Down syndrome (DS) is the most common genetic cause of intellectual disability and results from an extra chromosome 21 (Trisomy 21). Sleep issues and/or obstructive sleep apnea (OSA) are assumed to be part of the DS phenotype with a high prevalence but are often under recognized. This cross-sectional study of children with DS examines the caregiver reported sleep behaviors of 108 children with DS, ranging in age from 1.50 to 13.40 years (mean = 5.18 years) utilizing a standardized assessment tool, the Children's Sleep Habit Questionnaire (CSHQ). The CSHQ revealed 76% of children with DS had sleep problems, which began at a young age, and continue to persist and may recur with increasing age. Furthermore, children with DS who undergone adenoid and tonsillectomy for OSA continued to have sleep problems suggesting that ongoing monitoring of sleep issues is needed in this population. Implications of sleep problems and recommended anticipatory guidance and intervention are discussed.
INTRODUCTION
Down syndrome (DS, OMIM #190685) is the most common genetic cause of intellectual disability, with an incidence of 1 in 691 live births [Parker et al., 2010]. In general, sleep problems occur more frequently in children with intellectual disability compared to typically developing population [Didden et al., 2002; Harvey & Kennedy, 2002; Krakowiak et al., 2008]. In children with intellectual disability, sleep issues commonly noted are disturbances initiating and maintaining sleep, night awaking, excessive daytime sleepiness, and sleep disordered breathing. In addition to the intellectual disability, children with DS have craniofacial findings, including midfacial hypoplasia, narrow nasopharynx, micrognathia, small larynx, and hypotonia leading to floppiness of the upper airways, that increase their risk of having sleep disordered breathing issues, particularly obstructive sleep apnea (OSA) [Shott, 2006]. The prevalence of OSA is seen in only 1% to 4% of the general pediatric population, but increases to 30–63% among individuals with DS [Stebbens et al., 1991; Rosen et al., 2011].
Sleep is essential to the development in children and the parent-child dynamic. There is a link between sleep problems and/or OSA to health complications (e.g., obesity), parental stress, and neurocognitive, behavioral, and learning deficits in the typically developing population [Miano et al., 2011; Witmans et al., 2011]. These issues may become more pronounced for children with developmental and cognitive impairment such as DS. Yet, sleep issues are often under recognized by physicians and parents because they underestimate these problems and/or assume that they are part of the DS features. For instance, approximately 69% of parents reported that their child with DS did not have sleep issues, but 57% had an abnormal polysomnogram [Shott et al 2006]. For these reasons, the American Academy of Pediatrics Health Supervision for Children with DS currently recommends that physicians discuss symptoms of OSA that could be associated with poor sleep at every well-child visit, and all children with DS receive a polysomnogram by the age of 4 years [Bull, 2011].
Despite the high prevalence of sleep issues and/or OSA in children with DS, there have been limited studies examining the pattern of sleep problems in young children with DS. Furthermore, the extent of sleep problems and how they change with age have not been appropriately studied in children with DS due to small sample sizes, lack of standardized measures to evaluate sleep, and inadequate follow-up for children [Shott et al., 2006; Rosen, 2011; Rosen et al., 2011]. Our study examines the sleep profiles, across age groups as well as the impact of previous adenoid and tonsillectomy (AT) on sleep patterns in children with DS. Additionally, we also compare sleep problems in our DS cohort to a cohort of typically developing US children [Liu et al., 2005].
MATERIALS AND METHODS
Subjects
Subjects included 108 children with DS (52 boys, 56 girls) ranging in age from 1.50 to 13.40 years, with a mean age of 5.18 years (standard deviation [SD] = 2.47). Subjects consisted of 65% Caucasian, 25% African American, 6% Hispanic, and 4% Asian. No sleep medications were reported. The comparable cohort of typically developing US children is from the study by Liu et. al [2005].
Procedures
The study was a retrospective chart review of children with DS who were evaluated at the DS Clinic at Emory University from 2013–2014. As part of the clinic visit, parents typically complete the Children's Sleep Habit Questionnaire (CSHQ) [Owens et. al., 2000]. The abbreviated CSHQ consists of 33 questions that has been validated and widely used to assess sleep in children. The caregivers were instructed to reflect on the past week of sleep when completing the questionnaire. Eight subscales can be derived from the items: Bedtime Resistance, Sleep Onset Delay, Sleep Duration, Sleep Anxiety, Night Wakings, Parasomnias, Sleep Disordered Breathing, and Daytime Sleepiness. The internal consistency for each of the subscales was high with all α coeifficients above 0.98.
Each of the 33 items on the questionnaire were rated on a 3 point Likert scale: If the behavior occurred 0–1 nights per week, “rarely” would be marked and a score of (1) would be given for that behavior. Similarly, if the behavior described occurred 2–4 nights per week, “sometimes” would be marked and a score of (2) would be given, and if the behavior occurred 5–7 nights per week, “usually” would be marked and a score of (3) would be given. The scores for each of the 33 questions would be summed to find the total CSHQ score with the higher scores representing more sleep problems. A cut off score of 41 yields the best diagnostic confidence for identifying clinically significant sleep problems. In addition, we considered any item on the scale clinically significant if greater than 20% of responders marked a (2) or (3).
Statistics
Data from the demographic information were analyzed using descriptive statistics. For the different variables as indicated in Tables 1 and and2,2, descriptive statistics, including mean, range, and standard deviations of each continuous variables, were computed using SAS, version 9.1 software (Cary, NC), allowing for inspection and direct comparison. Pearson chi-square test was used to compare sample characteristics.
RESULTS
Sleep Problems and Impact of adenoid and tonsillectomy (AT)
A total of 108 CSHQ questionnaires were completed by parents. With a clinically significant cut off score of 41, the average CSHQ score was 50, with a range of 34–73. Clinically significant sleep problems were noted in 76% of our cohort including restless sleep and constant movement during sleep.
In our sample, 53 subjects did not have a diagnosis of OSA (DS-OSA) and 55 subjects had undergone previous AT for OSA (DS+OSA). The diagnosis of OSA was confirmed by the polysomnogram, and parents completed the CSHQ after the AT for the DS+OSA group. The average CSHQ score for the DS-OSA (mean age 4.41 years) was 49 with a range of 34–73 compared to an average score of 50 with a range of 38–73 for the DS +OSA group (mean age 5.94 years).
By using 41 as the cut off score for identifying children with clinically significant sleep problems, 78% of the DS+OSA group and 74% of the DS-OSA group had clinically significant sleep problems. As indicated in Table 1, 29 items were found to be significant (>20% of caregivers marking (2) or (3)) in the DS+OSA group, and 23 items were found to be significant in the DS-OSA group. By using the Pearson chi-square analysis,”falling asleep while watching television” was significantly different between the two cohorts with a p value of .005; only 15% of the DS+OSA group caregivers considered this a significant problem, while 35% of the DS-OSA group considered this a significant problem.
Age progression Analysis in Children with DS
The 108 participants were divided by age into 3 groups: ages 1–3 (n=32), ages 4–7 (n=58), and ages 8–13 (n=18). Pearson chi-sqaure analysis identified that with increasing age, children with DS begin to fall asleep in their own bed, p=.026, less frequently awakening, screaming/sweating, p=.024, and less frequently fall asleep when riding in a car p=.043, but become more afraid to sleep in the dark p=.013 (Table 2).
Additionally, there were 5 items that improved with increasing age. For example, children had less trouble sleeping away, p= .047 (Age group 1 to 2), less frequently awakened screaming and sweating, p = .006 (Age group 1 to 2), less frequently fell asleep while riding in a car, p=.018 (Age group 1 to 3) and p=.022 (Age group 2 to 3), and less frequently had trouble falling asleep in their own bed, p=.012 (Age group 2 to 3). Although the problem of bed wetting did improve, p=.038 (Age group 1 to 2) and p= .018 (Age group 1 to 3), the highest age range of 8+ still had bed wetting listed as a significant problem in 28% of the population which still represents a significant sleep problem.
Some items that do not significantly change between age groups included: “needs parent in room to sleep”, “moves to other's bed in the night”, “awakes one or more than once during the night”, “seems tired during the day”, and “others wake the child” (Table 2). Notably, items related to OSA including, “snoring”, “snorts and gasps”, and “restess” remain to be a persistent sleep problems as reported by the caregiver.
Sleep problems in a typical pediatric population compared to DS population
Liu et. al. [2005] used the CSHQ to compare sleep patterns in school-age children in China (N=292) and the United States (N=415). Only 13 significant items were found in the United States population, while 26 items were found to be significant in our DS participants (N=108). When we compared the 13 significant items in the typical pediatric population f