Hyperactivity

Hyperactivity is frequently reported in both children with habitual snoring,14,15,18,21,33,36–41 as well as those in whom SDB was formally diagnosed by polysomnography (PSG).^{17,20,35,42–47} Despite differences in definition of snoring or PSG-confirmed SDB, many studies support the relationship between snoring/SDB and hyperactive behaviors even when hyperactivity is measured with a range of parent-report tools, including the Conners’ Parent Rating Scales,^15,42,43,45 the Child Behavior Checklist,^{20,35,42,44,47} or the Behavioral Assessment Scale for Children.^23,46 In a survey of over 800 families using validated instruments,¹⁵ symptoms of SDB were associated with hyperactive behaviors with a trend toward a dose–response relationship between reported snoring frequency and behavior. Only a small number of studies have failed to find associations with SDB and hyperactive behaviors.^48–50

Inattention

Attention, which is a prerequisite to optimal learning, is a critical behavior arising from brain mechanisms and can be categorized as sustained, selective, and divided attention, thus representing a cluster of variables, each of which contributes to learning and memory. Inattentive behaviors identified by parental report have been observed in children with habitual snoring15,21,37,51 and PSG-defined SDB^{20,23,46–49} although this finding is not as robust as the associations with hyperactivity. Different categories of attention, for example, selective and sustained attention, can also be measured using objective assessments such as auditory or visual continuous performance tests (CPT) and therefore may provide more robust assessment than parental report. Such studies have shown that even children with mild SDB exhibit some deficits in attention compared to controls.^52–54

A small study of Australian children found that both selective and sustained attention measured objectively using the auditory CPT were found to be impaired in children with habitual snoring compared to controls.⁵⁵ Similarly, in New Zealand, Galland et al.⁵⁶ found that in comparison to a normal population, children with objectively confirmed SDB, compared to those without, had significantly higher scores on a visual CPT for inattention and impulsivity albeit within the average range of a normal non-clinical score.

Impaired auditory and visual attention has also been reported in children with objectively confirmed SDB compared to standardized norms.⁵⁷ Recently, event-related potential (ERP) recordings using a high-density array during an oddball attention task have shown objective evidence of impaired attention in children with SDB.⁵⁸ Since ERP patterns strongly correlate with learning, reading and school performance, the authors postulated that their findings suggest that brain changes associated with pediatric SDB have the potential to be used to determine which children might require earlier diagnosis or treatment. Nonetheless, some studies fail to observe differences in visual attention.^46,59 Emancipator et al.⁶⁰ proposed that the CPT might either not be sufficiently sensitive in children who are not obviously sleepy or possibly, with the increase in time children now spend playing video games, such CPT tools might be less discriminating.

Aggressive Behaviors

In addition to hyperactivity and inattention, aggressive and bullying behaviors are beginning to receive more attention in the SDB literature. Estimates suggest that up to 25% of children in elementary schools are affected,⁶¹ with a higher prevalence in boys.⁶² Aggressive behaviors present a major challenge not only for schools, which often have local, state, and national programs to address this issue, but also for society, as aggressive children are at high risk for future psychiatric symptoms, violence, substance abuse, and criminality,⁶³ while the victims of bullying also suffer. Clearly, the causes of aggressive behaviors are complex and include social, biological, and cultural factors; however, there is now emerging evidence that sleep problems might play a role.

In a large population-based study of over 3000 5-year-old children, those with symptoms of SDB were twice as likely to have parentally reported aggressive behaviors,³³ which is similar to a study that also adjusted for comorbid hyperactivity and stimulant use.⁶⁴ A recent report from our group, the first specifically designed to query parents of aggressive elementary school children as well as non-aggressive controls about symptoms of SDB, found that aggressive children were twice as likely to have SDB symptoms compared to non-aggressive children.²² Notably, daytime sleepiness rather than snoring appeared to drive the relationship with aggressive behavior, which suggests that other causes of daytime sleepiness, such as poor sleep hygiene, are also important.

Of note, short sleep duration – which perhaps might partially explain sleepiness – and sleep difficulties have been found to be associated with aggressive behaviors in young children65,66 and suicidal ideation in adolescents.⁶⁷ In children with objectively confirmed SDB, aggressive behaviors are more frequent than in children without SDB⁴⁸ even when SDB is mild.²⁰ Children with aggressive behaviors also have EEG slowing during wakefulness,⁶⁸ which might reflect deficient levels of arousal or excessive daytime sleepiness, likely mediated via the prefrontal cortex.⁶⁹

Parent versus Teacher Report of Behavior

There are conflicting findings on the association between SDB and behavior depending on whether behavioral reports are provided by parents or teachers. The literature regarding teacher reports is small compared to parental reports. Some studies report elevated hyperactive behaviors on both parent and teacher scales,14,17,46 while others found only elevated hyperactive behaviors on parent scales.^23,37 Our own data have shown that teacher-reported bullying behaviors did not show associations with symptoms of SDB but parent reports did.²²

Recently, Kohler et al.⁷⁰ performed a direct comparison of parent and teacher reports in children with SDB. They found that both parents and teachers report more problematic behavior, which is predominantly internalizing such as anxious and withdrawn behavior, somatic complaints, and social and affective problems. In addition, parents reported a greater severity and range of behaviors. Overall, the concordance for individual children was poor. The limited number of studies that have collected teacher reports, the different tools used, and the sample sizes involved make it difficult to reach conclusions regarding classroom behavior. However, despite the inconsistencies, the teacher-report studies published to date appear to support a role for SDB in at least some areas of behavioral regulation.

SDB and ADHD

Attention deficit hyperactivity disorder (ADHD) is the most commonly diagnosed pediatric mental health disorder in North America, and sleep problems are one of the most frequently reported comorbidities in these children. The major features of ADHD (e.g., inattention, hyperactivity, and impulsivity) are also frequent manifestations of childhood SDB and, conversely, comorbid sleep problems are highly prevalent in ADHD. Therefore it is unsurprising that the relationship between SDB and ADHD is of great interest. Multiple studies have shown that children with ADHD demonstrate a number of parentally reported sleep problems,71–74 with a frequency up to five times greater than that of otherwise healthy children.⁷⁵ Children with ADHD have been reported to snore more than their peers,^76,77 with some studies suggesting that snoring is more common in those with the hyperactive/impulsive subtype of ADHD.⁷⁸ However, polysomnographic data are less clear in terms of an association between SDB and ADHD.⁷⁹

Methodological issues may be at least in part related to such inconsistencies, particularly since the majority of studies did not use criteria from the diagnostic and statistical manual of mental disorders (DSM) for ADHD but instead relied on parental report of hyperactivity symptoms. In one study of school-aged children that did use formal criteria, diagnoses of ADHD were found in almost a third of children.⁵³ However, a recent meta-analysis,²⁵ which included studies utilizing rigorous criteria for ADHD, suggested that the apnea–hypopnea index (AHI) in the three objective studies retained in the meta-analysis^80–82 were not very elevated (1.0, 5.8, and 3.57, respectively). Nonetheless, using a pediatric AHI threshold of 1,⁸³ these values suggest that SDB may indeed be more frequent in children with ADHD compared to controls. Interestingly, children with ADHD and an AHI between 1 and 5 have been reported to improve more following adenotonsillectomy than after stimulant treatment.^26,84 This raises questions about appropriate screening and intervention in these children. A recent working group report has suggested key areas for future research in this field.⁸⁵

Intelligence

The intelligence quotient (IQ) is often reported in studies of SDB although findings are not consistent. Lower IQ scores have been reported in children with SDB compared to controls, although these scores are typically still within the normal range.19,21,49,50,55,59,92–96 One study in children awaiting adenotonsillectomy found that compared to healthy non-snoring children, the snoring children had a 10-point reduction in IQ.⁹⁵ Of course, the clinical significance of this remains to be shown for a high-functioning child, but a 10-point IQ difference could be rather significant in children performing at a lower level. Several studies fail to support findings of differences in full-scale IQ,^44,91,97,98 although some have found lower scores for verbal IQ (language skills) in children with SDB.^44,97

Lack of robust findings between SDB and IQ is perhaps not surprising given that measurement of IQ is complex and, in essence, measures performance across several tasks rather than a focus on a particular area of cognition. Standardized vocabulary tests, as a proxy measure of IQ and an excellent predictor of cognition and academic success,⁹⁹ have demonstrated that the difference in scores between children with and without SDB may be equivalent to the impact of lead exposure.¹⁰⁰ These findings clearly have great clinical significance for a child’s future if indeed they are supported by additional studies. Nonetheless, there are many other factors which clearly impact a child’s IQ and which require consideration in studies of SDB, including genetics, parental education level, as well as biological and environmental factors.^23,101,102

Memory

Multiple studies have failed to find evidence for memory impairment in children with SDB.21,49,103 In those that have reported memory deficits^50,55,104 only one has reported a dose–response effect.¹⁰⁵ Inconsistencies in memory findings are likely related to the type of memory measured (such as verbal memory or working memory). In addition, many reports provide only a cumulative memory score rather than address specific processes involved in memory acquisition.

A recent study of memory recall in children with and without SDB¹⁰⁴ found that memory recall to a picture was impaired in children at both an immediate memory assessment as well as a follow-up assessment the following day. Notably, the children with SDB also demonstrated declines in recall performance, which suggests that children with SDB require more time and additional learning opportunities to reach immediate and longer-term recall performance and that these children may have slower information processing and/or secondary memory problems perhaps due to inefficient encoding.¹⁰⁶ Indeed, changes in basic perceptual processes that underlie higher-order functions have been shown to be impaired in pediatric SDB even when performance is not altered on standard measures of memory.¹⁰⁷

Academic Performance

Good academic performance is often essential for future career success and many studies have reported on deficits in academic performance in children with SDB. The term ‘academic performance’ encompasses a range of achievements/abilities and it can be assessed by various means including mathematical abilities, spelling, reading, writing, and overall school grade. In a landmark study of first-grade children, Gozal¹⁶ found a 6–9-fold increase in gas-exchange abnormalities in the lowest-performing tenth percentile.

Several studies have reported lower grades in mathematics, spelling, reading, and science51,108,109 in children with SDB compared to controls, even when intermittent hypoxemia is absent,¹¹⁰ which suggests that primary snoring may impact academic achievement. It is also possible that the presence of hypoxemia may affect the threshold of respiratory events associated with performance deficits, as the threshold for respiratory disturbances associated with learning problems may be lower in the presence of hypoxemia.¹¹¹ For example, in the absence of hypoxemia, a respiratory disturbance index >5 has been associated with parent-report of learning problems in young children; however, when the presence of hypoxemia was used to define the respiratory event, a respiratory disturbance index >1 was associated with learning problems.¹¹¹ Children with SDB have also been found to perform lower than controls on a phonological processing test,^49,112 which measures phonological awareness, a skill that is critical for learning to read. In addition, processes mentioned earlier^106,107 that may underlie memory encoding and storage will also impact academic performance.

We should also be reminded that the vast majority of current literature focuses on young school-aged children and does not include adolescents, a unique developmental stage where challenges differ considerably from young children and where any SDB-associated behavioral difficulties may result in significant impairment in school performance at a critical time for future success.²³ In addition to verbal problems, poor academic achievement may also be affected by inattention difficulties due to the complex brain associations involved. Measurement of school performance is inherently difficult, and the role of SDB difficult to tease out, as it really represents a number of factors, which include age, SES, home environment, genetics, behavior, and cognition.²³ Unsurprisingly, when some of the latter variables are accounted for, a number of studies fail to find evidence of an association between SDB and academic performance.^{50,60,113,114}