Singer EV, Shelton AR, Malow BA. Can we predict which children with autism will have obstructive sleep apnea? J Clin Sleep Med. 2019;15(10):1389–1390.
Children with autism spectrum disorder (ASD) are known to have more sleep problems than typically developing (TD) children.1–4 Furthermore, they are at higher risk for behavioral problems during the daytime,5 for which obstructive sleep apnea (OSA) can be a contributing factor.6 Children with sleep disordered breathing have more attention problems, aggression, lower social competency, worse communication and diminished adaptive skills.7 These problems can augment the difficulties that children with ASD already face in social interaction and communication. As in the general pediatric population,8 adenotonsillectomy may improve daytime behaviors in children with ASD and OSA.9 Therefore, being able to predict which children with ASD are at risk for OSA is not only important for improving the health of children with ASD but also their daytime functioning and development in social communication.
The gold standard for the diagnosis of OSA is polysomnography (PSG), with clinical evaluation, audio/video recordings, and questionnaires being unreliable in distinguishing OSA from primary snoring in children.10 However, examining demographic features and clinical characteristics that predict OSA, including severity of OSA, in children referred for PSG may be valuable.
In this issue of Journal of Clinical Sleep Medicine, Tomkies and colleagues retrospectively examined the demographic and clinical characteristics of children, ages 2–14 years, with ASD referred for PSG and further used this data to examine predictors of OSA in this population.11 These children were referred for PSG due to suspicion for OSA. Demographic and clinical variables, including age, sex, race, body mass index (BMI), tonsil size, history of allergies, asthma, gastroesophageal reflux disease, seizure disorder, cerebral palsy, or attention deficit hyperactivity disorder did not differ between children with ASD who had PSG-documented OSA and those without OSA. However, increasing weight, but not any of the above variables, was found to be a predictor of severe OSA (apnea-hypopnea index ≥ 10). While Hispanic ethnicity and African-American race were associated with severe OSA, their risk was related to higher weight. In final models that included weight, only weight independently predicted severe OSA.
One strength of this study is the analysis of a wide variety of factors. While some of these factors, such as medications that contribute to weight gain or hypotonia, were not able to be analyzed due to the small sample, this study lays the groundwork for future research to better define predictors of OSA in this population. This article also highlights important racial and ethnic differences in sleep for patients with ASD, and more broadly, neurodevelopmental delay (NDD). Diverse racial and ethnic groups are included, with 49% of the study sample Hispanic, 27% African American, and 22% Caucasian (2% other). The African American and Hispanic participants were more likely to have severe OSA due to increased weight. This finding is consistent with the literature documenting a higher prevalence of obesity among African-American and Hispanic children.12 In the last decade there has been a “push” to reduce racial disparities in health care.13 However, to reduce racial disparities, we must first know what those racial disparities are. The literature exploring sleep disorders in children and adolescents has mostly included a small and homogenous study population (predominantly white).14–16 Research evaluating racial and ethnic disparities in children with NDD is even more limited. Race and ethnicity are important biopsychological determinants of sleep.17 Racial identification influences physiological processes that can lead to fragmented sleep.17–19 Many studies that do look at racial/ethnic differences mainly have evaluated sleep duration and insomnia symptoms and mostly used self-reported data such as parental report.20 While the authors point out that the diversity in their study is due to local demographics, to our knowledge, this is the first study showing racial difference in OSA in the NDD population. To delineate racial disparities, sleep practitioners and researchers must be intentional about including different ethnicities in our study samples. Once the disparities are recognized, the next step is to assess “the why” including genetic and environmental factors. Studies that combine objective data with patient self-report are a promising methodological way to discover inconsistencies and cross-validate.20 We can then determine the points of intervention to reduce sleep disparities and implement them successfully.
The lack of clinical or demographic predictors of OSA supports the use of PSG to identify OSA in children with ASD. Obtaining PSG in the ASD population is feasible as only 1 of 60 children failed to complete the PSG. A family-centered care approach to PSG, including psychological preparation, coping strategies for the child and family, continuous praise and reassurance for the child, and respect for the family, can facilitate success (in any child).21 Given the increasing prevalence of ASD in children (currently 1 in 59),22 attention to identifying and treating OSA is timely.
All authors have seen and approved the manuscript. Work for this study was performed at Vanderbilt University, Nashville, TN. The authors report no conflicts of interest.