Hindawi Publishing Corporation
Journal of Nutrition and Metabolism
Volume 2012, Article ID 134202, 8 pages
IndraNarangand Joseph L.Mathew
Division of Respiratory Medicine, The Hospital For Sick Children, 555 University Avenue, Toronto, ON, Canada M5G 1X8
Correspondence should be addressed to Indra Narang, email@example.com
Received 29 May 2012; Accepted 31 July 2012
Academic Editor: Dominique Bougl´ e
Copyright © 2012 I. Narang and J. L. Mathew. This is an open access article distributed under the Creative Commons Attribution
License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly
The global epidemic of childhood and adolescent obesity and its immediate as well as long-term consequences for obese
risk of adult obesity and clinically significant consequences affecting the cardiovascular and metabolic systems. Importantly,
obesity is additionally complicated by obstructive sleep apnea (OSA), occurring in up to 60% of obese children. OSA, which is
diagnosed using the gold standard polysomnogram (PSG), is characterised by snoring, recurrent partial (hypopneas) or complete
(apneas) obstruction of the upper airway. OSA is frequently associated with intermittent oxyhemoglobin desaturations, sleep
disruption, and sleep fragmentation. There is emerging data that OSA is associated with cardiovascular burden including systemic
hypertension, changes in ventricular structure and function, arterial stiffness, and metabolic syndromes. Thus, OSA in the context
as early recognition and treatment of OSA in obese children are likely to result in the reduction of cardiometabolic burden in obese
children. This paper summarizes the current state of understanding of obesity-related OSA. Specifically, this paper will discuss
epidemiology, pathophysiology, cardiometabolic burden, and management of obese children and adolescents with OSA.
The epidemic of pediatric obesity has caused serious concern
all over the world as the prevalence has increased alarmingly
over time, not only in developed countries but also in
developing countries [1, 2]. Furthermore, there is increasing
recognition that childhood obesity is occurring at progres-
sively younger ages . Recent publications have highlighted
the challenge of defining childhood obesity in a manner that
is both evidence based as well as uniformly applicable across
different settings . In general, a statistical definition using
BMI for age is used wherein >85th percentile is defined as
overweight and >95th percentile as obesity . In contrast,
the WHO defines obesity as BMI for age Z-score >3 and
overweight as Z-score >2. A large-scale multicentric study
calculated BMI in children and adolescents and extrapolated
the cut-off values for adult obesity (BMI > 30) and over-
weight (BMI > 25), to the corresponding values in childhood
and adolescence. Based on this, they were able to tabulate
age- and gender-specific cut-off values for children and ado-
lescents . At the present time, waist circumference is not
used routinely to define obesity in children and adolescents.
interplay of genetic, environmental (life-style), socioeco-
nomic, cultural, and psychological factors which are beyond
the scope of this paper. However, interestingly the pattern
of in utero growth may program the pattern of subsequent
body fat deposition and neuroendocrine interactions that
promote eating behavior. Specifically, there is an observed
increase in childhood obesity with increasing birth weight
. Counterintuitively, infants with low birth weight and an
early adiposity rebound are also predisposed to higher rates
of obesity in later childhood .
of childhood obesity. Specific morbidities associated with
obesity include hypertension, left ventricular abnormalities,
insulin resistance, type 2 diabetes, dyslipidemia, nonalco-
holic fatty liver disease, and obstructive sleep apnea .
Further, in obese individuals, the clustering of dyslipidemia,
hypertension, and impaired glucose tolerance/insulin resis-
In a followup of over 200,000 Norwegian adolescents, the
relative risk of death due to ischaemic heart disease for those
2 Journal of Nutrition and Metabolism
with a BMI > 85 percentile was 2.9 for males and 3.7 for fem-
ales when compared to those with lower BMI percentiles
. These may underrepresent the true cardiovascular
burden in adulthood given the knowledge that 75% of obese
children will become obese adults [12, 13]. Additional com-
plications of obesity include menstrual problems and poly-
cystic ovarian disease, gallstones, orthopedic issues, and psy-
During sleep in normal individuals, there is reduction in
the tone of airway musculature; however pharyngeal dilator
activity keeps the airway patent. Therefore, although normal
children can have occasional pauses in breathing for up to
10–15 seconds, there is no significant airflow limitation.
Therefore paO2 may fall only by 2–4mmHg, and end-
tidal CO2 may increase marginally by 3-4mmHg. More
importantly, there is no arousal from sleep [16, 17].
Obstructive sleep apnea (OSA) is part of the spectrum
of clinical conditions comprising sleep-disordered breathing
upper airway obstruction. In children, obstructive apnea is
defined by the absence of nasal airflow despite the presence
of chest wall and abdominal wall movements, for a duration
of at least two breaths. In contrast, the term “obstructive
hypopnea” refers to decrease in nasal airflow by 50% from
the baseline accompanied by fall in oxygen saturation of 3%
and/or arousal. The number of apneic and hypopneic events
per hour of sleep is expressed as apnea/hypopnea index
(AHI) on polysomnography . In adults, AHI < 5/hour is
is regarded abnormal. In general, the same criteria can be
used for adolescents in the age group 12–15 years. For those
beyond 18 years, it is recommended that adult criteria be
used. In children, AHI is also used to categorize the severity
of OSA; AHI up to 1.5events/hr is classified as mild, 1.5–
5.0events/hr as moderate, and >5.0events/hr as severe. This
is in contrast to adults, where the corresponding values are
5–15events/hr, 15–30events/hr, and >30events/hr.
OSA is characterised by snoring, recurrent partial (hypo-
pneas), or complete (apneas) obstruction of the upper
airway. OSA is associated with intermittent oxyhemoglobin
desaturation, sleep disruption, and fragmentation .
tual snoring, whereas only 1–3% has OSA [20, 21]. However
Habitual snorers typically do not have obstructive apnea,
hypopnea, respiratory effort-related arousals, or abnormal
gas exchange. This is because neuromuscular compensation
in these children prevents significant airway obstruction.
In OSA, the episodes of airway obstruction can be related
to increased airway collapsibility on account of mechanical
and neuronal factors. The most common mechanical factor
in children is hypertrophy of adenoids and/or tonsils nar-
rowing the airway lumen . Approximately 2% of oth-
erwise healthy children have large tonsils and adenoids that
mechanically obstruct airways [20, 23]. However, OSA is a
balance of mechanical obstruction and decreased activity of
pharyngeal dilator muscle activity. During sleep, children
with OSA have reduced airway muscle tone which critically
narrows and obstructs the airway, resulting in upper airway
obstruction. The consequent hypoxemia results in arousal
with restoration of airway tone and relief of the obstruction.
The frequency of episodic apnea determines the diagnosis
and severity of OSA.
OSA can be suspected by the presence of both nocturnal as
well as day-time symptoms. The most common nighttime
symptoms are snoring during sleep; sometimes parents are
able to describe characteristic episodic pauses in breathing
despite movement of the chest or abdomen. Other descrip-
tions include gasping, restlessness during sleep, nighttime
sweating, sleeping in unusual positions, parasomnias (sleep
terrors, sleep walking), and secondary nocturnal enuresis.
The daytime symptoms are based on functional conse-
quences of disturbed sleep and/or hypoxemia/hypercarbia.
These include early morning headache and sometimes
nausea or vomiting, excessive daytime sleepiness, and
fatigue. Recent reports have also highlighted neurocognitive
consequences of OSA including decreased concentration,
diminished memory, difficulty in making decisions, learning
difficulties, and also behavioural manifestations such as
even social withdrawal.
Children with OSA are often mouth breathers and some-
times have hyponasal speech. Children with severe OSA
can also have growth stunting. Clinical examination usually
reveals a crowded oropharynx, enlarged tonsils, and reduced
peritonsillar space. Sometimes, a large tongue may also con-
tribute to airway obstruction. Endoscopic examination iden-
tifies hypertrophied adenoids.
Currently, polysomnography (PSG) or a sleep study is
the gold standard for a specific diagnosis of OSA .
Given the complexity of this investigation in terms of skill,
resources, and time, some investigators have tried to use
alternate approaches such as clinical questionnaires, objec-
tive and/or subjective measures of daytime sleepiness, over-
night oximetry, audio recording, video recording, and nap
methods can identify children with OSA, they have poor
negative predictive value .
There is now ample data confirming that OSA associated
with obesity is highly prevalent in children and adolescents.
The association between obesity and OSA emerges from two
sets of observations; the first is the observed high preval-
ence of OSA among obese children and adolescents, and the
second is the higher proportion of children with OSA who
are obese. Thus it appears that both conditions can coexist
and yet potentiate the adverse impact of each. It is believed
that the prevalence of OSA among obese children and
adolescents can be as high as 60% . In one study of obese
Journal of Nutrition and Metabolism3
children undergoing polysomnography, it was observed that
46% had OSA . Likewise, in another study OSA was
observed in 59% of obese children undergoing evaluation
study reported that 55% children scheduled for bariatric
surgery for morbid obesity had OSA . In fact, in many
children and adolescents with OSA, the severity of OSA
parallels the severity of obesity . A recent population-
based study involving 400 children between 2 and 8 years of
age found that obesity was the most significant risk factor for
analysis suggested that for each unit increase in BMI, there
was a 12% higher risk of OSA .
5.Mechanismsfor IncreasedRisk of OSA in
There are multiple factors that interact to significantly in-
Similar to nonobese children, airway obstruction by
adenotonsillar hypertrophy is a fairly common cause of OSA
among obese children [31–33] affecting approximately 45%
of all obese children with OSA . However, alarmingly,
following adenotonsillectomy, OSA persists in about 50%
of obese children  which is significantly higher than
the observed persistence rate of 10–20% amongst nonobese
children [36, 37]. Another additional interesting observation
is that the prevalence of adenotonsillar hypertrophy among
obese children is higher than among nonobese children,
which indirectly suggests that adenotonsillar hypertrophy in
obese children could be a consequence of another distinct
mechanism. Possible explanations include endocrine medi-
ated somatic growth that results in larger and/or heavier fat
pads, soft palate, and tongue among adults with obesity .
adolescents as well.
Functional factors that operate to promote upper airway
obstruction OSA in obese individuals during sleep include
altered neuromuscular tone resulting in greater upper airway
collapsibility during sleep. Indeed, measurements of airway
flow and mechanics have shown that in obese children, there
is a positive critical closing pressure of the pharynx causing
the airway to collapse during sleep with even mild negative
inspiratory pressure .
Additional mechanical factors that predispose to func-
tional abnormalities include central adiposity and an excess
mechanical load on the chest wall. These factors interact
and result in decreased chest wall excursion and decreased
diaphragmatic excursion causing a reduction in, chest wall
compliance with reduced functional residual capacity and
tidal volumes. As a result, hypoventilation, atelectasis, and
ventilation/perfusion mismatch may ensue resulting in in-
creased work of breathing resulting in fatigue, all of which
may be exacerbated during sleep and could further predi-
ing physiologies are not well understood, they could in part
explain why adenotonsillectomy is not curative in all obese
children with hypertrophied adenoids and tonsils.
inObese Childrenand Adolescents
6.1. Excessive Daytime Sleepiness (EDS). EDS is prevalent
among obese children with and without OSA ; specif-
ically EDS increased progressively and significantly with
increasing BMI. Prepubertal obese subjects with OSA have
more EDS than non-obese subjects with OSA of similar
6.2. Quality of Life (QOL). Multiple published studies de-
monstrate reported poor QOL among overweight and obese
children and adolescents  and those with OSA . In
one study with 151 children, with a mean age of 12 years, the
presence of OSA was a predictor of poor QOL in overweight
6.3. Neurocognitive Function. OSA is associated with cog-
nitive, behavioral, and functional deficits in young children
function, it is believed that sleep fragmentation associated
with OSA is a key determinant of behavioral alterations
in pediatric OSA subjects. A recent study with 52 children
reported improvement in both neurobehavioral function
and daytime sleepiness in children who used an average of 3-
hour positive airway pressure (PAP) at night . In another
small study with 6 obese adolescents, even modest level of
formance whereas a similar group of 7 nonadherent adoles-
cents showed academic decline . Resolution of OSA is
associated with improvement in neurocognitive status.
in obese children and those with OSA . Increased phy-
sical activity may not only promote weight loss but also, sec-
ondary to weight loss, may improve the severity of OSA .
6.5. Cardiovascular Burden. Multiple adult studies indicate
in the context of obesity . A similar evaluation of child-
hood obesity-related OSA on cardiovascular structure and
function is currently not available. However, indirect mea-
surements that reflect blood pressure regulation, cardiac
function, autonomic dysfunction, and endothelial properties
suggest a similar pattern in obese children and adolescents
[48, 51, 52]. The precise mechanisms linking cardiovascular
disease both to OSA and obesity are not completely under-
stood. However, a common mechanism is activation of the
sympathetic nervous system. Specifically, repetitive arousals,
episodic hypoxaemia, hypercapnia, and changes in intra-
thoracic pressures lead to sympathetic activation via chemo-
receptor activation, impaired baroreflex sensitivity, and
increased adiposity elevates levels of free fatty acid (FFA)
which with increased levels of leptin promote sympathetic
activation. Chronic sympathoactivation instigates dyslipi-
daemia, left ventricular modelling, endothelial dysfunction
and arterial stiffness, inflammation with high levels of
4 Journal of Nutrition and Metabolism
hs-CRP, and insulin resistance with resultant glucose intoler-
ance. All of these factors are inextricably linked and together
sodic hypoxemia in children with OSA causes pulmonary
vasoconstriction and ultimately pulmonary artery hyperten-
sion [49, 53].
Hypertension. In children, obesity is a risk factor for high BP,
and OSA is independently associated with increased BP [51,
54]. In a recent study, among prepubertal, non-obese child-
ren, the presence of OSA was associated with an elevation
in BP by 10–15mmHg independently of BMI during both
wakefulness and sleep when compared to nonsnoring con-
trols (ZA). In a separate study of 140 children, children
with severe OSA when compared with controls with no
OSA showed significantly increased mean arterial BP during
awakefulness and sleep, increased diastolic BP during wake-
fulness and sleep, and increased systolic BP during sleep.
Almost one-third of the patients with severe OSA showed a
mean 24-hour systolic BP > 95th percentile. Similarly, obese
children with moderate-to-severe OSA had a significantly
OSA, suggesting that OSA may be a trigger for hypertension
in obese children . These findings are of significance as
a recent longitudinal study has shown that elevated BP in
childhood tracks into adult life and is associated with an
increased risk of hypertension and metabolic syndrome later
in life .
and cardiovascular mortality. Pilot data shows a significantly
higher left ventricular mass index (LVMI) with reduced
diastolic and systolic function among obese children without
documented OSA compared with lean controls . In non-
obese children with OSA, abnormalities in LVMI correlate
with both the presence and severity of OSA . One study
reported that subjects with severe OSA had an odds ratio
of 11.2 for LVMI > 95th percentile , while another
showed that relief of OSA following an adeno-tonsillectomy
resulted in measured cardiac variables in the same range
as controls . Furthermore, in non-obese children with
OSA, improvements in LV diastolic function  and the
right ventricular myocardial performance index have been
observed after resolution of OSA . Thus OSA in the con-
text of obesity is likely to exacerbate abnormalities of LV
structure and function.
Endothelial Function. OSA is also involved in causing endo-
thelial dysfunction, mediated by reduced levels of nitric
oxide and increased levels of mediators like endothelin-1 and
Cardiac Autonomic Activity. Cardiac autonomic activity is
usually measured using indices of heart rate variability
(HRV). Low HRV signifies sympathetic overdrive and has
been consistently associated with the risk of incident cardio-
vascular disease in adults. In obese children, HRV was lower
than non-obese children with body weight as the strongest
predictor for lower HRV . However, in non-obese child-
compared to those without OSA .
6.6. OSA and the Metabolic Syndrome. There is emerging
hypertension, and inflammation. These occur through sym-
pathetic hyperactivity, intermittent hypoxemia, and sleep
be a cause of obesity and not a consequence alone. This con-
towards the contribution of OSA to various components of
the metabolic syndrome and perhaps, more importantly, the
reversibility with treatment of OSA.
insulin resistance) in adolescents with OSA . In younger
children, adeno-tonsillectomy is associated with improve-
ment in lipid profile, insulin sensitivity, and inflammatory
markers in some studies [64, 65].
6.7. Contribution of OSA to Obesity. OSA is associated with
inadequate sleep quantity and quality in children as well as
adults. A recent systematic review  examining the rela-
tionship between sleep duration and the development of
of sleep was associated with increased risk of obesity (odds
a direct impact by worsening obesity.
It is clear that childhood obesity and OSA can present to a
wide range of professional disciplines on account of the mul-
tisystem manifestations. Therefore successful management
depends on concerted effort by a multidisciplinary team of
professionals including sleep physician, ENT surgeon, respi-
rologist, child nutritionist, child psychologist, cardiologist,
and social worker, working together with the obese child
and his/her family. The goals of management are enhanced
quality of life and prevention of short- and long-term com-
8.Management of Obesity
A detailed discussion on the various modalities for weight
reduction and management of obesity is outside the scope of
this paper; however a brief review of the current recommen-
dations is presented. Although the body of evidence in chil-
dren and adolescents is still being generated, it is generally
recommended they should perform at least 60 minutes of
moderately intense physical activity daily, in order to prevent
obesity or maintain weight. This should be encouraged even
if it does not result in weight loss, on account of the general
health benefits of exercise. They should also be advised to eat
meals at regular times and preferably free from distractions.
Pharmacological interventions are generally not recom-
mended for children below 12 years, barring exceptional cir-
In addition, the rare decision to use pharmacotherapy does
Journal of Nutrition and Metabolism5
Some experts maintain that medication is better utilized to
maintain weight loss, rather than induce it. Currently, Orli-
since Sibutramine has been withdrawn.
Bariatric surgery is also rarely recommended in children,
unless they are morbidly obese (BMI ≥ 40kg/m2) or 35–
40kg/m2with coexisting diseases that could be improved by
loss of weight. Even then, surgery is considered only after
nonsurgical measures have been tried without success.
9.Management of OSA
Based on the observation that almost half of all obese child-
ren with OSA have adeno-tonsillar hypertrophy, the Ameri-
can Board of Pediatrics  recommends adeno-tonsillect-
omy as the first step in management. Although adeno-tonsil-
lectomy results in improvement in 80% cases and improves
obstructive symptoms in 80% of cases of otherwise normal
children with OSA, children with morbid obesity are more
likely to fail treatment than normal children. In some series,
almost 50% continue to have OSA .
Therefore Positive Airway Pressure (PAP) therapy has
become the standard of care, usually in conjunction with
weight loss strategies. In adult patients, PAP therapy results
in dramatic improvement in OSA symptoms. In addition,
there are encouraging reports of improvement in cardiovas-
cular status including reduction in systolic and BP, LV func-
tion , markers of endothelial function , and cardiac
autonomic activity . In addition, withdrawal of PAP
for two weeks was associated with systolic and diastolic BP
increase of 4–6mmHg . PAP therapy can be administ-
ered either as continuous PAP (CPAP) or the more physio-
logical bilevel PAP (BiPAP). In children also, the symptoms
of OSA improve with PAP therapy; however there is limited
data evaluating its efficacy in improving clinical outcomes
In one study, non-obese children who had resolution of
OSA, 6 months following adeno-tonsillectomy, had a reduc-
tion in diastolic BP of the order of 5mmHg . The impor-
tance of this study cannot be overemphasised in the con-
text of recent findings of a large meta-analysis that lowering
systolic SBP by 10mmHg or diastolic BP by 5mmHg in
adults (regardless of the baseline BP) reduced fatal and non-
fatal cardiac events by approximately 25% .
Other treatment options that are sometimes useful in
adult patients with OSA include oral appliances and devices
that expand the upper airway space. However these require
skilled construction and are generally efficacious in mild
OSA only. However it is a viable option for those who cannot
or will not use CPAP. These appliances have limited value in
children on account of less developed dentition . Some
adults use simple devices to prevent sleeping in the supine
position. These devices work by promoting sleep in the lat-
eral or prone position.
Surgical management options include uvulo-palato-
pharngoplasty wherein bulky soft tissues that obstruct the
airway can be trimmed or excised to create a larger airway
space. It has also been used to strengthen and support
hypotonic pharyngeal muscles in those children where redu-
ced neuromuscular tone is responsible for airway floppiness
and obstruction. Some centres use the procedure for obese
children with severe OSA, to reduce redundant oropharyn-
geal tissue bulk.
Presently, there is no randomized trial comparing the
various modalities in children adolescents, to estimate the
superiority of one over the other.
Childhood and adolescent obesity have reached epidemic
OSA significantly complicates obesity and is an inde-
pendent risk factor for cardiovascular, metabolic, neuro-
cognitive burden as well as negative impact on the quality
of life in obese children.
All disciplines involved in the well-being of obese child-
ren must be involved in sleep surveillance strategies to high-
nition and treatment of OSA, in addition to weight loss stra-
tegies, could provide an opportunity for cardiovascular and
metabolic risk reduction in childhood which would posi-
tively impact the health of these children not only in child-
hood but also in adulthood.
The Journal of the American Medical Association, vol. 299, no.
20, pp. 2401–2405, 2008.
 L. Wang, L. Kong, F. Wu, Y. Bai, and R. Burton, “Preventing
chronic diseases in China,” The Lancet, vol. 366, no. 9499, pp.
Dietz, “Early adiposity rebound and the risk of adult obesity,”
Pediatrics, vol. 101, no. 3, article E5, 1998.
 C. L. Ogden and K. M. Flegal, “Childhood obesity: are we all
 The Expert Committee on Clinical Guidelines for Overweight
in Adolescent Preventive Services, “Guidelines for overweight
in adolescent preventive services: recommendations from an
expert committee,” American Journal of Clinical Nutrition, vol.
59, no. 2, pp. 307–316, 1994.
lishing a standard definition for child overweight and obesity
worldwide: international survey,” British Medical Journal, vol.
320, no. 7244, pp. 1240–1243, 2000.
 J. J. Reilly, J. Armstrong, A. R. Dorosty et al., “Early life risk
factors for obesity in childhood: cohort study,” British Medical
Journal, vol. 330, no. 7504, pp. 1357–1359, 2005.
of adult obesity: a review,” Obesity Reviews, vol. 13, no. 4, pp.
 J. Chaicharn, Z. Lin, M. L. Chen, S. L. D. Ward, T. Keens, and
M. C. K. Khoo, “Model-based assessment of cardiovascular
autonomic control in children with obstructive sleep apnea,”
Sleep, vol. 32, no. 7, pp. 927–938, 2009.