of interests related to the content of this study. B.G.S. serves in a non-paid capacity on the Board of Directors for the Band of Angels Foundation, a non-profit organization, and on the Medical and Science Advisory Board for the Massachusetts Down Syndrome Congress. He is a non-paid clinical advisory to the National Center for Prenatal and Postnatal Down Syndrome Diagnoses Resources. B.G.S. occasionally gets remunerated from Down syndrome non-profit organizations for speaking engagements about Down syndrome. He receives research support from Hoffmann-La Roche, Inc. B.G.S. receives annual royalties from Woodbine House, Inc., for the publication of his book Fasten Your Seatbelt: A Crash Course on Down Syndrome for Brothers and Sisters. B.G.S. is occasionally asked to serve as an expert witness for legal cases where Down syndrome is discussed. B.G.S. has a sister with Down syndrome. E.A.M. receives research support from Biotie Therapies, Inc. and serves on Data and Safety Monitoring Boards for Acorda Therapeutics and Shire Human Genetic Therapies.

© 2017 Wiley Periodicals, Inc.

Figures

FIGURE 1

The 15 most relevant variables for predicting moderate or severe OSA. srbd03: While sleeping, does your child snore loudly?, cshqwt: wake time; cshq13: child sleeps too little; cshq30: child awakens alarmed by a frightening dream; cshq10: child struggles at bedtime (cries, refuses to stay in bed, etc.); srbd18: this child often has difficulty organizing tasks and activities; srbd13: this child often has difficulty organizing tasks and activities; cshq09: child resists going to bed at bedtime; srbd02: while sleeping, does your child always snore?; srbd04: while sleeping, does your child have heavy or loud breathing?; cshq02: while sleeping, does your child have heavy or loud breathing?; phys_hypertension_pct: hypertension percentile; cshq39: adults or siblings wake up child; cshq29: child awakens during the night screaming, sweating, and inconsolable; cshq08: child is ready to go to bed at bedtime. Variables further defined in e-Spreadsheet 1 in Supplementary Materials. Relevance is a sensitivity-weighted measure of increase in false positive rate when a given predictor is dropped from the model defined mathematically in e-Section 2 in Supplementary Materials