S l e e p a n d Re s p i r a t o r y

P h y s i o l o g y i n C h i l d ren
Kristie R. Ross, MD, MS*, Carol L. Rosen, MD

KEYWORDS
 Respiration  Reference values  Oxygen saturation  Carbon dioxide  Sleep/physiology  Child
 Infant  Adolescent

KEY POINTS
 The maturation of respiratory physiology during sleep contributes to the unique features of child-
hood sleep disorders.
 Ventilation decreases during sleep in children as it does in adults, with variability related to sleep
state.
 Knowledge of the range of normal values of respiratory parameters during sleep, including respira-
tory rates, oxygen saturation, measures of carbon dioxide, and number and patterns of apneas, is
crucial for the physician to evaluate common sleep disorders in children.

INTRODUCTION and mechanical factors.1 Chemical information is

relayed through peripheral chemoreceptors in the
Sleep is a critical determinant of health. As a child carotid and aortic bodies. The carotid chemore-
develops from infancy through adolescence, ceptors, although small, have a uniquely high blood
important changes in respiratory physiology occur. supply, and sense the arterial concentration of O2
Maturational changes of breathing during sleep and relay the information to the medulla. The ca-
contribute to the unique features of childhood sleep rotid body chemoreceptors are responsible for
disorders. The clinician’s ability to evaluate com- about 90% of the ventilatory response to hypox-
mon disorders related to sleep in children, including emia. Chemosensory function in the carotid bodies
apnea of prematurity, sudden infant death syn- is immature at birth and increases with age. Carbon
drome (SIDS), apparent life threatening events dioxide (CO2) is also sensed in the carotid body,
(ALTE), obstructive sleep apnea syndrome, and accounting for 20% to 50% of the response to arte-
other forms of sleep-disordered breathing relies rial hypercapnia. The remaining response to hyper-
on an understanding of normal patterns of breath- capnia comes from central brainstem receptors in
ing during sleep across the ages. The purpose the medulla. The central chemoreceptors respond
of this article is to review respiratory physiology primarily to changes in pH, mediated by CO2 ten-
during sleep throughout childhood. sion, in the cerebrospinal fluid. Laryngeal chemo-
reflexes (LCR) located in the epithelium that
OVERVIEW OF NORMAL RESPIRATION surrounds the airway respond to acidic solutions
DURING SLEEP with reflexes that include startle, swallowing, laryn-
geal constriction, apnea, and bradycardia. Me-
Breathing during sleep is controlled by voluntary chanical information sensed through receptors in
and behavioral factors in addition to metabolic the lung and chest wall are also relayed to the
chestmed.theclinics.com

Disclosure: The authors have no disclosures relevant to this article.

Division of Pediatric Pulmonology and Sleep Medicine, Department of Pediatrics, Rainbow Babies and
Children’s Hospital, Case Western Reserve University School of Medicine, 11100 Euclid Avenue, RBC 3001,
Cleveland, OH 44106, USA
* Corresponding author.
E-mail address: [email protected]

Clin Chest Med - (2014) -–-

http://dx.doi.org/10.1016/j.ccm.2014.06.003
0272-5231/14/$ – see front matter Ó 2014 Elsevier Inc. All rights reserved.
2 Ross & Rosen

medulla via the vagal nerve. Higher central nervous DEVELOPMENTAL CHANGES IN RESPIRATORY
system centers can override the respiratory cen- CONTROLLERS
ters to control nonbreathing functions such as
speaking and laughing. These voluntary and Postnatally there is an increase in the hypoxic
behavioral controls are affected by sleep state. sensitivity (resetting) of both carotid and aortic
Clinical problems associated with disorders of res- chemoreceptors, and a diminishing influence of
piratory control are listed in Box 1. descending inhibitory effects on breathing in hyp-
In general, ventilation decreases during sleep oxia.3 Compared with the adult, peripheral chemo-
compared with wakefulness, but varies with receptors assume a greater role in the newborn.
sleep state. During non–rapid eye movement Although not essential for initiation of fetal respira-
(NREM) sleep, breathing is regulated primarily tory movements, animal studies show that periph-
by carbon dioxide and is characterized by the eral chemoreceptor denervation in the newborn
absence of behavioral controls. Breathing is period results in severe respiratory impairment
regular with reduced tidal volume and respiratory and a high probability of sudden death. In the
rate compared with wakefulness, resulting newborn, steady-state hypoxia produces a tran-
in decreased minute ventilation. This decline, in sient increase in ventilation followed by a decrease
combination with the supine position and back to or below baseline level. With maturation,
decrease in intercostal muscle tone, results in a this biphasic response to hypoxia changes to a
decrease in functional residual capacity. Upper sustained ventilatory response. By contrast, a
airway tone and lung volume also decrease dur- steady-state response to CO2 is present at all
ing sleep, resulting in increased upper airway ages from birth and increases with advancing post-
resistance. Compared with the regular breathing natal age. There are only limited data on ventilatory
seen during NREM sleep, breathing during rapid responses in different sleep states in infants, but
eye movement (REM) sleep is irregular in terms the directionality is similar to the findings in adults,
of both respiratory rate and tidal volume. Short with responses to hypoxia and hypercapnia that
central respiratory pauses are common during are reduced in REM compared with NREM sleep.
REM sleep in children. Inhibition of tonic activity In the newborn, hypercapnia and hypoxic ventila-
of the intercostal muscles during REM results in tory responses interact to augment respiratory
a further decline in functional residual capacity. responses. With increasing age, peripheral chemo-
At the same time, activity of the diaphragm re- receptors undergo progressive decrement in their
mains stable, and this incoordination between relative sensitivity.4 LCR responses are signifi-
the intercostal muscles and diaphragm results cantly more active in the immediate postnatal
in paradoxic chest and abdominal movement period compared with later in life, and the pattern
during REM sleep that usually resolves by the of the response also changes with maturation.
age of 3 years. A relative decrease in upper The predominant LCR response in the newborn in-
airway muscle tone when diaphragmatic cludes swallowing and apnea, and differs from the
contractions remain unchanged can predispose LCR responses seen in the older infant or adult
to obstructive apnea, especially when the airway (cough and the expiration reflex).5,6
is already small or narrow. Finally, hypoxic and
hypercapnic ventilatory drives decrease during VENTILATION, RESPIRATORY PATTERNS, AND
sleep. Therefore, normal children experience a APNEAS
small increase in the partial pressure of CO2
Normative data on tidal volume and minute ventila-
and a small decrease in arterial oxyhemoglobin
tion in children are not readily available, as most
saturation (SpO2) during sleep. The magnitude
studies have focused on describing respiratory
of these changes has not been systematically
rates and patterns, gas exchange, and the fre-
studied in large pediatric samples of healthy
quency and type of apneas seen in healthy children
children, but is believed to average 2% for
of various ages. More data are available for preterm
SpO2 and 4 to 6 mm Hg for CO2.2 These sleep-
and full-term infants in the first months of life than
related changes in ventilation, upper airway sta-
for children and adolescents, but several new re-
bility, and gas exchange can be exaggerated in
ports of normative sleep and breathing data in
children with underlying pulmonary, upper
these age groups have been published.7–9
airway, and neuromuscular problems, resulting
in increased vulnerability to sleep-disordered
Respiratory Frequency
breathing. Differences between newborn and
adult respiratory systems that make the infant As would be predicted from knowledge about the
more vulnerable to ventilatory failure are summa- inverse relationship between respiratory rate and
rized in Box 2. body size in other mammals,10 respiratory rates
 Sleep and Respiratory Physiology in Children 3

Box 1 Cerebral palsy

Clinical problems associated with disorders of Depressant drugs
respiratory control in children
Craniofacial syndromes
Psychogenic Obesity hypoventilation syndrome
Hyperventilation Trisomy 21
Habitual cough
Vocal cord dysfunction
Reflexive
are highest in newborns and decrease until they
Breath-holding spell reach normal adult values in middle childhood.11,12
Expiratory apnea The changes in respiratory frequency with matura-
tion are shown in Table 1. Infants have higher res-
Neurometabolic or Genetic Disorders
piratory rates per unit body weight than older
Central congenital hypoventilation syndrome children, and this relationship is not proportional
Central hypoventilation associated with hypo- to body size.13 During both sleep and waking
thalamic or endocrine dysfunction states, respiratory rates decrease during the first
Rett syndrome 3 years of life with increasing body weight, but
this decrease is not directly proportional to body
Leigh disease
size. Boys have a higher respiratory rate for a given
Prader-Willi syndrome body weight than girls.13 The mean difference be-
Conditions Associated with Brainstem Malfor- tween the waking and sleep respiratory rate is
mation or Brainstem Lesions greatest in infants, and decreases to 1 to 2 breaths
per minute in adolescents. Respiratory rate varies
Arnold-Chiari malformation
with sleep state in infants, children, and adoles-
Dandy-Walker malformation cents, with changes being most prominent and
Möbius syndrome clinically relevant in infants. In general, the respira-
Hypoxic-ischemic encephalopathy tory frequency is higher during REM than during
NREM sleep in newborns and infants.14 In a small
Tumors sample of healthy children aged 9 to 13 years, res-
Vascular abnormality, infarction, or hemorrhage piratory rates during sleep were variable but were
Other Central Nervous System Disorders
highest during waking and light N1 sleep, and
lowest during stage 2 sleep later in the night. Rates
Spinal muscular atrophy
Polio
Box 2
Familial dysautonomia
Characteristics of the respiratory system in
Peripheral Nervous System Disorders infants that increase vulnerability to
respiratory failure
Myasthenia gravis
Guillain-Barré syndrome Difficulty transitioning to mouth breathing dur-
ing nasal obstruction, particularly during rapid
Muscular dystrophy
eye movement (REM) sleep
Myotonic dystrophy
Laryngeal reflexes trigger more profound
Thoracic Cage, Lung, or Airway Disease cardiorespiratory depression
Kyphoscoliosis Increased chest-wall compliance
Asphyxiating thoracic dystrophy (Jeune disease) Barrel-shaped chest wall with more horizontal
position of the ribs
Interstitial lung disease
Increased REM sleep time with loss of stabilizing
Asthma
intercostal muscle tone
Conditions that Affect Multiple Respiratory Loss of expiratory braking in REM sleep
Systems
Lower specific lung compliance
Sudden infant death syndrome
Higher metabolic demand
Obstructive sleep apnea syndrome
Immaturity of chemoreceptors
Apnea of prematurity
4 Ross & Rosen

Table 1
Respiratory rates during sleep from birth through adolescence

Newborn Infant Preschool School Age Adolescent

77
Mean (SD) 29 (6) 15.3 (1.6) NREM16
32.3 (4.9)48 16.4 (1.5) REM16
Median (IQR) 40 (34–52)80 2981 20 (18–21)82 18 (17–19)82 18 (17–18)82
Range 21–6281 14–2682 15–2582 15–2282

Abbreviations: IQR, interquartile range; NREM, non–rapid eye movement sleep; REM, rapid eye movement sleep; SD,
standard deviation.

were slightly higher in boys than in girls in all Measurement conditions and methodology vary
stages of sleep.15 In a small sample of healthy ad- from study to study. Some studies have included
olescents, respiratory rate and minute ventilation children with snoring while others have not. Find-
decreased by 8% from wakefulness to NREM ings by age groups are discussed in this section.
sleep, and increased by 4% from NREM to REM
sleep.16 Apnea in full-term infants
Central apneas are respiratory events defined by
Tidal Volume and Minute Ventilation the absence of both airflow and respiratory effort.
Minute ventilation is the product of respiratory rate Apnea in infants is most commonly defined as a
and tidal volume, and is related to metabolic rate. pause in breathing for at least 20 seconds, or a
To respond to increased metabolic demand, min- shorter pause that is associated with bradycardia
ute ventilation can be increased by increasing the or oxygen desaturation.7 Although the term “pro-
respiratory rate, increasing the tidal volume, or longed” is often used to describe central apneas
both. In the newborn and young child, increasing with a duration of at least 20 seconds, clinicians
the respiratory rate (rather than tidal volume) is should be aware that apnea durations of 20 sec-
the most energy-efficient strategy to cope with onds or longer occur occasionally, and those of
higher ventilatory needs.13,17 This strategy of 30 seconds or longer rarely, in healthy term infants.
changing respiratory frequency rather than tidal Shorter (<20 seconds) central pauses are
volume is consistent with data in resting humans commonly seen in REM sleep, after a sigh breath
showing that both tidal volume and dead space or a body movement, and during transition from
per body weight remains essentially unchanged wakefulness to sleep. Determining normative
from birth to adulthood (about 6 mL/kg for tidal values for central apneas across different age
volume and 2.2 mL/kg for dead space).18 There groups can be challenging because studies have
are limited data on normal tidal volume and minute used different technologies, time frames, and res-
ventilation in healthy children from newborn and piratory event definitions, and there is some debate
adolescent age groups.14,16 In general, minute about the evidence base for the currently accepted
ventilation is slightly higher in REM than in NREM definition.24
sleep, consistent with the higher respiratory rates Nevertheless, in full-term infants central respira-
in REM. As expected, minute ventilation de- tory pauses are common, occur frequently after
creases with age (from 250 mL/kg/min in new- body movements, and are more frequent during
borns to 100 mL/kg/min in adolescents), and active sleep and REM sleep.25 Most events
parallels the maturational changes in respiratory meeting the aforementioned criteria for apnea in
frequency and metabolic needs. term infants are central in nature,26 with substantial
evidence that obstructive and mixed apneas are
rare in healthy infants.26–28 In the first 6 months of
Apnea Type, Duration, and Frequency
life, central apneas are very frequent with median
Numerous studies have investigated the fre- apnea indices of approximately 5 per hour and
quency and duration of apnea in newborns and in- the highest values in the first 5 weeks of life, 8/h.
fants in the first year of life, with fewer studies in In the second 6 months of life, the median central
older children and adolescents.7–9,19–23 Compari- apnea indices of 6.4/h in REM and 1.7/h in NREM
sons among the different studies are difficult for sleep are reported. The frequency of central ap-
a variety of reasons. Definitions of respiratory neas declines after the first year of life. When
events are not standardized, in terms of both recorded, obstructive apneas occur mainly during
length of event and how they are classified. REM sleep.29 Mechanisms for the greater
 Sleep and Respiratory Physiology in Children 5

respiratory instability in infants during REM breathing is more frequent in preterm infants, varies
compared with NREM sleep include immaturity of across studies in full-term infants, and decreases
central respiratory control and phasic inhibitory- during the first 2 years of life. Episodes of periodic
excitatory mechanisms inherent to REM sleep.30 breathing can be seen in 80% to 100% of 1-
The sleeping position (prone or supine) does not week-old term infants. In term infants, the upper
alter the incidence, duration, or type of apnea in limits for sleep time spent in periodic breathing
healthy infants.28 The incidence of both central are 5% to 10% at 1 month of age, decreasing to
and obstructive apneas decreases with increasing 2% at 3 months of age. In preterm infants, the upper
postmenstrual age.26,31 Although apnea was at one limits are higher: 15% to 20% at 1 month of age and
time hypothesized to be the pathophysiologic pre- 5% to 10% at 3 months of age.48–50 In otherwise
cursor to SIDS, extensive research has failed to asymptomatic preterm infants with long episodes
support this concept.32–35 of periodic breathing that result in reduced satura-
tion values, administration of supplemental O2 is
Apnea in preterm infants associated with decrease in time spent in periodic
Apnea of prematurity usually refers to the sudden breathing.51 The pathophysiology of periodic
cessation of breathing that lasts for at least 20 sec- breathing is thought to be due to increased sensi-
onds or, if of shorter duration, is accompanied by tivity of the peripheral chemoreceptors.45 This
bradycardia or oxygen desaturation in an infant hypothesis is supported by work in animal models,
younger than 37 weeks gestational age.36 Apnea and findings that periodic breathing does not occur
of prematurity generally resolves when the child in the first 48 hours of life when the hypoxic
reaches term, but may persist for several addi- response of peripheral chemoreceptors is sup-
tional weeks in the most premature infants.37–39 pressed.52 In addition, compared with older chil-
Extreme episodes lasting at least 30 seconds usu- dren and adults, infants breathe very close to their
ally cease at approximately 43 weeks postmenst- CO2 apneic threshold. The average threshold of
rual age.40 In contrast to term infants in whom eupneic partial pressure of CO2 (PCO2) in neonates
obstructive events are rare, apnea of prematurity (1.1  0.2 mm Hg above the apneic threshold) is
is characterized by a combination of obstructive, much lower than the adult threshold (3.4 
central, and mixed apneas,41 although there is little 0.4 mm Hg). This closeness of the eupneic and
consensus regarding the frequency of these apneic CO2 thresholds creates greater vulnerability
events. The difficulty of distinguishing central ap- in the respiratory control system for infants such
neas (no effort, no airflow) from obstructive apneas that minor oscillations in breathing may bring
(effort, no airflow) is technically challenging in eupneic PCO2 values below threshold, causing ap-
these fragile infants.42 Continuous positive airway nea. Lower lung volumes and faster desaturation
pressure has been shown to reduce apnea in pre- rates also contribute to this instability.53 Older
term infants, suggesting that upper airway studies suggesting that increased periodic breath-
obstruction is an important contributor to apnea ing was a marker for increased risk of SIDS have
of prematurity.43 Obstructive apnea decreases been challenged.50,52,54
with increasing postmenstrual age,31 which may
be related to the improvement in extrathoracic CARDIORESPIRATORY EVENTS IN INFANCY
airway stability with maturation.44 In addition, the
degree to which laryngeal chemoreflexes trigger Bradycardia, apnea, and hypoxemia are closely
cardiorespiratory depression seems to be related in preterm infants.2 Heart rate varies with
increased in prematurity.6 Chronic hypoxemia sleep state. In a cohort of healthy term infants
related to immature lungs appears to enhance aged 1 to 4 months, a decrease in heart rate was
the ventilatory response of the peripheral chemo- more likely to be associated with NREM sleep than
receptors to hypoxemia, which may also explain with REM sleep.55 The precise mechanisms under-
the increased frequency of apnea in prematurity.45 lying these relationships, and the thresholds that
In healthy preterm infants, apnea triggered by de- define abnormal heart rate and SpO2 in this popula-
saturation decreases over time because of devel- tion, remain controversial. In preterm infants, 83%
opmental improvements in chest-wall stability46 of bradycardic episodes were associated with ap-
and ventilation-perfusion matching.47 nea and 86% were associated with desaturation.56
To test the hypothesis that cardiorespiratory events
Periodic breathing including apnea and bradycardia are more common
Periodic breathing, a common respiratory pattern in infants at increased risk for SIDS, researchers in 5
in infants, is defined as 3 or more episodes of apnea cities conducted a 6-month longitudinal cohort
lasting 3 or more seconds, separated by conti- study of in-home cardiorespiratory event recording
nued respiration of 20 seconds or less. Periodic in a total of 1079 infants. The sample included
6 Ross & Rosen

healthy term infants and infants considered to be at in Table 2, decreases from infancy through
increased risk for SIDS for a variety of reasons, adolescence.58 Obstructive apneas of any length
including a history of ALTE, having a sibling who are rare in both normal full-term infants and chil-
had died of SIDS, or prematurity (<34 weeks’ gesta- dren.7–9,20,22,26,58 Obstructive apneas occur
tion with birth weight <1750 g).40 Surprisingly, ap- mainly in REM and lighter NREM sleep. Despite
neas using conventional criteria of an event lasting the variation in measurement techniques, respira-
20 or more seconds, or of shorter duration but asso- tory event definitions, scoring approach, and sam-
ciated with bradycardia, were common even in the ple population, these studies are remarkably
healthy term infants. Forty-three percent of healthy similar in their findings: (1) obstructive apneas
term infants had at least 1 event that met criteria are extremely rare; and (2) central pauses are
for a “conventional” apnea. There was no difference seen frequently in healthy infants, including
in the frequency of these conventional apneas in pauses that last up to 30 seconds, and decrease
the healthy term infants and the term infants at in frequency but are still present throughout
increased risk for SIDS. Only preterm children had childhood.
an increased risk of conventional apneas. The inves-
tigators also examined the occurrence of “extreme” GAS EXCHANGE
apneas, defines as apnea longer than 30 seconds or
an age-adjusted bradycardia threshold of a specific There are several changes in gas exchange during
duration. Only 2% of the healthy term infants expe- sleep in normal adults. The partial pressure of CO2
rienced 1 or more extreme apneas. Similar to the increases 3 to 7 mm Hg the partial pressure of
conventional apneas, only preterm children had a arterial O2 (PaO2) decreases 3 to 9 mm Hg, and
significant increase in the risk of experiencing the SpO2 decreases 2% in comparison with wake-
extreme apneas. This increased risk persisted until fulness.1 Similar changes are seen in children. As
about 43 weeks postmenstrual age. Of concern, a in adults, these changes can be exaggerated in
follow-up study of developmental outcomes in this children with lung disease or upper airway
same cohort in the second year of life showed that obstruction. Most information about gas exchange
5 or more cardiorespiratory events per hour was during sleep in children comes from the use of
associated with lower adjusted mean differences noninvasive monitoring techniques including pulse
in the mental development index of the Bayley oximetry measurements of SpO2, transcutaneous
Scales of Development in both term and preterm measurement of CO2 (tcCO2), and end-tidal CO2
infants.57 (EtCO2) measurement. Normal values from infancy
to adolescence are summarized in Table 3.
RESPIRATORY EVENTS DURING SLEEP
BEYOND THE FIRST YEAR OF LIFE OXYGEN
The first study to establish normal values for respi- The ventilatory response to changes in PaO2 is
ratory events using polysomnography during sleep exponential. There is little increase in ventilation
in children was published in 1978, and included 22 until PaO2 falls below 60 mm Hg. The response to
children aged 9 to 13 years.15 Short respiratory hypoxemia is augmented by hypercapnia, and de-
pauses, more frequent during stage 1 and REM creases with age and training. Normative data for
sleep, were reported. No obstructive events were noninvasive measures of oxygenation (SpO2 by
described. In 1992, Marcus and colleagues20 pub- pulse oximetry) from infancy through adolescence
lished a more comprehensive study of 50 children are summarized in Table 3. Data from the Collab-
aged 1 to 17 years. A limitation to this study was orative Home Infant Monitoring Evaluation study
the lack of neurophysiologic confirmation of sleep. published in 1999 provides the most comprehen-
Obstructive apneas were found to be rare in chil- sive study of SpO2 during infancy.60 Their report
dren, occurring in only 18%, with an average index of longitudinal data of SpO2 during the first 25 post-
of 0.1  0.5 per hour (range 0–3.1) and no obstruc- natal weeks in 64 healthy term infants provided
tive apneas longer than 10 seconds. Central ap- valuable information not addressed in earlier
neas were more common, with 30% of children studies, which were generally limited to 1 to 2
in this study having a central apnea longer than nights of monitoring, or brief recording periods
10 seconds. Subsequent studies have confirmed during respiratory events spread out over
that short central pauses (generally 10 seconds weeks.61–63 The median baseline SpO2 in these
or less) are part of the normal pattern of breathing healthy infants was 97.9% (10th percentile SpO2
during sleep in children and adolescents,21,58 and 95.2%) and did not change with age or sleep posi-
occur more often in REM than in NREM sleep.59 tion. Most infants in this study had at least one
The frequency of respiratory pauses, summarized 3-minute epoch with SpO2 lower than 90%, and
 Table 2
Central apnea indices in healthy children from infancy through adolescence

12.5 y 15.2 y 16.9 y

8 y (6.3–10.1) 11.5 y (9.7–13.2) (11.4–17.2) (13.4–17.2) (15.8–17.9)
1.4 y (1.1–1.9) 3 y (2.2–3.5) 5 y (4.3–5.8) Tanner I Tanner II Tanner III Tanner IV Tanner V
Central apnea 2.8 (1.0–4.3) 1.5 (0.7–6.9) 1.1 (0.5–3.2) 0.9 (0.3–2.7) 0.4 (0.0–1.2) 0.5 (0.1–1.8) 0.1 (0.0–0.8) 0.1 (0.0–1.3)
index
(episodes per

Sleep and Respiratory Physiology in Children

hour of total
sleep time)
Maximum 11.7 (10.1–14.5) 13 (9.7–18.8) 14.1 (10.1–17.4) 15.5 (9.7–20.1) 13.3 (9.0–18.2) 13.0 (9.9–16.3) 12.5 (10.9–20.3) 11.2 (8.8–24.0)
apnea
duration (s)
SpO2 nadir (%) 92.0 (87.6–95.0) 92.0 (88.8–95.0) 94.0 (83.6–96.0) 93.5 (87.0–96.0) 94.0 (91.4–96.0) 93.5 (90.5–96.0) 93.5 (91.0–96.0) 94.0 (91.5–96.0)

Data are presented as median (10th–90th percentile). These indices are based on central apneas that are scored without requiring an associated desaturation or arousal.
Data from Scholle S, Wiater A, Scholle HC. Normative values of polysomnographic parameters in childhood and adolescence: cardiorespiratory parameters. Sleep Med
2011;12(10):988–96.

7
8 Ross & Rosen

Table 3
Oxygen saturation and carbon dioxide normal values in infancy through adolescence

Infant Preschool School Age Adolescent

60,67,77 8,9,82 8,9,71,72,82
SpO2 mean or median 97–99 99 96–99 97–9916,68,82
SpO2 lower limit of normal (5th 94–9558,64,74 958,9,82 958,9,71,72,82 9512,55,65
percentile or 2 SD below mean)
Desaturation index (4) 0.39 0.5–1.29,71,72
Decreases in SpO2 to less than 90% (per hour)
Median (IQR) 0.0 (0.0–0.2)82 0.0 (0.0–0.3)82 0.0 (0.0)82
Range 0.0–2.7 0.0–0.6 0.0–0.3
tcCO2 mean or median 40–4149,77,80 42
Interquartile range 37–4449,80 40–46
EtCO2 mean (SD) 40.6 (4.6)9 40.7 (4.5)9 40.7 (4.5)9
Percent sleep time with 20.4 (28)9 1.6 (3.8)7 1.6 (3.8)7
EtCO2 >45 mm Hg, mean (SD) 6.9 (19.1)20 6.9 (19.1)20
21.6 (29)9

acute decreases in SpO2 occurred in most infants. the severity of resulting hypoxemia, the increase
These transient acute decreases improved with in the stability of oxygenation with age is likely
age and reduced the frequency of periodic breath- due to increased and more stable lung volumes
ing pattern, and based on this and earlier work the relative to oxygen consumption in the older child.
investigators concluded that they are part of the
normal breathing and oxygenation behavior in CARBON DIOXIDE
infants.60,61
Gestational age and sleep state influence the Arterial CO2 is maintained with relatively little vari-
development of stability in arterial oxygenation. ation. Ventilation increases linearly in response to
Although healthy preterm infants have baseline increasing CO2 production in wakefulness and
SpO2 values in the same range as full-term infants, sleep. This response is augmented by hypoxia;
the variability about the baseline is greater in pre- the slope of the increase in ventilation in response
term infants. The frequency of transient desatura- to a given pressure of arterial CO2 (PaCO2) is
tion episodes to 80% or less varies considerably increased in the presence of a lower PaO2.
with age and between individual patients, with Although most healthy children maintain a resting
rates being highest in preterm infants, lower in PaCO2 near 40 mm Hg, there is substantial varia-
term infants,60,61,64–67 and lowest in older children tion in the set point and sensitivity. Under normal
and adolescents. During the regular breathing of conditions during NREM sleep, changes in PaCO2
quiet or NREM sleep, most infants do not have ep- of 2 mm Hg or less are promptly recognized by
isodes of desaturation, or when episodes do occur chemoreceptors. The rapid negative feedback
they are brief. By contrast, during active or REM system between the brain and lungs maintains
sleep, the brief apneic pauses are more likely to PaCO2 within a small range between the eupnea
be associated with desaturation. and apnea thresholds.
In healthy children 2 to 16 years of age, using a Arterial CO2 can be estimated noninvasively us-
desaturation criterion of 90% or less, most chil- ing transcutaneous tcCO2 and EtCO2; each method
dren do not show any episodes of desatura- has advantages and disadvantages.76 Normative
tion.7–9,16,20,21,68–72 This increased stability in data using these noninvasive measurements are
oxygenation may be partially explained by devel- summarized in Table 3. In a sample of healthy
opmental changes in the relationship between term infants during the first 9 months of life, tcCO2
lung volume and oxygen consumption. Infants values averaged 40.5  2.25 mm Hg, with no dif-
have a highly compliant chest wall, resulting in a ferences between active and quiet sleep.77 tcCO2
functional residual capacity that is significantly values during sleep change little with postnatal
lower than that of adults when compared on the age during the first 2 years of life.77–79 Normal chil-
basis of metabolism.73 Lung volume in infants de- dren show an increase in EtCO2 values of 4 to
creases even further during apneic pauses74 and 10 mm Hg during sleep, spending an average of
during REM sleep.75 Because lung volume at the 6.9 1/ 19.1% of sleep time with EtCO2 values
onset of breath-holding is a major determinant of above 45 mm Hg.20 In a sample of 50 healthy
 Sleep and Respiratory Physiology in Children 9

children and adolescents, children had EtCO2 in breathing during sleep. New York: Informa
values above 45 mm Hg for an average of 6.9% Healthcare USA, Inc; 2008. p. 19–46.
 19.1% of sleep time. In this same population, 7. Uliel S, Tauman R, Greenfeld M, et al. Normal poly-
peak EtCO2 values ranged from 38 to 53 mm Hg, somnographic respiratory values in children and
with a mean of 46  4 mm Hg.20 CO2 values are adolescents. Chest 2004;125(3):872–8.
statistically lower in REM than in NREM sleep, 8. Traeger N, Schultz B, Pollock AN, et al. Polysomno-
but the difference (1–2 mm Hg) is clinically trivial. graphic values in children 2-9 years old: additional
The limited normative data for CO2 values data and review of the literature. Pediatr Pulmonol
measured during polysomnography from infancy 2005;40(1):22–30.
to childhood are summarized in Table 3. A recent 9. Montgomery-Downs HE, O’Brien LM, Gulliver TE,
study9 describes much higher values during sleep et al. Polysomnographic characteristics in normal
in children than previously reported.7,20 preschool and early school-aged children. Pediat-
rics 2006;117(3):741–53.
10. Crosfill ML, Widdicombe JG. Physical characteris-
SUMMARY tics of the chest and lungs and the work of breath-
ing in different mammalian species. J Physiol 1961;
Sleep-related changes in respiratory physiology in
158:1–14.
children vary with age, but generally include a
11. Iliff A, Lee VA. Pulse rate, respiratory rate, and
reduction in respiratory rate and minute ventila-
body temperature of children between two months
tion, mild reductions in SpO2, mild increases in
and eighteen years of age. Child Dev 1952;23(4):
PCO2, the presence of short central respiratory
237–45.
pauses, and rare obstructive respiratory events.
12. Bardella IJ. Pediatric advanced life support: a review
Knowledge of the maturation of breathing patterns
of the AHA recommendations. American Heart Asso-
from infancy through adolescence is crucial to the
ciation. Am Fam Physician 1999;60(6):1743–50.
ability of the physician to evaluate children for
13. Gagliardi L, Rusconi F. Respiratory rate and body
common disorders of sleep.
mass in the first three years of life. The working
party on respiratory rate. Arch Dis Child 1997;
76(2):151–4.
REFERENCES
14. Gaultier C. Respiratory adaptation during sleep
1. Kreiger J. Breathing during sleep in normal sub- from the neonatal period to adolescence. In:
jects. In: Kryger M, Roth T, editors. Principles and Guilleminault C, editor. Sleep and its disorders in
practices of sleep medicine. Philadelphia: Elsevier children. New York: Raven Press; 1987. p. 67–98.
Saunders; 2005. p. 232–44. 15. Carskadon MA, Harvey K, Dement WC, et al.
2. Rosen CL. Maturation of breathing during sleep. Respiration during sleep in children. West J Med
In: Marcus CL, Carroll JL, Donnelly DF, et al, edi- 1978;128(6):477–81.
tors. Sleep and breathing in children, second edi- 16. Tabachnik E, Muller NL, Bryan AC, et al. Changes
tion, developmental changes in breathing during in ventilation and chest wall mechanics during
sleep. New York: Informa Healthcare USA, Inc; sleep in normal adolescents. J Appl Physiol Respir
2008. p. 117–30. Environ Exerc Physiol 1981;51(3):557–64.
3. Carroll JL, Donnelly DF. Postnatal development of 17. Mortola JP. Some functional mechanical implica-
carotid chemoreceptor function. In: Marcus CL, tions of the structural design of the respiratory sys-
Carroll JL, Donnelly DF, et al, editors. Sleep and tem in newborn mammals. Am Rev Respir Dis
breathing in children, second edition, develop- 1983;128(2 Pt 2):S69–72.
mental changes in breathing during sleep. New 18. Polgar G, Weng TR. The functional development
York: Informa Healthcare USA, Inc; 2008. p. 47–73. of the respiratory system from the period of gesta-
4. Marcus CL, Glomb WB, Basinski DJ, et al. Devel- tion to adulthood. Am Rev Respir Dis 1979;120(3):
opmental pattern of hypercapnic and hypoxic 625–95.
ventilatory responses from childhood to adulthood. 19. Acebo C, Millman RP, Rosenberg C, et al. Sleep,
J Appl Physiol (1985) 1994;76(1):314–20. breathing, and cephalometrics in older children
5. Thach BT. Maturation and transformation of re- and young adults. Part I – normative values. Chest
flexes that protect the laryngeal airway from liquid 1996;109(3):664–72.
aspiration from fetal to adult life. Am J Med 2001; 20. Marcus CL, Omlin KJ, Basinki DJ, et al. Normal pol-
111(Suppl 8A):69S–77S. ysomnographic values for children and adoles-
6. Praud JP, Samson N. Laryngeal function and cents. Am Rev Respir Dis 1992;146(5 Pt 1):1235–9.
neonatal respiration. In: Marcus CL, Carroll JL, 21. Tang J, Rosen CL, Larkin EK, et al. Identification of
Donnelly DF, et al, editors. Sleep and breathing in sleep-disordered breathing in children: variation
children, second edition, developmental changes with event definition. Sleep 2002;25(1):72–9.
10 Ross & Rosen

22. Verhulst SL, Schrauwen N, Haentjens D, et al. 36. Committee on Fetus and Newborn. American
Reference values for sleep-related respiratory Academy of Pediatrics. Apnea, sudden infant
variables in asymptomatic European children and death syndrome, and home monitoring. Pediatrics
adolescents. Pediatr Pulmonol 2007;42(2):159–67. 2003;111(4 Pt 1):914–7.
23. Witmans MB, Keens TG, Ward SL, et al. Obstruc- 37. Henderson-Smart DJ. The effect of gestational age
tive hypopneas in children and adolescents. Am on the incidence and duration of recurrent apnoea
J Respir Crit Care Med 2003;168(12):1540. in newborn babies. Aust Paediatr J 1981;17(4):
24. Elder DE, Campbell AJ, Galletly D. Current defini- 273–6.
tions for neonatal apnoea: are they evidence 38. Eichenwald EC, Aina A, Stark AR. Apnea
based? J Paediatr Child Health 2013;49(9): frequently persists beyond term gestation in in-
E388–96. fants delivered at 24 to 28 weeks. Pediatrics
25. Ellingson RJ, Peters JF, Nelson B. Respiratory 1997;100(3 Pt 1):354–9.
pauses and apnea during daytime sleep in normal 39. Darnall RA, Kattwinkel J, Nattie C, et al. Margin of
infants during the first year of life: longitudinal ob- safety for discharge after apnea in preterm infants.
servations. Electroencephalogr Clin Neurophysiol Pediatrics 1997;100(5):795–801.
1982;53(1):48–59. 40. Ramanathan R, Corwin MJ, Hunt CE, et al. Cardio-
26. Guilleminault C, Ariagno R, Korobkin R, et al. Mixed respiratory events recorded on home monitors:
and obstructive sleep apnea and near miss for comparison of healthy infants with those at
sudden infant death syndrome: 2. Comparison of increased risk for SIDS. JAMA 2001;285(17):
near miss and normal control infants by age. Pedi- 2199–207.
atrics 1979;64(6):882–91. 41. Albani M, Bentele KH, Budde C, et al. Infant sleep
27. Flores-Guevara R, Plouin P, Curzi-Dascalova L, apnea profile: preterm vs. term infants. Eur J Pe-
et al. Sleep apneas in normal neonates and infants diatr 1985;143(4):261–8.
during the first 3 months of life. Neuropediatrics 42. Upton CJ, Milner AD, Stokes GM. Upper airway
1982;13(Suppl):21–8. patency during apnoea of prematurity. Arch Dis
28. Kahn A, Groswasser J, Sottiaux M, et al. Clinical Child 1992;67(Spec No 4):419–24.
symptoms associated with brief obstructive sleep 43. Miller MJ, Carlo WA, Martin RJ. Continuous positive
apnea in normal infants. Sleep 1993;16(5):409–13. airway pressure selectively reduces obstructive
29. Kahn A, Groswasser J, Rebuffat E, et al. Sleep and apnea in preterm infants. J Pediatr 1985;106(1):
cardiorespiratory characteristics of infant victims of 91–4.
sudden death: a prospective case-control study. 44. Duara S, Silva Neto G, Claure N, et al. Effect of
Sleep 1992;15(4):287–92. maturation on the extrathoracic airway stability of
30. Haddad GG, Epstein RA, Epstein MA, et al. Matu- infants. J Appl Physiol (1985) 1992;73(6):2368–72.
ration of ventilation and ventilatory pattern in 45. Gauda EB, McLemore GL, Tolosa J, et al. Matura-
normal sleeping infants. J Appl Physiol Respir En- tion of peripheral arterial chemoreceptors in rela-
viron Exerc Physiol 1979;46(5):998–1002. tion to neonatal apnoea. Semin Neonatol 2004;
31. Hoppenbrouwers T, Hodgman JE, Cabal L. 9(3):181–94.
Obstructive apnea, associated patterns of move- 46. Heldt GP. Development of stability of the respira-
ment, heart rate, and oxygenation in infants at low tory system in preterm infants. J Appl Physiol
and increased risk for SIDS. Pediatr Pulmonol (1985) 1988;65(1):441–4.
1993;15(1):1–12. 47. Woodrum DE, Oliver TK Jr, Hodson WA. The effect
32. Franks CI, Watson JB, Brown BH, et al. Respiratory of prematurity and hyaline membrane disease on
patterns and risk of sudden unexpected death in oxygen exchange in the lung. Pediatrics 1972;
infancy. Arch Dis Child 1980;55(8):595–9. 50(3):380–6.
33. Southall DP, Richards JM, Rhoden KJ, et al. Pro- 48. Kelly DH, Riordan L, Smith MJ. Apnea and periodic
longed apnea and cardiac arrhythmias in infants breathing in healthy full-term infants, 12-18 months
discharged from neonatal intensive care units: fail- of age. Pediatr Pulmonol 1992;13(3):169–71.
ure to predict an increased risk for sudden infant 49. Schluter B, Buschatz D, Trowitzsch E. Polysomno-
death syndrome. Pediatrics 1982;70(6):844–51. graphic reference curves in the first and second
34. Southall DP, Richards JM, Stebbens V, et al. year of life. Somnologie 2001;5:3–16.
Cardiorespiratory function in 16 full-term infants 50. Glotzbach SF, Baldwin RB, Lederer NE, et al. Peri-
with sudden infant death syndrome. Pediatrics odic breathing in preterm infants: incidence and
1986;78(5):787–96. characteristics. Pediatrics 1989;84(5):785–92.
35. Gordon D, Southall DP, Kelly DH, et al. Analysis of 51. Simakajornboon N, Beckerman RC, Mack C, et al.
heart rate and respiratory patterns in sudden infant Effect of supplemental oxygen on sleep architec-
death syndrome victims and control infants. Pediatr ture and cardiorespiratory events in preterm in-
Res 1986;20(7):680–4. fants. Pediatrics 2002;110(5):884–8.
 Sleep and Respiratory Physiology in Children 11

52. Barrington KJ, Finer NN. Periodic breathing and 68. Chipps BE, Mak H, Schuberth KC, et al. Nocturnal
apnea in preterm infants. Pediatr Res 1990;27(2): oxygen saturation in normal and asthmatic chil-
118–21. dren. Pediatrics 1980;65(6):1157–60.
53. Khan A, Qurashi M, Kwiatkowski K, et al. Measure- 69. Owen G, Canter R. Analysis of pulse oximetry data
ment of the CO2 apneic threshold in newborn infants: in normal sleeping children. Clin Otolaryngol Allied
possible relevance for periodic breathing and ap- Sci 1997;22(1):13–22.
nea. J Appl Physiol (1985) 2005;98(4):1171–6. 70. Rosen CL, Larkin EK, Kirchner HL, et al. Preva-
54. Finer NN, Barrington KJ, Hayes B. Prolonged peri- lence and risk factors for sleep-disordered breath-
odic breathing: significance in sleep studies. Pedi- ing in 8- to 11-year-old children: association with
atrics 1992;89(3):450–3. race and prematurity. J Pediatr 2003;142(4):383–9.
55. Haddad GG, Bazzy AR, Chang SL, et al. Heart rate 71. Urschitz MS, Wolff J, Von Einem V, et al. Reference
pattern during respiratory pauses in normal infants values for nocturnal home pulse oximetry during
during sleep. J Dev Physiol 1984;6(4):329–37. sleep in primary school children. Chest 2003;
56. Upton CJ, Milner AD, Stokes GM. Apnoea, brady- 123(1):96–101.
cardia, and oxygen saturation in preterm infants. 72. Moss D, Urschitz MS, von Bodman A, et al. Refer-
Arch Dis Child 1991;66(Spec No 4):381–5. ence values for nocturnal home polysomnography
57. Hunt CE, Corwin MJ, Baird T, et al. Cardiorespira- in primary schoolchildren. Pediatr Res 2005;58(5):
tory events detected by home memory monitoring 958–65.
and one-year neurodevelopmental outcome. 73. Cook CD, Cherry RB, O’Brien D, et al. Studies
J Pediatr 2004;145(4):465–71. of respiratory physiology in the newborn infant. I.
58. Scholle S, Wiater A, Scholle HC. Normative values Observations on normal premature and full-term in-
of polysomnographic parameters in childhood and fants. J Clin Invest 1955;34(7 Pt 1):975–82.
adolescence: cardiorespiratory parameters. Sleep 74. Olinsky A, Bryan MH, Bryan AC. Influence of lung
Med 2011;12(10):988–96. inflation on respiratory control in neonates. J Appl
59. Gaultier C. Apnea and sleep state in newborns and Physiol (1985) 1974;36(4):426–9.
infants. Biol Neonate 1994;65(3–4):231–4. 75. Henderson-Smart DJ, Read DJ. Reduced lung vol-
60. Hunt CE, Corwin MJ, Lister G, et al. Longitudinal ume during behavioral active sleep in the newborn.
assessment of hemoglobin oxygen saturation in J Appl Physiol Respir Environ Exerc Physiol 1979;
healthy infants during the first 6 months of age. 46(6):1081–5.
Collaborative Home Infant Monitoring Evaluation 76. Morielli A, Desjardins D, Brouillette RT. Transcuta-
(CHIME) Study Group. J Pediatr 1999;135(5): neous and end-tidal carbon dioxide pressures
580–6. should be measured during pediatric polysom-
61. Stebbens VA, Poets CF, Alexander JR, et al. Oxy- nography. Am Rev Respir Dis 1993;148(6 Pt 1):
gen saturation and breathing patterns in infancy. 1599–604.
1: full term infants in the second month of life. 77. Horemuzova E, Katz-Salamon M, Milerad J. Breath-
Arch Dis Child 1991;66(5):569–73. ing patterns, oxygen and carbon dioxide levels in
62. Levene S, McKenzie SA. Transcutaneous oxygen sleeping healthy infants during the first nine
saturation in sleeping infants: prone and supine. months after birth. Acta Paediatr 2000;89(11):
Arch Dis Child 1990;65(5):524–6. 1284–9.
63. Hunt CE, Hufford DR, Bourguignon C, et al. Home 78. Hoppenbrouwers T, Hodgman JE, Arakawa K,
documented monitoring of cardiorespiratory et al. Transcutaneous oxygen and carbon dioxide
pattern and oxygen saturation in healthy infants. during the first half year of life in premature and
Pediatr Res 1996;39(2):216–22. normal term infants. Pediatr Res 1992;31(1):73–9.
64. Poets CF, Stebbens VA, Alexander JR, et al. Oxy- 79. Schafer H, Koehler U, Ewig S, et al. Obstructive
gen saturation and breathing patterns in infancy. sleep apnea as a risk marker in coronary artery dis-
2: preterm infants at discharge from special care. ease. Cardiology 1999;92(2):79–84.
Arch Dis Child 1991;66(5):574–8. 80. Schafer T, Schafer D, Schlafke ME. Breathing,
65. Poets CF, Stebbens VA, Alexander JR, et al. Arterial transcutaneous blood gases, and CO2 response
oxygen saturation in preterm infants at discharge in SIDS siblings and control infants during sleep.
from the hospital and six weeks later. J Pediatr J Appl Physiol (1985) 1993;74(1):88–102.
1992;120(3):447–54. 81. Poets CF, Stebbens VA, Southall DP. Arterial oxy-
66. Richard D, Poets CF, Neale S, et al. Arterial oxygen gen saturation and breathing movements during
saturation in preterm neonates without respiratory the first year of life. J Dev Physiol 1991;15(6):
failure. J Pediatr 1993;123(6):963–8. 341–5.
67. Poets CF, Stebbens VA, Lang JA, et al. Arterial 82. Poets CF, Stebbens VA, Samuels MP, et al. Oxygen
oxygen saturation in healthy term neonates. Eur J saturation and breathing patterns in children. Pedi-
Pediatr 1996;155(3):219–23. atrics 1993;92(5):686–90.