The impact of obstructive sleep apnea screening guidelines in a population-based, midwestern cohort of children with Down Syndrome

Abstract

Objective/Background:

The high rate of obstructive sleep apnea (OSA) in Down Syndrome (DS) is well described in the literature. The impact of the 2011 screening guidelines has not been fully evaluated. The objective of this study is to evaluate the impact of the 2011 screening guidelines on the diagnosis and treatment of obstructive sleep apnea (OSA) in a community cohort of children with Down Syndrome.

Patients/Methods:

This is a retrospective, observational study conducted on 85 individuals with DS born between 1995 and 2011 in a nine-county region of southeast Minnesota. The Rochester Epidemiological Project (REP) Database was used to identify these individuals.

Results/Conclusions:

Sixty-four percent of the patients with DS had OSA. Post guideline publication, the median age at OSA diagnosis was higher (5.9 years; p=0.003) and polysomnography (PSG) was used more often to establish the diagnosis. Most children underwent first line therapy with adenotonsillectomy. There was a high degree of residual OSA after surgery (65%). There were trends post guideline publication towards increased PSG use and for consideration of additional therapy beyond adenotonsillectomy. The use of PSG before and after first line treatment for OSA in children with DS is needed due to the high rate of residual OSA. Unexpectedly, in our study, the age at OSA diagnosis was higher after guideline publication. Continued assessment of clinical impact and refinement of these guidelines will be of benefit to individuals with DS given the prevalence and longitudinal nature of OSA in this population.

Introduction

The well documented increased risk of obstructive sleep apnea (OSA) in children with Down Syndrome (DS)^1–3 has led to the development of screening guidelines for identification and treatment of OSA. In 2001, the American Academy of Pediatrics (AAP) published screening guidelines for children with DS which recommended discussing OSA symptoms (snoring, restless sleep, sleep positions) and referring to a specialist as indicated ⁴. The guidelines published in 2011 recommend assessing for signs of sleep apnea at health maintenance visits and referral to a pediatric sleep specialist for polysomnogram by age 4 years ⁵. These guidelines were updated in 2022 to recommend a polysomnogram between 3 and 4 years of age⁶; after age 4 years, periodic assessment of OSA symptoms is recommended.

In addition, there are guidelines for diagnosis and treatment of OSA that include a PSG prior to adenotonsillectomy (AT) put forth by the American Academy of Sleep Medicine (AASM) and the American Academy of Otolaryngology-Head and Neck Surgery (AAO-HNS) (published in 2011^7,8, updated in 2019⁹). The AASM guidelines recommend repeat PSG after AT to evaluate for residual OSA.

The impact of these guidelines on referral and treatment patterns is not well documented. Giménez and colleagues surveyed caregivers of children with DS and found that 58.7% were aware of the AAP screening guidelines and the median age at OSA diagnosis was lower after the guideline publication⁹. Esbensen and colleagues looked at practice patterns and sleep evaluations/interventions in children with DS and found that despite high rates of reported sleep problems, less than half of those children underwent PSG¹⁰. Knollman and colleagues compared the percentage and age of children with DS who underwent PSG before and after the publication of the 2011 AAP screening guidelines and found no difference in the percentage receiving PSG pre vs post guideline publication but children in the post guideline cohort were more likely to have an initial PSG by 4 years of age¹¹. Literature supports assessment of guidelines as well as evaluating barriers for implementation as full guideline adoption can take significant time¹²

The goal of our study was to assess the impact of the current guidelines on a midwestern cohort of pediatric DS patients. We conducted a retrospective, population-based observational study to document referral patterns and use of PSG in the diagnosis and treatment of OSA in children with DS before and after guideline publication. We hypothesized that post-guideline publication OSA diagnosed by PSG would increase and age at diagnosis would decrease.

Material and Methods

The Rochester Epidemiology Project (REP) database was used to obtain our study population. This is a medical linkage system that links medical records of residents of southeast Minnesota and Western Wisconsin¹³. Participants in the REP database have consented to research. Our study was approved by our institution’s IRB (20–009698).

We identified patients with DS who were born between 1/1/1995 and 12/31/2011 to mothers who were residents of a 9-county region of southeast Minnesota (Dodge, Fillmore, Freeborn, Goodhue, Mower, Olmsted, Steele, Wabasha and Waseca counties). Children with a diagnosis of DS documented in their medical record were included. Genetic testing was not verified. Inclusion criteria involved an evaluation by a licensed provider in a Mayo Clinic facility (to allow for internal chart review) and having consented to research participation.

Data collection was performed by author NP. Each patient record was manually reviewed through 7/1/2020 to confirm the diagnosis of DS, document sleep disorders and subsequent treatments. Consultations, operative reports, Primary Care and Sleep Medicine notes were reviewed. The diagnosis of OSA was determined either by review of the physician interpretation of the PSG report (“PSG confirmed OSA”) or by review of the Otolaryngology clinic and/or operative notes indicating that the AT indication was symptoms of OSA (“clinically suspected OSA”).

Statistical Analysis

Categorical data was summarized as total (%), while continuous variables were summarized as mean ± standard deviation if normally distributed, or median (interquartile range) if non-normally distributed. A chi-square test was used to evaluate whether mode of diagnoses (PSG confirmed or clinically suspected), or ultimate therapy for OSA was independent from the publication of the guidelines. We compared age at diagnosis before and after the publication of the guidelines, and how the age at diagnoses differed across birth years (by quartiles) using a Mann-Whitney test. We regarded p<0.05 as statistically significant (Wizard 2, version 2.0.4, Karelia Software).

Results

Ninety-four patients met the inclusion criteria. Eight patients were excluded due to death prior to age one year. One patient was excluded due to a confidential chart. The final study population consisted of 85 patients.

Demographics of this cohort are displayed in Table 1. OSA was identified in 64% of the study population. Pre-guideline publication, the median age at diagnosis was 3.3 years and post- guideline publication, the median age at diagnosis was 5.9 years (p = 0.003). There was a trend towards increasing referrals to Sleep Medicine post-guideline publication. Eleven children with DS were evaluated before 2011 and 26 children were evaluated after 2011 in the Sleep Medicine Clinic.

Table 1.

Demographic Data of the Community Cohort with Down Syndrome.

	Number	% of total (N=85)
Sex Male Female	35  50	41  59
Race White Black Asian Other Unknown/undisclosed	67  2  1  9  6	79  2  1  11  7

Diagnosis was further divided between clinically suspected OSA and PSG proven OSA. Clinically suspected OSA included those individuals who underwent AT for symptoms of sleep disordered breathing without a formal PSG. The diagnosis of OSA was more often established by PSG than by clinical suspicion post-guideline publication (65% vs. 35%, p=0.035). In the patients who were treated surgically with AT, there was no significant difference in the proportion of patients undergoing PSG prior to surgery (34% pre-guidelines, 36% post- guidelines, p=0.928). Eighty-nine percent of patients with OSA underwent first line treatment with AT. Sixty-five percent of those patients who had a post-operative PSG continued to have clinically significant OSA requiring further treatment with anti-inflammatory therapy, continuous positive airway pressure (CPAP) or revision adenoidectomy. These patients likely had persistent OSA symptoms, influencing the decision for post operative PSG.

Those receiving an OSA diagnosis via initial PSG were more likely to receive non-surgical treatment, though this difference was not statistically significant (41% vs 24%, p=0.201). Forty-eight of 54 patients underwent surgical treatment with AT. Following guideline publication many patients did not undergo PSG post operatively (46% with follow-up PSG vs 54% without). Seventeen (65%) of the 26 patients who underwent post-surgical PSG had clinically significant OSA requiring further treatment. Five patients were treated with anti-inflammatory therapy, nine were treated with CPAP and three were treated with revision adenoidectomy.

“Final OSA therapy” referred to the final treatment for OSA and included AT as well as treatments post AT due to residual OSA (anti-inflammatory therapy with intranasal fluticasone and/or montelukast, CPAP, tracheostomy). Some patients did not have a PSG after AT. Regarding final OSA therapy pre and post guidelines, there were some small changes, most notably an increase in AT with subsequent post operative PSG (Figure 1). Changes in the individual therapies were not statistically significant, likely due to the small sample size (p=0.307).

Figure 1.

Final OSA Therapy Pre and Post Guideline Publication

Discussion

A strength of our study is the population-based design. The results did not support our hypothesis that the diagnosis of OSA in children with DS would occur at a younger age following guideline publication. Instead, we found the median age at OSA diagnosis was 5.9 years after versus 3.3 years prior to guideline publication (p = 0.003).

This lack of an earlier diagnosis highlights the importance of assessing for barriers of implementation and including key stakeholders in the formation of guidelines. The referral of children with DS for diagnosis and treatment of OSA requires the identification by the primary care provider/developmental pediatrician and a multidisciplinary approach to diagnosis and treatment involving Pediatric Sleep Medicine and Pediatric Otolaryngology. Additional subspecialists are also frequently involved in the care of DS patients and access the guidelines. Access to Pediatric Sleep Medicine is a concern for many geographic areas but did not apply to our study cohort’s geographic area. Pediatric PSG is time intensive, expensive and can be difficult to tolerate for children and caregivers. There is increasing literature on the difficulties preparing children, especially those with special needs, for PSG set-up^14–16and the recommendation for PSG in a DS patient may be met with caregiver resistance. We know from other work we have done that caregivers of children with special needs can be active in parent groups and this may influence their perception of the PSG experience¹⁴. Parts of the guidelines rely on symptomatic assessment and when adherence to general guidelines was evaluated, it was found that screening based on symptomatic assessment was less likely to have optimal adherence¹⁷. Studies have shown poor correlation between parental reports of OSA symptoms and PSG findings, therefore symptomatic screening guidelines may have a disadvantage in the DS population ¹⁸.

Post- guideline practice included a trend toward increased referral to Sleep Medicine and an increase in adherence to the recommendation for a PSG prior to AT in DS. Treatment options may have been impacted as there was a trend for nonsurgical treatment post-guideline publication. Despite the high percentage of DS patients with residual OSA after AT, only 45.5% of our study population underwent PSG after AT.

A limitation of our study is the age range of the cohort studied. Our cohort was born between 1995 and 2011. The AAP guidelines were published in 2011, so our patients were 1–16 years of age at the time of the guideline publication. Only those patients born between 2007 and 2011 would have been 4 years of age or younger at the time of guideline publication. This is a small percentage of our total patients that presented at the recommended age for a screening PSG. Despite this design limitation, our patient population reflects the broader age range of children with Down Syndrome presenting to medical attention in the setting of recently published screening guidelines and reflects the real-world challenges of how to address the needs of children presenting later than ideal for specialized care.

Additional limitations of our study include the retrospective design, small sample size and predominantly White ethnicity.

Conclusions

Barriers remain in implementing OSA screening guidelines in DS. Identification and treatment of OSA at a young age can significantly impact sleep quality, daytime functioning, and prevent future comorbidities. Identification of barriers encountered by physicians and caregivers, multidisciplinary practice improvements and continued guideline refinement are key to obtaining optimal benefit from the guidelines and will benefit patients with Down Syndrome.