290903

37 AM
PEDIATRIC SLEEP DISORDERS
Neurobehavioral Morbidity Associated With Disordered Breathing During Sleep in

Children: A Comprehensive Review
Dean W. Beebe, PhD
Cincinnati Childrens Hospital Medical Center, Department of Pediatrics, University of Cincinnati College of Medicine, Cincinnati, OH
predictors of morbidity. Short-term SDB treatment outcome studies are

encouraging, but the long-term outcomes are not known. Failing to treat
SDB appears to leave children at risk for long-term neurobehavioral decits.
Conclusions: Childhood SDB is associated with neurobehavioral morbidity. Applying commonly used guidelines for causal inference, even in the
absence of a much-needed randomized clinical trial, there is strong evidence of association, consistent ndings, and specicity of effect. There
is suggestive evidence that this association ts the expected temporal
pattern and that SDB is a biologically plausible cause of neurobehavioral
decits. Clinicians should be alert to the coexistence of SDB symptoms
and concerns about a childs academic progress, attention, arousal, or
behavior or emotion regulation.
Keywords: Pediatrics, obstructive sleep apnea, sleep-disordered breathing, neuropsychology, cognitive function, behavioral functioning
Citation: Beebe DW. Neurobehavioral morbidity associated with disordered breathing during sleep in children: a comprehensive review. SLEEP
2006;29(9):1115-1134.
Study Objectives: To comprehensively review research on the association between childhood sleep-disordered breathing (SDB) and neurobehavioral functioning.
Design: Qualitative and quantitative literature review.
Setting: N/A.
Patients or Participants: N/A.
Interventions: N/A.
Measurements and Results: The ndings of 61 studies of the relationship between childhood SDB and neurobehavioral functioning were critically evaluated and synthesized. There is strong evidence that childhood
SDB is associated with decits in behavior and emotion regulation, scholastic performance, sustained attention, selective attention, and alertness.
There is also evidence that SDB has minimal association with a childs
typical mood, expressive language skills, visual perception, and working
memory. Findings have been insufcient to draw conclusions about intelligence, memory, and some aspects of executive functioning. Mechanisms by which SDB might result in neurobehavioral morbidity are being
explored, but clinical symptoms such as chronic snoring remain the best
ing, but not limited to, OSA.

Some of the earliest reports on childhood SDB suggested an
association with behavioral or learning deficits. For example,
William Hill proposed in 1889 that symptoms of what would later
be called SDB could cause backwardness and stupidity in children.9 The topic of childhood SDB then went largely ignored in
the published English literature for nearly a century. When it resurfaced in the 1970s, again academic and behavioral difficulties
were highlighted in case series.10,11 Interest in these cognitive and
behavioral effects, which are commonly called neurobehavioral because they are presumed to be mediated by the brain,12 has
since accelerated dramatically. The number of relevant articles
published in the past 5 years rivals or exceeds the volume that
had been cumulatively published before. This explosion of interest has yielded important findings, but it has also made it difficult
to stay abreast of the field. Review articles rapidly become dated,
and, in many cases, the literature covered by recent research articles has been incomplete or selective in nature.
The goal of this article is to present a comprehensive, up-todate review of the published literature on the association between
childhood SDB and neurobehavioral functioning. It will cover
the major domains of neurobehavioral functioning, highlighting
relevant findings, methodologic and conceptual issues, and directions for future work. This will be followed by a summary of the
several mechanisms that have been proposed by which childhood
SDB might result in neurobehavioral deficits. Finally, it will discuss the limited data that are available on temporal relationships
between SDB and neurobehavioral functioning.
Table 1 summarizes methodologic features of the major studies involved in this review.2,3,6,8,13-70 Studies listed here were com-
INTRODUCTION
BREATHING DURING SLEEP FALLS ON A SPECTRUM
THAT RANGES FROM REGULAR UNOBSTRUCTED RESPIRATION TO SEVERE POLYSOMNOGRAPHY (PSG)-verified
obstructive sleep apnea (OSA). Based upon conventional definitions, the population base rate of childhood OSA is believed to be
about 1% to 3%, but clinical symptoms of less-severe breathing
obstruction, such as chronic snoring, may be present in more than
10% of children.1-3 The point on this spectrum at which breathing
abnormalities, particularly breathing obstruction, can be considered pathologic remains open to debate, but many researchers and
clinicians agree that any such threshold would be best determined
via a link to medical or behavioral morbidity (e.g., see reference
1). Based in part upon studies of behavioral morbidity, several
authors have suggested that nocturnal breathing disturbance can
be detrimental to children at levels much below conventional PSG
thresholds for defining OSA.4-8 As a result, the term sleep-disordered breathing (SDB) has been used to refer to the pathologic
end of the spectrum of nocturnal respiratory functioning includDisclosure Statement
This is not an industry supported study. Dr. Beebe has indicated no nancial
conict of interest.
Submitted for publication December 1, 2005
Accepted for publication March 14, 2006
Address correspondence to: Dean W. Beebe, Division of Behavioral Medicine and Clinical Psychology (MLC 3015), Cincinnati Childrens Hospital
Medical Center, 3333 Burnet Avenue, Cincinnati, OH 45229; Tel: (513) 6363489; Fax: (513) 636-7756; E-mail: dean.beebe@cchmc.org.
SLEEP, Vol. 29, No. 9, 2006
1115
Neurobehavioral Morbidity in Childhood SDBBeebe
SLEEP, Vol. 29, No. 9, 2006
1116
Comm (Gen Peds)
Smedje, 199922
Clinical (Sleep)
Clinical (Sleep)
Clinical (ENT)
Comm (Gen)
Harvey, 199965
Owens, 200023
Goldstein, 200025
Ferreira, 200026
988
36 with SDB, no controls

(compared to published norms)
29 neurologically normal
children with OSA
18 with OSA, no controls
1844
Clinical (Psyc) vs
27 w/ ADHD from psychiatric
Comm (Gen Peds) clinics, 43 children without ADHD
from psychiatry clinics, 73 controls
from general pediatric clinics
Comm (Gen)
297
Gozal, 199821
Cross-sect follow-up
of 1993 cohort
Longitudinal
Cross-sectional
Pre-post treatment
Cross-sectional
Cross-sectional
Design
6-7
4-5
2-14
4-18
61
Age, y
Cross-sectional
Pre-Post treatment
Cross-sectional
Pre-Post treatment
Cross-sectional
Pre-Post treatment
Cross-sectional
?
none
None
Behavioral
Exclusions
Mental
retardation
Snoring
O2 desat
SDB and
restlessness
symptoms
OSA
OSA
SDB symptoms
SDB symptoms
Snoring
Q-aire
PSG (AHI > 5)

Oximetry
(>4% desat)
Q-aire
Q-aire, clinical
referral
Q-aire
Nature of SDB Source of SDB

information
Snoring
Q-aire
Memory vocabulary
Intelligence, vigilance,
impulsivity, parent and
teacher report on validated
behavior questionnaire
Clinical diagnosis of ADHD,
parent report on behavior
questionnaire
Parent and teacher report

on validated behavior
questionnaire
Neurobehavioral
Outcomes
Single-item parent report
of school and behavioral
functioning.
Clinically derived
psychiatric diagnosis
Single-item parent
report of behaviors
Q-aire, oximetry (O2

Official school
desat events, low O2
grades
nadir, or high CO2)
5-7
None
Snoring
Q-aire
Single-item parent report of
very active temperament
2.9 1.3
Neuro
OSA
PSG (cutoff
Intelligence, parent report
abnormality
not specified) on nonvalidated questionnaire
5-12 Stimulant use
OSA
PSG (AHI >1 plus Intelligence, verbal skills,
daytime symptoms)
visual skills, selective
attention, memory, finger
tapping, parent report on
validated behavior
questionnaires
2-18
Behavioral
SDB symptoms Q-aire, referral
Parent report on validated
disorders, learning
questionnaire.
disabilities, psychiatric
diseases
6-11
?
Snoring
Q-aire
Parent report of school
achievement and of behavior
on validated questionnaire
1st grade
2-18 Controls excluded

if had ADHD
Across both data collections, 242

6-7 at
never snored, 79 snored at baseline
follow-up
but not follow up, 62 snored at follow
up but not baseline, 124 snored at both.
Clinical (Obesity)
5 OSA, 9 control
Cross-sectional
13 1.8
Clinical
12 OSA, 11 BS, 10 controls
Pre-Post treatment 5.8-12.5
Of 782 subjects, 66 subjects

selected to show high risk or
low risk of SDB
507
Chervin, 199720
Rhodes, 199519
Ali, 199618
Ali, 20002
Ali, 199417
Sample Size
71 children with behavioral,

developmental, or academic problems
vs 355 controls.
Clinical (Psyc)
150 children with various
psychiatric diagnoses
Clinical (ENT)
61 SDB, 31 healthy
Sample
Source
Comm (Gen Peds)
Oxford, England Study

Ali, 199316
Comm (Gen)
Stradling, 199015
Simonds, 198414
Weissbluth, 198313
First Author, y
Table 1Summary of Methods Used in Studies of the Association between SDB and Neurobehavioral Functioning
SLEEP, Vol. 29, No. 9, 2006
1117
Sample
Source
Comm (Gen)
Comm (Gen Peds)
Gozal, 200132
Stein, 200133
77 OSA; 72 controls
Kaemingk, 200342
Cross-sectional
Longitudinal
872
229 from 2002 sample who
were contacted 4 years later
235
Cross-sectional
Pre-post treatment
Cross-sectional
Cross-sectional
Cross-sectional
Cross-sectional
Cross-sectional
Cross-sectional
Cross-sectional
Cross-sectional
Design
866
235
Comm (Gen)
Comm (Gen Peds)
64 SDB, no controls
28 OSA, 10 controls
147 SDB, 243 controls
895
797 with grades in lowest

quartile, 791 in highest quartile
472
59 OSA, 54 without
45 SDB, no controls
13 snorers, 13 nonsnoring controls

(subsample of Blunden, 2000)
16 snorers, 16 nonsnoring controls
Sample Size
Goodwin, 200341
TuCASA Study
Goodwin, 200340
Chervin, 200337
Chervin, 200538
Michigan Study
Chervin, 200236
Clinical
Clinical (Sleep) vs.

comm. (Gen)
Lewin, 200235
Goldstein, 200224
Clinical Orthodontic
Nelson, 200166
Comm (Gen)
Clinical (Sleep)
Chervin 200131
Brunetti, 200134
Clinical (ENT)
Richards 200030
Kennedy, 200429
South Australia Study

Blunden, 200028
Clinical (ENT)
First Author, y
Table 1Continued
10 3
6-11
Mental
retardation
None
none
Neurobehavioral
Outcomes
OSA
Snoring
OSA

behavior questionnaires
Parent report of learning

problems on single item
Unattended PSG
Intelligence, verbal skills,
(AHI > 5)
nonverbal skills, academic
testing, memory, parent report
questionnaire
Q-aire
Unattended PSG
(varying cutoffs)
Q-aire
SDB and restlessness

symptoms
ADHD,
mental retardation

behavior questionnaire.
Parent report of behaviors and
school achievement on
questionnaires
Parent report on unvalidated
single hyperactivity item
OSA
PSG (OSA divided in Intelligence, verbal skills,
two based on AHI > 10 visual skills, parent report
and nadir O2 < 90%)
questionnaire
SDB
Q-aire, referral
symptoms
Snoring daytime
Q-aire
sleepiness
Snoring, with a Q-aire screen
subgroup having followed by
OSA
PSG (AHI > 3)
Snoring
Q-aire
Snoring

information
behavioral
disorders, LD,
psychiatric diseases
2-13.99
2-18
4-12
6-17
3-11
4-12
Behavioral
Exclusions
Q-aire, referral
visual skills, vigilance,
memory, parent report on
validated behavior
questionnaire.
Snoring
Referral, Q-aire.
Snorers did not have
marked OSA.
2.5 - 15.5
None
SDB symptoms Retrospective
chart review
achievement
2-18
None
OSA
PSG (AHI > 5 or
Parent report of ADHD
esophageal pressure diagnosis, stimulant use,
< -20)
and behavior on validated
questionnaire.
13-14 Current snoring Snore at age 2-6
Q-aire
Objective school performance
5-10
Age, y
SLEEP, Vol. 29, No. 9, 2006
1118
Sample
Source
Comm (Gen)
Comm (Gen)
Clinical (ENT) vs
Comm (Gen)
Comm (At-risk)
154 SDB, 293 controls
19 OSA, 19 nonsnoring
controls (subset of 2003 sample)
39 SDB, 20 control
Cleveland Childrens Sleep and Health Study

Rosen, 20047
Comm (Gen)
829 (40 OSA, 122 primary
snorers, 667 neither)
Emancipator, 200670
835 (164 SDB, 671 controls)
Castronovo, 200368
MontgomeryDowns, 200453
Friedman, 200354
1010
87 Primary Snorers 31 Controls
OBrien, 20046
Louisville preschool study

MontgomeryComm (At-risk)
Downs, 200351
35 OSA, 35 controls
44 significant ADHD, 27
mild ADHD, 39 controls
205 (subset of 2003 sample

that had more thorough testing)
3019
99 habitual snorers, 352 nonsnorers

995 (re-analysis of 2003 sample, dropping
children with congenital heart disease)
1144
403
480 (115 SDB)
Sample Size
OBrien, 200450
Louisville First-grade study

OBrien, 200349
Comm (Gen)
Gottlieb, 200448
Boston Study
Gottlieb, 200347
Urschitz, 200444
Urschitz, 200546
Hannover, Germany Study

Urschitz, 200345
Comm (Gen)
Comm
Mulvaney, 200643
Goodwin, 200539
First Author, y
Table 1Continued
Cross-sectional
Cross-sectional
Pre-post treatment
Pre-post treatment
Cross-sectional
Cross-sectional
Cross-sectional
Longitudinal
Cross-sectional
Cross-sectional
Design
ADHD
Mental retardation
none
Mental
retardation
8-11
none
ADHD, other
psychiatric
disorders
4.1 0.5
?
5-9
4 0.5
5-7
9.6 0.7
Snoring,
OSA
OSA or
clinical
diagnosis
Snoring
OSA
Snoring
Snoring,
not OSA
OSA
OSA
Q-aire, oximetry
(nadir O2 < 90%)
Parent report of behavior on

unvalidated questionnaire,
actual school grades
Neurobehavioral
Outcomes
Parent-reported learning
Parent-report on single-item
behavioral scales that were
validated in a subset of subjects
Intelligence, vigilance, selective
attention, executive functioning,
parent report on validated
Parent-reported preschool
performance and behaviors on
validated questionnaire
Intelligence
Q-aire, Unattended Parent report on validated

PSG (AHI>5 or AI>1) behavior questionnaires
academic testing, vigilance,
abstract reasoning
PSG (AHI >1) or

Verbal skills, intelligence
symptoms kissing
tonsils
Q-aire
Parent report on unvalidated
daytime behavior questionnaire
PSG (AHI > 5)
Q-aire
PSG (AHI and PLM Parent-report on validated

treated continuously) behavior questionnaire
compared to PSG and parent
report of snoring
PSG (pooled AHI, Intelligence, verbal skills,
RAI, and O2 nadir)
executive functioning,
validated behavior
questionnaires.
Q-aire, PSG
verified no OSA
Q-aire
Oximetry (several O2
Parent report of
-related indexes)
mathematics performance
SDB symptoms
Hypoxia
Snoring, O2
saturation
Behavioral
Exclusions
information
Mental retardation SDB/OSA Q-aire, Unattended
PSG (AHI > 1)
OSA
Unattended PSG
(15th %ile on AHI)
9.6 0.7
Age, y
SLEEP, Vol. 29, No. 9, 2006
1119
Sample
Source
Clinical (sleep)
Clinical
(Peds sick visits)
Blunden, 200527
Cross-sectional
Cross-sectional
54
35
117 kids with adenotonsillar
hypertrophy, 104 controls
Clinical (ENT)
Clinical (ENT)
Mitchell, 200569
Roemmich, 200667
MontgomeryComm (Gen)
Downs, 200662
Kurnatowski, 200663 Clinical (ENT)
Pre-post treatment
Pre-post treatment
Cross-sectional
Behavioral
Neurobehavioral
Exclusions
information
Outcomes
8-12 Medication use,
OSA
PSG (AHI 1-10) Intelligence, academic testing,
mental retardation
working memory, vigilance,
memory
2-18 Medications that
Suspected
Q-aire, referral
could affect alertness
SDB
6-12 Mental retardation,
OSA
PSG (simple snorers
Intelligence, visual
nonstimulant psychiatric
AHI < 1, mild OSA scanning, vigilance, working
medications
AHI 1-5, mod-sev memory, memory, mental
AHI>5).
flexibility, problem-solving,
parent and teacher report on
validated behavior
questionnaires
5-14
?
SDB symptoms
referral
Vigilance, parent report on
validated behavior questionnaire
6-12
?
OSA,
PSG (AHI, PLMI) Vigilance, parent report on
restlessness
0-16
any illness,
Snore, BSP
Q-aire
physical or mental?
impulsivity, working memory,
2-14
learning
OSA
PSG
disabilities, behavioral
disorders
Grade 1-4
?
SDB
Q-aire
Performance on broad
cognitive screener
7-14
ADHD
Snoring
Q-aire
disabilities and school failure,
parent-report on validated
3-11
?
Snoring, OSA
Q-aire, PSG
(AHI >3)
performance
2-18
Psychiatric
OSA
PSG (AHI > 5)
disorders, Dev delay
6-12
Psychiatric
OSA
PSG (AHI > 1)
medications
Infants
none
SDB and restlessness
PSG
General cognitive development,
8.2 0.4 mo
motor development
6-13
none
OSA
PSG (AI >1 &
Verbal skills, visual skills,
SpO2 < 92%)
memory
Age, y
Clinical sample sources included children seen in sleep, psychiatry or psychology (Psyc), otolaryngology (ENT), and dermatology (Derm) clinics. Community sample sources derived from general
pediatrics offices (Gen Ped), children in special preschool programs (at risk), or other community sources, such as birth or school records (Gen). SDB refers to sleep-disordered breathing; OSA,
polysomnographically defined obstructive sleep apnea; PSG, polysomnography; AHI, apneahypopnea index; AI, apnea index; desat; desaturation; ADHD, attention-deficit/hyperactivity disorder;
Q-aire, questionnaire; LD, learning disabilities. Shading indicates that multiple publications from the same group shared a given methodological feature.
52
1215
Comm (Gen)
Sogut, 200561
Cross-sectional
Comm (Gen)
Arman, 200560
Cross-sectional
Pre-post treatment
Cross-sectional
Cross-sectional
Comm (Gen)
41 OSA, 41 controls undergoing

elective surgeries other than
adenotonsillectomy
79 SDB, 468 non-respiratory
sleep disorders, 633 controls
96 snorers, 190 controls
11 snorers, 13 behavioral sleep

problems, 9 both, 13 neither
(controls)
88 ADHD, 27 controls
Pre-Post treatment
Cross-sectional
6 moderate-severe OSA, 9 mild

OSA, 17 simple snorers, 17
nonsnoring community controls
19 OSA, no controls
Cross-sectional
Cross-sectional
Design
108 snorers, 72 controls
12 with OSA, no controls

Sample Size
Carvalho, 200564
Clinical (ENT vs.

other clinics)
Clinical (Psyc)
Huang, 200458
Tran, 200559
Clinical (ENT)
Avior, 200457
Melendres, 200456 Clinical (sleep) vs.

mix (Peds & Derm)
Beebe, 20048
Clinical (Sleep) vs
Comm (Gen)
Archbold, 200455
First Author, y
Table 1Continued
piled via computerized literature searches, manual review of the

references listed for recent studies, and direct contact with key
investigators in the field. Only studies that used a standardized
reproducible approach to the assessment of both SDB and neurobehavioral functioning are listed. In some cases, multiple studies were published off of the same or overlapping data sets; these
have been clustered in the table to emphasize that findings from
overlapping papers may not be independent. Because specific
neurobehavioral outcome measures have varied across studies,
only general outcome domains are listed in Table 1. The specific
measures used will be detailed later within each neurobehavioral
domain, accompanied by domain-specific figures and tables. It
should be noted that although some tables present findings in binary form for simplicity (significant vs nonsignificant), overall
conclusions considered statistical effect sizes, rather than simple
study counts.
During the course of this review, I will periodically return to a
pivotal question: what is the evidence that pediatric SDB causes
neurobehavioral deficits? The answer may have important implications for the prevention and amelioration of neurobehavioral
deficits in a large number of children. This answer, however, is
not as straightforward as it might seem. Philosophers of science
emphasize that, although inferences of causation are informed
by scientific evidence, they remain best guesses based upon
the state of the field and the dominant epistemology.c.f.,71-74 The
limits of causal inference are especially evident in epidemiological, neuropsychological, and medical research, in which random
assignment to conditions is impossible or unethical.71,73-77 Even
so, guidelines have been developed to facilitate the consideration
of causal inferences. These guidelines, which will be further defined over the course of this paper, include strength of association,
consistency of findings, specificity of effect, biologic credibility,
dose-response relationships, and temporal sequence.77-80
Potential Causes for Illusory or Inflated Associations

Statistical associations may be artificially inflated when SDB
samples are drawn from clinical populations, especially if they are
compared with volunteer control groups drawn from nonclinical
sources. Because pediatric specialty clinics are typically located
in tertiary care centers, clinical samples are likely to overrepresent
children whose parents are most concerned about their functioning. Especially in light of popular press on a suspected link between SDB and daytime behaviors (e.g. see reference 82), parents
may come to the clinic with primary concerns about their childs
daytime behaviors and the hope that a treatable sleep disorder
might underlie behavior problems. The result is that children seen
in sleep and otolaryngology clinics may have behavior problems
more often than do children with identical sleep problems in the
general community. Bias may be compounded when controls are
recruited using dramatically different techniques, such as newspaper postings, that tend to attract superhealthy individuals.8
A heavy reliance on parent report may also inflate findings. Insofar as parents come to the clinic with preexisting beliefs about
a relationship between sleep and neurobehavioral functioning, or
are led to have such beliefs (e.g., via suggestion that a childs
behavior will improve after adenotonsillectomy), this may introduce placebo effects and other biases in parent ratings. Further,
when information on both SDB and behavioral functioning are
obtained from the same parent, this can lead to shared method
variancethe tendency for spurious or inflated correlations
between 2 variables that are gathered using the same methods.83
This is not to say that parent report should be dismissedit is
likely the best source for some informationbut SDB-behavior
associations based solely upon parent report can be assumed to
yield upper-boundary estimates of associations.
Potential Causes of Suppressed or Attenuated Associations
ASSOCIATIONS OF CHILDHOOD SDB WITH NEUROBEHAVIORAL

FUNCTIONING
In contrast, statistical findings will be biased towards nonsignificance if either the sample is too small (statistically underpowered) or if effects are washed out because of the use of inadequate tools to determine the presence of SDB or neurobehavioral
deficit. Clinical symptoms of SDB have been criticized for a lack
of sensitivity39 and specificity to PSG-defined OSA (e.g., see references 84-86). If one assumes that more-severe PSG-defined
OSA is associated with greater morbidityan assumption that
shall be questioned laterthe reliance on symptom-based screenings might be expected to attenuate group comparisons. Similarly,
the use of unvalidated outcome measures or single-item neurobehavioral outcome scales, which tend to be less reliable than multiitem measures,87 would be expected to artificially limit group
comparisons. Though internally reliable, teacher-report scales
also probably yield lower-boundary estimates of statistical effects
because of variability in how well teachers know each child. As
discussed later, even highly reliable office-based tests may be ill
suited to assess for some real-world deficits, potentially limiting
statistical effects.
Finally, as indicated in Table 1, several studies have excluded
children with a history of psychiatric difficulties, including attention-deficit/hyperactivity disorder (ADHD). The rationale for
such a priori exclusion is typically unstated but may stem from
concerns about the potential effect of psychotropic medications
on sleep. Behavioral exclusions also may relate to a widespread
belief that ADHD is a unique disorder with an etiology that is unre-
Conventional guidelines for inferring causality start with the

basic premise that a cause and an effect should be consistently
associated.77-80 Given this, it is noteworthy that some authors have
been critical of the research that has associated childhood SDB
with neurobehavioral deficits. In particular, Ebert and Drake81 reviewed 17 articles and highlighted a number of shortcomings, including inadequate description of sampling methods and probable
sample biases, inconsistent measurement tools, possible reporting
biases, lack of consideration of possible confounds, and limited
statistical treatment. They concluded that only 2 studies20,32 were
of sufficient quality to assess for selection bias or the effects of
confounders. Although they found little evidence that these studies came to biased conclusions, for the most part Ebert and Drake
dismissed the rest of the research literature in their conclusions.
Although Ebert and Drakes81 paper acts as an important signpost towards increased methodologic rigor, their conclusions
may have been overly conservative. The definitive study is rare
in any aspect of science. Progress often is made via successive
approximations, as data accumulate across studies that have differing strengths and limitations. This warrants a less dismissive
approach, in which there is careful consideration of the potential
impact of methodologic flaws on the fields overall conclusions.
SLEEP, Vol. 29, No. 9, 2006
1120
lated to sleep. Unfortunately, this belief may be mistaken. ADHD

is diagnosed entirely on behavioral, not etiologic, grounds,88 and
children with ADHD may disproportionately display frank sleep
pathology.89 Regardless of reasoning, excluding children with
ADHD or other psychiatric disturbances from analyses may control away interesting findings. For example, an early publication
on the Tucson Childrens Assessment of Sleep Apnea study found
SDB to be associated with few behavioral deficits after excluding
children reported to have ADHD from analyses.42 However, associations between SDB and inattention, aggression, oppositional
behaviors, and poor social functioning emerged in a larger sample
that retained children with ADHD.43
1.4
Effect Size (d)
1.2
0.8
0.6
0.4
0.2
0
Ali et al, 1993
Rosen et al,
2004
Attention
Interpretive Approach
Beebe et al,
2004
O'Brien et al,
2004
Hyper/Impulsive
Mulvaney et
al, i.p.
POOLED
Externalizing
Figure 1Effect of sleep-disordered breathing on Parent-Reported

Attention, Hyperactivity/Impulsivity, and Other Externalizing Behaviors. Shown are standardized effect sizes derived from publications that included detail on all 3 behavioral domains. By convention,
an effect size of 0.2 is considered small, 0.5 is considered medium,
and 0.8 is considered large.94 The pooled effect sizes were computed
using a random-effects model,93 with the standard error of measurement represented by the line above each bar. See text for further information on effect size and pooled effect-size computations.
Although this list of methodologic issues is not exhaustive, it

does highlight the most common issues that could bias conclusions.
As a result, these potential sources of bias will be considered when
reviewing each domain of neurobehavioral functioning below.
Behavioral and Emotional Functioning in Applied Settings
The overwhelming majority of published studies have found
significantly more parent-reported behavioral or emotional
concerns among children with SDB than among other children of
the same age. Of the more than 30 studies in Table 1 that included
behavioral outcomes, only 3 failed to yield at least 1 behavioral
finding.42,50,59 Of those 3, 2 studies excluded children with
ADHD or other psychiatric disorder from analyses, suggesting a
possible artificial suppression of statistical effects. Thus, there is
strong evidence of some associations between SDB and daytime
behaviors; the question is which behaviors.
2 reported null findings. The relative severity of each type

of behavior problem could be directly examined in 6 studies
that did not prescreen for ADHD and used either a variant on
the Conners Parent Rating Scale91 or the Behavior Assessment
System for Children.92 Effect sizes for each study are shown in
Figure 1, along with the pooled effect size and standard error of
the measurement for that effect size based upon a random-effects
meta-analytic model.93
For readers who are less familiar with the concept of effect
size, this is a standardized measure of the strength of a statistical
relationship. The most commonly used index, d, generally refers
to the arithmetic difference between the mean scores of 2 groups,
divided by either the population standard deviation (if known, as
in IQ or T-scores) or an estimate thereof based on the sample data
(in this paper, I used the control-group standard deviation as the
basis for estimates). In other words, it indicates how many standard
deviations apart the centers of 2 groups fall. For consistency, only
the d statistic is reported here, with other effect-size metrics (e.g.,
odds ratios) converted to d using formulae provided by Lipsey
and Wilson.93 By convention, an effect size of 0.2 is considered
small, 0.5 is considered medium, and 0.8 is considered large.94
Meta-analysis combines (pools) these effect sizes, with findings
from larger studies weighted more strongly than those from
smaller studies. In the studies shown in Figure 1, the pooled effect
sizes were moderate in size, with hyperactive/impulsive and other
externalizing behaviors slightly, but not reliably, more affected
than attention.
It appears unlikely that the association between SDB and
behavior problems can be attributed to methodologic bias.
The supporting studies have varied methodologically. Several
have minimized the likelihood of inflated effect sizes by using
community recruitment, controlling for demographics, or
employing PSG-based definitions of SDB (e.g., see references
7, 43, and 44). Four have complemented parent-report data with
that obtained from teachers. As expected, the effect sizes were
smaller than with parent report, though 2 studies continued to
find significant associations between behavioral disturbances
and SDB.8,16 Of the 2 null studies that employed teacher-report
Behavior Regulation and Inattention

The most common finding is an elevated frequency of overt
behavior problems, such as hyperactivity, rebelliousness,
and aggression, in children with parent-reported SDB symptoms,2,7,13,16,17,22,26,36,45,56,61 those who have been referred for adenotonsillectomy because of suspected SDB,15,25 and those with
PSG-defined OSA.7,8,23,35,43,45,54,56,59,61 Inattentive behaviors have
also been reported, although in some cases the published findings do not allow for differentiation of impulsive versus inattentive behaviors (e.g., the attention subscale of the Child Behavior
Checklist90). Table 2 lists relevant studies that have yielded quantitative data on inattention, hyperactive/impulsive behaviors, or
other externalizing behaviors (e.g., aggression, oppositionality,
irritability).
When specific information on parent-reported hyperactivity/
impulsivity could be culled from publications, 12 studies
reported positive associations between SDB and hyperactivity/
impulsivity, 5 reported mixed findings across scales or analyses,
and 3including 1 study that excluded children with behavioral
disorders and another that used a nonvalidated single-item
scalereported null findings. When specific information on
parent-reported externalizing behavior was available, 15 studies
noted positive associations with SDB, 4 noted mixed findings,
and 4 noted null findings. Data were less consistently supportive
for a specific association with attention. Seven studies reported
positive associations between parent-reported inattention and
SDB, 4 reported mixed findings across scales or analyses, and
SLEEP, Vol. 29, No. 9, 2006
1121
Table 2Findings on Hyperactivity/Impulsivity, Externalizing Behaviors, and Inattention Among Children with Sleep-Disordered Breathing
Study
Stradling, 199015
Measure
Unvalidated single items
Parent-report findings
Teacher-report findings
HI, EXT elevated pre-treatment, somewhat
improved posttreatment
Ali199618
Conners
No elevations on HI, EXT, or ATN pretreatment,
No elevations pretreatment,
though improvements on all 3 post-treatment
no changes posttreatment
Smedje, 199922
Unvalidated single item
HI elevated
Owens, 200023
BASC
On BASC, externalizing composite and ATN both elevated.
Goldstein, 200025
CBCL
Externalizing and attention scales (hyperactivity & ATN not
differentiated) elevated pretreatment. ATN scale improved posttreatment.
Ferreira, 200026
Unvalidated single-item
EXT elevated.
Blunden, 200028
CBCL
EXT not elevated.
Chervin, 200131
Conners
HI elevated only when both SDB and periodic limb movement
Stein, 200133
CBCL
present. SDB associated with many CBCL scales, but only
unique association was with EXT
Brunetti, 200134
HI elevated.
Nelson, 200166
HI not elevated.
Lewin, 200235
CBCL
EXT elevated only in mild (not severe) OSA.
Goldstein, 2002 24
CBCL
Externalizing but not attention scale elevated.
No posttreatment gains in either.
Urschitz, 200444
Unvalidated items
HI and ATN elevated
Rosen, 20047
CBCL, Conners
On CBCL, EXT but not attention subscale elevated
(HI and ATN not differentiated). On Conners,
HI and EXT elevated, but ATN was not elevated.
Melendres, 200456
Conners
Composite ADHD index elevated, but not broken down into HI, ATN.
Beebe, 20048
BASC, BRIEF
EXT, HI elevated on both Q-aires, ATN elevated only
EXT clearly elevated, HI not
on BRIEF. Worst behaviors in milder SDB.
elevated, ATN inconsistent
Blunden, 200527
CBCL
EXT elevated.
CBCL
No difference between groups pre- or posttreatment on
Tran, 200559
EXT or attention scales.
Michell, 200569
BASC
HI and ATN elevated, but not EXT
Arman, 200560
Conners, DSM checklist
EXT, HI, ATN all elevated on one Q-aire but only
EXT, HI, ATN not different
across groups on either Q-aire.
EXT on both. ADHD cases were excluded.
Oxford, England Study
Conners
HI, EXT, ATN all elevated.
HI, ATN both elevated,
Ali, 199316
but EXT not elevated.
Conners
HI elevated
Ali, 199417
Conners
HI higher in chronic SDB than transient SDB,
Ali, 20002
which was higher than in no SDB.
Michigan Study
SDB elevated among subjects with
Chervin, 200236 Conners, DSM checklist
high HI, especially young boys.
Conners
EXT elevated even after controlling for HI.
Chervin, 200337
TuCASA Study
Conners
EXT, HI, ATN not elevated, but subjects with ADHD
Kaemingk, 200342
were a priori excluded.
CBCL, Conners
With ADHD subjects included, CBCL EXT and Attention problems
Mulvaney, 200643
scales elevated. On Conners, EXT, ATN elevated to a similar degree;
HI had mixed data.
Boston Study
Validated single items
EXT, HI, ATN all elevated.
Gottlieb, 200347
Conners
ATN elevated, but HI not.
Gottlieb, 200448
Louisville 1st-grade study
Conners
ADHD symptoms (HI & ATN not differentiated) associated monotonically
OBrien, 200349
with snoring, but only mild ADHD related to PSG-verified SDB.
CBCL, Conners
Both controls and SDB subjects had high HI, ATN, and
OBrien, 200450
EXT scores, but no group differences.
CBCL, Conners
On Conners, HI (not EXT, ATN) elevated. On CBCL, EXT
OBrien, 20046
and attention scale (HI and ATN not differentiated) elevated.
ADHD cases excluded.
HI refers to hyperactivity/impulsivity; EXT, externalizing behaviors (aggression/ oppositionality/conduct problems); (ATN) inattention; BASC,
Behavioral Assessment System for Children; BRIEF, Behavior Rating Inventory of Executive Functioning; Q-aire, questionnaire; CBCL, Child
Behavior Checklist; Conners, a validated variant on the Conners Rating Scales. The CBCL attention and Conners attention-deficit/hyperactivity
disorder (ADHD) subscales are heterogeneous, including items related to inattention, hyperactivity, and impulsivity.
SLEEP, Vol. 29, No. 9, 2006
1122
Table 3Findings on Emotional Functioning of Children With Sleep-Disordered Breathing

Study
Simonds, 198414
Owens, 200023
Measure
Clinical psyc diagnosis
BASC, Conners
Goldstein, 200025
Ferreira, 200026
Blunden, 200028
Stein, 200133
Lewin, 200235
CBCL
CBCL
CBCL
CBCL
Goldstein, 200224
OBrien, 200450
OBrien, 20046
Rosen, 20047
CBCL
CBCL, Conners
CBCL, Conners
CBCL, Conners
Beebe, 20048
BASC, BRIEF
Blunden, 200527
Tran, 200559
Mitchell, 200569
Arman, 200560
Mulvaney, 200643
CBCL
CBCL
BASC
Conners
CBCL, Conners
Parent-report findings
Affective disorders not associated with snoring
BASC Internalizing scale elevated but somatic not pulled out. In small subset of subjects who
had the Conners (n=9), anxiety appeared elevated, but less so than somatic.
Anxious/depressed subscale not elevated
Irritability elevated
Internalizing behavior elevated, but somatic subscale not pulled out
Anxious/depressed subscale associated with SDB, but so were almost all CBCL subscales.
Internalizing elevated, but mostly due to somatic concerns. Mild but not severe OSA had
elevated anxiety/depression scale.
Anxious/Depressed scale not elevated
Both controls and SDB had high anxious/depressed scores, but no group effects were found.
No effect on Conners anxious/shy subscale, but elevated CBCL anxious/depressed scale
CBCL anxious/depressed subscale not elevated, but Conners emotional lability subscale
sharply elevated
Marked effect on BRIEF emotional control subscale, but little effect on BASC anxiety or
depression subscales
Internalizing behavior elevated but somatic subscale not pulled out
No difference between groups on anxious/depressed subscale pre- or post-treatment
Anxiety subscale not elevated, depression subscale somewhat elevated, but not as markedly as
somatization subscale.
Anxiety not elevated
Neither anxious/depressed subscale nor internalizing composite was elevated
BASC, refers to the Behavioral Assessment System for Children; BRIEF, Behavior Rating Inventory of Executive Functioning; CBCL, Child
Behavior Checklist; Conners, a validated variant on the Conners Rating Scales; psyc, psychiatric; OSA, obstructive sleep apnea; SDB, sleep-disordered breathing.
data, 1 excluded children with ADHD from analyses,60 and the

other utilized very small samples and had an apparent floor
effect on some scales.18
to community controls; Rosen et al studied a large community

cohort.
Of note, data on the emotional functioning of children with
SDB were based upon parent report. To my knowledge, no childhood SDB study has published results from self-report questionnaires, which may be more sensitive to subjective mood. One
study gathered teacher-report data, noting negligible associations
between SDB and a childs typical mood but stronger associations with emotional stability.8
Mood and Emotional Control

Table 3 presents studies that investigated the emotional functioning of children with SDB using parent-report questionnaires.
Several publications reported only composite internalizing
scores that conflated relatively pure symptoms of anxiety and
depression with physical or somatic complaints. This is important
because such somatic complaints are common among children
with a variety of medical conditions, including SDB,24,35 and may
not be a valid reflection of emotional functioning in such children.95 Thus, there is reason to question the validity, or at least the
implications, of reports that children with SDB have elevated internalizing composites.23,27,28,33,35 Only 2 studies have reported that
children with SDB had clear elevations on more-direct measures
of anxiety or depression,23,33 whereas 2 had mixed results,6,35 and
8 found no significant associations.7,8,24,25,43,50,59,60 This preponderance of nonsignificant results appears legitimate, as it was obtained using a variety of research methods, some of which might
be expected to inflate effect sizes (e.g., clinical samples) or to impart considerable statistical power (e.g., large community samples
and PSG documentation of OSA).
An important distinction can be drawn between an individuals
typical mood and their degree of emotional stability.8,96,97 The
above findings suggest that SDB has a small to negligible association with childrens typical mood. However, as shown in Figure 2,
Beebe et al8 and Rosen et al7 found substantially stronger associations between SDB and parent-reported emotional instability. The
effect sizes were larger overall in the Beebe et al study, but that
may reflect the fact that it compared clinically referred patients
SLEEP, Vol. 29, No. 9, 2006
Sleepiness
How sleepiness is manifested during childhood remains a
point of considerable debate. Familiar outward signs of sleepiness in adults may not be evident in young children, leading some
to suggest that problems with behavior regulation may be a better
reflection of sleep deprivation or disruption in children.cf.98 Objective testing of sleep propensity using the Multiple Sleep Latency
Test99 has yielded evidence of increased physiologic sleepiness in
children with SDB, but these children rarely fall in the range considered pathologic in adults.55,100 Still, parents have consistently
identified children with SDB as being sleepier than controls on
questionnaire items.2,16,17,26,34,40,44,47,48,51,56
There appears to be a temptation in the field to account for
the wide range of neurobehavioral deficits associated with SDB
by attributing them to sleepiness (e.g., children act out because
they are sleepy). This temptation should be resisted until there
is clear evidence of physiologic mechanisms by which sleepiness
and neurobehavioral deficits occur in this population. Conventionally defined sleepiness and other aspects of neurobehavioral
functioning all fall at the same behavioral level of measurement
and are typically measured concurrently. Given this, it is risky to
use one to explain the other. Readers from the disciplines of
1123
R
di
at
r ic
s
et
B
lu
al
,2
nd
Em
00
en
5
et
an
al
ci
,2
pa
00
to
re
5
ta
l,
20
06
JS
04
04
m
er
yD
ow
ns
Pe
20
l,
20
l,
et
a
et
a
tg
o
00
.,
2
rie
n
O
'B
rie
n
G
ot
tli
e
et
al
al
,2
et
al
,2
et
ee
b
B
00
3
00
00
3
,2
al
et
k
in
g
m
an
,2
00
00
al
et
SDB
M
on
Figure 2Effect of Sleep-Disordered Breathing on Parent-Reported

Mood and Emotional Stability. Shown are standardized effect sizes.
For Beebe et al,8 the typical mood score represents the mean effect
size from the anxiety and depression subscales of the BASC, whereas
the emotional stability score represents the effect size of the BRIEF
Emotional Control subscale. For Rosen et al,7 the typical mood score
represents the mean effect size for the Child Behavior Checklist Anxiety/Depression subscale and the Conners Anxiety subscale, whereas
the emotional stability score represents the effect size of the Conners
Emotional Lability subscale. By convention, an effect size of 0.2 is
considered small, 0.5 is considered medium, and 0.8 is considered
large.94 See text for further information on effect size and pooled effect size computations.
Control
Figure 3Association between Sleep-Disordered Breathing (SDB)

and Overall IQ. The population mean is dened as 100, with a standard deviation of 15. An additional study by Kennedy et al29 is not
included because it used a substantial subsample of those reported on
by Blunden et al28 and yielded very similar results.
these, 2 studies reported no group differences on overall IQ,42,70

and 4 found relative IQ deficits among children with SDB compared with controls.6,48,50,52 Interestingly, all 4 of these recruited
young children, ranging from preschool through first grade,
whereas the other studies shown in Figure 3 focused on schoolaged children. There may be a moderating effect of age, in which
young children are at greater risk of morbidity.12 Other risk factors may also be important, as 1 study found that children who
had been born preterm showed a marginal association between
SDB status and IQ, whereas those born full-term showed no association.70 Alternatively, the apparently greater vulnerability of
the younger samples may reflect the fact that different tests are
used with preschool children than with school-aged children.
This raises a fundamental point about intelligence testing: a
composite IQ score reflects a childs combined performance
across a variety of specific tasks. Insofar as the skill set assessed
by 1 intelligence test differs from that assessed by another, findings may differ. Moreover, the use of composite IQ scores has
been criticized by some leaders in neuropsychology because such
scores tend to obscure the significant skill variability observed
in neurologically impaired populations, and they fail to address
important domains of cognitive functioning, such as memory.96,104
Instead, these authors argue for a more domain-specific assessment of functioning. Table 4 summarizes the findings of studies
of specific cognitive skills in children with SDB. Whenever possible, specific subtest findings are provided, although a number of
articles included only composite scores.
As shown in Table 4, findings have been extremely mixed
with respect to tests of language functioning. Most studies have
reported nonsignificant differences between children with SDB
and controls on measures of basic vocabulary and other expressive language tasks. This does not appear to be a methodologic
artifact. Nonsignificant findings were evident in relatively largesample studies,6,42 those that compared clinical groups to community controls,8,23,35 and those that defined SDB by conventional
PSG cutoffs.6,12,23,35,42,50 There has been mixed support for deficits in comprehension of spoken instructions/commands, with 1
author reporting clear deficits among clinically referred children
with OSA,63 another reporting mixed results across similar instruments in a community sample of primary snorers,6 and a third
finding null results across 2 instruments in a smaller community
psychiatry and neurology may be familiar with other examples of

this issue. Clinical depression often presents with a loss of interest in previously pleasurable activities (anhedonia) and difficulty
concentrating, but it would be erroneous to assume that anhedonia
causes concentration difficulties or vice versa. Rather, both are
believed to be part of a larger syndrome that has a complex root
psychophysiologic cause.101 Similarly, childhood epilepsy can be
associated with interictal learning problems that are not caused by
seizures; rather, both learning problems and seizures are symptoms of an underlying neural aberration.102 Likewise, behaviorally
defined sleepiness and other neurobehavioral deficits may share a
common cause (e.g., sleep deprivation or disruption) rather than
explain each other. This would be consistent with data from Golan
and colleagues,103 who found that children with ADHD also had
elevated sleep propensity; in their sample, children with ADHD
who did not show evidence of a primary sleep disorder (SDB or
periodic limb movement disorder) had evidence of other sleep
problems, such as short sleep time and poor sleep efficiency.
Cognitive and Academic Functioning
Figure 3 illustrates the major findings comparing the intellectual ability of children with SDB to controls. Groups of school-aged
children with SDB often score lower than controls but commonly
perform near the population average of 100. In most cases, any
significant difference between SDB and control groups has been
due to the controls being of above average intellect. This may
reflect a recruitment bias. Families of children who are not receiving clinical care (ie, who do not have some potential investment
in participating) but who nevertheless volunteer to participate in
research may be atypical of the general population, resulting in
super-normal mean intelligence scores that can inflate group
comparisons.8
Several recent studies have screened large community samples
and used identical recruitment techniques for children in the SDB
and control groups, thereby reducing the potential for bias. Of
SLEEP, Vol. 29, No. 9, 2006
w
in
O
'B
Emotional Stability
Fr
ie
d
Beebe et al, 2004
Le
ta
l,
li
e
A
Typical Mood
lu
n
Rosen et al, 2004
et
0.2
al
,2
19
96
0.4
ae
m
0.6
de
n
Effect Size (d)
0.8
130
120
110
100
90
80
70
Overall IQ
1124
Table 4Domain-by-Domain Findings on Cognitive Tests for Children With Sleep-Disordered Breathing
First author, y
Rhodes, 199519
Ali, 199618
Owens, 200023
Verbal Skills Visual Perception and Construction

Memory
(+) WISC-III Vocab
(+) WRAML Verb Mem
(+/-) WRAML Vis Mem
(-) PPVT
(-) McCarthy Verbal
(-) WISC-III Vocab
Blunden, 200028
(+) WPPSI-R &

WISC-III VIQ
Lewin, 200235
(-) DAS Verbal Cluster
Kaemingk, 200342
(-) WASI VIQ
Gottlieb, 200448
(-) Hooper
(-) McCarthy Visual
(-) WISC-III Block Des.
(-) VMI
(-) WRAML Verbal Lrn

(-) WRAML Story Mem.
(-) WRAML Pict. Mem.
(-) WRAML Des. Mem.
(+) WPPSI-R &

WISC-III PIQ
(-) DAS Visual Cluster
(+/-) WASI PIQ
(+) WRAML memory

screening index
OBrien, 200450
(-) DAS Verbal Comp.

(-) DAS Picture Sim.
(-) DAS Naming
(-) DAS Copying
(-) NEPSY Sp. Naming
(-) DAS Matching Letter(-) NEPSY Comp. of
Like Forms
Instructions
(+) DAS Pattern Constr.
(+) NEPSY Phonological Proc. (-) NEPSY Design Copy
(-) NEPSY Arrows
OBrien, 20046
(+) DAS Verbal Com.
(-) DAS Picture Sim.
(+) DAS Naming
(-) DAS Copying
(-) NEPSY Sp. Naming
(-) DAS Matching Letter(-) NEPSY Comp. of Instructions
Like Forms
(+) NEPSY Phonological Proc. (+) DAS Pattern Constr.
(-) NEPSY Design Copy
(+) NEPSY Arrows
Friedman, 200354 (-) K-ABC Vocab.
(-) K-ABC Gestalt Closure
(-) K-ABC Triangles
(+) K-ABC Matrices
(+) CAVLT
(+) NEPSY Memory domain
(-) NEPSY Memory for faces
(-) NEPSY Memory for Names
(-) NEPSY Narrative Memory
(-) CPT
(-) MFFT
(-) WISC-III Digit Span
(-) Verbal Fluency
(+/-) Cancellation Tests
(+) CPT
(+) DAS Processing Speed
(-) CPT
(+/-) WPPSI-R Animal Pegs
(+) NEPSY Attn/Exec domain
(-) DAS Recall of Digits
(-) NEPSY Sentence Rep.
(+) NEPSY Tower
(+) NEPSY Auditory CPT
(+) NEPSY Vis. Attention
(-) NEPSY Memory for faces

(-) NEPSY Memory for Names
(-) NEPSY Narrative Memory
(-) DAS Recall of Digits

(-) NEPSY Sentence Rep.
(-) NEPSY Tower
(-) NEPSY Auditory CPT
(+) NEPSY Vis Attention
(+) K-ABC word order

(-) K-ABC Spatial memory
(-) K-ABC Number Recall
(-) WASI Matrixes
(-) CMS Dot locations
(-) WISC-III Vocab.
(-) WISC-III Block Des.
(-) WRAML Verbal Learning
(+) WASI VIQ
(-) WASI PIQ
(-) CMS Word Pairs
(+) Rey Complex Figure Copy
(+) Luria verbal learning

(+) Rey Figure Delay
(+) CPT
(-) CMS numbers
(-) CMS sequences
(-) Ch. Category Test
(-) WISC-III D. Span
(-) Stroop
(-) WCST
(+/-) CPT
(+) NEPSY Vis. Attention
(+) NEPSY Verb. Fluency
(+) CPT
(+) CPT
(-) CMS numbers
(+) CPT
(+) TEA-Ch code transmisn.
(+/-) CPT
(+) K-ABC Riddles
(+) K-ABC Triangles
Archbold, 200455
Beebe, 20048
Attention and Executive Functioning
Avior, 200457
Huang, 200458
Blunden, 200527
Emancipator, 200670
(+) PPVT-R
Kurnatowski, 200663 (+) Token Test
(+) indicates deficits shown in children with sleep-disordered breathing; (-) indicates null group findings; (+/-) indicates mixed findings; WISC-III
refers to the Wechsler Intelligence Scale for Children (3rd edition); VIQ, Verbal IQ, PIQ, Performance IQ; PPVT, Peabody Picture Vocabulary Test;
WPPSI-R, Wechsler Preschool and Primary Scales of Intelligence-Revised; WASI, Wechsler Abbreviated Scales of Intelligence; DAS, Differential Abilities Scale; VMI, Beery Test of Visual-Motor Integration; K-ABC, Kaufman Assessment Battery for Children; CMS, Childrens Memory
Scale; CPT, Continuous Performance Test; MFFT, Matching Familiar Figures Test; vis, visual; mem, memory; verb, verbal; Pict, pictorial; vocab,
vocabulary.
SLEEP, Vol. 29, No. 9, 2006
1125
data were statistically significant only when SDB was viewed on

a rough continuum, rather than in discrete severity groups.
On office-based tests, there is little evidence of deficits in
short-term working-memory functioning among children with
SDB, as evidenced in a consistent lack of findings on measures
that require immediate recall of strings of rote verbal or visual
material, even when clinical or large samples are used.6,8,23,27,42,5
0,54,55
The ability to sustain attention or vigilance over time and
to inhibit impulsive responding during continuous performance
tests116 has been reported to be deficient in children with SDB
by most authors,8,27,28,50,55,57,58 though not all. Positive results from
continuous performance tests do not appear to be the result of
inflationary bias, as several studies used community-recruited
samples or defined SDB on clinical symptoms alone. Similarly
diverse studies have found SDB-associated deficits on cancellation tests that require rapid visual scanning and screening for target figures.6,8,23,35,48,50 As shown in 4, few other office-based tasks
have been used across multiple studies, and findings have been
especially limited on tests that assess mental flexibility, planning,
abstract reasoning, and concept-formation.
The importance of considering both office-based test results
and more applied outcome data is also highlighted when
examining scholastic functioning. Multiple studies have found that
childhood SDB is associated with poor academic performance,
as indexed in school records21,32,44,45 or parent report.13,30,34,40,41,51,61
Indeed, PSG-defined SDB may be more strongly associated with
learning problems than with male sex, obesity, or parent-reported
sleepiness, and the presence of both parent-reported snoring
and learning problems is strongly predictive of abnormal PSG
results.39 However, both studies that used office-based academic
achievement testing found that children with SDB and controls
had almost identical scores in core academic skills.42,70 Because a
childs classroom functioning can not be distilled into an officebased test, it is likely that the lower academic grades obtained
by children with SDB are the result of a combination of factors,
perhaps related more to behavioral functioning than to knowledge.
Indeed, children with SDB may be particularly vulnerable to
the effects of behavioral disturbance on academic performance.
Urschitz et al44 reported that poor academic performance was
much more closely related to hyperactivity, concentration deficits,
and daytime sleepiness among habitual snorers than among
nonsnorers. Although such poor academic performance may
not be immediately related to core academic skill deficits, over
time it may impede skill development, especially in vulnerable
subgroups.70
sample with PSG-defined OSA.50 Finally, 2 studies have associated SDB with deficits in early phonologic skills, which are considered precursors to reading development.6,50 The latter 2 samples
were carefully culled from a larger community recruitment and
used rigorous methods but came from the same research group
and consequently warrant independent replication.
There is no consistent evidence of a marked deficit in visual
perception, even in studies that might have been biased toward
finding results due to their use of clinical samples.8,23,35,54 There
has been some suggestion of a constructional deficit, particularly
on tests that require the child to build or construct geometric designs with colored tiles or blocks,6,28,50 but several authors have
not replicated this deficit.8,23,27 The ability to copy line drawings
has generally not been found to be deficient in children with
SDB.6,23,35,50 However, 1 group suggested that copying a highly
complex figure, which requires higher level organization and
planning skills,105 may be vulnerable.63
As shown in Table 4, 6 studies have failed to find evidence
of problems with new learning or memory among children with
SDB, as compared to 4 studies that found such evidence and 1 that
yielded mixed results. None of the biasing factors discussed earlier varied systematically with whether significant memory effects
were found. Although similarly mixed results have been reported
in the adult OSA literature,106 the lack of consistency across studies in the memory domain is somewhat surprising because both of
the proposed mechanisms by which SDB might impact neurobehavioral functioningsleep disruption and intermittent hypoxia
(see below)would be expected to differentially affect memory systems.107-109 It may be that some aspects of memory (e.g.,
memory acquisition with repetition42) are especially vulnerable.
Sleep-deprivation studies have found that overnight consolidation
of information, particularly memory for procedures, is most affected.107,108 This is impossible to test in children who are seen for
assessment in a single-day session, and procedural memory (as
opposed to declarative memory, which can be formally stated)
is rarely tested in children, even in experimental protocols.104
Attention and executive functioning, which have been proposed
by several authors to be vulnerable to OSA,106,110-112 are comprised
of a heterogeneous set of skills that are believed to be important in
higher-level, goal-directed adaptive functioning.97,105,113-115 Because
the conceptual boundary between primary attention processes and
higher-level executive functioning is fuzzy, they are considered
together here. Perhaps more than in any other neurobehavioral
domain, how attention and executive functioning are assessed
warrants careful consideration. Office-based assessments are objective, highly standardized, norm-referenced, and able to parse
out relatively fine-grained distinctions between cognitive skills,
all of which are advantages over third-party report methods (e.g.,
parent-report questionnaires). However, the rigidly standardized
procedures and office-based setting likely limit the aspects of
functioning that can be assessed, particularly in the domains of attention and executive functioning.97,104 In these cases, third-party
report or direct observation in a childs natural environment may
provide important information that complements office-based
findings.97,104 Difficulties regulating behaviors, emotions, and attention in the natural setting have been detailed above. Beebe et
al8 also found that parents of children with SDB reported problems flexibly adapting to changing demands, poor work initiation
and self-monitoring, and diminished planning and organization.
Teachers reported similar difficulties, although in most cases the
SLEEP, Vol. 29, No. 9, 2006
Summary of Findings and Implications for Causality

Research to date on childrens behavioral and emotional
functioning in the natural setting provides strong evidence that
SDB is associated with problems with attention and behavior
regulation. The most consistent finding has been an association
with impulsivity/hyperactivity and other externalizing behaviors
such as aggression, but there is also evidence for attention
problems of similar magnitude, falling in the moderate range on
conventional criteria. Moderate to large effect sizes also are evident
on measures of emotional stability, even though the association
between SDB and a childs typical mood appears limited. Daytime
sleepiness is not as obvious in childhood SDB as in adult SDB,
and sleepiness may not explain other neurobehavioral deficits.
1126
tant implications for mechanistic models of SDB-related morbidity. In the meantime, findings suggest that SDB is associated with
adverse neurobehavioral outcomes in several domains, which is
necessary but not sufficient to infer causality. Furthermore, such
correlational associations say little about the potential direction
of causality. Support for inferences about a causal direction must
come from other sources, including evidence of plausible biologic mechanisms of effect.
However, daytime sleepiness is often present to some degree and

represents an important outcome for clinical care and research
use.
Office-based cognitive testing suggests that most children with
SDB function within the average range of overall intellectual
ability. However, there is increasing evidence of reduced
performance on intelligence tests among individuals with SDB,
particularly during the preschool years. This may represent a
unique developmental vulnerability or the fact that somewhat
differing subtests comprise intelligence tests at different ages.
There is little evidence for expressive language problems in
children with SDB, although there is some evidence of problems
following oral instructions and difficulties with early phonologic
processing skills. Although there is some evidence of problems
reproducing complex drawings, data generally do not support
a link between SDB and visual perception or construction.
Memory-test findings have been surprisingly mixed, though it
may be that some memory processes are more vulnerable than
others. There is evidence of an association between SDB and
poor performance on continuous performance tests that rely upon
sustained attention and inhibition, as well as on cancellation tests
that require speed, visual scanning, and selective attention. In
contrast, there is little evidence of a primary short-term working
memory deficit, and inconsistency across measures of higherlevel reasoning skills precludes summary statements on those
skills at this time. Although the results of office-based testing
are important, one should be mindful that these may not predict
real-life performance. For example, poor academic performance
is associated with childhood SDB, but this may be less related to
core knowledge than to the interference of behavioral symptoms
in the classroom.
These findings bear on 3 of the conventional guidelines for
inferring causality: strength of association, consistency of findings,
and specificity of effect. 77-80 Causal inferences are reinforced by
strong associations that are replicable across multiple studies using
different populations, methods, and measures. Strong consistent
associations minimize the chances that unappreciated confounding
factors can account for the observed relationships. SDB has such
associations with behavioral and emotional dysregulation, poor
academic performance, and poor performance on attention tests.
Impressively, these associations have been remarkably robust
across studies from diverse national and cultural groups. Beyond
statistical effect size and consistency, when research indicates that
a purported cause is associated with one plausible set of outcomes
and not another, this specificity of effect further argues against
confounding factors (e.g., reporting bias) that might otherwise
affect a wide range of unrelated outcomes. Because it is consistent
with the theoretically posited impact of SDB on prefrontal
functioning,110 the fact that questionnaires indicate much greater
evidence of SDB-associated emotional dysregulation than adverse
typical mood fits the specificity guideline.
Nevertheless, there is considerable room for further research.
Given that much of a childs day is spent outside of the home,
there is a need to replicate behavioral findings using sources other
than the parent, such as objective measures in naturalistic settings
(e.g., standardized observations92). Moreover, developing a better
understanding of how SDB might be related to academic functioning could lead to more-effective preventative and remedial efforts. Finally, more-sophisticated investigations into the memory
and executive functioning of children with SDB will have imporSLEEP, Vol. 29, No. 9, 2006
PROPOSED MECHANISMS
According to the biologic plausibility guideline, causal inference is bolstered when it makes sense for an agent to cause
an outcome based upon what is known about human physiology.
Two major features of OSA act as starting points for the mechanistic pathways that have been most often proposed to explain the
associations between SDB and neurobehavioral deficits: intermittent hypoxia and sleep disruption.12,110,117,118 Anoxia or chronic
hypoxia is known to have adverse cerebrovascular outcomes, and
some have suggested that this may underlie the neurobehavioral
deficits observed in adults with OSA.119 However, it is appears
unlikely that such a cause, which presumably is applicable most
to adults with severe or long-standing disease, would come into
play for the large majority of children with SDB who show only
brief and relatively mild hypoxic episodes. Cerebrovascular effects aside, Drs. David Gozal and Barry Row have developed an
animal model that more directly links intermittent hypoxia to a
cascade of events that include increased oxidative stress, inflammatory processes, and structural and neurochemical changes in
the hippocampus and prefrontal cortex that ultimately lead to
functional deficits.109,120-125 Although it is an open question how
well these findings translate to humans, it is noteworthy that
adults and children with OSA have been found to have elevated
inflammatory markers, as well as precursors of such inflammation.126-128 In addition, a recent review found that children with a
wide variety of conditions involving hypoxia present with significant neurobehavioral deficits.129
Even so, intermittent hypoxia is probably not the sole starting
point for neurobehavioral deficits in childhood SDB. The severity of hypoxic events generated in animal models is considerably
greater than that typically seen in children with SDB, and, indeed,
neurobehavioral deficits have been documented in children with
negligible degrees of hypoxia.6,7,44,45,56 As reviewed by Blunden
and Beebe,117 SDB-related sleep disruption may be sufficient to
result in neurobehavioral deficits in these children. These authors
noted that deficits have been observed in children with nonSDBrelated sleep disruption130-134 and that experimental sleep restriction in healthy children can induce some of these deficits.133,135,136
Moreover, when combined with SDB, the chronic sleep restriction or fragmentation that can result from other dyssomnias may
have an additive or synergistic effect on neurobehavioral functioning.27,31 Interestingly, sleep deprivation and intermittent hypoxia may have parallel or shared mechanistic pathways. Experimental sleep deprivation results in elevated peripheral markers
for inflammation in otherwise healthy humans.137, 138 In rodents,
sleep deprivation results in oxidative stress,139,140 inflammatory
processes,141 inhibition of hippocampal long-term potentiation
and neurogenesis,142,143 and dysfunction of neural systems that are
involved in learning, motivation, goal direction, reward, and attentional capacity.142-144 Despite these leads, however, research lit1127
normalities that have been observed after intermittent hypoxia or

sleep deprivation in rodents have been subtle and best appreciated by direct inspection of brain parenchyma.109,120-125,142,143 From
a neurodevelopmental perspective, significant long-term deficits
can occur even without morbid tissue loss. The human brain undergoes considerable maturation during childhood, with anterior
regions coming to maturation in the second or third decades of
life.155,156 This maturation reflects an ongoing interaction between
genetic programming and the physiologic environment at the
cellular level (e.g., neuroelectrical/chemical stimulation, neurotrophic, and cerebral metabolic factors).102 Intermittent hypoxia
and sleep disruption may cause aberrations in that cellular environment directly (e.g., through alterations in neurotransmitter
concentrations120) or indirectly via the childs interactions with
the outside world (e.g., missed learning opportunities, negative
reactions to a childs disruptive behaviors157). As adverse conditions persist, the fundamental organization of neural tissue may
develop quite differently from that which might have occurred
under more favorable conditions, setting the stage for long-term
functional deficits.158 Given this, and the fact that numerous cognitive skills are developing rapidly during childhood, it is not
surprising that some authors have asserted that children may be
particularly vulnerable to the adverse effects of SDB.12,110,118,159
Indeed, there is evidence that intermittent hypoxia results in the
greatest neuronal and neurochemical abberrations in rats of an
age roughly analogous to human childhood120,125 and that, in other
animals, sleep deprivation can derail neural and skill progression
at critical points in early development.160,161
erature has been mixed on the presence and significance of sleep

disruption in children with SDB, possibly due to the misapplication of adult criteria to children.117
Correlations between neurobehavioral functioning and conventional PSG indexes have provided little insight. More often than
not, statistically significant findings are nested within a large
matrix of nonsignificant findings. For example, Kaemingk et al42
presented a matrix of 165 correlations, only 19 of which were
significant at the p < .05 level. Similarly, Beebe et al8 reported that
only 9 of nearly 750 correlations were significant at the p<.005
level. This would not be a major concern if multiple studies had
reported similar findings. However, different authors have implicated hypoxia,29,70 nocturnal movements,31 arousals,50 rapid eye
movement latency and percentage,6,49 percentage of slow-wave
sleep,8 percentage of stage 1 sleep,42 and number of obstructive
events per hour35,40,70 or counted in total19 in the daytime effects
of childhood SDB. None of these indexes has received support in
even half of the relevant studies.
There also has been little evidence for a threshold effect on
any PSG-based index, with the best evidence to date suggesting that any such threshold might be best detected using clinical
symptoms (e.g., snoring) rather than PSG. In some cases, children with relatively mild pathology on PSG have been reported to
have worse neurobehavioral deficits than severe cases and healthy
controls alike.8,35,49 Several recent studies have indicated that parent-reported chronic snoring in children is associated with poor
neurobehavioral outcome regardless of the presence of PSG-defined OSA.6-8,56 These studies have been derived both from clinical samples (which may introduce bias but ensure the presence of
relatively severe cases)8,56 and community samples (which minimize bias and increase generalizability).6,7
The neural regions that have been proposed to be most susceptible to SDB vary, with major candidates including the prefrontal cortex,110 subcortical gray matter,119 hippocampus,145 and
white matter more broadly.119 It is important to keep in mind that
these tissues are richly interconnected, and pathology at multiple
points in a system may lead to functional deficits. Also, it remains
possible that selective aspects of SDB may differentially impact
different structures or systems (e.g,. sleep disruption more heavily
affecting prefrontal cortex, intermittent hypoxia more heavily affecting the hippocampus), but this has not yet been demonstrated
in humans with SDB.
Whether the neurobehavioral deficits observed in childhood
SDB are associated with pathology that can be observed on neuroimaging remains to be seen, as no relevant studies had been
published as of early 2006. Structural magnetic resonance imaging studies of adults with OSA have yielded inconsistent and
sometimes null findings,146-149 but magnetic resonance spectroscopy studies have pointed toward abnormalities in the hippocampus
and frontal white matter but not the prefrontal cortex or parietal
white matter.150-153 Functional magnetic resonance imaging of activation patterns during a working memory task has further indicated poor activation of the dorsolateral prefrontal cortex in adults
with OSA.154 Many of these techniques can be applied to children,
and it will be intriguing to observe the findings of similar studies
that are in the works at several pediatric centers.
As we await these findings, it is important to note that the functional deficits observed in children with SDB may result from
subtle neuropathology and neurochemical abnormalities that are
not readily observed on conventional structural imaging. The abSLEEP, Vol. 29, No. 9, 2006
Summary of Findings and Implications for Causality

Whether we have evidence of biologic plausibility appears to
be in the eye of the beholder. Several mechanisms have been proposed by which SDB might result in neurobehavioral morbidity.
However, none has yet proven to be necessary and sufficient for
generating neurobehavioral deficits in children with SDB. In future work, it will be important to focus on translation of animal
models to human children, focusing on noninvasive markers for
processes that have been causally explicated in other species. Following up on animal findings (e.g., see reference 162), identification of vulnerable subgroups of humans promises to allow for
more targeted screening and intervention efforts.12 Explication of
possible developmental vulnerability is of particular interest to
those who work with pediatric samples.
Findings have also been mixed with respect to the biologic
gradient or dose-response guideline. This guideline suggests that
causal inferences are best supported by an incremental relationship between degrees of exposure to the purported causal agent
and its outcome (e.g., smoking frequency and lung cancer risk).
Children who snore are at increased risk for adverse neurobehavioral outcomes as compared with those who do not snore, but linear and ordinal associations between PSG-based indexes of SDB
severity and neurobehavioral functioning have been inconsistent
and small. It may be that there is a nonlinear threshold effect to
which conventional PSG indexes are insensitive; authors disagree
on whether the dose-response guideline applies to nonlinear relationships.73,75 Alternatively, in line with research on other public
health concerns (e.g., smoking, childhood lead exposure), more
robust associations may result from other definitions of exposure
(e.g., duration of SDB, severity during a sensitive developmental
1128
period). Examination of such possibilities will require carefully

designed prospective studies that are based upon theoretic models
of effect. Such prospective studies might also provide evidence
of the temporal relationship between SDB and neurobehavioral
deficits.
Recove ry ?
Ability
RELATIONSHIP BETWEEN SDB AND NEUROBEHAVIORAL FUNCTIONING OVER TIME

Although not sufficient in isolation to infer causality, documentation that a proposed cause was present prior to its purported consequence rules out a reversed causal sequence. The few studies
on the natural history of SDB and its relationship to neurobehavioral functioning over time have yielded data that are consistent
with this temporal sequence guideline. The first involved a 2-year
follow up on an epidemiologic sample of children from Oxford,
England, who were initially seen at 4 to 5 years of age.2,16,17 Although the rate of snoring stayed constant in the group as a whole,
only about half of the children who initially snored were reported
to do so at follow-up. Compared with those who snored at neither
time point, those who snored at either point had twice the rate
of reported hyperactivity at follow-up. Those who snored at both
points had more than triple the rate of reported hyperactivity at
follow-up. These data suggest that early snoring, even when it
resolves, increases long-term risk for daytime behavioral disturbance, and persistent snoring further increases that risk. Similar
findings were reported in a 1-year follow-up of a group of primary-school children from Hannover, Germany,44 though there was
evidence of spontaneous improvement in behavioral disturbances among children whose snoring resolved over time. Recently,
Chervin et al38 reported that, in a general pediatric sample from
Michigan, snoring at age 2 to 13 years was predictive of emergent
or exacerbated parent-reported hyperactivity 4 years later, after
controlling for baseline hyperactivity and for snoring at the follow-up. The longest-term follow-up data on children with SDB
was collected via retrospective parent report. Gozal and Pope32
surveyed the parents of high-performing and low-performing
students aged 13 to 14. After excluding students reported to currently snore from analyses, low-performing students were disproportionately reported to have snored at ages 2 to 6 years and to
have undergone adenotonsillectomy for snoring, whereas equal
numbers across groups had undergone adenotonsillectomy for recurrent infections.
Further evidence of temporality comes from SDB treatment trials. Adenotonsillectomy is often effective in treating OSA, and
short-term neurobehavioral gains have been reported in nonrandomized trials. Uncontrolled trials have reported posttreatment
improvements in reported child behavior25,57,67,69 and overall IQ
and sustained attention on office-based tests.52,57 However, without
a control group, parent expectancy and practice/exposure effects
complicate interpretation of these findings. A few studies have
also tracked non-SDB control groups over time, observing significantly greater gains in the treated-SDB group in parent-reported
inattention, hyperactivity, and aggression,15,18,59 though again it
is not known how much parents might expect behavioral gains
following adenotonsillectomy. More impressively, office-based
testing has suggested greater gains in working memory, sustained
attention, and visual scanning/attention18,54 but not overall IQ65 in
children treated with adenotonsillectomy than in control groups.
In the longest posttreatment follow-up available, Gozal21 reported
SLEEP, Vol. 29, No. 9, 2006
a
He
SDB
y
lt h
Delay?
Deficit?
SDB treatment
Time
Figure 4Hypothetical Posttreatment Developmental Trajectories.
In the diagram, sleep-disordered breathing (SDB) is assumed to cause
a period of stagnation in the acquisition of a skill, such as impulse
control. Following treatment, there are several possible outcomes, including a steep recovery until the child catches up with his or her
peers; resumption of development at a normal pace, which results in
a delay in development; or resumption of development at a somewhat
diminished pace, resulting in a long-term skill decit.
longitudinal data on a group of poorly performing first-grade students. He found that those who underwent adenotonsillectomy
for SDB went on to improve their grades the following school
year, whereas untreated children with or without SDB showed
little change in their grades over time.
No published study has yet examined long-term outcome after treatment for SDB, nor have even short-term neurobehavioral
data been published from a truly randomized clinical intervention
trial. Both are needed to better inform clinical care. In the meantime, current findings invite speculation regarding the possible
trajectory of neurobehavioral development following SDB treatment. As schematically illustrated in Figure 4, full recovery of a
skill due to a period of rapid catch-up development might be
optimistically predicted based upon treatment outcome findings,
but to date these studies have been short term in nature. Longer-term naturalistic studies suggest that SDB symptoms remit
spontaneously in many children. However, snoring during childhood, even if it remits, appears to be a risk factor for neurobehavioral deficits 4 years later and for academic deficits a decade
later. These findings suggest a risk for trajectories that result in
long-term delay or deficit. If indeed SDB causes neurobehavioral
deficits, determination of the modal recovery trajectories for key
neurobehavioral deficits will be important for intervention planning and prioritization of early detection and treatment efforts.
CONCLUSIONS
The current research literature allows for several reasonable
conclusions. First, there is strong evidence, even in studies that
used research designs that would minimize bias or tend to suppress effects, that childhood SDB is associated with deficits in
behavior regulation, perhaps paralleled by deficits in emotion
regulation as well. There also is strong evidence for inverse associations between SDB and sustained attention, selective attention, and alertness. Because the office-based testing setting may
be ill-suited to detect such effects, it is essential to consider outcome measures that reflect the childs functioning in their natural
setting. Indeed, applied outcomes such as academic grades have
shown some of the most robust associations with childhood SDB.
In contrast, there is good evidence, even in studies that used research designs that maximize statistical power or tend to inflate
1129
thanks Sarah Blunden, PhD, two anonymous reviewers, and the

Cincinnati Childrens Neuropsychology Group for feedback on
an earlier draft of this paper.
effects, that SDB has a small to negligible effect on childrens

typical mood, expressive language skills, visual perception, and
working memory skills. Finally, there are several domains of
neurobehavioral functioning in which findings have been mixed
or otherwise insufficient to draw firm conclusions. Recent welldesigned studies suggest the presence of a subtle effect of SDB
on overall intellectual functioning, particularly in young children.
However, in most cases, the overall IQ of children with SDB has
clustered around the population average. Until the inconsistency
in findings across studies is better understood, conclusions regarding intellectual outcome should be made with caution. Similarly, there is theoretical reason to suspect deficits in memory and
some higher-level aspects of executive functioning (e.g., problem
solving) in children with SDB, but the clinical research data are
simply inadequate to draw firm conclusions at this time.
Several mechanisms by which SDB might result in neurobehavioral morbidity are being actively explored. Ironically, despite
the application of sophisticated methods to the problem, oldfashioned clinical symptoms of SDB such as chronic snoring
remain the best predictors of neurobehavioral deficits. The longterm neurobehavioral outcome of children treated for SDB is not
known, though short-term treatment outcome studies are encouraging. It does appear that failing to treat SDB leaves children at
risk for long-term neurobehavioral deficits. For clinicians, these
findings should build confidence in the now-familiar call for early
screening for SDB in primary care settings. Straightforward questions about sleep (e.g., Owens BEARS approach) have proven
well-suited for this purpose.163 Similarly, simple questions about a
childs daytime behaviors can yield important information about
whether he or she is progressing academically or is experiencing problems regulating his or her behaviors, emotions, attention,
or arousal. The convergent presence of symptoms of SDB and
neurobehavioral deficits should be a red flag that a referral for
specialist follow-up is needed.
This paper considered several commonly applied guidelines
for causal inference. It appeared that guidelines related to strength
of association, consistency of findings, and specificity of effect
were well supported by the extant research. In contrast, there has
been little support for a dose-response relationship between SDB
and neurobehavioral functioning, though it is not clear whether
this might be related to a threshold effect, insensitive measures, or
examination of the wrong dose variables. There is evidence that
SDB fits the expected temporal pattern and is biologically plausible as a cause of neurobehavioral deficits, but different readers
may draw different conclusions about whether these guidelines
have been satisfied. Indeed, because of the inherently subjective
nature of causal inference,72,77,78 it is expected that some readers
may remain skeptical about whether SDB causes daytime deficits.
In making the dialogue more explicit, it is hoped that this paper
will facilitate debate on whether, and how, SDB might contribute
to neurobehavioral deficits in children. Given the relatively high
base rate of SDB in the general population, the answers may have
major public health implications.
REFERENCES
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
ACKNOWLEDGMENTS
Preparation of this article was supported in part by grant #K23
HL075369 from the National Institutes of Health. The author
SLEEP, Vol. 29, No. 9, 2006
19.
1130
American Academy of Pediatrics Section on Pediatric Pulmonology. Clinical practice guideline: Diagnosis and management of childhood obstructive sleep apnea syndrome. Pediatrics 2002;109:704712.
Ali NJ, Stradling JR. Epidemiology and Natural History of Snoring and Sleep-Disordered Breathing in Children. In: Loughlin GM,
Carroll JL, Marcus CL, eds. Sleep and Breathing in Children: a
Developmental Approach. Vol. 147. New York: Marcel Dekker;
2000:555-74.
Rosen CL. Obstructive sleep apnea syndrome in children: Controversies in diagnosis and treatment. Pediartr Clin North Am
2004;51:153-67.
Guilleminault C, Khramtsov A. Upper airway resistance syndrome
in children: A clinical review. Semin Pediatr Neurol 2001;8:20715.
Guilleminault C, Winkle R, Korobkin R, Simmons B. Children
and nocturnal snoring: Evaluation of the effects of sleep related
respiratory resistive load and daytime functioning. Eur J Pediatr
1982;139:165-71.
O'Brien LM, Mervis CB, Holbrook CR, et al. Neurobehavioral implications of habitual snoring in children. Pediatrics 2004;114:449.
Rosen CL, Storfer-Isser A, Taylor HG, Kirchner HL, Emancipator
JL, Redline S. Increased behavioral morbidity in school-aged children with sleep-disordered breathing. Pediatrics 2004;114:1640-8.
Beebe DW, Wells CT, Jeffries J, Chini B, Kalra M, Amin R. Neuropsychological effects of pediatric obstructive sleep apnea. J Int
Neuropsychol Soc 2004;10:962-75.
Hill W. On some causes of backwardness and stupidity in children:
And the relief of these symptoms in some instances by nasopharyngeal scarifications. Br Med J 1889;2:711-2.
Stool SE, Eavey RD, Stein NL, Sharrar WG. The "chubby puffer"
syndrome. Upper airway obstruction and obesity, with intermittent
somnolence and cardiorespiratory embarrassment. Clin Pediatr
1977;16:43-50.
Guilleminault C, Eldridge FL, Simmons FB, Dement WC. Sleep
apnea in eight children. Pediatrics 1976;58:23-31.
Beebe DW. Neurobehavioral effects of obstructive sleep apnea: An
overview and heuristic model. Curr Opin Pulm Med 2005;11:494500.
Weissbluth M, Davis AT, Poncher J, Reiff J. Signs of airway obsrtruction during sleep and behavioral, developmental, and academic problems. Dev Behav Pediatr 1983;4:119-21.
Simonds JF, Parraga H. Sleep behaviors and disorders in children
and adolescents evaluated at psychiatric clinics. Dev BehavPediatr
1984;5:6-10.
Stradling JR, Thomas G, Warley ARH, Williams P, Freeland A.
Effect of adenotonsillectomy on nocturnal hypoxaemia, sleep disturbance, and symptoms in snoring children. Lancet 1990;335:24953.
Ali NJ, Pitson DJ, Stradling JR. Snoring, sleep disturbance, and behaviour in 4-5 year olds. Arch Dis Child 1993;68:360-6.
Ali NJ, Pitson DJ, Stradling JR. Natural history of snoring and related behaviour problems between the ages of 4 and 7. Arch Dis
Child 1994;71:74-6.
Ali NJ, Pitson DJ, Stradling JR. Sleep disordered breathing: Effects
of adenotonsillectomy on behaviour and psychological functioning.
Eur J Pediatr 1996;155:56-62.
Rhodes SK, Shimoda KC, Waid R, et al. Neurocognitive deficits
in morbidly obese children with obstructive sleep apnea. J Pediatr
1995;127:741-4.
20. Chervin RD, Dillon JE, Bassetti C, Ganoczy DA, Pituch KJ. Symptoms of sleep disorders, inattention, and hyperactivity in children.
Sleep 1997;20:1185-92.
21. Gozal D. Sleep disordered breathing and school performance in
children. Pediatrics 1998;102:616-20.
22. Smedje H, Broman J-E, Hetta J. Parents' reports of disturbed sleep
in 5-7-year-old Swedish children. Acta Paediatrica. 1999;88:858865.
23. Owens J, Spirito A, Marcotte AC, McGuinn M, Berkelhammer L.
Neuropsychological and behavioral correlates of obstructive sleep
apnea syndrome in children: A preliminary study. Sleep Breath
2000;4:67-8.
24. Goldstein NA, Fatima M, Campbell TF, Rosenfeld RM. Child behavior and quality of life before and after tonsillectomy and adenoidectomy. Arch Otolaryngol Head Neck Surg 2002;128:770-5.
25. Goldstein NA, Post JC, Rosenfeld RM, Campbell TF. Impact of tonsillectomy and adenoidectomy on child behavior. Arch Otolaryngol,
Head Neck Surg 2000;126:494-8.
26. Ferreira AM, Clemente V, Gozal D, et al. Snoring in Portuguese
primary school children. Pediatrics 2000;106:1128-9.
27. Blunden S, Lushington K, Lorenzen B, Martin J, Kennedy D. Neuropsychological and psychosocial function in children with a history
of snoring or behavioral sleep problems. J Pediatr 2005;146:780-6.
28. Blunden S, Lushington K, Kennedy D, Martin J, Dawson D. Behavior and neurocognitive performance in children aged 5-10 years who
snore compared to controls. J Clin Exp Neuropsychol 2000;22:55468.
29. Kennedy JD, Blunden S, Hirte C, et al. Reduced neurocognition in
children who snore. Pediatr Pulmonol 2004;37:330-7.
30. Richards W, Ferdman RM. Prolonged morbidity due to delays in the
diagnosis and treatment of obstructive sleep apnea in children. Clin
Pediatr 2000;39:103-8.
31. Chervin RD, Archbold KH. Hyperactivity and polysomnographic
findings in children evaluated for sleep-disordered breathing. Sleep
2001;24:313-20.
32. Gozal D, Pope D. Snoring during early childhood and academic performance at ages thirteen to fourteen years. Pediatrics
2001;107:1394-9.
33. Stein MA, Mendelsohn J, Obermeyer WH, Amromin J, Benca R.
Sleep and behavior problems in school-age children. Pediatrics
2001;107:e60.
34. Brunetti L, Rana S, Lospalluti ML, et al. Prevalence of obstructive
sleep apnea syndrome in a cohort of 1,207 children of southern Italy.
Chest 2001;120:1930-5.
35. Lewin DS, Rosen RC, England SJ, Dahl RE. Preliminary evidence
of behavioral and cognitive sequelae of obstructive sleep apnea in
children. Sleep Med 2002;3:5-13.
36. Chervin RD, Archbold KH, Dillon JE, et al. Inattention, hyperactivity, and symptoms of sleep-disordered breathing. Pediatrics
2002;109:449-56.
37. Chervin RD, Dillon JE, Archbold KH, Ruzicka DL. Conduct problems and symptoms of sleep disorders in children. J Am Acad Child
Adolesc Psychiatry 2003;42:201-8.
38. Chervin RD, Ruzincka DL, Archbold KH, Dillon JE. Snoring predicts hyperactivity four years later. Sleep 2005;28:885-90.
39. Goodwin JL, Kaemingk KL, Mulvaney SA, Morgan WJ, Quan SF.
Clinical screening of school children for polysomnography to detect
sleep-disordered breathinghe Tucson Children's Assessment of
Sleep Apnea Study (TuCASA). J Clin Sleep Med 2005;1:247-54.
40. Goodwin JL, Kaemingk KL, Fregosi RF, et al. Clinical outcomes associated with sleep-disordered breathing in Caucasian and Hispanic
childrenthe Tucson Children's Assessment of Sleep Apnea study
(TuCASA). Sleep 2003;26:587-91.
41. Goodwin JL, Babar SI, Kaemingk KL, et al. Symptoms related to
sleep-disordered breathing in white and hispanic children. Chest
2003;124:196-203.
SLEEP, Vol. 29, No. 9, 2006
42. Kaemingk KL, Pasvogel AE, Goodwin JL, et al. Learning in children and sleep-disordered breathing: findings of the Tucson Children's Assessment of Sleep Apnea (TuCASA) prospective cohort
study. J Int Neuropsychol Soc 2003;9:1016-26.
43. Mulvaney SA, Goodwin JL, Morgan WJ, Rosen GR, Quan SF,
Kaemingk KL. Behavior problems associated with sleep disordered
breathing in school-aged children--the Tucson Children's Assessment of Sleep Apnea study. J Pediatr Psychol 2006; 31:322-330.
44. Urschitz MS, Eitner S, Guenther A, et al. Habitual snoring, intermittent hypoxia, and impaired behavior in primary school children.
Pediatrics 2004;114:1041-8.
45. Urschitz MS, Guenther A, Eggebrecht E, et al. Snoring, intermittent
hypoxia, and academic performance in primary school children. Am
J Respir Crit Care Med 2003;168:464-8.
46. Urschitz MS, Wolff J, Sokollik C, et al. Nocturnal arterial oxygen
saturation and academic performance in a community sample of
children. Pediatrics 2005;115:e204-9.
47. Gottlieb DJ, Vezina RM, Chase C, et al. Symptoms of sleep-disordered breathing in 5-year-old children are associated with sleepiness and problem behaviors. Pediatrics 2003;112:870-7.
48. Gottlieb DJ, Chase C, Vezina RM, et al. Sleep-disordered breathing
symptoms are associated with poorer cognitive function in 5-yearold children. J Pedatr 2004;145:458-64.
49. O'Brien L, Holbrook CR, Mervis CB, et al. Sleep and neurobehavioral characteristics in 5-7-year-old children with parentally reported symptoms of attention-deficit hyperactivity disorder. Pediatrics
2003;111:554-63.
50. O'Brien LM, Mervis CB, Holbrook CR, et al. Neurobehavioral
correlates of sleep-disordered breathing in children. J Sleep Res
2004;13:165-72.
51. Montgomery-Downs HE, Jones VF, Molfese VJ, Gozal D. Snoring
in preschoolers: associations with sleepiness, ethnicity, and learning. Clin Pediatr 2003;42:719-26.
52. Montgomery-Downs HE, Crabtree VM, Gozal D. Cognition, sleep
and respiration in at-risk children treated for obstructive sleep apnoea. Eur Respir J 2005;25:336-42.
53. Montgomery-Downs HE, O'Brien LM, Holbrook CR, Gozal D.
Snoring and sleep-disordered breathing in young children: subjective and objective correlates. Sleep 2004;27:87-94.
54. Friedman BC, Hendeles-Amitai A, Kozminzki E, et al. Adenotonsillectomy improves neurocognitive function in children with obstructive sleep apnea syndrome. Sleep 2003;26:999-1005.
55. Archbold KH, Giordani B, Ruzicka DL, Chervin RD. Cognitive executive dysfunction in children with mild sleep-disordered breathing. Biol Res Nurs 2004;5:168-76.
56. Melendres CS, Lutz JM, Rubin ED, Marcus CL. Daytime sleepiness and hyperactivity in children with suspected sleep-disordered
breathing. Pediatrics 2004;114:768-75.
57. Avior G, Fishman G, Leor A, Sivan Y, Kaysar N, Derowe A. The effect of tonsillectomy and adenoidectomy on inattention and impulsivity as measured by the Test of Variables of Attention (TOVA) in
children with obstructive sleep apnea syndrome. Otolaryngol Head
Neck Surg 2004;131:367-71.
58. Huang YS, Chen NH, Li HY, Wu YY, Chao CC, Guilleminault C.
Sleep disorders in Taiwanese children with attention deficit/hyperactivity disorder. J Sleep Res 2004;13:269-77.
59. Tran KD, Nguyen CD, Weedon J, Goldstein NA. Child behavior
and quality of life in pediatric obstructive sleep apnea. Arch Otolaryngol Head Neck Surg 2005;131:52-7.
60. Arman AR, Ersu R, Save D, et al. Symptoms of inattention and
hyperactivity in children with habitual snoring: evidence from
a community-based study in Istanbul. Child Care Health Dev
2005;31:707-17.
61. Sogut A, Altin R, Uzun L, et al. Prevalence of obstructive sleep apnea syndrome and associated symptoms in 3--11-year-old Turkish
children. Pediatr Pulmonol 2005;39:251-6.
62. Montgomery-Downs HE, Gozal D. Snore-associated sleep frag1131
63.
64.
65.
66.
67.
68.
69.
70.
71.
72.
73.
74.
75.
76.
77.
78.
79.
80.
81.
82.
83.
84.
85.
86.
87.
mentation in infancy: mental development effects and contribution

of secondhand cigarette smoke exposure. Pediatrics 2006; 117:
e496-502.
Kurnatowski P, Putynski L, Lapienis M, Kowalska B. Neurocognitive abilities in children with adenotonsillar hypertrophy. Int J Pediatr Otorhinolaryngol 2006;70:419-24.
Carvalho LB, Prado LF, Silva L, et al. Cognitive dysfunction in children with sleep-disordered breathing. J Child Neurol 2005;20:4004.
Harvey J, O'Callaghan MJ, Wales PD, Harris MA, Masters IB. Sixmonth follow-up of children with obstructive sleep apnoea. J Paediatr Child Health 1999;35:136-9.
Nelson S, Kulnis R. Snoring and sleep disturbance among children
from an orthodontic setting. Sleep Breath 2001;5:63-70.
Roemmich JN, Barkley JE, D'Andrea L, et al. Increases in overweight after adenotonsillectomy in overweight children with obstructive sleep-disordered breathing are associated with decreases
in motor activity and hyperactivity. Pediatrics 2006;117:e200-8.
Castronovo V, Zucconi M, Nosetti L, et al. Prevalence of habitual
snoring and sleep-disordered breathing in preschool-aged children
in an Italian community. J Pediatr 2003;142:377-82.
Mitchell RB, Kelly J. Child behavior after adenotonsillectomy for
obstructive sleep apnea syndrome. Laryngoscope 2005;115:20515.
Emancipator JL, Storfer-Isser A, Taylor HG, et al. Variation of cognition and achievement with sleep-disordered breathing in full-term
and preterm children. Arch Pediatr Adolesc Med 2006;159:1-8.
Rothman KJ, Greenland S. Modern Epidemiology Philadelphia:
Lippincott Williams Wilkins; 1998.
Kaufman JS, Poole C. Looking back on "Causal Thinking in the
Health Sciences". Annu Rev Public Health 2000;21:101-19.
US Surgeon General. (US Public Health Service). The Health Consequences of Smoking. 2004.
Ris MD. Causal inference in lead research: Introduction to the special section on the neurobehavioral effects of environmental lead.
Child Neuropsychol 2003;9:1-9.
Rothman KJ, Greenland S. Causation and causal inference in epidemiology. Am J Public Health 2005;95:S144-50.
Hennekins CH, Buring JE. Epidemiology in Medicine Philadelphia:
Lippincott Williams Wilkins; 1987.
Susser M. Causal Thinking in the Health Sciences: Concepts and
Strategies of Epidemiology New York: Oxford University Press;
1973.
Hill AB. The environment and disease: Association or causation?
Proc R Soci Med 1965;58:295-300.
Advisory Committee to the Surgeon General of the Public Health
Service. (US Department of Health, Education, and Welfare). Smoking health. 1964.
Susser M. What is a cause and how do we know one? A grammar for
pragmatic epidemiology. Am J Epidemiol 1991;133:635-48.
Ebert CS, Jr., Drake AF. The impact of sleep-disordered breathing
on cognition and behavior in children: a review and meta-synthesis
of the literature. Otolaryngol Head Neck Surg 2004;131:814-26.
Study links hyperactivity and snoring. The Washington Post
2003;A10.
Campbell D, Fiske D. Convergent and discriminative validation by
the multi-trait - multimethod matrix. Psychol Bull 1959;56:81-105.
Rosen CL. Obstructive sleep apnea syndrome (OSAS) in children:
diagnostic challenges. Sleep 1996;19:S274-7.
Marcus CL. Evaluation of pulmonary function and polysomnography
in obese children and adolescents. Pediatr Pulmonol 1996;21:17683.
Carroll JL, McColley SA, Marcus CL, Curtis S, Loughlin GM. Inability of clinical history to distinguish primary snoring from obstructive sleep apnea syndrome in children. Chest 1995;108:610-8.
Anastasi A. Psychological Testing New York: Macmillan Publishing; 1988.
SLEEP, Vol. 29, No. 9, 2006
88. Diagnostic and Statistical Manual of Mental Disorders. 4th ed.

Washington: American Psychiatric Association; 1994.
89. Owens JA. The ADHD and sleep conundrum: a review. J Dev Behav Pediatr 2005;26:312-22.
90. Achenbach TM, Edelbrock C. Manual for the Child Behavior
Checklist and Revised Child Behavior Profile. Burlington: Univerisity of Vermont; 1981.
91. Conners CK. Conners' Ratings Scales-Revised: Technical Manual
.North Tonawanda, NY: Multi-Health Systems; 1997.
92. Reynolds CR, Kamphaus RW. BASCBehavioral Assessment
System for Children Manual Circle Pines, MN: American Guidance
Services; 1992.
93. Lipsey MW, Wilson DB. Applied Social Research Methods. Practical Meta-Analysis. Vol. 49 Thousand Oaks, CA: Sage; 2001.
94. Cohen CK. Statistical Power Analysis for the Behavioral Sciences,
2nd ed Hillsdale, NJ: Lawrence Erlbaum Associates; 1988.
95. Perrin EC, Stein REK, Drotar D. Cautions in using the Child Behavior Checklist: Observations based on research about children
with a chronic illness. J Pediatr Psychol 1991;16:411-21.
96. Lezak MD, Howieson DB, Loring DW. Neuropsychological Assessment, 4th ed New York: Oxford University Press; 2004.
97. Gioia GA, Isquith PK, Guy SC, Kenworthy L. BRIEFBehavior
Rating Inventory of Executive Function. Odessa, FL: Psychological
Assessment Resources; 2000.
98. Fallone GP, Owens JA, Deane J. Sleepiness in children and adolescents: clinical implications. Sleep Med Rev 2002;6:287-306.
99. Carskadon MA, Dement WC, Mitler MM, Roth T, Westbrook PR,
Keenan S. Guidelines for the Multiple Sleep Latency Test (MSLT):
a standard measure of sleepiness. Sleep 1986;9:519-24.
100. Gozal D, Wang M, Pope DW. Objective sleepiness measures in pediatric obstructive sleep apnea. Pediatrics 2001;108:693-7.
101. Nemeroff CB, Vale WW. The neurobiology of depression: inroads to
treatment and new drug discovery. J Clin Psychiatry 2005;66(Suppl
7):5-13.
102. Anderson V, Northam E, Hendy J, Wrennall J. Developmental
Neuropsychology: a Clinical Approach. Philadelphia: Psychology
Press; 2001.
103. Golan N, Shahar E, Ravid S, Pillar G. Sleep disorders and daytime
sleepiness in children with attention-deficit/hyperactive disorder.
Sleep 2004:261-6.
104. Baron IS. Neuropsychological Evaluation of the Child. New York:
Oxford University Press; 2004.
105. Denckla MB. Measurement of Executive Function. In Lyon, GR,
ed. Frames for Reference of the Assessment of Learning Disabilities. Baltimore, MD: Brookes Publishing, 1994:117-142.
106. Beebe DW, Groesz L, Wells C, Nichols A, McGee K. The Neuropsychological Effects of Obstructive Sleep Apnea (OSA): a
meta-analysis of norm-referenced and case-controlled data. Sleep
2003;26:298-307.
107. Walker MP. A refined model of sleep and the time course of memory formation. Behav Brain Sci 2005;28:51-104.
108. Stickgold R, Walker MP. Memory consolidation and reconsolidation: what is the role of sleep? Trends Neurosci 2005;28:408-15.
109. Payne RS, Goldbart A, Gozal D, Schurr A. Effect of intermittent
hypoxia on long-term potentiation in rat hippocampal slices. Brain
Res 2004;1029:195-9.
110. Beebe DW, Gozal D. Obstructive sleep apnea and the prefrontal
cortex: Towards a comprehensive model linking nocturnal upper
airway obstruction to daytime cognitive and behavioral deficits. J
Sleep Res 2002;11:1-16.
111. Decary A, Rouleau I, Montplaisir J. Recovery of executive functions in OSAS patients after CPAP treatment. Sleep 2000;23:A80.
112. Naegele B, Pepin JL, Levy P, Bonnet C, Pellat J, Feuerstein C.
Cognitive executive dysfunction in patients with obstructive sleep
apnea syndrome (OSAS) after CPAP treatment. Sleep 1998;21:3927.
113. Denckla MB. A theory and model of executive function: a neuro1132
psychological perspective. In: Lyon GR, Krasnegor NA, eds. Attention, Memory, and Executive Function. Baltimore: Paul H. Brookes
Publishing; 1996.
114. Eslinger PJ. Conceptualizing, describing, and measuring components of executive function, a summary. In Lyon, GR, Krasnegor,
NA , eds. Attention, Memory, and Executive Function. Baltimore,
MD: Brookes Publishing, 1996:367-95.
115. Pennington B, Bennetto L, McAleer O, Roberts RJ. Executive
functions and working memory, theoretical and measurement issues.
In Lyon, GR, Krasnegor, NA , eds. Attention, Memory, and Executive
Function. Baltimore, MD: Brookes Publishing, 1996:327-348.116.
Riccio CA, Reynolds CR, Lowe PA. Clinical Applications of
Continuous Performance Tests New York: Wiley; 2001.
117. Blunden S, Beebe DW. The contribution of intermittent hypoxia,
sleep debt and sleep disruption to daytime performance deficits
in children: consideration of respiratory and non-respiratory sleep
disorders. Sleep Med Rev 2006:In Press.
118. Halbower AC, Mahone EW. Neuropsychological morbidity linked
to childhood sleep-disordered breathing. Sleep Med Rev 2006:In
Press.
119. Aloia MS, Arnedt JT, Davis JD, Riggs RL, Byrd D.
Neuropsychological sequelae of obstructive sleep apnea-hypopnea
syndrome: a critical review. J Inter Neuropsychol Soc 2004;10:77285.
120. Kheirandish L, Gozal D, Pequignot JM, Pequignot J, Row BW.
Intermittent hypoxia during development induces long-term
alterations in spatial working memory, monoamines, and dendritic
branching in rat frontal cortex. Pediatr Res 2005;58:594-9.
121. Xu W, Chi L, Row BW, et al. Increased oxidative stress is associated
with chronic intermittent hypoxia-mediated brain cortical neuronal
cell apoptosis in a mouse model of sleep apnea. Neuroscience
2004;126:313-23.
122. Li RC, Row BW, Kheirandish L, et al. Nitric oxide synthase and
intermittent hypoxia-induced spatial learning deficits in the rat.
Neurobiol Dis 2004;17:44-53.
123. Row BW, Liu R, Xu W, Kheirandish L, Gozal D. Intermittent
hypoxia is associated with oxidative stress and spatial learning
deficits in the rat. Am J Respir Crit Care Med 2003;167:1548-53.
124. Row BW, Kheirandish L, Neville JJ, Gozal D. Impaired spatial
learning and hyperactivity in developing rats exposed to intermittent
hypoxia. Pediatr Res 2002;52:449-53.
125. Gozal E, Row BW, Schurr A, Gozal D. Developmental differences
in cortical and hippocampal vulnerability to intermittent hypoxia in
the rat. Neurosci Lett 2001;305:197-201.
126. Mills PJ, Dimsdale JE. Sleep apnea: a model for studying cytokines,
sleep, and sleep disruption. Brain Behav Immun 2004;18:298-303.
127. Larkin EK, Rosen CL, Kirchner HL, et al. Variation of C-reactive
protein levels in adolescents: association with sleep-disordered
breathing and sleep duration. Circulation 2005;111:1978-84.
128. O'Brien LM, Gozal D. Autonomic dysfunction in children with
sleep-disordered breathing. Sleep 2005;28:747-53.
129. Bass JL, Corwin M, Gozal D, et al. The effect of chronic or
intermittent hypoxia on cognition in childhood: a review of the
evidence. Pediatrics 2004;114:805-16.
130. Minde K, Faucon A, Falkner S. Sleep problems in toddlers: Effects
of treatment on their daytime behavior. J Am Acad Child Adolesc
Psychiatry 1994;33:114-21.
131. Picchietti DL, Underwood DJ, Farris WA, et al. Further studies on
periodic limb movement disorder and restless legs syndrome in
children with attention-deficit hyperactivity disorder. Mov Disord
1999;14:1000-7.
132. Steenari MR, Vuontela V, Paavonen EJ, Carlson S, Fjallberg
M, Aronen E. Working memory and sleep in 6- to 13-year-old
schoolchildren. J Am Acad Child Adolesc Psychiatry. 2003;42:8592.
133. Sadeh A, Gruber R, Raviv A. The effects of sleep restriction and
extension on school-age children: What a difference an hour makes.
SLEEP, Vol. 29, No. 9, 2006
Child Dev 2003;74:444-55.

134. Sadeh A, Gruber R, Raviv A. Sleep, neurobehavioral functioning,
and behavior problems in school-age children. Child Dev
2002;73:405-17.
135. Randazzo AC, Muehlback MJ, Schweitzer PK, Walsh JK. Cognitive
function following acute sleep restriction in children ages 10-14.
Sleep 1998;21:861-8.
136. Fallone GP, Acebo C, Arnedt JT, Seifer R, Carskadon MA. Effects
of acute sleep restriction on behavior, sustained attention, and
response inhibition in children. Percept Mot Skills 2001;93:21329.
137. Meier-Ewert HK, Ridker PM, Rifai N, et al. Effect of sleep loss on
C-reactive protein, an inflammatory marker for cardiovascular risk.
J Am Acad Cardiol 2004;43:678-83.
138. Vgontzas AN, Zoumakis E, Bixler EO, et al. Adverse effects
of modest sleep restriction on sleepiness, performance, and
inflammatory cytokines. J Clin Endocrinol Metab 2004;89:211926.
139. Everson CA, Laatsch CD, Hogg N. Antioxidant defense responses
to sleep loss and sleep recovery. Am J Physiol Regul Integr Comp
Physiol 2005;288:R374-83.
140. Silva RH, Abilio VC, Takatsu AL, et al. Role of hippocampal
oxidative stress in memory deficits induced by sleep deprivation in
mice. Neuropharmacology 2004;46:895-903.
141. Hu J, Chen Z, Gorczynski CP, et al. Sleep-deprived mice show
altered cytokine production manifest by perturbations in serum IL1ra, TNFa, and IL-6 levels. Brain Behav Immunol 2003;17:498504.
142. McDermott CM, LaHoste GJ, Chen C, Musto A, Bazan NG, Magee
JC. Sleep deprivation causes behavioral, synaptic, and membrane
excitability alterations in hippocampal neurons. J Neurosci
2003;23:9687-95.
143. Hairston IS, Little MTM, Scanlon MD, et al. Sleep restriction
suppresses neurogenesis induced by hippocampus-dependent
learning. J Neurophysiol 2005;94:4224-33.
144. Hanlon EC, Andrzejewski ME, Harder BK, Kelley AE, Benca RM.
The effect of REM sleep deprivation on motivation for food reward.
Behav Brain Res 2005;163:58-69.
145. Gozal D, Daniel JM, Dohanich GP. Behavioral and anatomical
correlates of chronic episodic hypoxia during sleep in the rat. J
Neurosci 2001;21:2442-50.
146. Macey PM, Henderson LA, Macey KE, et al. Brain morphology
associated with obstructive sleep apnea. Am J Respir Crit Care Med
2002;166:1382-7.
147. Gale SD, Hopkins RO. Effects of hypoxia on the brain: neuroimaging
and neuropsychological findings following carbon monoxide
poisoning and obstructive sleep apnea. J Int Neuropsychol Soc
2004;10:60-71.
148. O'Donoghue FJ, Briellmann RS, Rochford PD, et al. Cerebral
structural changes in severe obstructive sleep apnea. Am J Respir
Crit Care Med 2005;171:1185-90.
149. Robbins J, Redline S, Ervin A, Walsleben JA, Ding J, Nieto J.
Associations of sleep-disordered breathing and cerebral changes on
MRI. J Clin Sleep Med 2005;1:159-65.
150. Bartlett DJ, Rae C, Thompson CH, et al. Hippocampal area
metabolites relate to severity and cognitive function in obstructive
sleep apnea. Sleep Med 2004;5:593-6.
151. Alchanatis M, Deligiorgis N, Zias N, et al. Frontal brain lobe
impairment in obstructive sleep apnoea: a proton MR spectroscopy
study. Eur Respir J 2004;24:980-6.
152. Kamba M, Inoue Y, Higami S, Suto Y, Ogawa T, Chen W. Cerebral
metabolic impairment in patients with obstructive sleep apnoea:
an independent association of obstructive sleep apnoea with white
matter change. J Neurol Neurosurg Psychiatr. 2001;71:334-9.
153. Kamba M, Suto Y, Ohta Y, Inoue Y, Matsuda E. Cerebral metabolism
in sleep apnea. Am J Respir Crit Care Med 1997;156:296-8.
154. Thomas RJ, Rosen BR, Stern CE, Weiss JW, Kwong KK. Functional
1133
imaging of working memory in obstructive sleep-disordered

breathing. J Appl Physiol 2005;98:2226-34.
155. Anderson V, Levin HS, Jacobs R. Executive functions after frontal
lobe injury: A developmental perspective. In: Stuss DT, Knight RT,
eds. Principles of Frontal Lobe Function. New York: Oxford; 2002.
156. Diamond A. Normal development of the prefrontal cortex from
birth through young adulthood: Cognitive functions, anatomy, and
biochemistry. In: Stuss D, Knight RT, eds. Principles of frontal lobe
function. New York: Oxford; 2002.
157. Michel GF. A developmental-psychobiological approach to
developmental neuropsychology. Dev Neuropsychol 2001;19:1132.
158. Courchesne E, Townsend J, Christopher C. Neurodevelopmental
principles guide research on developmental psychopathologies. In:
Cicchetti D, Cohen DJ, eds. Developmental Psychopathology, Vol.
1: Theory and Methods. Oxford: Wiley and Sons; 1995:195-226.
159. Gozal D, OBrien LM. Snoring and obstructive sleep apnoea in
children: why should we treat? Paediatr Respir Rev 2004;5(Suppl
A):S371-6.
160. Mirmiran M, Ariagno RL. Role of REM sleep in brain development
and plasticity. In: Maquet P, Smith C, Stickgold R, eds. Sleep and
Brain Plasticity. New York: Oxford University Press; 2003:181-7.
161. Frank MG, Stryker MP. The role of sleep in the development of
central visual pathways. In: Maquet P, Smith C, Stickgold R, eds.
Sleep and brain plasticity. New York: Oxford University Press;
2003:190-206.
162. Kheirandish L, Row BW, Li RC, Brittian KR, Gozal D.
Apolipoprotein E-deficient mice exhibit increased vulnerability
to intermittent hypoxia-induced spatial learning deficits. Sleep
2005;28:1412-7.
163. Owens JA, Dalzell V. Use of the BEARS sleep screening tool in
a pediatric residents continuity clinic: a pilot study. Sleep Med
2005;6:63-9.
SLEEP, Vol. 29, No. 9, 2006
1134

290903

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

290903

Uploaded by

Copyright:

Available Formats

37 AM

PEDIATRIC SLEEP DISORDERS

Neurobehavioral Morbidity Associated With Disordered Breathing During Sleep in

predictors of morbidity. Short-term SDB treatment outcome studies are

ing, but not limited to, OSA.

Neurobehavioral Morbidity in Childhood SDBBeebe

SLEEP, Vol. 29, No. 9, 2006

Neurobehavioral Morbidity in Childhood SDBBeebe

Comm (Gen Peds)

36 with SDB, no controls

PSG (AHI > 5)

Nature of SDB Source of SDB

Parent and teacher report

Q-aire, oximetry (O2

2-18 Controls excluded

Across both data collections, 242

Of 782 subjects, 66 subjects

71 children with behavioral,

Oxford, England Study

SLEEP, Vol. 29, No. 9, 2006

Neurobehavioral Morbidity in Childhood SDBBeebe

Comm (Gen Peds)

Comm (Gen Peds)

147 SDB, 243 controls

797 with grades in lowest

13 snorers, 13 nonsnoring controls

16 snorers, 16 nonsnoring controls

Clinical (Sleep) vs.

South Australia Study

Parent report on validated

Parent report of learning

SDB and restlessness

Parent report on validated

Nature of SDB Source of SDB

SLEEP, Vol. 29, No. 9, 2006

Neurobehavioral Morbidity in Childhood SDBBeebe

154 SDB, 293 controls

Cleveland Childrens Sleep and Health Study

87 Primary Snorers 31 Controls

Louisville preschool study

205 (subset of 2003 sample

99 habitual snorers, 352 nonsnorers

480 (115 SDB)

Louisville First-grade study

Hannover, Germany Study

Parent report of behavior on

Q-aire, Unattended Parent report on validated

PSG (AHI >1) or

PSG (AHI > 5)

PSG (AHI and PLM Parent-report on validated

SLEEP, Vol. 29, No. 9, 2006

Neurobehavioral Morbidity in Childhood SDBBeebe

41 OSA, 41 controls undergoing

11 snorers, 13 behavioral sleep

6 moderate-severe OSA, 9 mild

108 snorers, 72 controls

12 with OSA, no controls

Clinical (ENT vs.

Melendres, 200456 Clinical (sleep) vs.

piled via computerized literature searches, manual review of the

Potential Causes for Illusory or Inflated Associations

ASSOCIATIONS OF CHILDHOOD SDB WITH NEUROBEHAVIORAL

Conventional guidelines for inferring causality start with the

SLEEP, Vol. 29, No. 9, 2006