Egyptian_Pediatric yahoo group Egyptian_Pediatric yahoo group

http://health.groups.yahoo.com/group/ http://health.groups.yahoo.com/group/
egyptian_pediatric/ egyptian_pediatric/
 Symposium: respiratory medicine
Paediatric applied
respiratory physiology –
the essentials
Kevin Madden
Robinder G Khemani
Christopher JL Newth
Abstract
General paediatricians will encounter myriad respiratory abnormalities
during their careers. A basic knowledge of essential respiratory physi­
ology, its subsequent derangements due to disease states and how
to assess these abnormalities will help in the proper care of children.
This paper will begin with an overview of normal respiratory physiology
and how to monitor the efficiency of gas exchange. It will also discuss
common methods of non-invasive monitoring including pulse oximetry,
carbon dioxide monitoring, pulmonary function tests and respiratory
impedance plethysmography. Finally, paediatric disease states will be
used to illuminate the intersection between pathophysiology, clinical
symptoms and monitoring capabilities.
Keywords lung volume measurements; oximetry; paediatrics; plethys­
mography; pulmonary ventilation; respiratory function tests
Introduction
Bearing in mind the diversity and prevalence of respiratory ill-
nesses in children, paediatricians should understand the basics
of respiratory physiology and how to monitor respiratory func-
tion. This discussion will review normal respiratory physiology
and explore non-invasive forms of respiratory monitoring. With
this foundation, the paediatrician can accurately diagnose and
assess the severity of illness.
Kevin Madden MD is at the Department of Anesthesia and Critical Care
Medicine, Children’s Hospital Los Angeles, Los Angeles, USA.
Robinder G Khemani MD MSCI is Assistant Professor of Pediatrics,
University of Southern California, Keck School of Medicine, Children’s
Hospital Los Angeles, Department of Anesthesia and Critical Care
Medicine, Los Angeles, USA.
Christopher JL Newth MD FRCPC FRACP is Professor of Pediatrics, University
of Southern California, Keck School of Medicine, Children’s Hospital
Los Angeles, Department of Anesthesia and Critical Care Medicine, Los
Angeles, USA.
PAEDIATRICS AND CHILD HEALTH 19:6
Brief overview of normal respiratory physiology
Muscles of respiration
The most important and powerful muscle during the inspiratory
phase of respiration is the diaphragm, a dome-shaped musculofi-
brous septum that separates the thorax from the abdominal cav-
ity. When the diaphragm contracts, abdominal contents move
downward and the lung expands in the vertical and horizon-
tal planes. During normal tidal breathing the diaphragm moves
approximately 1 cm, but with forced inspiration and exhalation,
it can move up to 10 cm.
During inspiration, external intercostal muscles elevate and
move the ribs forward. This increases the lateral and anteropos-
terior diameters of the thoracic cavity. The two most common
accessory muscles of inspiration are the sternocleidomastoid and
scalenes. The sternocleidomastoid raises the sternum while sca-
lenes elevate the first two ribs. During normal respiration these
muscles do not participate in inspiration, but during exercise
or in pathological processes they can play an important role in
maintaining normal alveolar ventilation.
While expiration is normally passive due to the elastic proper-
ties of the lungs and chest wall, both exercise and certain patho-
physiological conditions invoke both the internal intercostal and
abdominal muscles including the internal and external obliques,
the transversus abdominis and the rectus abdominis. These mus-
cles work to decrease the thoracic volume and assist in forcing
air from the lungs.
Both the lungs and chest wall are elastic, and each compo-
nent has a natural propensity; the lung to collapse inward and
the chest wall to ‘spring’ outward. The equilibrium point of lung
volume where these forces are balanced is the functional residual
capacity (FRC).
Lung volumes and capacities
The various lung volumes and capacities can be measured during
different phases of the respiratory cycle and change under differ-
ent pathophysiological conditions. Spirometry is used to record
the volume of air moved during respiration.
The are four lung volumes which, when added together, equal
the total lung capacity (TLC) (Figure 1):
• tidal volume (VT) – volume of air inspired or expired during a
normal breath
• inspiratory reserve volume (IRV) – the additional volume of
air that can be inspired in addition to a tidal breath
• expiratory reserve volume (ERV) – the additional volume of
air that can be expired after the end of a tidal breath
• residual volume (RV) – volume of air remaining after the most
forceful expiration.
In evaluating a patient’s clinical status and understanding patho-
physiology, it is sometimes advantageous to consider two or
more lung volumes together as capacities. The four lung capaci-
ties (Figure 1) are:
• inspiratory capacity (IC) – equals tidal volume plus inspira-
tory reserve volume
• functional residual capacity (FRC) – equals expiratory reserve
volume plus residual volume, or the volume of air remaining
at the end of a normal expiratory breath
• vital capacity (VC) – equals inspiratory reserve volume plus
tidal volume plus expiratory reserve volume
249 © 2009 Elsevier Ltd. All rights reserved.
 Symposium: respiratory medicine
IRV
IC
VC
TLC
VT
ERV
FRC
RV
Figure 1 The various lung volumes and capacities obtained by
spirometry. IC, inspiratory capacity; VC, vital capacity; VT, tidal volume;
TLC, total lung capacity; RV, residual volume; IRV, inspiratory reserve
volume; ERV, expiratory reserve volume; FRC, functional residual
capacity.
• total lung capacity (TLC) – equals vital capacity plus residual
volume.
Note that since by definition residual volume cannot be exhaled
from the lungs, it cannot be measured by spirometry. FRC and
TLC also cannot be measured by spirometry alone but instead
are measured by gas-dilution techniques or plethysmography.
Gas exchange
The primary purpose of the respiratory system is gas exchange
to maintain cellular homeostasis. The two principal components
are delivery of oxygen and removal of carbon dioxide.
Oxygenation
The respiratory system helps to extract oxygen from the atmo-
sphere and deliver it to mitochondria. The partial pressure of
oxygen in the alveolus (PAO2) is a primary determinant of arterial
oxygen tension (PaO2).
PA O2 = [(Pb − PH2O )∗ FiO2 ] − PA CO2 / RQ
where Pb is barometric pressure, PH O is the partial pressure of
2
water vapour and FiO2 is the fraction of inspired oxygen. PACO2
is the partial pressure of carbon dioxide in the alveolus and RQ
is the respiratory quotient. For most purposes RQ is assumed to
be 0.8.
Substituting normal values for an individual breathing room
air at sea level, the PAO2 is approximately 100 mmHg. As oxygen
crosses the alveolar membrane into the pulmonary capillary net-
work a negligible amount of oxygen tension is lost (around 10
mmHg). Thus, the PaO2 of a normal individual is approximately
90 mmHg. By examining the alveolar gas equation closely, one
can see that three conditions can cause a decrease in PAO2:
altitude (low Pb), hypoxic gas mixture (low FiO2) and hypoven-
tilation (high PACO2). The most common aetiology of hypox-
aemia is neither altitude nor hypoventilation, however, but a
PAEDIATRICS AND CHILD HEALTH 19:6
­ isturbance in the number of alveoli participating in matched
d
ventilation/perfusion (V′/Q′). With either compromised ventila-
tion with adequate perfusion (shunt) or adequate ventilation
with compromised perfusion (dead space), oxygen residing in
the alveolus cannot move into the pulmonary capillary network
and hypoxaemia will ensue. A wide variety of clinical condi-
tions can cause derangements in V′/Q′, and interventions such
as continuous positive airway pressure (CPAP), biphasic posi-
tive airway pressure (BiPAP) or endotracheal intubation with
mechanical ventilation correct hypoxaemia by restoring normal
lung volumes, assisting cardiac function and improving the
V′/ Q′ ­relationship.
Ventilation
The respiratory system also eliminates carbon dioxide from
the blood through the alveolus. Arterial carbon dioxide ten-
sion (PaCO2) is directly proportional to minute ventilation (VE)
where:
VE = respiratory rate × tidal volume
VE = fVT
It is worth noting that VT comprises dead-space volume (VD)
and alveolar volume (VA).
Dead-space volume is the portion of the tidal breath that does
not participate in gas exchange with pulmonary capillaries. Dead
space comprises anatomical dead space and non-anatomical
dead space. Anatomical dead space is found within the conduct-
ing airways – nose, mouth, oropharynx, trachea, bronchi and
bronchioles – and accounts for approximately 20–30% of a tidal
breath, although it is relatively larger in infants. Non-anatomi-
cal dead space approaches zero in healthy individuals, as most
lung units are equivalently ventilated and perfused. In infants
and children with respiratory disease, however, total dead space
may approach 60–70% of a tidal breath.
In contrast, alveolar volume is the portion of inspired breath
that arrives at the alveoli and participates in gas exchange with
pulmonary capillaries. Therefore, it is more accurate to state that
PaCO2 is proportional to alveolar minute ventilation (MVA):
MVA = f ( VT − VD )
Hence, an elevated PaCO2 can arise from a decrease in respiratory
rate, a decrease in tidal volume, or an increase in dead space.
Methods of assessing respiratory function in non-
intubated patients
Pulse oximetry
Cyanosis is the hallmark clinical sign of hypoxaemia, but it can
only be recognized confidently when the oxygen saturation is
below 75% and cannot be recognized if the haematocrit is less
than 15%. Pulse oximetry allows for non-invasive and continu-
ous monitoring of arterial oxygen saturation (SaO2). The basic
principles of pulse oximetry are that oxygenated haemoglobin
(HbO2) absorbs mostly infrared light while deoxygenated hae-
moglobin (Hb) absorbs mostly red light. Pulse oximeters exploit
the pulsatile nature of arterial blood and successfully ‘ignore’ the
250 © 2009 Elsevier Ltd. All rights reserved.
 Symposium: respiratory medicine
contributions of the skin, tissues and venous blood. Each stroke
volume ejected from the heart corresponds to an increase in arte-
rial blood volume across the measuring site. During systole light
absorption peaks, while during diastole it nadirs. The amplitude
of the waveform generated is proportional to the amplitude of
arterial blood volume. Pulse oximeters, therefore, can detect
pulsus paradoxus, an important finding in certain pulmonary
­dysfunctions.
Pulse oximeters can produce spuriously high or low readings
when a patient has a dyshaemoglobinaemia. In carbon monox-
ide intoxication, carboxyhaemoglobin (HbCO) absorbs light in
the red wavelength and is interpreted by the pulse oximeter as
oxygenated haemoglobin, falsely elevating the SpO2. Methae-
moglobin (Hbmet) is absorbed at a wavelength between red and
infrared light, leading to a falsely lowered SpO2 that generally
plateaus between 85 and 88%. The partial pressure of oxygen,
however, is not changed by the various dyshaemoglobinaemias.
Co-oximetry performed on arterial blood samples will accurately
measure SaO2 as this technique can measure the proportion of
other forms of haemoglobin. Under normal conditions, however,
pulse oximetry measurements correlate well with co-oximetry
when the saturation is between 70 and 100%.
Ventilation
The ability to monitor the efficiency of ventilation by non-
­invasive methods provides the clinician with critical information.
Often the underlying disease masks signs of CO2 retention and
clinical diagnosis can be extremely difficult. The features only
appear when the PaCO2 is above 100 mmHg and can occur only
when the inspired air is enriched with oxygen, as the highest
PCO2 possible while breathing room air is about 90 mmHg. This
degree of hypercapnia by itself does not appear to be dangerous,
and death usually results from the associated hypoxia.
Transcutaneous CO2 monitors (TCOMs) offer a reliable, non-
invasive method for estimating arterial carbon dioxide. A small
adhesive patch, usually placed on the abdomen, warms the skin
and allows CO2 to diffuse readily where it is read by the TCOM.
In general, TCOM monitoring correlates well with PaCO2,
although overestimations can occur due to heat and local CO2
production. While the accuracy of TCOMs has been historically
poor in obese patients, newer-generation TCOMs have improved
reliability.
Nasal cannula end-tidal CO2 detection: an easy and non-inva-
sive way of measuring end-tidal CO2 in spontaneously breathing,
non-intubated patients is by using divided nasal cannulas that
simultaneously deliver oxygen via one prong and sample exhaled
gas through the other. Studies have shown that the end-tidal CO2
correlates highly with PaCO2. Such monitoring is useful for pro-
cedural sedation or to monitor children at risk for impending
respiratory failure. It is physiologically impossible for end-tidal
CO2 monitors to read greater than PaCO2, so any elevated value
should be taken seriously and further investigated by obtaining
an arterial blood gas.
Pulmonary function tests
In addition to measuring lung volumes and capacities, spirom-
etry can be utilized with a forced expiratory manoeuvre that
PAEDIATRICS AND CHILD HEALTH 19:6
can help clinicians determine mechanical disorders of the respi-
ratory tract, quantify the degree of disorder and classify it as
obstructive, restrictive or mixed in aetiology. Using a variety of
techniques, these can be done on even uncooperative (albeit
sometimes sedated) infants and young children.
Forced expiratory volume: as discussed earlier, vital capacity
(VC) is the lung volume that can be exhaled after a full inspira-
tion. In contrast, forced vital capacity (FVC) is the lung volume
that can be exhaled after a full inspiration as quickly and forc-
ibly as possible, and is highly reproducible. When compared to
forced expiratory volume (FEV1), which by convention is the
amount of gas forcibly exhaled in one second, one can quickly
classify the nature of respiratory disease. In cooperative children
over the age of 5 years the FEV1/FVC ratio is normally 0.8; values
less than 0.8 represent obstructive lung disease, whereas restric-
tive diseases have proportional decreases in FEV1 and FVC, pre-
serving the ratio of 0.8. Forced expiratory volumes provide an
objective standard to monitor response to treatment or clinical
improvement.
Flow–volume curves: by plotting expiratory and inspiratory
flows against lung volume instead of time, the clinician can
obtain more information about underlying lung or airway dys-
function. Unfortunately, forced expiratory spirometry provides
information solely about expiratory dysfunction. Flow–volume
curves, on the other hand, can reveal both inspiratory and expi-
ratory dysfunction and aid in the classification of the disease as
obstructive or restrictive.
Respiratory inductance plethysmography
The inspiratory cycle begins as the diaphragm moves downward.
The abdomen and rib cage then expand in concert, known as
thoraco-abdominal synchrony. However, various respiratory con-
ditions will alter the timing of these events, resulting in thoraco-
abdominal asynchrony (TAA). The asynchrony can be quantified
using respiratory inductance plethysmography, and the degree
of asynchrony correlates well with the amount of respiratory
dysfunction.
Two elastic belts into which a wire is sewn are worn around
the chest and abdomen. A current is passed through the belts,
generating a magnetic field. The act of breathing changes the
cross-sectional area of the abdomen (ABD) and the rib cage
(RC), altering the shape of the magnetic field generated by the
belts and inducing a measurable opposing current. A computer
analyses the current produced and generates (1) the phase angle,
which is the degree of synchrony of the ABD and RC and (2) the
directional loop.
In a normal child, the phase angle is typically less than 22° and
the directional loop is anticlockwise. As the asynchrony worsens,
the phase angle gets larger, although maintaining the anticlock-
wise direction. When there is complete asynchrony the phase
angle is 180°. In the case of bilateral diaphragmatic paralysis,
the directional loop becomes clockwise (Figure 2). In unilateral
paralysis, a ‘figure 8’ is created, and no phase angle can be mea-
sured. The clinical correlate is known as Hoover’s sign, which is
the paradoxical inward movement of the lower lateral rib cage
during inspiration or in the neonate an abnormal movement
of the umbilicus to the contralateral side.
251 © 2009 Elsevier Ltd. All rights reserved.
 Symposium: respiratory medicine
Rib cage
RC
Abdomen
Synchronous
Rib cage
RC
Abdomen
Phase shift
Asynchronous
Rib cage
RC
Abdomen
Paradoxical
Pressure–rate product
A small water- or air-filled catheter of similar size to a nasogastric
feeding tube can be inserted with its tip in the mid-oesophagus
to measure, relatively non-invasively, both the peak-to-trough
swings in oesophageal pressure with respiration, and the respi-
ratory rate. The product of the two is called the pressure–rate
product and is an excellent surrogate for work of breathing
­measurements.
Case studies
Obstructive airways disease
The primary pathological defect in obstructive airways disease
is airflow limitation. This limitation can occur during expira-
tion, inspiration or both. Both large airways obstruction (croup,
epiglottitis, foreign-body aspiration, laryngomalacia, tracheoma-
lacia) and medium (asthma) and small (bronchiolitis) airways
obstruction have hallmark findings on physical examination and
in the tests discussed above. Knowledge of these patterns assists
the physician to localize the obstruction, determine disease sever-
ity, monitor disease progression and determine treatment.
Extrathoracic obstructions can be clearly differentiated from
intrathoracic obstructions based on characteristic patterns in
flow–volume curves. The primary abnormality of an extratho-
racic obstruction is with inspiratory flow. There is a flatten-
ing of the inspiratory flow limb with a fairly normal expiratory
flow pattern. Intrathoracic obstructions can be subdivided into
PAEDIATRICS AND CHILD HEALTH 19:6
m/s = 0
φ = 0°
ABD
m
m/s = 0.71
φ = 45°
s
Figure 2 Respiratory plethysmography with
ABD the Respitrace. On the left-hand side, rib
cage (RC) and abdomen (ABD) movements
are plotted as a function of time, and the
right-hand side graphically represents the
phase angles and direction of the loop. Note
m/s = 0 the usual situation where the diaphragm
φ = 180° leads inspiration. As breathing moves from
synchronous to asynchronous, the size of
the phase-angle loop widens, although
preserving anticlockwise rotation. However,
ABD
with diaphragm paralysis (paradoxical), the RC
and ABD are exactly 180° out of phase, with
clockwise movement.
­ ariable or fixed lesions. A variable lesion such as intrathoracic
v
tracheomalacia will have abnormalities with expiratory flow and
produce a flattening of the expiratory limb, while the inspira-
tory limb will appear normal. Fixed lesions such as foreign-body
aspiration, external compression from extratracheal masses or
tracheal stenosis have inspiratory and expiratory flow limitation
manifested as a flattening of both the inspiratory and expiratory
limbs of the flow–volume curve (Figure 3).
Croup
Physical examination – the major abnormality in croup (an
extrathoracic obstruction) is inspiratory airflow limitation. Air-
flow is generated by lowering intrathoracic and intratracheal
pressures below extrathoracic atmospheric pressure. According
to Poiseuille’s law, the change in pressure is inversely propor-
tional to the fourth power of the radius of the airway. Therefore,
decreasing a child’s airway radius from 5 mm to 2.5 mm leads to
a 16-fold increase in the pressure for airflow. To accomplish this,
children will increase their work of breathing by using accessory
muscles (i.e. supraclavicular retractions). Moreover, children’s
subglottic submucosa is non-fibrous, and the mucous membrane
is attached more loosely than in adults, allowing for oedema to
accumulate more easily. The soft supporting cartilage of the lar-
ynx and the narrow radius of the child’s airway also allow for
dynamic collapse of airways during inspiration. This manifests
clinically with the classic inspiratory stridor that often accompanies
viral croup.
252 © 2009 Elsevier Ltd. All rights reserved.
 Symposium: respiratory medicine

in amplitude during inspiration known as pulsus paradoxus

50 (­Figure 4). The magnitude of depression during inspiration has
40 been consistently correlated with the severity of extrathoracic
Expiration
airway obstruction and can be used to monitor disease progres-
30
sion or resolution.

Flow (litres/minute)
20
Flow and pressure curve against time – the limitation of
10 inspiratory airflow can be visualized when flow and pressure are
0
simultaneously plotted against time (Figure 5).
500 400 300 200 100
Respiratory inductance plethysmography – phase angles
–10
increase significantly in viral croup and continuous phase-angle
–20 monitoring can demonstrate clinical improvement as resolution
Inspiration –30 of the obstruction correlates with decreasing phase angles and
improving pulmonary function.
–40
Pressure–rate product – this value increases significantly
–50 in croup and can be used continuously to follow response to
Tidal volume (mL)
­therapy.
Figure 3 Flow–volume curve for fixed intrathoracic airway obstruction.
Asthma
(Normal curves are dotted, diseased curves are solid.) There is flow
Physical examination – the major abnormality in asthma is
limitation during inspiration and expiration manifested as a flattening
expiratory airflow limitation. Underlying airway hypersensitivity
on both limbs of the flow–volume curve.
causes reversible narrowing and an increase in airways resis-
tance up to 500% of normal values. This increased resistance
Pulsus paradoxus – as intrathoracic pressure becomes more causes distal air trapping, ultimately leading to elevated end-
negative, afterload on the left ventricle increases and cardiac expiratory lung volumes and a prolonged expiratory phase with
output decreases. Since the pulse oximeter waveform ampli- wheezing. On visual inspection, patients with poorly controlled
tude is proportional to arterial blood volume, there is a decrease asthma often have a barrel-chest deformity due to the chronically

E
F

Plethysmographic wave

A
B
100

C
D
0
Blood pressure (mm Hg)
Inhale

Exhale

Breathing cycle

Figure 4 Pulsus paradoxus in a child with croup. Points A and B represent systolic blood pressure, while points C and D represent diastolic blood
pressure. Points E and F represent the peak amplitude of the pulse-oximeter waveform. Note that at the peak of inspiration (bottom panel) there is
a concomitant decrease in both systolic and diastolic pressure (middle panel, points D and B) and magnitude of the pulse oximeter waveform (top
panel, point F); this is known as pulsus paradoxus.

PAEDIATRICS AND CHILD HEALTH 19:6 253 © 2009 Elsevier Ltd. All rights reserved.
 Symposium: respiratory medicine

pulsus paradoxus correlates with the severity of the asthma

Pre-epinephrine nebulization ­exacerbation.
100 Forced expiratory spirometry and flow-volume curves –
80 ­spirometry represents the most accessible way to gauge the
60
severity of obstruction and monitor the effects of therapy. The
40
measured FEV1/FVC is less than 0.8 since the reduction in air-
20
flow (FEV1) is greater than the reduction in lung volume (FVC),
0
with the degree of reduction correlating with disease severity.
0.08
0.2
0.32
0.44
0.56
0.68
0.80
0.92
1.02
1.14
1.26
1.38
1.50
1.62
1.74
1.86
1.98
–20
–40 The expiratory limitation is readily visible on flow–volume curves
–60 as reduced peak expiratory flow and a characteristic concave
–80 appearance on the expiratory limb. Note the change in shape in
Time (seconds)
the expiratory limb after bronchodilator therapy from concave to
Post-epinephrine nebulization straight, the increase in peak expired flow and FEV1 (Figure 6).
100
80 Restrictive lung disease
60 Whereas obstructive airways diseases are rooted in airflow limi-
40 tation, the hallmark of restrictive lung disease is decreased TLC.
20
The aetiologies are diverse – increased elastic recoil (interstitial
0
inflammation/fibrosis), decreased outward recoil of the chest wall
0.06
0.20
0.34
0.48
0.62
0.76
0.90
1.02
1.16
1.30
1.44
1.58
1.72
1.86
2.00
2.14

–20
(scoliosis), respiratory muscle weakness (spinal muscular atro-
–40
–60
phy), alveolar destruction (pneumonia), thoracic space-occupying
–80
lesions (tumour, blood, air, effusion, cysts), significant abdominal
Time (seconds) distension (acute abdomen, intra-abdominal masses) – but the
consequences are the same: decreased TLC with preservation of
Flow (ml/s) Pressure (cmH2O)
normal airflow.

Figure 5 Flow–pressure curve for a child with croup. Flow and pressure
Scoliosis
curves plotted against time. Top panel: before epinephrine – traces
Physical examination – deformities of the thoracic cage in
demonstrating inspiratory flow limitation. During initial inspiration
scoliosis lead to decreased lung capacities. In fact, the degree of
there is a positive increase in inspiratory flow associated with a change
restrictive pulmonary disease is clearly correlated with the sever-
in oesophageal pressure (left of solid arrow). To the right of the solid
ity of scoliosis, and those with severe thoracic cage deformities
arrow, until inspiratory effort decreases (open arrow), there is virtually
are prone to more rapid decompensation when afflicted with
no further increase in inspiratory flow despite a large increase in
respiratory infections. Initial attempts at compensation prompt
negative pressure. During expiration, there is also evidence of flow
patients to breathe with faster respiratory rates and deeper than
limitation, with oesophageal pressure higher and flows lower than
those seen in the patient after epinephrine inhalation. Lower panel:
after epinephrine. From the beginning of inspiration, flow increases
50
markedly up to the point marked by the solid arrow as pressure
40
decreases. Thereafter, there is a further small decline in oesophageal
pressure to its minimum but modest negative value (open arrow). Expiration 30
During expiration, very modest increases in pressure are associated
Flow (litres/minute)
20
with relatively high flows.
10

0
500 400 300 200 100

elevated end-expiratory lung volumes. During an acute asthma –10

exacerbation, children will breathe at a lower ­ respiratory rate –20

than normal to minimize work of breathing and will commonly
Inspiration –30
have pursed lip breathing. The increased resistance prevents the
–40
dynamic collapse of the airways on expiration. As the airway
resistance increases further and initial compensatory mechanisms –50
Tidal volume (mL)
fail, patients will become tachypnoeic and use accessory muscles
to generate more negative inspiratory pressure to overcome the Diseased After response Normal
increased elastic recoil of the lungs and chest wall from hyperin- to bronchodilator
flation and increased lung volumes. Eventually, life-threatening
respiratory muscle fatigue leads to hypoventilation, hypercarbia Figure 6 Flow–volume curve for a child with asthma. The expiratory
and hypoxaemia. limitation is manifested as reduced peak expiratory flow and the
Pulsus paradoxus – the elevated airways resistance raises characteristic concave appearance on the expiratory limb. Note the
intrathoracic pressure and, much as in croup, causes a transient change in shape of the expiratory limb after bronchodilator therapy
decrease in cardiac output during inspiration. The ­consequential from concave to straight and the increase in peak expiratory flow.

PAEDIATRICS AND CHILD HEALTH 19:6 254 © 2009 Elsevier Ltd. All rights reserved.
 Symposium: respiratory medicine
their normal tidal volume. As they begin to fatigue, baseline end-
expiratory lung volumes decrease. Alveolar collapse, worsening
V′/Q′ mismatch and hypoxia soon follow.
Forced expiratory spirometry and flow-volume curves – given
the insidious nature of scoliotic deformities, measurement of lung
volumes and capacities is an effective way to gauge the result-
ing decline in pulmonary function over time. It has been shown
that postoperative pulmonary complications increase with dete-
rioration in pulmonary function tests. Since restrictive diseases
produce a proportional reduction in airflow and lung volume,
the FEV1/FVC remains unchanged or even slightly greater than
normal. However, the lung volumes obtained will be lower than
predicted.
Neuromuscular disorder (e.g. spinal muscular atrophy, infant
botulism)
Physical examination – most children with neuromuscular
disorders do not exhibit prominent respiratory symptoms. Simi-
larly to patients with scoliosis, their respiratory abnormalities
usually become apparent when they are afflicted by an infection
of the respiratory tract. Unlike patients with scoliosis, however,
their restrictive lung disease is further complicated by underlying
muscular weakness. If disease progression is severe enough, they
will manifest other signs on physical exam unique to neuromus-
cular disorders. The diaphragm is initially seemingly ‘spared’
compared with the weaker intercostal muscles and children will
use their accessory muscles to breathe at higher respiratory rates
than normal individuals to minimize work of breathing. Eventu-
ally this will progress to paradoxical breathing, usually apparent
clinically and on respiratory inductance plethysmography.
Respiratory inductance plethysmography – patients with spi-
nal muscular atrophy will often progress to complete thoraco-
abdominal asynchrony. The rib cage is neither stabilized nor
expanded during inspiration and the resulting negative intratho-
racic pressure causes the rib cage to collapse. The phase angle
will be above normal or up to 180° and the directional loop
­anticlockwise (Figure 2).
Conclusion
There is a wide array of non-invasive tools available to the pae-
diatrician for diagnosis and management of various respiratory
diseases. With a basic knowledge of respiratory physiology and
how pathophysiological states can be monitored, the clinician
can optimize therapeutic interventions and readily track disease
progression. ◆
Further reading
Argent AC, Hatherill M, Newth CJ, Klein M. The effect of epinephrine
by nebulization on measures of airway obstruction in patients with
acute severe croup. Intensive Care Med 2008; 34: 138–47.
Barker SJ, Tremper KK. The effect of carbon monoxide inhalation on
pulse oximetry and transcutaneous PO2. Anesthesiology 1987; 66:
677–9.
Barker SJ, Tremper KK, Hyatt J. Effects of methemoglobinemia on pulse
oximetry and mixed venous oximetry. Anesthesiology 1989; 70:
112–7.
PAEDIATRICS AND CHILD HEALTH 19:6
Corkey CW, Barker GA, Edmonds JF, et al. Radiographic tracheal
diameter measurements in acute infectious croup: an objective
scoring system. Crit Care Med 1981; 9: 587–90.
Frey B, Butt W. Pulse oximetry for assessment of pulsus paradoxus:
a clinical study in children. Intensive Care Med 1998; 24: 242–6.
Gagnon S, Jodoin A, Martin R. Pulmonary function test study before and
after spinal fusion in young idiopathic scoliosis. Spine 1989; 14:
486–90.
Hammer J, Newth CJ. Infant lung function testing in the intensive care
unit. Intensive Care Med 1995; 21: 744–52.
Hammer J, Eber E. Paediatric pulmonary function testing. Basel: Karger,
2005.
Hedstrand U. Ventilation, gas exchange, mechanics of breathing and
respiratory work in acute bronchial asthma. Acta Soc Med Ups
1971; 76: 248–70.
Hoover CF. The functions of the intercostal muscles. JAMA 1919; 73:
17–20.
Kim YJ, Lenke LG, Bridwell KH, et al. Pulmonary function in adolescent
idiopathic scoliosis relative to the surgical procedure. J Bone Joint
Surg Am 2005; 87: 1534–41.
Lenke LG, Bridwell KH, Blanke K, Baldus C. Analysis of pulmonary
function and chest cage dimension changes after thoracoplasty in
idiopathic scoliosis. Spine 1995; 20: 1343–50.
Maniscalco M, Zedda A, Faraone S, et al. Evaluation of a
transcutaneous carbon dioxide monitor in severe obesity. Intensive
Care Med 2008; 34: 1340–4.
Mithoefer JC, Bossman OG, Thibeault DW, Mead GD. The clinical
estimation of alveolar ventilation. Am Rev Respir Dis 1968; 98:
868–71.
Newth CJL, Hammer J. Measurements of thoraco-abdominal asynchrony
and work of breathing in children. In: Hammer J, Eber E, eds.
Paediatric pulmonary function testing. Basel: Karger, 2005.
Nichols MM. Shifting umbilicus in neonatal phrenic palsy (the belly
dancer’s sign). Clin Pediatr 1976; 15: 342–3.
Nichols DG. Respiratory muscle performance in infants and children.
J Pediatr 1991; 118: 493–502.
Numa AH, Newth CJ. Anatomic dead space in infants and children.
J Appl Physiol 1996; 80: 1485–9.
Perez A, Mulot R, Vardon G, et al. Thoracoabdominal pattern of
breathing in neuromuscular disorders. Chest 1996; 110: 454–61.
Permutt S. Some physiological aspects of asthma: bronchomuscular
contraction and airways calibre. Ciba Found Study Group 1971; 38:
63–85.
Rayner J, Trespalacios F, Machan J, et al. Continuous noninvasive
measurement of pulsus paradoxus complements medical decision
making in assessment of acute asthma severity. Chest 2006; 130:
754–65.
Sivan Y, Deakers TW, Newth CJ. Thoracoabdominal asynchrony in acute
upper airway obstruction in small children. Am Rev Respir Dis 1990;
142: 540–4.
Sivan Y, Eldadah MK, Cheah TE, Newth CJ. Estimation of arterial
carbon dioxide by end-tidal and transcutaneous PCO2
measurements in ventilated children. Pediatr Pulmonol 1992;
12(3): 153–7.
Stalcup SA, Mellins RB. Mechanical forces producing pulmonary edema
in acute asthma. N Engl J Med 1977; 297: 592–6.
Steele DW, Santucci KA, Wright RO, et al. D. Pulsus paradoxus: an
objective measure of severity in croup. Am J Respir Crit Care Med
1998; 157: 331–4.
255 © 2009 Elsevier Ltd. All rights reserved.
 Symposium: respiratory medicine
Tobias JD, Flanagan JF, Wheeler TJ, et al. Noninvasive monitoring of
end-tidal CO2 via nasal cannulas in spontaneously breathing children
during the perioperative period. Crit Care Med 1994; 22: 1805–8.
West JB. Mechanics of breathing. In: Respiratory physiology:
the essentials. Baltimore: Lippincott, Williams and Wilkins,
2003. Yelderman M, New W Jr. Evaluation of pulse oximetry.
Anesthesiology 1983; 59: 349–352.
Zhang JG, Wang W, Qiu GX, et al. The role of preoperative pulmonary
function tests in the surgical treatment of scoliosis. Spine 2005; 30:
218–21.
PAEDIATRICS AND CHILD HEALTH 19:6
Practice points
• Pulse oximetry, TCOMs and nasal cannula end-tidal CO2
devices are easy and reliable ways to monitor gas exchange
• Characteristic changes in pulmonary function assist in
classifying underlying respiratory dysfunction
• Ongoing use of pulmonary function tests allow for monitoring
of disease progression and efficacy of treatment
256 © 2009 Elsevier Ltd. All rights reserved.
 Symposium: respiratory medicine
The aetiology of childhood
asthma
Warren Lenney
Abstract
Asthma is an historic disease recorded clearly in ancient times by the
Chinese and the Greeks. In developed countries the prevalence in chil-
dren is very high, but the prevalence has been rising rapidly in the devel-
oping countries over the last 30 years. The aetiology is multifactorial and
is a combination of genetic and environmental influences. Genetically,
the mother plays a greater part than the father, but no single gene or
polymorphism can account for symptoms and exacerbations in isola-
tion. It is becoming evident that numerous environmental factors are
crucial in symptom development, and include rhinoviral infections, diet,
pets, endotoxins, seasonality, pollution, parental smoking, and obesity.
Research helps us understand some of the mechanisms involved, but
these mechanisms will vary hugely from one patient/family to another.
Understanding this variation is key to personal management and the
development of asthma control in this very common global disease.
Keywords aetiology; childhood asthma; food allergy; genetics; obesity;
pets; pollutants; tobacco smoke; virus infections
Introduction
Children with asthma demonstrate episodic symptoms and
variable airflow obstructions which occur spontaneously or in
response to various environmental trigger factors. There is also a
strong genetic influence which is determined more by the child’s
mother than by the father. The Oxford English Dictionary defines
aetiology as ‘the assignment of a cause’ and ‘that part of medi-
cal science which investigates the causes of disease’. The latter
definition dates back to 1684, but there has been a consider-
able increase in our understanding of the causations of asthma
since then. In 1860, Henry Hyde Salter published his ‘Treatise
on Asthma’. Allergy and atopy had not been propounded at that
time, but he was clearly aware of the hereditary nature of the
disease and of the association with contact and ‘emanations’
from cats and dogs. In 1892, Sir William Osler drew attention
to the wide variety of pathological changes occurring in asthma,
such as mucosal oedema, inflammation, and the production
of ‘gelatinous’ mucus. He also noted the relationship between
these changes and their corresponding symptoms demonstrated
during exposure to dust, cats, foods, infections, and emotional
extremes. He was well informed that asthma usually commenced
Warren Lenney MBChB DCH MRCP MD FRCP FRCPCH is Professor of Respiratory
Child Health, Keele University and Academic Department of Child
Health, University Hospital of North Staffordshire, Stoke-on-Trent, UK.
PAEDIATRICS AND CHILD HEALTH 19:6
in childhood and that it appeared to run in families. Today this
exaggerated response to various stimuli is accepted as one of the
key characteristics seen in both children and adults with asthma,
namely bronchial hyper-responsiveness (BHR). The main thrust
of this review is to highlight the main genetic and environmental
influences which result in the production of symptoms and exac-
erbations in children with asthma: in other words, the relevant
factors to consider in the aetiology of the disease.
Genetic influences
Although asthma has been known to run in families for centu-
ries, twin studies have shown that BHR is inherited indepen-
dently. It is the BHR mechanisms which are critical to asthma
aetiology, as these are what truly determine the symptoms, the
expression of the disease. The pathological expression has been
shown as mucosal oedema, inflammation, hypertrophy of the air-
way smooth muscles, smooth-muscle shortening, and increased
muscle contractions.
The complexity of the above processes suggests there may
be hundreds of potentially attractive candidate genes associated
with asthma, and this has been confirmed by the large num-
bers of genetic mapping studies which have been published over
recent years.
Some areas of the human genome have more consistently
been associated with many of the recognized asthma pheno-
types. These are 6p21-24 (the major histocompatibility complex),
11q13-21 (clara cell secretory protein, IgE high-affinity receptor
and glutathione S-transferase genes) and 20p13 (ADAM 33). It
was particularly exciting in relation to the identification of ADAM
33 that two subsequent candidate gene studies confirmed an
association between BHR and ADAM 33 polymorphism.
Groups of candidate genes which have been shown to be
relevant to asthma are those encoding cytokines and chemo-
kines, those encoding receptors associated with the T Helper cell
2 (TH2) response, and those genes related to oxidative stress.
These associations are complex, however, and can be influenced
by many simple factors such as the size and age of the child. If
BHR is not corrected for such parameters, the associations can
change dramatically and even disappear completely. The risk of
transmission of asthma is greater through the child’s mother, and
this has been recently confirmed in relation to polymorphism
in glutathione S-transferase (GSTP1), one of the genes respon-
sible for the detoxification of drugs and products of oxidative
stress. Some studies have found correlations with the father’s
genetic make-up in relation to BHR, and one requires an in-depth
knowledge of asthma genetics to fully interpret all the published
findings.
One of the issues in attributing specific genetic relationships
to asthma in children can best be summarized by Figure 1. These
relationships depend on the many and complex interactions
between in-utero, neonatal and external environmental factors,
all of which vary depending on the specific asthma phenotype,
and even in a single patient these do not remain constant but
vary over time. The phenotype may well vary between the sea-
sons, some worsening in the spring and early summer, others
worsening in the winter. Asthma symptoms in children are more
variable than those in their adult counterparts, and they may
well change significantly from one year to the next. Long-term
257 © 2009 Elsevier Ltd. All rights reserved.
 Symposium: respiratory medicine
Mother’s genes
In-utero environment Fetal genes
Neonatal immunophenotype
External environment Child’s genes
Asthma phenotype
Figure 1 Proposed model of gene–environment interactions in
determining asthma phenotype.
symptomatology and severity are very hard to predict clinically
in the individual young patient.
The above clearly shows that the genetic associations are
complex, asthma aetiology is multifactorial, and despite the
insistence of the media that one day we will find the gene for
asthma, there is no such thing.
Food allergy and diet
In surveys of patients attending asthma clinics, over two-thirds
think their asthma is triggered by specific foods. When food
allergy challenge studies are undertaken, however, the figure is
much lower, usually less than 25%. The reason for the high fig-
ure in epidemiological surveys is probably the inclusion of foods
such as ice-cream and fizzy drinks, where the causation of the
bronchospasm is more likely to be related to very low tempera-
tures or a low/acidic pH. Foods implicated most commonly in
causing asthma symptoms, following challenge studies, are pea-
nut, milk, egg, and tree nuts.
Both food allergy and asthma are considered to have an
important allergic component mediated through immunoglobulin
E (IgE), but as with genetics, the relationships are not straight-
forward. When different countries are surveyed to establish the
prevalence figures for atopy and asthma, even if the prevalence
of atopy is similar in each country the prevalence of asthma can
vary enormously. Conversely, over a number of years, the preva-
lence of asthma can rise steadily with no similar change in atopy
prevalence.
The role of diet has been investigated extensively in relation
to asthma prevalence over recent years. Anti-oxidants such as
vitamin C and E, fatty acids, and magnesium and potassium have
been studied, as have flavourings, colouring agents, and addi-
tives such as monosodium glutamate, tartrazine and sulphites.
A Cochrane review of fish-oil 3 fatty acid supplementation
could not find any consistent beneficial effect when compared
to placebo. The evidence for the effect of the other compounds
(colouring agents and additives etc) is also not clear, although
some patients can be adversely affected by them. As with other
PAEDIATRICS AND CHILD HEALTH 19:6
factors related to the development of asthma in children, the role
of our diet has to be considered as a trigger which may influ-
ence prevalence and/or severity, but which needs careful assess-
ment in each patient or in each patient group, depending on the
asthma phenotype.
Viral infections
Respiratory viruses are the most common cause of asthma exacer-
bations in children and in adults with asthma. Viruses are respon-
sible for 85% of all childhood exacerbations. The most important
virus is the rhinovirus (RV), which is now known to infect the
lower as well as the upper airway. The innate immune response
to RV infection was postulated as being deficient in asthmatic
patients. Production of interferon β (IFN-β) was investigated and
shown to be low. Patients with asthma therefore probably have
decreased immunity to RV, resulting in increased cell lysis and
more severe lower-airway symptoms following infection. RV
infection may also act synergistically with allergic inflammation,
and the finding of RV together with high allergen exposure in
children results in increased risk of hospitalization with asthma.
Respiratory syncytial virus (RSV) has traditionally been con-
sidered the most important virus for severe wheezing in the first
2 years of life, and indeed it is still the commonest pathogen
to produce acute viral bronchiolitis (AVB) in this very young
age group. In the UK alone RSV is responsible for the major-
ity of the 20,000 infants admitted to our paediatric wards with
AVB each winter. However, in elegant prospective studies in the
USA, Lemanske et al found that RV isolated during wheezing ill-
nesses in infancy were the strongest predictor of wheezing in the
third year of life. We must be careful to try to separate recurrent
respiratory symptoms of cough and wheezing following AVB
from wheezing in association with allergic sensitization and the
development of true atopic asthma. In practice, however, in the
clinic or surgery in primary or secondary care such differential
diagnoses are almost impossible to make, and final differentia-
tion can only be confirmed in the fullness of time. The reality for
clinicians is much less obvious than many epidemiological stud-
ies would have us believe.
Clearly, RSV and other respiratory viruses may be found in
infants with a future diagnosis of atopic allergic asthma, but the
research evidence at present points to RV as the key virus in
the aetiology of true symptoms of atopic asthma. Whether it is
the genetic make-up of the child which is the real determinant
of asthma status remains an unanswered question, but this may
well be the most important factor in the future progression of
asthma symptoms. By that I mean that RV or any other virus
may produce symptoms in infants, but those infected who then
develop atopic asthma are those with the susceptible genetic
make-up in the first place.
There is increasing interest in bacterial infections, such as with
Mycoplasma pneumoniae and Chlamydia pneumoniae, which are
known to cause respiratory infections in all ages. In one study,
20% of 2–15-year-old children admitted to hospital with asthma
were found to have mycoplasma present as assessed by culture
through nasopharyngeal aspirates and paired serum titres for
mycoplasma antibodies. We need more information on these
organisms as potential infective agents which may be factors in
the development of asthma in both children and adults.
258 © 2009 Elsevier Ltd. All rights reserved.
 Symposium: respiratory medicine
Pests, pets and endotoxins
In the 1970s it was generally believed that the aetiology of asthma
was mainly determined by the house-dust mite and, in particu-
lar, its allergenic faeces. Cats were considered to frequently con-
tribute to symptoms, and the answer seemed simple – get rid
of the family cat and take specific measures to keep levels of
house-dust mite low. Remove carpets, cover the mattress with
polythene, regularly damp-dust the bedroom and living rooms,
and buy non-allergenic bedding and pillows. Studies from moun-
tainous regions of the world, however, informed us that at levels
over 8000 feet above sea level, such as in the Andes, house-
dust mites could not survive, but asthma prevalence was high.
In countries where cats were not domesticated the prevalence
of asthma was as high as in other countries where cats were
the commonest household pet. Numerous cohort studies have
now evaluated the relevance of dust, house-dust mites, furry
and feathered pets and endotoxins (inflammatory lipopolysac-
charides from gram-negative bacteria that are ubiquitous in the
indoor environment) in the development of allergic sensitization
and childhood asthma. Some studies, but not others, have shown
a dose–response relationship for mite allergen exposure and sen-
sitization. In the study by Simpson et al, the higher the endotoxin
allergen exposure the greater the risk of sensitization, but when
the results were re-analysed by genotype only the children with a
particular genotype of the CD14 gene confirmed the relationship,
emphasizing the complex interaction between environmental
and genetic backgrounds.
Perhaps the most convincing evidence of environmental influ-
ences in relation to asthma prevalence is the consistent finding
that children reared on farms have a lower incidence of asthma.
The most likely explanation for these findings is that exposure
to irritant, allergenic and infectious factors in early childhood
somehow encourages tolerance to common aeroallergens, per-
haps by promoting T-helper-cell type 1 (TH1) as opposed to TH2-
type immunological responses early in life. Much research has
taken place over the past 20 years, often involving large cohort
studies, to assess the importance of aeroallergens in the develop-
ment of asthma in children. We now understand the mechanisms
involved far better, but in practical terms we are little closer to
knowing whether we can alter/manipulate the indoor and out-
door environments to influence asthma prevalence.
Obesity
The incidence of asthma and that of obesity have increased in
parallel over the last four decades. There are mechanical, immu-
nological, hormonal and inflammatory effects of obesity that
may well play a role in the persistence of asthma. Although the
relationship is not completely clear, many prospective studies
suggest that increased body mass index (BMI) is associated with
an increase in asthma. However, few studies have adjusted for
physical activity, and in some the association is seen only in
males, in others it is seen only in females. Whilst diet and BMI
are relatively easy to measure, the assessment of physical activ-
ity is more challenging, and it could be that lack of activity is the
primary explanation for the increased levels of asthma symp-
toms, and not the obesity itself. There is no doubt that obesity
is a major present-day health hazard, but whether it is directly
PAEDIATRICS AND CHILD HEALTH 19:6
related to increased levels of asthma in childhood populations
requires further study.
Tobacco smoke
Cigarette smoke contains particulate matter and more than 2000
chemical compounds. Tobacco smoke in the environment is
incompatible with good respiratory health. A systemic review of
parental smoking revealed an odds ratio for asthma of 1:21 and
for wheeze of 1:4. Although maternal smoking is not consistently
associated with an increase in asthma prevalence, it is associ-
ated with more severe disease, and childhood asthma symptoms
often improve when the mother stops smoking. Similarly, BHR is
increased in children whose parents smoke, and increased BHR
at 1 month of age has been shown to predict lower lung function
at 6 years of age. Wheezing and doctor-diagnosed asthma are
more closely related to passive/environment tobacco smoke in
the preschool age range than in schoolchildren 5–16 years of age,
but such findings should not detract from the statement that the
single most important measure to improve the health of children
would be the exclusion of tobacco smoke from their indoor and
outdoor environments.
Other pollutants
It is known that nitrogen dioxide, sulphur dioxide, ozone, diesel
exhaust and particulate matter can all increase BHR and inflam-
mation. The mechanisms may involve oxidative stress, epithelial
damage, and an increase in pro-inflammatory mediators. As with
other triggers, such factors are unlikely to work alone, and an
interesting study of children 8–11 years old who wore nitrogen
dioxide monitors throughout the study showed that respiratory
symptoms increased and lung function fell as nitrogen dioxide
levels increased. These changes occurred in the week before a
respiratory infection became clinically obvious, once again dem-
onstrating the complex interactions which may be necessary in
the patients’ environment if they are to influence the develop-
ment of asthma symptoms.
Other issues
Asthma is a syndrome with different triggers, different outcomes
and different responses to medications. Recent studies have iden-
tified factors such as paracetamol usage and rhinitis in children
as important in progression to adult asthma. Progress in under-
standing mechanisms responsible for asthma symptomatology is
slow and cure is still a pipedream.
Summary
Asthma is a classical multifactorial disease whose aetiology is
complex, involving both genetic and environmental triggers. The
prevalence has risen steeply over the last 30 years at a rate that
cannot be explained simply on genetic grounds. Clearly there
are important environmental factors which have influenced this
sharp rise, and I have tried to allude to a number of these. Each
patient is different, and trying to assess which environmental
factors are important for that individual may result in improve-
ment in asthma control if identified triggers can be eliminated
259 © 2009 Elsevier Ltd. All rights reserved.
 Symposium: respiratory medicine
or reduced. Of equal importance is the relationship between
such environment triggers and specific genetic polymorphisms.
Asthma aetiology typifies the comment ‘one glove does not fit
all’. There will not suddenly be a miracle cure for this common
yet very complex disease. Each patient behaves in a different
way, making asthma a fascinating disease in which to study the
true meaning of the word ‘aetiology’.
Conflict of interest
The author has no conflict of interests which may have any influ-
ence on the content of this article. ◆
Further reading
Bernstein JA, Alexis N, Barnes C, et al. Health effects of air pollution.
J Allergy Clin Immunol 2004; 114: 1116–23.
Biscardi S, Lorrot M, Marc E, et al. Mycoplasma pneumoniae and
asthma in children. Clin Infect Dis 2004; 38: 1341–6.
Chauhan AJ, Inskip HM, Linakar CH, et al. Personal exposure to nitrogen
dioxide (NO2) and the severity of virus-induced asthma in children.
Lancet 2003; 361: 1939–44.
Child F, Lenney W, Clayton S, et al. Correction of bronchial challenge
data for age and size may affect the results of genetic association
studies in children. Pediatr Allergy Immunol 2003; 14: 193–200.
Child F, Lenney W, Clayton S, et al. Maternal but not paternal variation
in GSTP1 is associated with asthma phenotypes in children. Respir
Med 2003; 97: 1247–56.
Daniels SE, Bhattacharrya S, James A, et al. A genome-wide search for
quantitative trait loci underlying asthma. Nature 1996; 383: 247–50.
Ernst P, Cormier Y. Relative scarcity of asthma and atopy among rural
adolescents raised on a farm. Am J Respir Crit Care Med 2000; 161:
1563–6.
ERS and The Lancet. A step-change in asthma. Lancet 2008; 372:
1009–119.
Friedlander SL, Busse WW. The role of rhinovirus in asthma
exacerbations. J Allergy Clin Immunol 2005; 116: 267–73.
Fryer AA, Bianco A, Hepple M, et al. Polymorphism at glutathione
S-transferase GSTP1 locus: a new marker for bronchial
hyperresponsiveness and asthma. Am J Respir Crit Care Med 2000;
161: 1437–42.
Graves PE, Kabesch M, Halonen M, et al. A cluster of seven tightly
linked polymorphisms in the IL-13 gene is associated with total
serum IgE levels in three populations of white children. J Allergy
Clin Immunol 2000; 105: 506–13.
Haagerup A, Bjerke T, Schiotz PO, et al. Asthma and atopy – a total
genome scan for asthma susceptibility genes. Allergy 2002; 57:
680–6.
Johnston SL, Pattemore PK, Sanderson G, et al. Community study of
role of viral infections in exacerbations of asthma in 9–11 year old
children. BMJ 1995; 310: 1225–9.
Lucas SR, Platt-Mills TAE. Paediatric asthma and obesity. Paediatr
Respir Rev 2006; 7: 233–8.
PAEDIATRICS AND CHILD HEALTH 19:6
Lemanske Jr RF, Jackson DJ, Gangnon RE, et al. Rhinovirus illnesses
during infancy predict subsequent childhood wheezing. J Allergy
Clin Immunol 2005; 116: 571–7.
Murray CS, Poletti G, Kebadze T, et al. Study of modifiable risk factors
for asthma exacerbations: virus infection and allergen exposure
increase the risk of asthma hospitalization in children. Thorax 2006;
61: 376–82.
Onorato J, Merland N, Terral C, et al. Placebo-controlled double-
blind food challenge in asthma. J Allergy Clin Immunol 1986; 78:
1139–46.
Osler W. Bronchial asthma. principles and practice of medicine.
New York: D Appleton, 1892.
Palmer LJ, Rye P, Gibson N, et al. Airway responsiveness in early
infancy predicts asthma, lung function and respiratory symptoms by
school age. Am J Respir Crit Care Med 2001; 163: 37–42.
Priftanji A, Strachen D, Burr M, et al. Asthma and allergy in Albania and
the UK. Lancet 2001; 358: 1426–7.
Salter HH. On asthma: its pathology and treatment. London: John
Churchill, 1860.
Simpson A, John SL, Jury F, et al. Endotoxin exposure, CD14 and
allergic disease: an interaction between genes and the environment.
Am J Respir Crit Care Med 2006; 174: 386–92.
Strachen DP, Cook DG. Parental smoking and childhood asthma;
longitudinal and case control studies. Thorax 1998; 53: 204–12.
Van Eerdewegh P, Little RD, Dupuis J, et al. Association of the ADAM33
gene with asthma and bronchial hyperresponsiveness. Nature 2002;
418: 426–30.
Woods RK, Thien FC, Abramson MJ. Dietary marine fatty acids (fish oil)
for asthma in adults and children. Cochrane Database Syst Rev
2002 (3): CD001283.
Practice points
• The symptoms seen in children with asthma are a direct
result of an exaggerated response to various stimuli
i.e. bronchial hyper-reactivity (BHR)
• The aetiology of asthma is multifactorial, involving both
genetic and environmental effects
• The rhinovirus is the single most important trigger to
generate an acute exacerbation of asthma
• Individual children may be adversely affected by furry or
feathery animals, but environmental issues are complex and
confusing
• Children brought up on farms where there are animals have
a lower prevalence of asthma, as shown in every reported
study worldwide
• Despite the fact that environmental tobacco smoke clearly
worsen asthma symptoms, clinicians seem unable to
influence families to stop smoking
260 © 2009 Elsevier Ltd. All rights reserved.
 Symposium: respiratory medicine
Advances in the
management of asthma
Andrew Bush
Cara Bossley
Louise Fleming
Nicola Wilson
Abstract
Exciting new data are increasing the evidence base for the management
of paediatric asthma. To inform the treatment of preschool wheeze, the
best current classification is into episodic (viral) and multitrigger wheeze,
rather than according to epidemiological pattern (transient versus per-
sistent) and the presence or absence of atopy. Episodic (viral) wheeze is
treated intermittently, with either inhaled bronchodilators or oral monte-
lukast at the time of viral colds. If this approach fails, intermittent high-
dose inhaled corticosteroids may be tried. Oral prednisolone is ineffective
in the treatment of all but the severest attacks of preschool episodic
(viral) wheeze, and is not a primary-care medication in this context. In
older children the role of long-acting β2 agonists has been explored. They
are not indicated as first-line prophylactic therapy. In children with more
severe symptoms, a single-inhaler strategy using budesonide/formoterol
should be considered. In children who do not respond to conventional
asthma therapy, the diagnosis and the way in which the prescribed treat-
ment is being used should be reviewed rather than more treatment being
blindly given. Most cases will improve with conventional management
which is properly undertaken, and will not require novel therapies.
Keywords adherence; atopy; beta agonist; inhaled corticosteroid; leu-
kotriene receptor antagonist; macrolide; prednisolone; viral infection;
wheeze
Introduction
The aim of this review is to highlight recent important clinical
trials, particularly those which should lead to a change in clinical
Andrew Bush MD FRCP FRCPCH is Professor of Paediatric Respirology at the
Imperial School of Medicine at the National Heart and Lung Institute,
and Royal Brompton Hospital, London, UK.
Cara Bossley MRCPCH is a Clinical Research Fellow at the Imperial
School of Medicine at the National Heart and Lung Institute, and Royal
Brompton Hospital, London, UK.
Louise Fleming MRCPCH is a Clinical Research Fellow at the Imperial
School of Medicine at the National Heart and Lung Institute, and Royal
Brompton Hospital, London, UK.
Nicola Wilson MD is an Honorary Consultant Paediatrician at the
Imperial School of Medicine at the National Heart and Lung Institute,
and Royal Brompton Hospital, London, UK.
PAEDIATRICS AND CHILD HEALTH 19:6
practice in the treatment of wheezing disorders throughout child-
hood. Familiarity with the updated version of the British Thoracic
Society/Scottish Intercollegiate Guidelines Network (BTS/SIGN)
Asthma guidelines is assumed. We aim to produce an update
of treatment for children seen in primary and secondary care in
particular, based on new evidence; space precludes a review of
the recent advances in the basic science of asthma.
Preschool wheeze
Phenotyping preschool wheezing syndromes
There are at least three ways in which wheeze phenotypes have
been characterized: by epidemiological descriptions of the change
in wheeze over time, by the presence or absence of atopy, and by
current symptom pattern, i.e. episodic (viral) versus multitrigger.
Epidemiology: the classic epidemiological phenotypes are
transient wheeze (first 3 years of life only); persistent wheeze
(throughout the first 6 years of life), and late-onset wheeze (start-
ing after the first 3 years of life), as described in the Tucson
study. These patterns are to some extent forced in that the chil-
dren were only seen three times in the first 6 years of life, so
no other phenotype could have been detected. The Avon lon-
gitudinal study obtained questionnaires annually, and used a
more objective mathematical technique – latent class analysis –
to determine from the data whether there were different phe-
notypes, a more sophisticated approach. They determined
phenotypes of (1) never or infrequent (59%); (2) transient (16%)
and (3) prolonged early wheeze (9%); (4) intermediate (3%),
(5) late (6%) and (6) persistent wheeze (7%). Although these
epidemiological phenotypes have value for the epidemiologist
probing mechanisms of disease, they are useless for the clinician,
because they can only be allocated retrospectively, and therefore
cannot guide contemporaneous therapeutic decisions. The basic
message, familiar from clinical practice, is that some wheezy pre-
school children will outgrow their symptoms, and some will not.
A number of predictive indices have been developed which gen-
erally have a good negative predictive value (will tell you who
will not have persistent symptoms) but a poor positive predictive
value (little better than flipping a coin at telling you who will
have persistent symptoms). So it is very difficult for the clinician
to determine whether a wheezy 2-year-old will still be wheezing
at the age of 6. The epidemiological studies have certainly taught
us a great deal about the natural history of wheezing disorders in
childhood, but are not useful in clinical practice.
Atopy: the second way of classifying wheeze is by the presence
or absence of atopy, determined either by skin-prick tests or by
the report of eczema or allergic rhinitis. This too is imperfect.
Firstly, just because the child is atopic does not mean that atopy
and wheeze are connected, any more than an ingrowing toenail
in such a child could rationally be described as an atopic digi-
topathy. Indeed, in very young children with wheeze, whether
atopic or not, airway wall histology is normal. In older children
with wheezing in response to viruses and many other triggers,
airway wall histology shows typical eosinophilic inflammation
and structural airway wall changes irrespective of the presence
of atopy. Atopy is also a dynamic process, with children mani-
festing it over time; so when assessing the current non-atopic
261 © 2009 Published by Elsevier Ltd.
 Symposium: respiratory medicine
wheezer, it is impossible to know whether atopy will appear
­subsequently.
Symptom pattern: the third approach, adopted by the recent
guidelines of the European Respiratory Society (ERS), switches
the focus to symptom pattern at the time of presentation as the
way to guide treatment. The history should determine whether
the child has episodic (viral) wheeze, characterized by wheeze
in association with usually clinically diagnosed viral infections
only, or multitrigger wheeze, characterized by wheezing with
viral infections, but also with wheeze to typical triggers such
as exercise, allergen and smoke exposure. The implications for
treatment are discussed below; in summary, episodic problems
should be treated with intermittent therapy, whereas children
with multitrigger symptoms should be considered for continu-
ous treatment, usually with low-dose inhaled corticosteroids
(ICS), using a three-stage protocol. It should be noted that this
classification accepts that phenotypes may be modified by time
or treatment; thus an infant with episodic (viral) wheeze may
evolve into a multitrigger picture, and the infant with multitrig-
ger wheeze, when treated with ICS, may be left with episodic
(viral) exacerbations, which are much more difficult to control
with prophylactic therapy. Also, although the tacit assumption
is sometimes made that the terms episodic (viral) and transient,
and also multitrigger and persistent are synonymous, there is
no evidence that this is the case. Episodic (viral) wheeze can
certainly persist even into adult life, and multitrigger wheeze can
remit in mid-childhood.
Implications for treatment of preschool wheeze
The ERS guidelines attempted an evidence-based approach to
treatment (hampered by the lack of much evidence!) of preschool
wheeze; subsequent to the publication of these guidelines, two
further papers on the acute management of preschool wheeze
have been published. The stimulus for the guidelines was to try
to rectify the disconnection between the epidemiological data
and the recommendations of many previous guidelines, and also
to be of practical value for the clinician. There might be two
reasons for treating preschool wheeze: (a) to modify the disease
and prevent persistence of symptoms and the development of
full-blown asthma in mid-childhood, and (b) for current relief
of symptoms.
Can treatment alter the outcome of preschool wheeze?: the
short answer is, no. Although parents want to know the prog-
nosis of the child, in practice it actually makes no difference
to therapy because we have no medication which prevents the
development of persistent asthma. Studies have shown that
early use of continuous or intermittent (at the time of symp-
toms) ICS have all shown that they have no effect on outcome.
A trial of cetirizine given to infants with atopic dermatitis to try
to prevent the development of wheeze showed no difference
for the treatment group as a whole, but some benefit, based
on fairly small numbers, in subgroups with radioallergosorbent
test (RAST) positivity to house-dust mite, pollen or both. This
hypothesis-generating trial was not confirmed by a repeat study
of l-cetirizine, in which no benefit was shown. So treatment for
preschool wheeze should be directed by present symptoms, not
future prognosis. It cannot be overstressed that, whatever the
PAEDIATRICS AND CHILD HEALTH 19:6
evidence in adults, the early use of inhaled corticosteroids in
preschool wheeze to optimize future outcome is not indicated.
Symptomatic treatment for preschool wheeze: given the
lack of any evidence that the prognosis of preschool wheeze
is affected by treatment, we suggest that a symptom-based
approach, as suggested by the ERS guidelines, is appropriate.
The validity of this approach appears to be borne out by a
limited amount of clinical trial data, and in practice we suggest
that in any case mothers don’t want to medicate a completely
well toddler.
Treatment of episodic (viral) wheeze: bronchodilators – the
first-line treatment of episodic (viral) wheeze is with inhaled
short-acting β2 agonist and/or inhaled anticholinergics just at
the time of symptoms, depending on the response. A mask and
spacer is used; the mask can be discarded in older children.
Treatment of episodic (viral) wheeze: leukotriene receptor
antagonists – if this is insufficient, the use of intermittent monte-
lukast, just at the time of viral wheezing, should be considered,
following the publication of the PRE-EMPT study. This group
randomized more than 200 children to receive placebo or mon-
telukast at the time of a virus-induced exacerbation of wheez-
ing. The number and duration of episodes was the same in both
groups, but in the montelukast group the children lost one-third
fewer days out of school or child care, and the carers lost one-
third fewer days from work. This strategy does not work for all
children, but is well worth considering if episodic (viral) wheeze
is seriously disruptive. The approach is more logical than using
montelukast daily for episodic (viral) wheeze, since the eleva-
tion of cysteinyl leukotrienes is present only during viral infec-
tions, and is more likely to be practical since parents are mostly
reluctant to medicate well children. It is of course possible to
use montelukast as a prophylactic medication, but no study has
shown that this approach is superior to low-dose inhaled corti-
costeroids at any age.
Treatment of episodic (viral) wheeze: steroids – there have
been important new studies addressing the role of inhaled and
oral corticosteroids in preschool episodic (viral) wheeze. A
Cochrane review had previously concluded that prophylactic
low-dose inhaled corticosteroids did not prevent viral exacerba-
tions of wheeze, but that there might be a role for intermittent
high-dose treatment. New evidence has been published on the
treatment of acute exacerbations of episodic wheeze in preschool
children. A Canadian study compared high-dose (1.5 mg daily)
inhaled fluticasone against placebo, given at the time of viral
exacerbations of wheeze, and demonstrated benefit in terms of
fewer prescribed courses of oral prednisolone. This has to be
considered a proof-of-concept study; the fluticasone-treated chil-
dren had a small reduction in linear growth over the 6–12-month
study period, and they could not exclude effects on adrenal func-
tion. It should be recalled that 10% of preschool children have
more than ten viral colds a year, and symptoms may last for at
least 2 weeks, so these high doses may lead cumulatively to a
big total dose of fluticasone. Dose-finding studies are needed,
with more detailed safety data, before this strategy can be
­recommended.
Oral prednisolone is the bedrock of treatment for acute exac-
erbations of asthma in older children and adults, and this med-
ication is widely prescribed for acute preschool viral wheeze.
262 © 2009 Published by Elsevier Ltd.
 Symposium: respiratory medicine
A previous study recruited preschoolers who had been admit-
ted to hospital for an exacerbation of wheeze, and randomized
them to a parent-initiated course of either prednisolone or pla-
cebo at the time of the next episode. This intervention had no
benefit. The group has now reported a randomized double-blind,
placebo-controlled trial in nearly 700 preschool children who
came to hospital with a presumed viral exacerbation of wheeze.
There was no benefit of oral prednisolone in this more severe
group either. Therefore one is forced to the conclusion that, in
preschool children with episodic (viral) wheeze, prednisolone
is indicated in secondary care only in children in whom a very
severe or prolonged course is anticipated. Prednisolone should
not be prescribed in primary care for these children. It may
be that prednisolone will benefit the highly atopic, multitrig-
ger wheezer who has a viral exacerbation, or children already
admitted to hospital and looking like needing admission to the
intensive care unit, but, in particular in young children, this is
probably the exception.
It may appear to be paradoxical that high-dose inhaled ste-
roids may work whereas oral steroids do not; the (speculative)
explanation may be that the steroid effect is by local vasocon-
striction as a topical effect, leading to a reduction in airway cali-
bre by lessening airway oedema.
Treatment of multitrigger wheeze: what works best – there
has been a single randomized controlled trial comparing the
addition of either oral montelukast or nebulized budesonide or
placebo to inhaled salbutamol in the setting of episodic (viral)
wheeze. Both budesonide and montelukast treatment led to mod-
est improvements in the severity but not frequency of attacks.
Given the side-effect profile, we would recommend intermittent
montelukast as the first line, with the use of intermittent ICS for
montelukast treatment failures.
Treatment of multitrigger wheeze – if a preschool child
is having symptoms several times per week, and symptoms
respond to inhaled bronchodilators, then a trial of ICS may
be merited. That some preschool children benefit from ICS is
indubitable, in particular those aged 3 years and over, who
are atopic, and who have multitrigger wheeze. However, ben-
efit in non-atopics is unlikely, and given the evidence from
bronchoalveolar lavage and endobronchial biopsy studies that
wheezing at age less than 3 years is usually not an eosinophilic
disease, the use of these medications in such children should
be very limited if they are used at all. If a trial is contemplated,
we recommend a three-stage protocol. The first step is to intro-
duce ICS for an arbitrary 2-month period in a moderately high
dose (for example, budesonide 400 μg twice daily). Symptoms
are assessed at stage two, and the medication discontinued. If
there has been no response, the symptoms are not steroid-sensi-
tive; if they have disappeared, it is not clear whether this is due
to the medication or the passage of time. If symptoms appear
to respond, but then recur on stopping the medication, then
treatment is restarted, titrating to the lowest dose required to
control symptoms. Only thus will the over-­diagnosis of asthma
due to an apparent response to steroid therapy be avoided.
Undoubtedly some preschool children will benefit, but equally
without doubt steroids are grossly over-prescribed to young
children.
Treatment of preschool wheeze: summary – it is clear that we
need radically to overhaul our current therapeutic practices with
PAEDIATRICS AND CHILD HEALTH 19:6
regard to the prescription of oral and inhaled corticosteroids, and
to curtail their use drastically in preschool children.
Asthma in the older child
Treatment of mild asthma in children: mission creep
Current guidelines recommend low-dose inhaled corticosteroids
as first-line therapy in children with asthma who merit prophy-
laxis. Long-acting β2 agonists are introduced at a later step, and
in children, the benefits are by no means clear-cut. Data from
Denmark has shown that, in primary care, more children are
prescribed combination inhalers (Seretide™) than inhaled corti-
costeroids. Over a period of less than 10 years the proportion of
school-age asthmatic children prescribed combination therapy
has risen from 16% to 44%, with no evidence base, and no new
trials and no new evidence-based guideline recommendations to
support such a change. One is forced to the conclusion that this
change is driven by aggressive marketing by the pharmaceutical
industry. This has implications of fiscal cost, but also, perhaps
more importantly, safety and efficacy. A recent meta-analysis
has highlighted the possible adverse effects of long-acting β2
agonists; we accept that the main concern is when they are
used as single agents, without concomitant inhaled corticoste-
roids. Efficacy may be an issue; the Pediatric Asthma Controller
Trial (PACT) was a three-way comparison between montelukast
5 mg at night versus fluticasone 100 μg twice daily versus fluti-
casone 100 μg once daily and salmeterol 50 μg twice daily. More
than 80 children were completions in each limb of the study.
Montelukast was significantly inferior to the other regimens
(which were equivalent) in the primary outcome of days when
asthma was controlled. However, the fluticasone-only regimen
was better than both the alternatives for some secondary end-
points. The number of children requiring rescue steroid therapy
showed a trend to be smaller in the fluticasone group (P < 0.10),
and exhaled nitric oxide levels and the improvement in pre-
bronchodilator first-second forced expired volume (FEV1) were
significantly better in the fluticasone group than either of the
other regimens. Thus, if anything, the evidence favours the
current guideline recommendations, that first-line prophylaxis
should be with inhaled corticosteroid monotherapy and not
combination treatment. Stand firm against the siren cry of the
marketers!
Treatment of more severe paediatric asthma: SMART or GOAL?
Two new treatment strategies have been proposed. The data
come largely from adult studies, but there are enough paediat-
ric data to challenge current dogma. The GOAL (Gaining Opti-
mal Asthma ControL) philosophy is that all symptoms can be
abolished by increasing medication (in this case Seretide) to
ever higher levels. The SIGN/BTS target of using short-acting
β2 agonist less than three times a week is replaced by the much
more stringent one of requiring no rescue medication at all.
Furthermore, in this study, the dosage having been racked up
to high levels, no attempt at reduction was made over the fol-
lowing 12 months. This strategy, seductive as it may be, seems
to us to have significant flaws. Firstly, it confuses control of
interval symptoms with prevention of exacerbations. Loss of
baseline control is not the same as an exacerbation. It is per-
fectly possible to be symptom-free between viral colds, but
263 © 2009 Published by Elsevier Ltd.
 Symposium: respiratory medicine
still have ­ significant virus-associated exacerbations. Indeed,
increasing the dose of prophylactic medications in an unavail-
ing attempt to prevent viral exacerbations may well have been
a significant contributor to the epidemic of severe hypoglycae-
mia secondary to adrenal suppression reported in children on
high-dose inhaled steroids. Secondly, given the tendency for
much childhood asthma to improve, and hence guideline rec-
ommendations to try to reduce therapy when asthma is well
controlled, the strategy of no dose reduction seems rather dan-
gerous. Finally, we do not have the data to know whether it is
better in terms of long-term respiratory outcome and systemic
side-effects such as growth failure to have a few symptoms
requiring short-­acting β2 agonists, with a lower dose of inhaled
corticosteroids, or to have a higher steroid dosage and no symp-
toms. In terms of respiratory outcomes, it should be noted that
in the Children’s Asthma Management Program (CAMP) study,
treatment with inhaled corticosteroids did not prevent around
25% of asthmatic children from having suboptimal increases in
spirometry with growth. At the moment, this approach cannot
be recommended in children.
The alternative approach, SMART (Symbicort Maintenance
and Reliever Therapy), is a one-inhaler strategy which takes
advantage of the rapid onset of action of the long-acting β2 agonist
formoterol. This is based on a study comparing three regimens:
budesonide 100 μg/formoterol 6 μg once daily plus extra doses of
the same inhaler as required (SMART regimen): the same inhaler
twice daily, with rescue terbutaline; or budesonide 400 μg twice
daily with rescue terbutaline. The children in the SMART regi-
men had fewer exacerbations. Perhaps surprisingly, there was
no dramatic increase in ICS use in the SMART regimen, with an
average of less than one extra inhalation per day. This regimen
has the advantage of simplicity, and might possibly increase the
adherence of recalcitrant adolescents, although as yet there is no
evidence for this. Of course it relies on the adequate perception
of symptoms, which is a problem in some asthmatics. Nonethe-
less, this is a regimen we have found to be useful in clinical
practice. There is an urgent need for a well-designed randomized
controlled trial comparing SMART and GOAL in children with
asthma.
Treatment of more severe paediatric asthma: get the
basics right
Perhaps one of the most worrying trends in asthma therapeu-
tics is the ever-higher doses of medication being prescribed in
primary care before a referral to hospital is made. Perhaps this
seems like special pleading by tertiary care, but we already know
of the hypoglycaemia and adrenal failure caused by high-dose
ICS prescribed inappropriately. The approach to a child with
asthma not responding to moderate doses of standard therapy is
not slavishly to increase the therapy, but to ask why there is no
response. The two usual reasons are that the diagnosis is wrong,
or that the child is not getting the treatment. This was under-
scored in a trial recruiting children with asthma uncontrolled on
a minimum of budesonide 400 μg twice daily plus salmeterol
50 μg twice daily, to determine whether adding azithromycin or
montelukast as supplementary therapy was beneficial. Neither
was effective. However, the main significance of the trial was
that of 292 children evaluated, only 55 went into the trial; the
commonest reasons for non-randomization were non-adherence
PAEDIATRICS AND CHILD HEALTH 19:6
to treatment and improved control with proper supervision. In
another study of inner-city asthma, attention to detail in the run-in
period led to a marked attrition of symptomatic asthmatics. So
the really important take-home message is that often therapy-
resistant asthma is no such thing.
The standard definitions of severe asthma incorporate per-
sistent symptoms or severe exacerbations or both, despite high-
dose standard therapy, usually combinations of high-dose ICS,
long-acting β2 agonists and leukotriene receptor antagonists. We
suggest that children meeting this definition require systematic
evaluation rather than escalating trials of treatment. A full work-
up for alternative diagnoses is essential. If the diagnosis is truly
asthma, then we have suggested the term ‘problematic severe
asthma’ for this group. A full multidisciplinary assessment,
including a nurse-led home and school visit, is undertaken to
further categorize these children. Those with ‘difficult asthma’
include children who are non-compliant, have adverse environ-
mental factors, or have psychosocial issues. Their asthma may
still be difficult to treat, but they would not be candidates for
novel molecular-based therapies such as anti-IgE (Xolair™). The
second group have severe therapy-resistant asthma (sometimes
called ‘refractory asthma’ in adult practice), and are worked up
with a series of invasive tests, including bronchoscopy. Indeed,
we recommend this sort of detailed specialist evaluation, which
goes beyond even the NICE (National Institute for Clinical Excel-
lence) guidelines, before Xolair is prescribed. In our experience,
more than 50% of problematic severe asthmatics need attention
to the basics rather than expensive and potentially dangerous
therapies. ◆
Further reading
Bacharier LB, Phillips BR, Zeiger RS, et al. CARE Network. Episodic use
of an inhaled corticosteroid or leukotriene receptor antagonist in
preschool children with moderate-to-severe intermittent wheezing.
J Allergy Clin Immunol 2008; 122: 1127–35.
Bateman ED, Boushey HA, Bousquet J, et al. GOAL Investigators Group.
Can guideline-defined asthma control be achieved? The Gaining
Optimal Asthma ControL study. Am J Respir Crit Care Med 2004;
170: 836–44.
Bisgaard H, Hermansen MN, Loland L, et al. Intermittent inhaled
corticosteroids in infants with episodic wheezing. N Engl J Med
2006; 354: 1998–2005.
Bisgaard H, Le Roux P, Bjåmer D, et al. Budesonide/formoterol
maintenance plus reliever therapy: a new strategy in pediatric
asthma. Chest 2006; 130: 1733–43.
Brand PL, Baraldi E, Bisgaard H, et al. Definition, assessment and
treatment of wheezing disorders in preschool children: an evidence-
based approach. Eur Respir J 2008; 32: 1096–110.
Ducharme FM, Lemire C, Noya FJ, et al. Preemptive use of high-dose
fluticasone for virus-induced wheezing in young children. N Engl J
Med 2009; 360: 339–53.
Guilbert TW, Morgan WJ, Zeiger RS, et al. Long-term inhaled
corticosteroids in preschool children at high risk for asthma. N Engl
J Med 2006; 354: 1985–97.
Henderson J, Granell R, Heron J, et al. Associations of wheezing
phenotypes in the first 6 years of life with atopy, lung function and
airway responsiveness in mid-childhood. Thorax 2008; 63: 974–80.
264 © 2009 Published by Elsevier Ltd.
 Symposium: respiratory medicine
Murray CS, Woodcock A, Langley SJ, et al. Secondary prevention of
asthma by the use of inhaled fluticasone dipropionate in wheezy
infants (IWWIN): double-blind, randomised controlled study. Lancet
2006; 368: 754–62.
Oommen A, Lambert PC, Grigg J. Efficacy of a short course of
parent-initiated oral prednisolone for viral wheeze in children
aged 1–5 years: randomised controlled trial. Lancet 2003; 362:
1433–8.
Panickar J, Lakhanpaul M, Lambert PC, et al. Oral prednisolone for
preschool children with acute virus-induced wheezing. N Engl J Med
2009; 360: 329–38.
Robertson CF, Price D, Henry R, et al. Short-course montelukast for
intermittent asthma in children: a randomized controlled trial. Am J
Respir Crit Care Med 2007; 175: 323–9.
Sorkness CA, Lemanske Jr. RF, Mauger DT, et al. Childhood Asthma
Research and Education Network of the National Heart, Lung,
and Blood Institute. Long-term comparison of 3 controller
regimens for mild–moderate persistent childhood asthma: the
Pediatric Asthma Controller Trial. J Allergy Clin Immunol 2007; 119:
64–72.
Strunk RC, Bacharier LB, Phillips BR, et al. CARE Network. Azithromycin
or montelukast as inhaled corticosteroid-sparing agents in
moderate-to-severe childhood asthma study. J Allergy Clin Immunol
2008; 122: 1138–44.
Szefler SJ, Mitchell H, Sorkness CA, et al. Management of asthma based
on exhaled nitric oxide in addition to guideline-based treatment for
inner-city adolescents and young adults: a randomised controlled
trial. Lancet 2008; 372: 1065–72.
PAEDIATRICS AND CHILD HEALTH 19:6
Practice points
• Preschool wheeze is most usefully described on the basis of
current symptoms –episodic (viral) versus multitrigger –
rather than by the presence or absence of atopy
• Epidemiological phenotypes (transient versus persistent)
can only be applied retrospectively, and are of little use in
planning treatment
• Episodic (viral) wheeze should be treated with intermittent
therapy
• Oral prednisolone should not be used in any but the very
severest attack of episodic (viral) wheeze; it is NOT a
primary-care drug in this context
• If inhaled corticosteroids are used in preschool children, a
three-step protocol, including a trial of discontinuing the
medication, should prevent incorrect diagnostic labelling of
children as ‘asthmatic’
• First-line prophylactic treatment of school-age children with
asthma remains low-dose inhaled corticosteroids
• In more severely asthmatic children, consideration of fixed-
dose combination therapy with budesonide/formoterol as a
single inhaler regimen could be considered
• Most children with asthma not responding to simple
treatment do NOT require more treatment, but a review of
the diagnosis and investigation into how well the prescribed
treatments are actually being used
265 © 2009 Published by Elsevier Ltd.
 Symposium: respiratory medicine
Management of bronchiolitis
Madeleine Adams
Iolo Doull
Abstract
Bronchiolitis is the commonest cause of hospital admission in infancy.
Severity varies from mild and self-limiting through to respiratory failure
requiring intensive care and ventilation. Many viruses cause bronchioli-
tis, the commonest being respiratory syncytial virus (RSV). Supportive
care is the mainstay of treatment, with emphasis on fluid replacement
and oxygen therapy. Agents with evidence of no benefit in acute bron-
chiolitis include β2 agonists, ipratropium, montelukast, corticosteroids,
antiviral agents such as ribavirin or RSV immunoglobulin, physiotherapy,
nebulized deoxyribonuclease or antibiotics. It is possible that nebulized
epinephrine has a small short-term effect, and that nebulized 3% hyper-
tonic saline administered with a bronchodilator may decrease length of
stay in hospital. Preventative strategies such as RSV immunoglobulin
or the anti-RSV monoclonal antibody palivizumab can decrease disease
severity.
Keywords bronchiolitis; bronchodilators; corticosteroids; hypertonic
saline; RSV
Introduction
In 1963 EOR Reynolds concluded that ‘oxygen therapy is vitally
important in bronchiolitis and there is little convincing evidence
that any other therapy is consistently or even occasionally use-
ful’. It is arguable that there has been little progress in the sub-
sequent 46 years. Treatments that might be effective include
nebulized 3% hypertonic saline mixed with a bronchodilator,
and ventilatory support for respiratory failure. For the major-
ity of patients, however, supportive management is the main-
stay, with emphasis on treating insufficient fluid intake and
hypoxia.
Epidemiology
Bronchiolitis is the commonest cause of hospitalization in
infancy. It usually affects infants aged 1–6 months, although it
can occur up to 2 years of age, and is usually a mild self-limiting
illness that does not require medical intervention. It is a clinical
syndrome characterized initially by coryzal symptoms followed
Madeleine Adams MRCPCH is a Specialist Registrar in Paediatrics at the
Cystic Fibrosis/Respiratory Unit, Children’s Hospital for Wales, Cardiff, UK.
Iolo Doull DM FRCPCH is a Consultant Respiratory Paediatrician at the
Cystic Fibrosis/Respiratory Unit, Children’s Hospital for Wales, Cardiff, UK.
PAEDIATRICS AND CHILD HEALTH 19:6
by onset of harsh cough, tachypnoea and wheezing. On exami-
nation there may be chest hyperinflation with costal recession,
and fine inspiratory crackles and polyphonic expiratory wheeze
on auscultation.
A significant contributor to confusion over the management
of bronchiolitis is the absence of an internationally agreed com-
mon definition. In the United Kingdom, Australasia, and parts
of Europe, bronchiolitis is interpreted as the presence of tachy-
pnoea, hyperinflation of the chest, and characteristically wide-
spread fine end inspiratory crackles on auscultation. Wheeze is
commonly but not invariably present. The pattern of illness is
virtually always seen in the first year of life, most commonly in
the first 6 months of life. In contrast, in North America and other
parts of Europe, bronchiolitis is a term for any viral infection of
the lower respiratory tract in the first 2 years of life, and may
include children with recurrent wheeze. The confusion over defi-
nition is compounded by different types of study; many studies
in inpatients will be closer to the UK and Australasian definition,
while many studies in outpatients will include mostly children
corresponding to the North American and European definition.
Furthermore there is no widely accepted validated scoring system
to measure illness severity, making comparisons between studies
difficult. Commonly, outcome measures in interventional studies
will include summation scores of symptoms and signs, oxygen
saturations, respiratory rate, and length of stay in hospital.
Approximately 1–3% of infants will be hospitalized with
bronchiolitis, of whom 1–2% will require ventilation for either
respiratory failure or apnoeas. Hospitalization for bronchiolitis is
commoner in boys and in lower socioeconomic groups. Preterm
infants (less than 32 weeks) and infants with chronic lung dis-
ease, congenital cardiac abnormalities, or immune deficiencies
are at higher risk of severe disease. Mortality rates in hospitalized
infants are 0.5–1.7%. Respiratory syncytial virus (RSV) accounts
for 50–90% of cases of bronchiolitis resulting in approximately
20000 UK admissions per year. Annual epidemics of RSV occur
between late autumn and spring (October to March in the north-
ern hemisphere). The virus rapidly infects the respiratory tract
epithelium, causing epithelial necrosis and destruction of the
cilia. There is a florid inflammatory response with neutrophil and
lymphocytic cellular infiltration and oedema of the submucosa,
and an imbalance in cytokines with an excess of TH type II over
type I cytokines demonstrated by high interleukin 4 (IL4)/inter-
feron γ ratio. There is increased mucus production from goblet
cells which, in combination with the decimated epithelial cells,
results in mucus plugging. Mucus plugging causes obstruction
of bronchioles with resultant air trapping leading to areas of
hyperinflation and airway collapse. As the respiratory epithelium
regenerates, the new non-ciliated cells are poorly equipped to
clear the inflammatory debris. Hypoxaemia is primarily due to
ventilation/perfusion mismatch, and hypercapnoea is a relatively
late phenomenon.
After RSV, the next most frequent cause is probably human
metapneumovirus (hMPV) which causes a similar clinical pic-
ture as RSV, although dual infection with both RSV and hMPV
may cause more severe disease. Rhinovirus may be a commoner
cause in older infants, and less frequent causes include adenovi-
rus, parainfluenza, influenza, coronavirus, and the more recently
identified bocavirus. It is possible that bacterial superinfection is
associated with increased severity of disease.
266 © 2009 Elsevier Ltd. All rights reserved.
 Symposium: respiratory medicine
Management
Interventions can be classed as supportive, therapeutic or
­preventative.
Supportive management
The current management of bronchiolitis is primarily support-
ive, concentrating on the major effects of the condition, namely
inadequate feeding, respiratory distress and apnoeas. Inadequate
feeding is usually secondary to respiratory distress and the con-
sequent increased work of breathing. The infant will be tachy-
pnoeic, may have bouts of coughing with increased upper airway
secretions, and thus may struggle to feed adequately. Tachy-
pnoea increases fluid loss, and so the infant can easily become
dehydrated.
Infants with bronchiolitis are generally intolerant of inter-
ventions, and so minimal handling is recommended. For milder
cases giving small volumes of feed at regular intervals may be
sufficient. Administration of oxygen may be enough to decrease
work of breathing, so allowing regular fluid intake. If this proves
inadequate, a nasogastric tube can be passed to administer
enteral fluids – initially bolus feeds, but if this worsens respira-
tory distress, by continuous feeding. Ultimately if the infant is
unable to tolerate enteral feeds, fluids can be administered intra-
venously. RSV infection can result in inappropriate syndrome of
inappropriate antidiuretic hormone secretion (SIADH), and so
after determining serum electrolytes, intravenous fluids are usu-
ally restricted to 75% of maintenance requirements.
Respiratory support
The ventilation/perfusion mismatch characteristic of bronchiol-
itis results in hypoxia and increased work of breathing. Oxygen
is the mainstay of treatment for respiratory distress, and is usu-
ally administered via head-box to minimize handling. For higher
concentrations, a combination of head box and facemask or
rebreathing bag may be used, although this usually implies signif-
icant respiratory failure and often precedes respiratory collapse.
Increasing impairment of gas exchange will result in hypoxia
unresponsive to head-box oxygen and hypercarbia, eventually
leading to physical exhaustion and complete respiratory failure.
Ventilatory support
Between 2% and 5% of bronchiolitis admissions will have
respiratory failure, and increasingly non-invasive ventilation is
used to offer continuous positive airway pressure (CPAP) sup-
port, usually via nasal prongs to decrease the need for intubation
and sedation. It is thought that the action of CPAP is through
prevention of airway collapse during the respiratory cycle, so
preserving ventilation and decreasing ventilation/perfusion mis-
match. Observational studies suggest that in bronchiolitis CPAP
decreases the respiratory rate, pulse rate, and partial pressure of
carbon dioxide, and so decreases the intubation rate and conse-
quently the rate of ventilator-associated pneumonia. CPAP does
not appear to affect length of hospital stay or duration of ventila-
tion. A controlled study demonstrated that if initiated early, nasal
CPAP resulted in a significant decrease in PCO2 and was well
tolerated without any significant complications, suggesting that
earlier intervention would improve outcome. If respiratory failure
continues despite nasal CPAP, full intubation and ventilation is
PAEDIATRICS AND CHILD HEALTH 19:6
warranted. In those that are mechanically ventilated, administra-
tion of exogenous surfactant may decrease duration of mechani-
cal ventilation. If mechanical ventilation is ineffective, there may
be a role for extracorporeal membrane oxygenation (ECMO).
Therapeutic interventions
Bronchodilators
Interpretation of the evidence for the efficacy of bronchodilators
in bronchiolitis is hampered by the heterogeneity of the inclusion
criteria in different studies. Many studies, particularly if outpa-
tient-based, will include infants up to 2 years old with recurrent
wheeze, and it is arguable that many of these infants may have an
asthma phenotype. Three classes of bronchodilators have been
trialled in bronchiolitis: β2 agonists, ipratropium, and adrenergic
agents, all of which have shown benefit in asthmatics.
β2 agonists
Bronchodilators are the treatment of choice for acute asthma
in older children and adults, and there is some evidence that
bronchodilators may produce short-term improvement in clini-
cal scores in outpatient-based studies of bronchiolitis. An earlier
meta-analysis by Flores and Horwitz concluded that β2 agonists
had a statistically significant but clinically insignificant effect on
oxygen saturations and heart rate when used in milder cases in
an outpatient setting, but had no significant effect on hospital-
izations. A later systematic review of all pharmacological treat-
ments concluded that β2 agonists had no significant beneficial
effects. The most recent Cochrane review of 22 trials involving
1428 infants with bronchiolitis who received inhaled bronchodi-
lators (including β2 agonists, ipratropium and adrenergic agents)
reported a significant improvement in overall average clinical
score, but had no effect on either pulse oximetry measurements
or on risk of hospitalization. However, the inclusion criteria for
many of the studies allowed children with recurrent wheezing
up to 24 months of age, and subgroup analysis suggested that
benefit was primarily in outpatient studies of shorter duration,
suggesting that many of those who showed benefit may have had
recurrent wheeze rather than isolated bronchiolitis.
Ipratropium
An earlier systematic review identified four studies assessing
the use of nebulized ipratropium bromide in bronchiolitis and
concluded that there was no significant benefit. A subsequent
Cochrane analysis reached similar conclusions. It is likely that
ipratropium has a role in young infants with recurrent wheeze.
Adrenergic agents
Adrenergic agents such as epinephrine appear attractive as they
combine the β-adrenergic effects of bronchodilation with the
α-adrenergic effects of vasoconstriction of the bronchiolar vas-
culature. The vasoconstrictive action in particular could poten-
tially decrease the characteristic mucosal oedema and mucus
hyper-secretion. A meta-analysis of 14 studies concluded that
nebulized epinephrine had beneficial effects compared to either
placebo or salbutamol, but the effects were short-term. In studies
of inpatients, when compared to placebo epinephrine resulted in
a ­significantly better clinical score at 60 minutes, and when com-
pared to salbutamol resulted in a significantly lower respiratory
267 © 2009 Elsevier Ltd. All rights reserved.
 Symposium: respiratory medicine
rate after 30 minutes. Short-term benefits were more pronounced
in outpatient studies, but the benefits lasted less than 60 minutes
and had no effect on hospitalization rates.
Thus in summary there is very little evidence of benefit for
any bronchodilators in bronchiolitis. What evidence there is of
very short-term benefit from epinephrine, mostly in the outpa-
tient setting, which may reflect benefit in infants with recurrent
wheeze phenotypes.
Montelukast
The leukotriene receptor antagonist montelukast may be benefi-
cial in young children and infants when started early in acute
(mostly virus-induced) asthma/wheezing. However, a recent
placebo-controlled study of montelukast in 53 infants with a first
episode of bronchiolitis demonstrated absence of benefit, with no
significant differences in length of hospitalization, clinical sever-
ity score, or inflammatory mediators between the two groups.
Corticosteroids
Corticosteroids would appear to be an attractive therapy for
bronchiolitis; bronchiolitis is characterized by a florid inflamma-
tory response, and the anti-inflammatory actions of corticoste-
roids are beneficial in older children and adults with asthma.
Unfortunately there is a large body of evidence for their lack of
benefit in bronchiolitis. The most recent Cochrane review of 13
trials included nearly 1200 children aged 0–30 months who had
received the equivalent of 0.5–10 mg/kg of systemic prednisone
for 2–7 days. Although the length of hospitalization was 0.38
days shorter in the steroid-treated group, the difference was not
significant, and there was a similar lack of significant differences
in respiratory rate, oxygen saturations, need for supplemental
oxygen, need for supportive fluids, and need for bronchodilators.
Furthermore corticosteroids showed no significant benefits even
in subgroup analyses of infants who were less than 12 months
of age, had confirmed respiratory syncytial virus (RSV) infection,
or had no previous history of wheezing. It is interesting to note
that a recent large trial of oral corticosteroids in older preschool
children with acute wheeze also demonstrated no benefit.
Hypertonic saline
The mode of action of hypertonic saline (HS) in bronchiolitis is
unclear. In vitro, HS increases the airway surface liquid (ASL)
height in an epithelial cell line model, and in vivo HS increases the
mucociliary clearance of radiolabelled aerosol in both normal con-
trols and asthmatics. Hypertonic saline is increasingly utilized in
cystic fibrosis (CF) where it increases mucociliary clearance for at
least 8 hours in a dose-dependent manner, and increases sputum
expectoration. In a randomized placebo-controlled study 4 ml of
7% HS administered twice daily via a nebulizer resulted in signifi-
cant decreases in the number of both mild and severe respiratory
exacerbations, in the days antibiotics received per year; and in the
number of days absent from work or school. HS can induce bron-
chospasm even in normal adults, and so in CF it is recommended
that HS is only administered after a bronchodilator.
There is increasing evidence that disease pathogenesis in CF
is due to increased sodium absorption from the airway surface,
leading to airway dehydration and subsequent loss of the normal
ASL volume and thus decreased ciliary clearance. Administration
of HS is thought to attract water into the airway, so ­increasing
PAEDIATRICS AND CHILD HEALTH 19:6
ASL volume and ciliary clearance. Other agents that attract
water into the airway, such as the sugar alcohol mannitol, have
similar effects. It is unclear whether a similar mode of action is
applicable to bronchiolitis, and other postulated modes of action
include breakdown of ionic bonds within airway mucus, result-
ing in decreased mucus viscosity and so easing clearance of air-
way secretions, and increased prostaglandin E2 release which
stimulates ciliary beat frequency; or possibly HS simply induces
coughing, even in normal subjects, and so some of its action
could be simply through its tussive effect.
A Cochrane review of four trials of HS in 254 infants with
bronchiolitis demonstrated significantly decreased duration of
hospitalization. Three trials were from the same group of inves-
tigators, and used a similar treatment protocol of administering
4 ml of 3% HS every 8 hours, mixed with either 1.5 mg of epi-
nephrine or 5 mg of terbutaline. Although administration of a
bronchodilator was not essential in the other trial, the majority
of subjects concomitantly received a bronchodilator. The pooled
data suggested that 3% HS decreased the median duration of
stay by 0.94 days (95% confidence Interval: −1.48 to −0.4, P =
0.0006) compared to nebulized 0.9% saline, although there was
no significant effect on hospitalization rate.
Antiviral agents
A controlled trial of the antiviral agent ribavirin via aerosol in
infants receiving mechanical ventilation for severe RSV infec-
tion initially showed significant benefit, although there were
concerns over the choice of aerosolized water as the control. A
repeat study using aerosolized saline as control demonstrated no
benefit, and a subsequent meta-analysis of all controlled trials
concluded that there was no significant benefit from aerosolized
ribavirin, although there may be a slight reduction in the dura-
tion of mechanical ventilation. There may however be a role for
intravenous ribavirin in severe disease in immunocompromised
patients. Intravenous RSV immune globulin is ineffective in the
acute treatment of RSV lower respiratory tract infection.
Other treatments
Other treatments that have shown no benefit in acute bronchi-
olitis include nebulized deoxyribonuclease, chest physiotherapy,
and antibiotics.
Prevention
Infants known to be at risk of severe bronchiolitis can receive
passive immunization against RSV, either with RSV immunoglob-
ulin (RSVIG) by monthly intravenous injections or with monthly
intramuscular injections of the humanized monoclonal antibody
palivizumab. Neither has any risk of RSV transmission, and nei-
ther interferes with the routine childhood vaccination schedule.
RSV immunoglobulin
The PREVENT study was a randomized double-blind compari-
son of RSV immunoglobulin prophylaxis every 30 days versus
placebo in over 500 premature infants and infants with broncho-
pulmonary dysplasia; it demonstrated a significant reduction of
hospitalization for RSV and duration of hospitalization, but dem-
onstrated no significant effect on the need for mechanical venti-
lation, days on intensive care, or mortality. RSVIG is no ­longer
268 © 2009 Elsevier Ltd. All rights reserved.
 Symposium: respiratory medicine
used because of the need for monthly intravenous infusions and
the small theoretical risk of viral transmission.
Humanized monoclonal antibody
There is probably no agent that divides neonatologists and respi-
ratory paediatricians as much as palivizumab. Palivizumab is
a humanized immunoglobulin G1 that binds to the RSV fusion
protein, and in the cotton rat model it is 50–100 more potent
than RSV immunoglobulin. It prevents entry of RSV into the cells
lining the respiratory tract, and when administered intramuscu-
larly has a half life of 18–20 days. It is administered by monthly
intramuscular injections during the RSV season, and is currently
licensed for use in preterm infants (less than 35 weeks’ gestation)
who are less than 6 months old at the start of the bronchiolitis
season, any child under 2 years of age who has received treat-
ment for bronchopulmonary dysplasia in the previous 6 months,
or in children with haemodynamically significant heart disease.
The Impact study was a placebo-controlled trial of intramus-
cular palivizumab every 30 days in over 1500 high-risk infants;
the infants were either born at or earlier than 35 weeks’ ges-
tation and were less than 6 months of age, or were less than
2 years of age and had had a clinical diagnosis of broncho-
pulmonary dysplasia which required treatment (oxygen, cor-
ticosteroids, bronchodilators or diuretics) in the preceding
6 months. Palivizumab significantly decreased the risk of hos-
pitalization, duration of hospitalization, and the need for inten-
sive care, although it had no significant effect on either the need
for ventilation or death.
Although the relative effect was striking, the absolute effect
was much less so, and the calculated number needed to treat
(NNT) was 17. Due to its high cost, its economic value has been
questioned, with an estimated cost of preventing one hospital
admission of £43,000. In 2005 the Joint Committee for Vacci-
nations and Immunizations (JCVI) recommended that palivuz-
imab should be given to all children less than 2 years of age who
had previously been treated with home oxygen for chronic lung
disease, or those with haemodynamically significant congenital
heart disease, pulmonary hypertension, or severe congenital
immune deficiency. Despite these recommendations, the use of
Palivuzimab varies greatly in different areas of the UK.
Vaccination
At present there is no effective vaccine for bronchiolitis. Attempts
have been made to develop a vaccine against RSV; however,
when this vaccine was entered into clinical trials patients given
the vaccine went on to develop more severe symptoms than
those not vaccinated. Current efforts to develop a vaccine suit-
able for human trials focus on using non-animal-derived materi-
als to obtain viral plasmid DNA; at present the efficiency of viral
recovery is 30–100%.
Conclusions
Fluids and oxygen are the mainstays of management. There is
increasing evidence for the role of nebulized 3% hypertonic
saline administered with a bronchodilator. CPAP and if neces-
sary mechanical ventilation are effective for respiratory failure.
Palivizumab decreases the risk of hospitalizations in high-risk
infants, although there is controversy over its benefits. ◆
PAEDIATRICS AND CHILD HEALTH 19:6
Further reading
Amirav I, Luder AS, Kruger N, et al. A double-blind, placebo-controlled,
randomized trial of montelukast for acute bronchiolitis. Pediatrics
2008; 122: e1249–55.
Deshpande SA, Northern V. The clinical and health economic burden
of respiratory syncytial virus disease among children under 2 years
of age in a defined geographical area. Arch Dis Child 2003; 88:
1065–9.
Elkins MR, Robinson M, Rose BR, et al. A controlled trial of long-term
inhaled hypertonic saline in patients with cystic fibrosis. N Engl J
Med 2006; 354: 229–40.
Embleton ND, Harkensee C, McKean MC. Palivizumab for preterm
infants. Is it worth it? Arch Dis Child Fetal Neonatal Ed 2005; 90:
F286–9.
Gadomski AM, Bhasale AL. Bronchodilators for bronchiolitis. Cochrane
Database Syst Rev 2006(3): CD001266.
Handforth J, Friedland JS, Sharland M. Basic epidemiology and
immunopathology of RSV in children. Paediatr Respir Rev 2000; 1:
210–4.
Hartling L, Wiebe N, Russell K, et al. A meta-analysis of randomized
controlled trials evaluating the efficacy of epinephrine for the
treatment of acute viral bronchiolitis. Arch Pediatr Adolesc Med
2003; 157: 957–64.
Javouhey E, Barats A, Richard N, et al. Non-invasive ventilation as
primary ventilatory support for infants with severe bronchiolitis.
Intensive Care Med 2008; 34: 1608–14.
King VJ, Viswanathan M, Bordley WC, et al. Pharmacologic treatment
of bronchiolitis in infants and children: a systematic review. Arch
Pediatr Adolesc Med 2004; 158: 127–37.
Palivizumab, a humanized respiratory syncytial virus monoclonal
antibody, reduces hospitalization from respiratory syncytial
virus infection in high-risk infants. The IMpact-RSV Study Group.
Pediatrics 1998; 102: 531–7.
Patel H, Platt R, Lozano JM, Wang EE. Glucocorticoids for acute viral
bronchiolitis in infants and young children. Cochrane Database Syst
Rev 2004(3): CD004878.
Thorburn K, Harigopal S, Reddy V, et al. High incidence of pulmonary
bacterial co-infection in children with severe respiratory syncytial
virus (RSV) bronchiolitis. Thorax 2006; 61: 611–5.
Williams JV, Harris PA, Tollefson SJ, et al. Human metapneumovirus
and lower respiratory tract disease in otherwise healthy infants and
children. N Engl J Med 2004; 350: 443–50.
Zhang L, Mendoza-Sassi RA, Wainwright C, Klassen TP. Nebulized
hypertonic saline solution for acute bronchiolitis in infants.
Cochrane Database Syst Rev 2008(4): CD006458.
Practice points
• Although for most infants bronchiolitis is a mild, self limiting
illness, it is nevertheless the commonest cause of hospital
admission in infancy – 1–3% of infants will be hospitalized
• It is a clinical syndrome characterized by coryzal symptoms
followed by onset of harsh cough, tachypnoea and
269 © 2009 Elsevier Ltd. All rights reserved.
 Symposium: respiratory medicine
wheezing with chest hyperinflation with costal recession on
examination, and fine inspiratory crackles and polyphonic
expiratory wheeze on auscultation
• RSV accounts for 50–90% of cases of bronchiolitis – other
causative agents include Human Metapneumovirus (hMPV),
rhinovirus and less frequently adenovirus, parainfluenza,
influenza, coronavirus and the more recently identified
bocavirus
• Management of bronchiolitis is primarily supportive,
concentrating on the major complications of inadequate
feeding, respiratory distress and apnoeas. Fluids (either
PAEDIATRICS AND CHILD HEALTH 19:6
enteral or parenteral) and Oxygen are the mainstays,
although CPAP or mechanical ventilation are effective for
respiratory failure
• There is no role for bronchodilators, corticosteroids, antiviral
agents, physiotherapy, nebulized DNAse or antibiotics,
but nebulized 3% hypertonic saline administered with a
bronchodilator may decrease length of stay in hospital
• Preventative strategies such as RSV immunoglobulin or the
anti-RSV monoclonal antibody palivuzimab can decrease
disease severity and need for hospitalization, although there
is controversy over their benefits
270 © 2009 Elsevier Ltd. All rights reserved.
 Symposium: respiratory medicine
Obstructive sleep apnoea in
children
Serena Favaro
Tom Hilliard
John Henderson
Abstract
Habitual snoring is a very common problem in the paediatric population,
especially in preschool children. The differentiation of snoring children
into those with primary snoring and those with obstructive sleep apnoea
(OSA) is difficult on clinical grounds and the gold standard investigation
remains the overnight polysomnogram. The epidemiology of sleep-dis-
ordered breathing and OSA in children is changing, in part in relation
to the rising prevalence of overweight and obesity in young people.
There is also increasing recognition of the adverse neurodevelopmental
sequelae that can be associated with OSA, and possibly with milder
manifestations of sleep-disordered breathing such as primary snoring.
It is therefore important that paediatricians are aware of the problems
of sleep-disordered breathing in children, and that a detailed sleep his-
tory is part of routine clinical assessment when sleep-related breathing
disturbance is suspected or possible.
Keywords adenotonsillectomy; child; continuous positive airways
­pressure; obstructive sleep apnea syndrome; polysomnography; sleep
Definition
Obstructive sleep apnoea (OSA) is defined as a ‘disorder of
breathing during sleep characterized by prolonged upper airway
obstruction that disrupts normal ventilation during sleep and
normal sleep patterns’. Obstruction occurs intermittently and
may be partial or complete. It is related to increased upper air-
way resistance and may be associated with snoring. Persistent
snoring is often reported by parents as the first indication of air-
way obstruction during sleep, but not all individuals with OSA
have habitual snoring. A history of frank apnoeic events is less
frequently reported in children than in adults.
OSA is part of a complex of sleep-disordered breathing (SDB)
with a spectrum of clinical manifestations and severity. The first
Serena Favaro MD is a Clinical Fellow in Paediatric Respiratory Medicine
at the Department of Respiratory & Sleep Medicine, Bristol Royal
Hospital for Children, Bristol, UK.
Tom Hilliard MD FRCPCH is a Consultant in Paediatric Respiratory
Medicine at the Department of Respiratory & Sleep Medicine, Bristol
Royal Hospital for Children, Bristol, UK.
John Henderson MD FRCPCH is a Reader in Paediatric Respiratory
Medicine at the Department of Respiratory & Sleep Medicine, Bristol
Royal Hospital for Children, Bristol, UK.
PAEDIATRICS AND CHILD HEALTH 19:6
level of this continuum is primary snoring, that is snoring with-
out apnoeas, sleep disruption or gas exchange alteration. The
other extreme is frank OSA with obstruction that disrupts nor-
mal ventilation during sleep and normal sleep structure. Between
these extremes is a specific entity termed upper-airway resistance
syndrome (UARS). This is characterized by increasingly negative
intrathoracic pressures during inspiration causing arousal and
sleep fragmentation, but with preserved airflow and oxygenation.
Despite normal gas exchange, UARS can be associated with day-
time neurobehavioral symptoms that occur in association with
OSA. It is still unclear whether primary snoring has similar mor-
bid outcomes, especially involving cognitive impairment.
Epidemiology
The prevalence of OSA in childhood appears to be around 2–3%,
and it affects children of all ages. The peak prevalence is between
2 and 8 years when tonsillar and adenoidal size is greatest in
relation to upper-airway calibre.
Frequent snoring is reported by parents in 3–15% of children,
while the prevalence of reported apnoeic events is 0.2–4%. How-
ever, this figure is probably an underestimate of the true preva-
lence of OSA due to the exclusion of other conditions in the SDB
spectrum that are not associated with apnoeas, including UARS
and obstructive hypoventilation. An increased prevalence in
males compared with females has been reported, possibly related
to hormonal effects of testosterone on airway collapsibility and
central fat deposition. Postpubertal boys have also been reported
to have an increased length of the upper airway compared with
girls, another factor that might increase the susceptibility to air-
way obstruction in the adult male population.
In view of the reported association between OSA and obe-
sity in adults, this relationship has also been studied in children.
However, due in part to methodological differences between
studies and reliance on self-reported or parent-reported weight
and height, there are still uncertainties about the strength of the
relationship between body mass, body fat and sleep-disordered
breathing in children, although OSA is increasingly recognized in
the overweight and obese childhood population.
Other factors that have been identified to increase the risk of
OSA are those associated with increased resistance of the upper
airway, including craniofacial abnormalities and syndromic condi-
tions such as Down, Prader–Willi and Marfan syndromes, which
have all been reported to increase the risk of OSA. In addition,
neurodevelopmental abnormalities and neuromuscular diseases
can impair upper airway control with loss of airway patency.
The relatively high prevalence of habitual snoring and symp-
toms suggesting OSA in first-degree relatives of children with OSA
suggests a familial predisposition to the condition. Genetic stud-
ies have identified a chromosomal region containing the ApoE
gene that confers an increased risk of OSA in children, but is also
more frequently associated with neurocognitive impairment in
OSA than would be expected by chance, suggesting that it may
have a modifying effect on the neurological consequences of OSA.
Genome-wide studies have suggested shared genetic determinants
of obesity and OSA, and genetic determinants of obesity have been
estimated to account for one-third of the variance in severity of
OSA. It is likely that genomic approaches will be applied increas-
ingly to investigate the aetiology of OSA and its consequences.
271 © 2009 Elsevier Ltd. All rights reserved.
 Symposium: respiratory medicine
Pathophysiology
Control of airway patency
Several factors are involved in the control of airway patency
during sleep. Increased upper-airway resistance during sleep
can result from interactions between anatomical factors (includ-
ing craniofacial morphology and soft-tissue hypertrophy; both
of which lead to increased airway collapsibility), ventilatory
and neuromuscular control, and alterations in arousal thresh-
old. Analysis of breathing patterns during sleep in children not
affected by OSA shows that obstructive episodes, inspiratory
flow limitation, and arousals secondary to increased respiratory
effort are uncommon occurrences. However, apnoeas of 10–15
seconds’ duration and oxygen desaturations are relatively com-
mon, although the oxygen saturation rarely falls below 90% in
these so-called ‘physiological’ apnoeas. Finally, a reduced respi-
ratory rate with increased end-tidal CO2 has been documented
during non-REM sleep in otherwise healthy children.
Neuromuscular activity is an important factor in the control
of airway patency during sleep, particularly the actions of the
pharyngeal dilator muscles, which include the genioglossus,
hyoglossus and styloglossus. Activation of these muscles causes
forward movement of the tongue leading to increased pharyngeal
airway calibre and stiffness. Genioglossus activation decreases
at sleep onset, but in children with OSA it increases during non-
REM sleep compared with healthy children. In contrast, in REM
sleep there is reversal of this increased tone and pharyngeal
muscle activity is reduced, leading to an increased tendency to
apnoea. REM sleep is characterized by a general reduction in
postural muscle tone, including the intercostal muscles. This
leads to lower functional reserve capacity in REM sleep, which
exacerbates the abnormalities in gas exchange that occur in asso-
ciation with apnoea.
In children with primary snoring, neuromuscular compen-
sation appears to be sufficient to avoid apnoea, hypopnoea,
arousal, and alteration in gas exchange. However, even in the
presence of successful neuromuscular compensation, alterations
of breathing during sleep have been reported, including limita-
tion of flow, increased respiratory effort and respiratory rate, and
subtle disruption of sleep structure. In UARS there is no altera-
tion in gas exchange and no apnoeas, but arousals do occur in
association with a lower level of respiratory effort compared with
those in children with OSA, possibly due to differences in arousal
thresholds.
Anatomical considerations
The anatomical configuration of the head and neck can influ-
ence the calibre of the pharyngeal airway and thus contribute to
an increased risk of obstruction. Conditions associated with nar-
rowed pharyngeal airway include skeletal abnormalities, espe-
cially those associated with hypoplasia or retropositioning of the
mandible or maxilla, which may be isolated or associated with
specific skeletal dysplasias and dysmorphic syndromes.
Another important determinant of upper-airway calibre is the
soft tissue of the pharynx. The mass of adenoids, tonsils and soft
palate is greater in children with OSA than in healthy children,
and is greatest in relation to airway size in preschool children,
in whom OSA is highly prevalent. The oropharynx also includes
several soft-tissue structures – such as tonsils, tongue, uvula,
PAEDIATRICS AND CHILD HEALTH 19:6
soft palate and lateral pharyngeal wall – that can contribute to
OSA. The relatively greater importance of the enlargement of soft
tissues in young compared with older children is related to the
ratio between soft tissue and the stage of craniofacial develop-
ment. The facial skeleton grows rapidly during early childhood,
reaching about two-thirds of adult size by the age of 3 years.
A longitudinal cephalometric study showed that the growth of
soft tissues is more rapid than skeletal development in children
aged 3–5 years. This results in a relatively narrowed airway and
increased susceptibility to OSA in this age group.
Given the importance of soft-tissue enlargement in the patho-
genesis of OSA, a growing concern has been the increased
prevalence of obesity in children and the effects of this on sleep-
disordered breathing. Given the increasing recognition of the
contribution of obesity to OSA, a new classification of OSA based
on the underlying pathophysiology has been proposed. Type I is
associated with markedly increased lymphadenoid tissue mass in
the upper airways in the absence of obesity, and type II is char-
acterized by obesity with milder lymphoid tissue hypertrophy.
A third group that includes patients with OSA associated with
craniofacial and neuromuscular disorders has also been mooted,
although the evidence base for a type III in the OSA classification
is not yet well developed.
Nasal airway patency may also contribute to the aetiology of
OSA due to the contribution of the nasal airways to total airways
resistance. Children with allergic rhinitis, which causes dilata-
tion of turbinate vessels leading to oedema of the nasal mucosa,
can develop habitual mouth breathing. This results in elongation
of the face shape with caudal growth of the mandible resulting
in further narrowing of the airway. The combination of these
changes increases both the risk of OSA and its severity.
Effect of OSA on sleep structure
Sleep is a non-homogeneous, active, cyclic process that is divided
into non-REM and REM phases. REM sleep is a metabolically
active state and represents around 20–25% of sleep time in older
children and adults, gradually diminishing with increasing age.
The function of REM sleep remains unclear, but it is believed
to be important for learning and memory consolidation, and for
neuronal plasticity and development. In contrast, non-REM sleep
is believed to have a role in energy conservation and antioxidant
defences.
Arousal is a common finding in children affected by OSA.
Arousal is a ‘rescue’ mechanism which permits opening of the
collapsed airway and normalization of gas exchange. Increased
respiratory effort and increased carbon dioxide tension act as
triggers for this mechanism, but the arousal threshold varies with
sleep state, being highest in slow-wave sleep and lowest in REM
sleep. Children with OSA have been shown to have decreased
arousal responsiveness compared with a healthy population, but
the reason for this difference is still unclear. The combination of
apnoea, hypopnoea and arousal leads to disruption of the sleep
architecture in patients with OSA. Changes in sleep structure
include alteration in the usual cyclic pattern of sleep, increased
slow-wave sleep, and electroencephalographic (EEG) changes
related to respiratory effort.
The majority of obstructive apnoeas in children have been
shown to occur during REM sleep and to be more frequent during
the last third of the night, when the proportion of REM is greater
272 © 2009 Elsevier Ltd. All rights reserved.
 Symposium: respiratory medicine
compared with the first third. Therefore, most arousals will affect
REM sleep but, in a study of adults with OSA, no differences in
clinical presentation or daytime sleepiness were found between
those with events mostly in REM or non-REM sleep. However,
in children with OSA, sleep fragmentation predominating during
REM sleep could explain some of the neurocognitive sequelae
that have been described.
Clinical presentation
Daytime and night time symptoms
Children affected by sleep apnoea can present with a range of
symptoms that can manifest not only during sleeping time but
also in wakefulness. Greater morbidity, with increased health
care and drug utilization, appears to be present years before
OSA is diagnosed and effectively treated. The most common
symptoms, reported in more than 96% of cases, are snoring
and difficulty in breathing during sleep. Episodes of increased
respiratory efforts, with paradoxical abdominal movements, are
also described. These may be accompanied by gasping, frequent
awakenings or continuous movements during sleep. Nocturnal
sialorrhoea is another symptom reported by parents and is sec-
ondary to mouth breathing associated with nasal obstruction.
Nocturnal symptoms also include episodic enuresis, possibly
related to increased natriuretic peptide secretion. Patients with
adenotonsillar hypertrophy may not complain of difficulty in
breathing while awake but more commonly have daytime symp-
toms of increased frequency of upper respiratory tract infections,
sinusitis, sore throat and mouth breathing.
However, none of these symptoms is pathognomonic of OSA,
and a history of nocturnal snoring does not distinguish between
OSA and primary snoring.
In contrast with the OSA in adults, excessive daytime sleepi-
ness (EDS) is less commonly reported in children, being present
in around 7–10% of cases. It has been suggested that intrusive
daytime naps contribute to the prediction of clinically signifi-
cant OSA in children. Also, EDS is more prevalent in children
with more severe OSA and with higher body mass index. How-
ever, the identification of EDS in children is still difficult due
to its often subtle manifestations. The recent introduction of a
new questionnaire with higher specificity for daytime sleepi-
ness has suggested that the incidence of this complication is
substantially higher (40–50%) than previously reported. Chil-
dren with OSA can also present with daytime hyperactivity and
learning difficulties as a consequence of sleep disruption (see
below).
Sequelae of OSA
Neurocognitive impairment
In contrast to historical descriptions of florid physical manifesta-
tions of severe OSA, there is increasing recognition of the impor-
tance of neurocognitive consequences of even mild degrees of
OSA, possibly including primary snoring. Such children may
present with hyperactivity, inattentiveness, and poor school per-
formance. In addition, aggressive behaviour may be reported.
The basis of such reversible behaviour change is not completely
clear, but could be the result of chronic sleep disturbance and
episodic hypoxia.
PAEDIATRICS AND CHILD HEALTH 19:6
Using magnetic resonance spectroscopy, Halbower and others
have recently shown vulnerability of specific areas of the cortex
to sleep fragmentation and hypoxaemia resulting in the forma-
tion of free radicals and lactate in association with alterations of
cerebral blood flow. Therefore, children with OSA can develop
impairment of IQ, memory or executive functions. A relationship
between the degree of disruption of sleep architecture and the
severity of behavioural changes has also been reported. However,
several studies in the literature suggest that behavioural changes
may be reversible once OSA has been effectively treated.
The role of sleep fragmentation in the pathogenesis of neuronal
deficit has been suggested by studies showing that children with
primary snoring (without gas exchanges abnormalities) can also
be affected by neurological impairment, and children affected by
primary snoring or mild OSA can present with neurobehavioural
changes. In the light of this recent growing knowledge about the
relationship between mild SDB and neurobehavioural changes,
the importance of a careful assessment of snoring children should
be highlighted. There is also emerging evidence that other factors
such as obesity and short sleep duration may exacerbate adverse
neurocognitive effects of SDB.
Multisystem morbidity
OSA-related morbidity can affect other systems, including cardio-
vascular, metabolic and endocrine function. The basis of this may
be secondary to oxidative and systemic inflammatory processes.
The underlying mechanism leading to cardiovascular morbidi-
ties (such as blood-pressure dysregulation, systemic hyperten-
sion, left ventricular dysfunction) is most likely an inflammatory
response in the microvasculature. This leads to raised circulat-
ing C-reactive protein (CRP), a recognized risk factor for cardio-
vascular disease. In addition, increased sympathetic activity has
been demonstrated in the context of OSA. The combination of
OSA and obesity in adults is associated with metabolic syndrome
(insulin resistance, dyslipidaemia, hypertension), but this rela-
tionship is not well established in children. There is some recent
evidence that levels of circulating adipokines, such as leptin,
are raised in children with OSA, but the association is mainly
through increased body mass.
In the past, OSA has been associated with failure to thrive,
but this seems to be less common now. The pathophysiology
of growth failure in this context is a reduction of insulin-like
growth factor (IGF-1) or insulin-like growth factor binding pro-
teins (IGFBPs), which reverses following adenotonsillectomy.
Children with OSA and chronic snorers have disruption of slow-
wave sleep, which is characterized by secretion of growth hor-
mone (GH) and IGF-1. In addition to this mechanism, there may
also be increased energy expenditure and dysphagia leading
to decreased energy intake. Furthermore, patients successfully
treated for OSA, even if obese, show acceleration in weight gain
afterwards.
How to diagnose OSA
In accordance with the Clinical Practice Guidelines of the
­American Academy of Paediatrics, it is important to avoid
delayed diagnosis in order to avoid the potentially serious con-
sequences of untreated OSA. Therefore, it is necessary to review
273 © 2009 Elsevier Ltd. All rights reserved.
 Symposium: respiratory medicine
the most appropriate techniques for the diagnosis and the
­various ­treatment options.
History and physical examination
The first objective in assessing a child with suspected OSA is to
decide whether advanced evaluation is needed. A sleep history
and thorough clinical examination can help to determine the pos-
sibility of SDB. However, history is often unreliable. The loud-
ness of snoring does not correlate with the probability of OSA,
and obstructive events are more likely to occur in the later part
of the night when parental observation is less likely.
A common picture on examination of a child referred for sus-
pected OSA is the union of multiple physical findings, defined
as ‘long face syndrome’. Children with obstructive episodes
often mouth-breathe, with the final result of alteration in dento-
alveolar morphology characterized by high-arched palate, narrow
maxilla, retrognathia, and increased lower facial height. Evalu-
ation of the oral cavity should include assessment of the pala-
tal size and shape and classification of tonsillar size. Inspection
of the nose may reveal causes of obstruction to nasal patency,
such as polyps, oedema of the nasal mucosa, or septal defects.
In addition, an evaluation of the profile of the child is helpful to
identify micrognathia or retrognathia (especially in children with
a genetic condition associated with craniofacial abnormalities).
An alteration of the voice can be another mark of obstruction: a
hyponasal voice is usually secondary to adenoidal hypertrophy.
Although cardiovascular examination is mandatory, an abnormal
cardiac examination is rarely present in children with OSA.
Despite these clues, it should be noted that several studies
have shown that history and physical examination are not able
to discriminate between primary snoring and obstructive sleep
apnoea.
Screening questionnaires
A number of structured sleep questionnaires have been devised
to attempt to improve the ability to discriminate between primary
snoring and OSA. The 22-item sleep-disordered breathing scale of
the Pediatric Sleep Questionnaire (PSQ) has been recently evalu-
ated against polysomnography and found to have a sensitivity
of 81% and a specificity of 85%. However, questionnaires can
probably be best regarded as a screening tool to decide which
child should be referred for further investigations.
Further investigations
The gold standard in the diagnosis of OSA is the nocturnal poly-
somnogram (PSG). This laboratory-based test is suitable for chil-
dren of all ages and is the best technique for differentiating between
primary snoring and OSA and for evaluating the severity of OSA. A
full polysomnogram includes EEG, electro-oculogram (EOG), elec-
tromyogram (EMG) and electrocardiogram (ECG) leads (for the
identification of the sleep structure), abdominal and thoracic bands
(for the assessment of respiratory movements/effort), pulse oxim-
etry (a device to measure nasal and oral airflow, often a thermistor
or end-tidal CO2 monitor), and a sensor to detect changes of body
position. A snoring recording device can be also included, as well
as video recording. An overnight PSG can be used to analyse sleep
disturbance as well as ventilatory changes. Several measures can
be classified, including apnoeic episodes (type, number and calcu-
lation of the apnea/hypopnea index), arousal episodes secondary
PAEDIATRICS AND CHILD HEALTH 19:6
to respiratory events, and gas exchange abnormalities. Detailed
guidelines for interpreting PSG findings have been published but
are outside the scope of this review.
In view of the resources required to perform a PSG, sev-
eral abbreviated studies have been investigated as alternative
approaches. Nap PSG may be helpful if positive, but can result
in underestimation of OSA and a high false-negative rate. This
is attributable to the shorter total sleep time and lower propor-
tion of REM sleep included in a daytime nap. Simpler tests such
as unsupervised overnight oxygen saturation (oximetry) can
give information about desaturations during sleep, but cannot
exclude OSA if negative. The specificity and sensitivity of this
method to detect moderately severe OSA are only 60% and 67%
respectively. Capnography combined with oximetry can provide
additional information relating to cessation of airflow by measur-
ing end-tidal CO2 traces, but has not been fully evaluated for the
diagnosis of OSA in children.
Other tests designed for unsupervised use include audio and
video recordings, but the discordance among results reported
from different centres suggests that further evaluation of these are
required. For example, sensitivity varies from 71% to 94% and
specificity from 29% to 80% in different studies. Multichannel
devices designed for use in patients’ homes have been evaluated
in adults, and studies are under way in paediatric populations.
These may offer a cost-effective way of obtaining detailed sleep
information in an unsupervised setting.
Direct evaluation of the airway via flexible nasendoscopy can
be considered to investigate the level of obstruction. In compari-
son with lateral neck x-ray, endoscopy offers a more accurate
assessment of the postnasal space and provides a dynamic evalu-
ation of the airway. It is also important to remember that the
postnasal space can be normal in an awake child and become
smaller during sleep, particularly in REM.
Treatment
Surgical intervention
The recommended treatment for the majority of children with
OSA associated with adenotonsillar hypertrophy is adenotonsil-
lectomy, which is curative in over 80% of cases, although a small
proportion of patients have persistence or recurrence of symp-
toms postoperatively. Resolution of symptoms following surgery
is lower in obese subjects, those with specific underlying condi-
tions (such as craniofacial abnormalities), and in patients with a
family history of OSA. Therefore, a repeat sleep study 6–8 weeks
post-surgery is recommended for this population at high risk of
recurrence or persistence of OSA.
Although adenotonsillectomy is generally considered safe,
there are recognized complications, including bleeding, pain,
poor oral intake, vomiting and dehydration during the first week
post-surgery. Late complications include velopharyngeal incom-
petence and nasopharyngeal stenosis. Mortality is reported to
be around 1 in 10,000 cases, but the presence of risk factors
(severe OSA, especially with related gas exchanges anomalies,
other medical conditions, obesity) is associated with increased
perioperative morbidity and mortality.
Another surgical approach to children with OSA, particu-
larly in association with mandibular hypoplasia, is distraction
osetogenesis. Although the benefits of this procedure are well
274 © 2009 Elsevier Ltd. All rights reserved.
 Symposium: respiratory medicine
e­ stablished for craniofacial anomalies, its application to the treat-
ment of OSA compared with conventional approaches has yet to
be confirmed. Similarly, the place of orthodontic appliances in
the management of OSA in children remains unclear.
Non-surgical treatment
Several studies have shown the benefit of treatment with topi-
cal corticosteroids alone, leukotriene receptor antagonist, or a
combination of the two in the treatment of mild to moderate OSA
or residual symptoms post-tonsillectomy. The beneficial effect
is probably related to the expression of glucocorticoid receptors
in the upper-airway lymphoid tissue and production of leukotri-
enes, in addition to other inflammatory mediators, in the airways
of children with OSA. Therefore, medical therapy should be con-
sidered as the first-line option in children with mild OSA, but
also as an adjunct to the treatment of severe OSA, for example
while awaiting surgery or for persistent or recurrent symptoms
postoperatively.
For children with clinically significant OSA that persists or
recurs after adenotonsillectomy, the use of non-invasive continu-
ous positive airway pressure (CPAP) should be considered. This
has been shown to be highly efficacious in the treatment of OSA
in children, and is often used in cases complicated by obesity
or syndromes associated with craniofacial abnormalities. How-
ever, as in adults, the efficacy of this treatment may be limited
by suboptimal adherence. Therefore, adherence to treatment and
maintenance of optimal pressure to achieve clinical efficacy as
the child grows must be evaluated on a regular basis. Guide-
lines on titration of CPAP pressures have recently been published
and require access to a sleep laboratory with full PSG facilities.
Therefore, children that do not have resolution of symptoms fol-
lowing adenotonsillectomy and those with OSA complicated by
additional risk factors should be managed in a specialist centre
with these facilities available. ◆
Further reading
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.
Blunden SL, 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; 10: 109–18.
Bonuck K, Freeman K, Henderson J. Growth and growth biomarker
changes after adenotonsillectomy: systematic review and meta-
analysis. Arch Dis Child 2009; 94: 83–91.
Chervin RD, Weatherly RA, Ruzicka DL, et al. Subjective sleepiness
and polysomnographic correlates in children scheduled for
adenotonsillectomy vs other surgical care. Sleep 2006; 29: 495–503.
Chervin RD, Weatherly RA, Garetz SL, et al. Pediatric sleep
questionnaire: prediction of sleep apnea and outcomes. Arch
Otolaryngol Head Neck Surg 2007; 133: 216–22.
Halbower AC, Degaonkar M, Barker PB, et al. Childhood obstructive
sleep apnea associates with neuropsychological deficits and
neuronal brain injury. PLoS Med 2006; 3: e301.
Kushida CA, Chediak A, Berry RB, et al. Clinical guidelines for the
manual titration of positive airway pressure in patients with
obstructive sleep apnea. J Clin Sleep Med 2008; 4: 157–71.
PAEDIATRICS AND CHILD HEALTH 19:6
Lipton AJ, Gozal D. Treatment of obstructive sleep apnea in children: do
we really know how? Sleep Med Rev 2003; 7: 61–80.
Lumeng JC, Chervin RD. Epidemiology of pediatric obstructive sleep
apnea. Proc Am Thorac Soc 2008; 5: 242–52.
Marcus CL, Ward SL, Mallory GB, et al. Use of nasal continuous positive
airway pressure as treatment of childhood obstructive sleep apnea.
J Pediatr 1995; 127: 88–94.
Standards and indications for cardiopulmonary sleep studies in
children. American Thoracic Society. Am J Respir Crit Care Med 1996;
153: 866–78.
Cardiorespiratory sleep studies in children. Establishment of normative
data and polysomnographic predictors of morbidity. American
Thoracic Society. Am J Respir Crit Care Med 1999; 160: 1381–7.
Clinical practice guideline: diagnosis and management of childhood
obstructive sleep apnea syndrome. Pediatrics 2002; 109: 704–12.
Ovchinsky A, Rao M, Lotwin I, Goldstein NA. The familial aggregation of
pediatric obstructive sleep apnea syndrome. Arch Otolaryngol Head
Neck Surg 2002; 128: 815–8.
Palmer LJ, Buxbaum SG, Larkin E, et al. A whole-genome scan for
obstructive sleep apnea and obesity. Am J Hum Genet 2003; 72:
340–50.
Verhulst SL, Schrauwen N, Haentjens D, et al. Sleep-disordered
breathing in overweight and obese children and adolescents:
prevalence, characteristics, and the role of fat distribution. Arch Dis
Child 2007; 92: 205–8.
Practice points
• Obstructive sleep apnoea (OSA) in children is part of a
spectrum of sleep-disordered breathing that includes primary
snoring
• Frequent (or habitual) snoring is very common in children,
affecting up to 15% of the population, particularly in the
preschool age group
• Presentation of OSA with frank symptoms of sleep apnoea is
less common in the paediatric than in the adult population
• There is increasing recognition that even mild sleep-
disordered breathing, without gas-exchange abnormalities,
can be associated with adverse neurocognitive sequelae in
children
• A detailed sleep history should form a routine part of
the clinical assessment of a child when sleep-disordered
breathing is possible or suspected
• Screening tests, such as pulse oximetry, have relatively
low sensitivity for the detection of OSA, and overnight
polysomnography remains the gold standard for diagnosis
and classification of severity
• Adenotonsillar hypertrophy is an important cause of OSA
in children, and adenotonsillectomy is curative in a high
proportion of cases
• Anti-inflammatory treatment with topical corticosteroids is
effective in reducing symptoms in mild to moderate OSA
• Children with OSA that persists or recurs after
adeonotonsillectomy, or where it is complicated by obesity
or abnormalities of craniofacial development, should be
assessed in a specialist centre
275 © 2009 Elsevier Ltd. All rights reserved.
 Symposium: respiratory medicine

Management of upper- Causes of upper-airway obstruction in children

airway obstruction in Congenital Acquired

children Nasal Infectious

Nasal masses Peritonsillar abscess
Gabbi Parker Choanal atresia/stenosis Retropharyngeal abscess
Pyriform aperture stenosis Epiglottitis
Harish Vyas Pharyngeal Croup
Craniofacial anomalies Bacterial tracheitis
Laryngeal Non-infectious
Laryngomalacia Foreign body aspiration
Abstract
Vocal-cord paralysis Acquired vocal-cord paralysis
The upper airway extends from the nares and lip to the subglottis.
Subglottic haemangioma Vocal cord dysfunction
Obstruction can occur at any level, and may be congenital or acquired,
Congenital subglottic stenosis Acquired subglottic stenosis
acute or chronic. The level and severity of the obstruction determine the
Bifid epiglottis Adenotonsillar hypertrophy
clinical picture. Whilst a careful history and examination may reveal the
Laryngeal web/atresia Respiratory papillomatosis
diagnosis, investigations are frequently necessary and include physi-
Laryngeal lymphangioma Malignancy
ological studies, imaging and endoscopy. This review discusses the most
Congenital laryngeal Anaphylaxis & hereditary
commonly seen upper-airway disorders, including their presentation and
saccular cyst angioedema
management. An approach to the history and examination is covered,
Laryngospasm
as well as a discussion of the risks and benefits of the most commonly
Hypocalcaemia/hypokalaemia
used investigational strategies.
Caustic ingestions/thermal injury
Trauma
Keywords acquired; airway; congenital; endoscopy; investigations;
­obstruction
Table 1

girls. It frequently occurs with other congenital abnormalities,

Congenital upper-airway obstruction
Congenital upper-airway obstruction can present from birth to
the first few weeks of life. It is an important differential diagnosis
of severe respiratory distress in the neonatal period. The most
common anomalies are described below and have been classified
anatomically (Table 1).
most commonly the CHARGE association.
Whilst unilateral atresia may go undiagnosed until later in
childhood, bilateral atresia presents soon after birth with severe
respiratory distress and inability to pass a nasogastric tube. Stabi-
lization requires an oropharyngeal airway prior to surgical repair
via a transnasal or transpalatal approach.

Nasal obstruction
Bilateral nasal obstruction in neonates presents with upper-
­airway obstruction and cyanosis which improves on crying but
worsens on feeding. Nasal obstruction in older children generally
causes mouth breathing without airway obstruction.
Nasal masses: these range from cystic (e.g. meningoencephalo-
coele) to solid lesions (e.g. glioma, hamartoma, chordoma, tera-
toma and lymphangioma), and diagnosis is by a combination of
computed tomography and magnetic resonance imaging. Man-
agement is generally surgical, but may involve chemotherapy or
radiotherapy for malignant lesions.
Choanal atresis and stenosis: choanal atresia is lack of patency
of the posterior nasal aperture, and is seen more commonly in
Gabbi Parker MRCPCH is a Registrar in Paediatrics, Nottingham
University Hospitals, Queen’s Medical Centre Campus, Nottingham, UK.
Harish Vyas DM FRCPCH FRCP is a Professor in PICU and Respiratory
Medicine and Head of Service, Children & Young People, Nottingham
University Hospitals Queen’s Medical Centre Campus, Nottingham, UK.
Congenital pyriform aperture stenosis: this is rare and is asso-
ciated with holoprosencephaly. It is caused by bony overgrowth
of the nasal process of the maxilla, and is differentiated from
choanal atresia on computed tomography.
Treatment involves nasal stenting and drilling of the aperture
through a bucco-labial incision. The child has to be old enough
for this procedure to be performed. A tracheostomy may be
required in the intervening period.
Pharyngeal obstruction
Craniofacial anomalies: a number of craniofacial syndromes can
cause upper-airway obstruction of varying severity at the pharyn-
geal level. The most common of these is the Pierre–Robin sequence,
which consists of glossoptosis, micrognathia and cleft palate. Man-
agement is conservative, and includes insertion of a nasopharyngeal
airway, although insertion of a tracheostomy tube is occasionally
required. Upper-airway obstruction, if present, tends to resolve as
the child grows, often within the first 6 months of life.
Laryngeal obstruction
The laryngeal region consists of three regions:
• the supraglottis (epiglottis, aryepiglottic folds, false cords and
ventricles)

PAEDIATRICS AND CHILD HEALTH 19:6 276 © 2009 Published by Elsevier Ltd.
 Symposium: respiratory medicine
• the glottis (at the level of the vocal cords)
• subglottis (from lower margin of vocal cord and inferior limit
of cricoid cartilage).
Laryngomalacia: laryngomalacia is the commonest congenital
anomaly of the upper airway. The classical presentation is with
inspiratory stridor in an otherwise well child within the first few
weeks of life. Whilst this may worsen during the first 6 months,
resolution by 2 years is the norm. The severity of stridor is vari-
able, and is exacerbated by crying, feeding, lying supine, and
intercurrent respiratory tract infections. Flexible laryngoscopy
reveals an omega-shaped epiglottis that prolapses over the lar-
ynx during inspiration, and is recommended in all but the mild-
est cases. Microlaryngoscopy and bronchoscopy should also be
considered as in nearly 20% of cases there will be another airway
abnormality.
In virtually all cases, management is watchful waiting and
treatment of associated conditions (e.g. gastro-oesophageal
reflux). However, surgical intervention is warranted for severe
cases complicated by failure to thrive, obstructive sleep apnoea,
or cor pulmonale. Supraglottoplasty (aryepiglottoplasty) is cur-
rently the preferred operative procedure.
Vocal-cord paralysis: vocal-cord paralysis can be unilateral or
bilateral, congenital or acquired. The congenital form is usually
idiopathic, although it is associated with abnormalities of the
central nervous system (e.g. Arnold–Chiari malformation) or car-
diovascular anomalies, which cause stretching or compression of
the vagus or recurrent laryngeal nerve.
Bilateral paralysis presents with high-pitched stridor and sig-
nificant airway compromise, but near-normal voice as the cords
lie in the adducted position. Severe cases require intubation
followed by tracheotomy. As complete, spontaneous recovery
occurs in over half of the patients, surgical procedures – includ-
ing vocal cord lateralization, arytoidectomy and laser cordotomy
– are postponed until the patient is at least 11 years old, unless
decannulation is considered imperative at an earlier age.
In contrast to bilateral paralysis, congenital unilateral vocal-
cord paralysis may initially go unnoticed, and most children do
not require surgical intervention. The most common presenting
symptom is a weak, breathy cry. This is because the cord rests
in the paramedian position, allowing air escape during phona-
tion. Mild stridor may or may not be present. Aspiration is a
common complication as the cords cannot approximate to close
the glottis.
Subglottic haemangioma: subglottic haemangiomas are congen-
ital vascular lesions that have the potential to cause significant
airway obstruction. Most are not present at birth, but grow rap-
idly over the first few months of life, causing progressive bipha-
sic stridor, but tend to involute slowly after 1 year of age. Voice
quality may also be altered. More than 50% of patients also have
cutaneous haemangiomas, particularly of the head and neck.
Whilst the diagnosis can be made on rigid bronchoscopy, MRI is
required to demonstrate the extent of the lesion.
Most infants with subglottic haemangiomas require surgical
intervention. Traditionally this involved a tracheostomy whilst
awaiting spontaneous involution, although more recently other
procedures have been carried out to try to avoid this. These
PAEDIATRICS AND CHILD HEALTH 19:6
include laser ablation, intralesional injection of interferon/scle-
rosing agent, systemic steroids, or open surgical resection often
combined with a cricoids split.
Congenital subglottic stenosis: congenital subglottic stenosis is
defined as a diameter of 4.0 mm or less at the level of the cri-
coid region in a term infant. Presenting symptoms depend on the
degree of stenosis. Whilst severe airway obstruction at delivery
necessitating emergency tracheostomy can be seen, most chil-
dren are not symptomatic for the first few weeks to months. The
classical presentation is with recurrent croup associated with
prolonged biphasic stridor. This is because viral infections which
cause even mild subglottic swelling critically compromise the
upper airway, precipitating obstruction.
Diagnosis is by rigid bronchoscopy, although plain lateral or
anteroposterior radiographs may demonstrate the characteristic
narrowing at the level of the subglottis. Most cases resolve spon-
taneously by the age of 3–4 years. Mild congenital subglottic ste-
nosis can therefore be managed by observation alone, but more
severe cases require intervention. This ranges from laryngotra-
cheoplasty with anterior and/or posterior cartilage stenting to
partial cricotracheal resection. Placement of a tracheostomy tube
is required in less than 50% of cases.
Other laryngeal anomalies: these Includee a variety of other
causes of laryngeal obstruction, including congenital bifid epi-
glottides, congenital laryngeal saccular cysts, laryngeal webs and
atresias, and laryngeal lymphangiomas.
External tracheal compression
Vascular anomalies of mediastinal vessels may cause intratho-
racic tracheal compression resulting in symptomatic upper-airway
obstruction. Lymphatic lesions (e.g. cystic hygroma) can cause
obstruction by the same mechanism, usually at the thoracic inlet.
Acquired upper-airways obstruction
Acquired causes of upper-airways obstruction in children are
far more common than congenital causes, and are usually due
to infections (in which case appropriate antibiotics are crucial),
foreign bodies, or trauma. Steroids and nebulized adrenaline may
be helpful in reducing acute inflammation. Due to vaccination
programmes and improved anaesthetic skills, tracheostomies
are now rarely necessary for acute obstruction. They are largely
reserved for children who have severe, fixed obstruction that is
unlikely to improve for a prolonged period of time.
Infectious cause
Peritonsillar and retropharyngeal abscess: the pharynx is divi­
ded into potential spaces, where deep-neck abscesses can form;
two of these include the peritonsillar and retropharyngeal spaces.
Peritonsillar abscesses are the most common deep-neck abscesses
in children, and usually present in later childhood. The classical
presentation is with low-grade pyrexia, a severe sore throat, and
a muffled voice. Drooling and trismus may be seen. The diagnosis
can be made by inspection, if tolerated, of the oral cavity (reveal-
ing asymmetrical tonsillar swelling), but CT may be required. Man-
agement includes incision and drainage, and, as with all infectious
causes of upper-airway obstruction, appropriate antibiotic cover.
277 © 2009 Published by Elsevier Ltd.
 Symposium: respiratory medicine
Retropharyngeal abscesses, in contrast, tend to occur in pre-
school age children, and are due to the progression of bacterial
pharyngitis or pharyngeal trauma. Presentation is with fever,
drooling, asymmetrical neck swelling and neck extension. Man-
agement is with parenteral antibiotics with or without surgical
drainage.
Epiglottitis: epiglottitis is a life-threatening, rapidly progressive
condition characterized by local inflammation and oedema of the
supraglottis and epiglottis. Following the routine introduction of
Haemophilus influenzae type B vaccination in 1990, the incidence
of epiglottitis has dropped dramatically. Whilst traditionally seen
in children aged 2–7 years, most cases seen now occur in older
children, frequently caused by other pathogens.
Epiglottitis classically presents with abrupt onset of high
fever, dysphagia, drooling due to odonophagia, and quiet stridor
in a toxic-looking child who prefers to sit upright with the head
extended. Management is with intubation to protect the airway
and intravenous antibiotics. Intubation should be carried out in
theatre, and is generally only required for 24–48 hours.
Croup: viral croup, also known as acute laryngotracheobronchitis,
is the commonest cause of upper-airway compromise in children,
and is most common between the ages of 6 months and 3 years.
It is usually caused by the parainfluenza virus, and can usually
be diagnosed on clinical grounds. Presenting symptoms include a
barking cough, hoarse voice and stridor, usually on a background
of 1–2 days of coryzal symptoms. Treatment is with steroids with
or without adrenaline nebulization. Intubation is rarely required.
Spasmodic croup is also a recognized entity. It involves inter-
mittent stridor and barking cough in the absence of fever.
Bacterial tracheitis: this can occur de novo or on a background
of viral croup. The child generally has a high fever, appears
toxic, and may have rapidly progressive upper-airway obstruc-
tion. Unlike epiglottitis, ability to lie flat and control secretions
is generally not impaired. Treatment is with parenteral antibiot-
ics. Intubation is frequently required, and visualization reveals
oedema, ulceration, and copious secretions. The most common
causative organism is Staphylococcus aureus.
Non-infectious causes
Foreign body aspiration: inhaled foreign bodies can lodge at any
level of the upper airway from the supraglottis to the trachea,
and are a leading cause of accidental death in children under
5 years of age. Fortunately, most pass through the vocal cords to
the lower airways where they are less immediately life-threaten-
ing. The presentation varies depending on the level and degree
of obstruction. Oesophageal foreign bodies can compress the tra-
chea, causing partial airway obstruction.
Acquired vocal cord palsy: as with the congenital form, acquired
vocal-cord paralysis can be unilateral or bilateral. It is caused
by damage to the recurrent laryngeal nerve; the left is damaged
more frequently due to its length and course. There are various
aetiologies. The most common of these include birth trauma (due
to traction of the neck leading to stretch injury) and mediasti-
nal or neck surgery. Central causes (e.g. hydrocephalus) tend to
cause bilateral paralysis.
PAEDIATRICS AND CHILD HEALTH 19:6
Vocal cord dysfunction: functional upper-airway obstruction
due to paradoxical cord adduction during inspiration (or both
inspiration and expiration) is usually seen in older children, and
is often misdiagnosed as asthma. It is frequently precipitated by
psychological stress, but is also associated with gastro-oesopha-
geal reflux, which should be excluded. Visualization of the cords
during an acute episode confirms the diagnosis. Management is
best provided by a team that includes an ENT surgeon, speech
therapist, physiotherapist and clinical psychologist.
Acquired subglottic stenosis: this occurs much more frequently
than the congenital form, and usually follows prolonged neonatal
intubation, particularly if an oversized endotracheal tube, with
no leak, is used. The underlying cause is scarring following pres-
sure necrosis.
Adenotonsillar hypertrophy: adenotonsillar hypertrophy can
cause sleep-related upper-airway obstruction, which tends to
present with snoring, with or without apnoeas, and daytime
somnolence. Children who already have reduced pharyngeal
tone due to neuromuscular conditions are particularly at risk.
Alveolar hypoventilation may be severe enough to cause pulmo-
nary hypertension and cor pulmonale. Sleep studies can be used
to document the frequency and severity of obstructive respira-
tory events. Treatment is with adenotonsillectomy.
Acute tonsillitis on a background of tonsillar hypertrophy may
be sufficient to cause acute severe airway obstruction. A simi-
lar picture can also be caused by severe pharyngeal swelling in
Epstein–Barr virus infection.
Neoplastic disorders & extrinsic compression: both benign
and malignant neoplasms can occur within the upper airway,
leading to partial or complete obstruction. The most common
benign neoplasm is juvenile-onset respiratory papillomato-
sis, which is usually seen in children under 5 years of age.
It is caused by the human papilloma virus, which is thought
to be acquired in the peripartum period. Papillomas can occur
throughout the respiratory tract, but those in the upper air-
way can lead to significant obstruction. Malignant tumours are
extremely rare, and include rhabdomyosarcomas and squa-
mous-cell carcinomas.
Large mediastinal masses, such as those caused by lymphoma
or T-cell leukaemia, may cause upper-airway obstruction by
extrinsic compression of the trachea.
Other conditions: other conditions which can cause upper-
airway obstruction include anaphylaxis and hereditary angio-
oedema, hypocalcaemia or hypokalaemia, and thermal injury or
caustic ingestions leading to acute laryngeal swelling. Blunt or
penetrating trauma can cause obstruction at any level.
Clinical assessment of the child with upper-airway
obstruction
History
A careful paediatric history may be sufficient to make the diagno-
sis. Important details include;
• age
• speed of onset and any precipitating events
278 © 2009 Published by Elsevier Ltd.
 Symposium: respiratory medicine

• associated symptoms, including fever and drooling Severe, fixed obstruction causes biphasic stridor. A muffled
• past medical history detailing the neonatal period (birth trau- voice or weak voice indicates supraglottic pathology, whereas
ma or prolonged intubation), and any history of previous neck hoarseness indicates narrowing at the level of the glottis, thus
or chest surgery. affecting vocal cord function.
Palpation of the tongue, floor of the mouth, palate and neck is
Examination helpful in cases of congenital upper-airway obstruction, and may
Observation may reveal important clues to the diagnosis, for reveal a submucous cleft or cystic mass.
example craniofacial anomalies or cutaneous haemangiomas.
Chest deformities such as Harrison’s sulcus may be present
Investigation (Table 3)
and suggest long-standing pathology. Assessment of the work
of breathing and severity of obstruction are made at this stage; Investigations available for upper-airway disorders include phys-
children with severe progressive obstruction require immediate iological studies, imaging, and endoscopy
airway protection rather than a detailed physical examination Physiological studies include blood gases and spirometry. As
which, as well as taking time, may cause distress and precipitate hypoxia and hypercapnia occur late, blood gas determinants are
complete obstruction. Neck swelling associated with fever sug- of limited use. Spirometry is generally suitable only for patients
gests a peritonsillar or retropharyngeal abscess. over 6 years of age (as cooperation is required) with persistent
Auscultation is often the most useful part of the examina- stridor. It can demonstrate the site of the obstruction, as well as
tion (Table 2), and can be performed without a stethoscope. variability and response to bronchodilators. Flow-volume loops
Obstruction between the nares and the pharynx causes ster- are able to demonstrate three patterns:
torous ‘snoring-like’ breathing. Narrowing below this level • variable extrathoracic obstruction (flattening of the inspira-
leads to stridor, which is often the most prominent symptom tory limb)
of upper-airway obstruction. It is caused by the oscillation of • variable intrathoracic obstruction (flattening of the expiratory
a narrowed airway due to increased air turbulence. Milder limb)
degrees of narrowing may produce stridor only when airflow is • fixed upper-airway obstruction (flattening of both limbs caus-
increased (e.g. during crying), although quiet or absent stridor ing a ‘box-shaped’ loop).
may signifiy critical compromise causing decreased air move- Imaging includes radiography, computed tomography (CT),
ment and hence tissue vibration (see epiglottitis). Stridor is magnetic resonance imaging (MRI), and contrast studies.
predominantly an inspiratory sound, although an expiratory Radiographs are frequently used as a screening tool. Neck
component can be heard in lesions at or below the glottis. films may reveal changes associated with epiglottitis, croup, or
retropharyngeal abscess, but are frequently non-specific. How-
ever, it is important to avoid sending a child to an x-ray depart-
ment with potential airway obstruction. It is best carried out in
Auscultative findings and voice quality in relation to
the emergency department or intensive care unit. Chest films are
site of obstruction
useful in cases of suspected intrathoracic pathology, and may
Site of obstruction Noise during breathing Voice quality reveal changes associated with vascular rings (e.g. right-sided
cycle aortic arch). Foreign bodies will be seen on both provided that
they are radio-opaque.
Nasal Stertorous, snoring- Normal or hyponasal CT is indicated for evaluation of choanal atresia/pyriform
like aperture stenosis and retropharyngeal abscesses. CT is generally
Nasopharyngeal Predominantly readily available and has excellent spatial resolution. Whilst it is
inspiratory useful for imaging tumours and vascular malformations, it has
Oropharyngeal largely been replaced by MRI for these indications. Its main dis-
Supraglottis Inspiratory stridor Muffled advantage was poor tracheal imaging in the long axis, but with
Biphasic if critical three-dimensional reconstruction images this is now possible.
Glottis Inspiratory stridor ± Hoarse, breathy Whilst MRI does not involve radiation, it is a lengthy procedure
expiratory component frequently requiring sedation which carries additional risks in
Biphasic if severe/ Near normal situations where the airway is already compromised.
fixed phonation may be A barium swallow is diagnostic in the vast majority of cases of
seen in bilateral vascular rings, as compression of the oesophagus causes charac-
vocal cord palsy teristic filling defects. It is a simple and inexpensive procedure to
Subglottis Inspiratory stridor ± Normal or weak perform, although structures surrounding the oesophagus cannot
expiratory component be visualized. Angiography for vascular malformations has been
Biphasic if severe/ almost entirely replaced by MRI.
fixed Endoscopy remains the most important tool for evaluating
Trachea Expiratory stridor Normal the upper airway, and can be performed with a rigid or flexible
Biphasic if severe/ endoscope. It is indicated in persistent stridor of over 2 weeks’
fixed duration, and abnormal cry or hoarseness. Flexible endoscopy is
generally performed transnasally (transorally in young babies),
Table 2 allowing the entire upper airway to be visualized. Vocal-cord

PAEDIATRICS AND CHILD HEALTH 19:6 279 © 2009 Published by Elsevier Ltd.
 Symposium: respiratory medicine
Investigations available for upper-airway disorders
Investigation Indication
Neck radiograph Suspected epiglottitis
in stable child
Foreign body
Chest radiograph Foreign body
Intrathoracic pathology
Vascular ring
Barium swallow Vascular ring
Computed Choanal atresia
tomography (CT) Pyriform aperture stenosis
Retropharyngeal abscess
Tumours
Aberrant arteries/vascular rings
Acute trauma
Magnetic resonance Tumours
imaging (MRI) Aberrant arteries/vascular rings
Subglottic haemangioma
Tracheal stenosis
Flexible endoscopy Vocal-cord dysfunction
Vocal-cord paralysis
Bronchoscopy
Assessment of awake
airway dynamics
Rigid endoscopy Foreign body removal
Biopsy of lesion
Documenting posterior
glottis/subglottis/trachea
Vocal-cord paralysis
Spirometry Chronic stridor in older child
Response to bronchodilators
Assess variability
Table 3
function and airway dynamics are well assessed by this means,
as mild sedation only is required.
Rigid endoscopy provides less functional assessment as anaes-
thesia is required. However, it is required for removal of foreign
bodies, biopsy, and occasionally ventilation. It also allows the
posterior glottis, subglottis, and trachea to be visualized. Vocal
cord function can be assessed on waking.
Any acute illness in a child with chronic airways obstruc-
tion may lead to severe complications. Transudative pulmonary
oedema has been well documented in association with upper-
airway obstruction, due to the markedly negative intrathoracic
pressures generated.
Screening for complications
Severe, chronic upper-airway obstruction of any aetiology can
cause failure to thrive, feeding difficulties, pulmonary hyperten-
sion, and eventually cor pulmonale. All children should there-
PAEDIATRICS AND CHILD HEALTH 19:6
Disadvantages
Foreign body will be visualized only if radio-opaque
Changes associated with epiglottitis/croup are subjective
Risk if child is not carefully supervised during taking of radiograph
Two-dimensional imaging only (c.f. CT/MRI)
Foreign body not seen if radiolucent
May be normal with vascular ring
Involves radiation
Oesophageal compression visualized, not surrounding structures
Poor visualization of trachea in the long axis
Radiation involved (c.f. MRI)
Differentiation of soft tissues/vascular structures less good than MRI
Long duration of examination
Sedation frequently required
Spatial resolution less good than CT
Expensive
Requires experienced reader
Not possible to take biopsies/remove foreign bodies
Less good for visualizing subglottis/trachea (c.f. rigid endoscopy)
Requires anaesthesia
Vocal cord movement/airway dynamics can only be assessed on waking
Increased incidence of mechanical trauma (c.f. flexible endoscopy)
Requires cooperation
fore have their height and weight plotted and an ECG should be
­performed. ◆
Further reading
Cinnamond MJ. Congenital disorders of the larynx, trachea and
bronchi. In: Scott-Brown, ed. Paediatric otolaryngology. London:
Butterworths, 1997.
Daniel SJ. The upper airway: congenital malformations. Paediatr Respir
Rev 2006; 7S: S260–3.
Denoyelle FM, Mondain M, Gresillon N, et al. Failures and complications
of supraglottoplasty in children. Arch Otolaryngol Head Neck Surg
2003; 129: 1077–80.
Dinwiddie R. Congenital upper airway obstruction. Paediatr Respir Rev
2004; 5: 17–24.
Eber E. Evaluation of the upper airway. Paediatr Respir Rev 2004; 5:
9–16.
280 © 2009 Published by Elsevier Ltd.
 Symposium: respiratory medicine
Hammer J. Acquired upper airway obstruction. Paediatr Respir Rev
2004; 5: 25–33.
Isaacson G. Congenital anomalies of the nose. Also available at: www.
uptodate.com, 2008 (accessed 2 Feb 2009).
Kilham HA, McEniery JA. Acute upper airway obstruction. Curr Paediatr
1991; 1: 17–25.
Loftis L. Acute infectious upper airway obstructions in children. Semin
Pediatr Infect Dis 2006; 17: 5–10.
Loftis L. Emergent evaluation of acute upper airway obstruction in children.
Also available at: www.uptodate.com, 2006 (accessed 2 Feb 2009).
Meissner H-H, Robinson L, Dunbinett SM, et al. Pulmonary edema as
a result of chronic upper airway obstruction. Respir Med 1998; 92:
1174–6.
Quintero D, Fakhoury K. Assessment of stridor in children. Also
available at: www.uptodate.com, 2007 (accessed 2 Feb 2009).
Rutter M. Evaluation and management of upper airway disorders in
children. Semin Pediatr Surg 2006; 15: 116–23.
Trachsel D, Hammer J. Indications for tracheostomy in children. Paediatr
Respir Rev 2006; 7: 162–8.
PAEDIATRICS AND CHILD HEALTH 19:6
Practice points
• Careful history and examination alone frequently reveal the
diagnosis
• Auscultation is particularly informative
• Children with acute, severe, or rapidly progressive obstruction
need to be identified immediately so that immediate
management can be instituted
• Consider upper-airway obstruction in cases of unexpected
neonatal respiratory distress
• Height and weight should be documented, and an ECG
performed in cases of chronic obstruction
• Spirometry is a useful tool in children able to cooperate
• MRI is largely replacing CT as a second-line tool in evaluating
chronic obstruction
• Endoscopy remains the most important tool for assessing
stridor in children
281 © 2009 Published by Elsevier Ltd.
 Occasional review
When should CT be
performed in children with
lung disease?
Kate M Park
Catherine M Owens
Abstract
Computed tomography (CT) can provide invaluable information in many
paediatric chest diseases and is the most sensitive way of imaging the
lungs due to its high spatial resolution. With the advent of multidetec-
tor CT, high-speed isotropic imaging allows superb visualization of the
tracheobronchial tree and, following the administration of intravenous
contrast, of the pulmonary and systemic circulation, allowing even small
vessels to be visualized and in some cases negating the need for con-
ventional (more invasive) diagnostic angiography. However, the radiation
burden delivered by CT is the largest of any imaging modality, and it
is well established that the lifetime cancer mortality risk as a result of
radiation exposure is higher in children than in adults. It is essential,
therefore, that indications for CT are restricted to those that are of real
diagnostic benefit, and that the radiation dose is kept to a minimum. We
discuss the indications for CT in paediatric respiratory disease and the
optimal use of the different types of CT scan available.
Keywords lung diseases; paediatrics; tomography, x-ray computed
Introduction
Computed tomography (CT) can provide invaluable information
in many paediatric chest diseases and is the most sensitive way
of imaging the lungs due to its high spatial resolution. With the
advent of multidetector CT (MDCT), high-speed isotropic imag-
ing allows superb visualization of the tracheobronchial tree and,
following the administration of intravenous contrast, of the pul-
monary and systemic circulation, allowing even small vessels
to be visualized, and negating the need for angiogram in some
cases.
However, the radiation burden delivered by CT is the larg-
est of any imaging modality, and it is well established that the
lifetime cancer mortality risk as a result of radiation exposure is
higher in children than in adults. It is essential, therefore, that
indications for CT are restricted to those that are of real diagnos-
tic benefit, and that radiation dose is kept to a minimum.
Kate M Park BA BM BCh MRCP FRCR is a Radiology Fellow at the Department
of Imaging, Great Ormond Street Hospital for Children NHS Trust,
London, UK.
Catherine M Owens BSc MBBS MRCP FRCR is a Consultant Radiologist at the
Department of Imaging, Great Ormond Street Hospital for Children NHS
Trust, London, UK.
PAEDIATRICS AND CHILD HEALTH 19:6
Types of CT scan
Multidetector CT, the standard ‘volume’ scan
Using MDCT, a three-dimensional or isotropic set of data (in
which the resolution is equal in all three dimensions) of the
entire chest is acquired. This allows reconstruction of the images
in all planes, and is used for the majority of indications. Acquisi-
tion time is fast due to the simultaneous scanning of multiple
slices, which decreases the risk of movement artefact and hence
the need for sedation. The standard thicker collimated volumet-
ric scan (if more than 1.2 mm) does not, however, provide bona
fide high-resolution images of the lung parenchyma such as
those acquired in a traditional non-contiguous incremental high-
resolution CT (HRCT) e.g. scans performed at 1-mm thickness at
10-mm intervals.
MDCT – the ‘combiscan’
Again, an isotropic data set is obtained, but with narrower tube
collimation – that is, a narrower beam of radiation (e.g. 0.75–1
mm, compared to 1.5 mm for the volume scan on a 16-row scan-
ner) – in order to improve fine detail. There is a small increase
in radiation dose with the narrower collimation, but the greater
detail can be particularly useful for investigating short steno-
ses, small airways problems, and cardiovascular anomalies. A
CT angiogram can be performed as a variation of this scan for
assessment of the pulmonary and/or systemic circulation.
Post-processing of the acquired data
The dataset obtained from an MDCT scanner is isotropic and
can therefore be manipulated in three dimensions with the data
displayed to optimize visualization of a particular structure or
pathology. The images can be viewed in the conventional axial
plane, sagittal or coronal planes, or reconstructed to produce
three-dimensional or volumetric images. The latter two require
more time and post-processing skills and are therefore not used
in all cases, but can be particularly helpful, for example, in pre-
surgical planning, depicting anatomical structures and their rela-
tionships in a more ‘life-like’ fashion (Figure 1).
High-resolution computed tomography
In HRCT, 1-mm slices are obtained throughout the chest every
10–20 mm, and therefore three-dimensional reconstruction is not
possible as the dataset is not contiguous. High-resolution images
of the lung parenchyma are obtained at only a third of the radia-
tion dose of a conventional spiral CT. These images are therefore
ideal for follow-up of diffuse interstitial lung disease or airways
disease. In many centres, the initial diagnostic scan would be a
volumetric acquisition to exclude any associated pathologies such
as lymphadenopathy, mediastinal lesions or vascular ­anomalies.
Indications for CT
A referral for CT may be made on the basis of chronic symp-
toms, such as cough, wheeze, stridor or recurrent infections;
on the basis of unexpected or unexplained findings, such as a
mass/nodule or suspected lymphadenopathy on chest x-ray or
on another imaging modality such as an antenatal ultrasound
scan; or for follow-up of a known condition or following trauma.
Patients’ symptoms or imaging findings found on other imaging
282 © 2009 Elsevier Ltd. All rights reserved.
 Occasional review

a SVC Trachea

Right aortic arch Left aortic arch

b Right aortic arch Compressed trachea

Carina Left aortic arch

c Tracheal stenosis

d Left pulmonary Left aortic

artery arch

Right aortic arch

Right pulmonary
artery

Heart

Descending aorta

Figure 1 Computed tomography (CT) scan of a balanced double aortic arch causing tracheal stenosis: a axial maximum intensity projection (MIP),
b coronal MIP, c CT scan volume rendered for air, and d volume rendered for contrast.

PAEDIATRICS AND CHILD HEALTH 19:6 283 © 2009 Elsevier Ltd. All rights reserved.
 Occasional review
modalities may be suggestive of a possible pathology and direct
the radiologist in terms of which CT protocol to use.
Ideally the protocol used for imaging the chest should provide the
most relevant information at the lowest radiation burden. Factors
such as the patient’s age, allergy history, ability to lie still/breath-
hold,and need for contrast must be taken into account,and the
question of whether CT is the most appropriate modality to answer
the clinical question should always be addressed (Figure 2).
Indications for a ‘standard’ volume scan
Suspected upper airway stenosis: presentation of tracheal or
main bronchial narrowing is typically with stridor, but can pres-
ent with a seal-like cough or recurrent infections. Classically,
inspiratory stridor is thought to suggest extrathoracic upper-
­airways problems and biphasic (inspiratory and expiratory)
stridor intrathoracic narrowing. Expiratory stridor or wheeze
suggests lower-airway narrowing. Tracheal stenosis is most com-
monly secondary to extrinsic obstruction, and in older children
dysphagia may be a coexisting feature.
CT can often identify the cause of the disorder, delineate the
anatomical relationships of the airways to the major vessels,and
identify any more distal sequelae of the disorder or associated
abnormalities.
Extrinsic compression – in the neonate and infant, extrinsic
compression is most commonly secondary to vascular anoma-
lies, and when suspected should be imaged using a relatively
thin collimation (combiscan or CT angiogram) protocol which
will be discussed below.
The more common non-vascular causes of extrinsic compression
include lymphoma, bronchogenic cyst (Figure 3), oesophageal dupli-
cation cyst, mediastinal teratoma, lymphadenopathy, thyroid and
thymic lesions and lymphatic and venous ­malformations.
Intrinsic compression – tracheal stenoses can occur congeni-
tally and may be short- or long-segment resulting from the pres-
ence of focal or diffuse complete tracheal cartilage rings. They
are often associated with cardiac defects, particularly pulmonary
artery sling (Figure 4). Post-intubation stenosis develops in 0.7–
Is CT the best modality
Yes
What is the indication?
• Strictures • Cardiovascular anomalies
• Tumour masses • Short focal tracheobronchial stenoses
• Metastases
• Tracheomalacia
Routine Combiscan/
volume scan CT angiogram
Figure 2 Indications for the different types of computed tomography (CT) scan.
PAEDIATRICS AND CHILD HEALTH 19:6 284
8% of patients in whom prolonged endotracheal intubation is
required. Stenoses can also occur following trauma or ingestion
of caustic substances. Intrinsic tracheal masses or inflammation
are rare but can be seen with subglottic haemangiomata, Wegen-
er’s granulomatosis and laryngotracheal papillomatosis.
Suspected tracheobronchomalacia: tracheobronchomalacia
(TBM) refers to a weakness of the central airway walls second-
ary to deficient elastic or cartilaginous components, resulting
in expiratory airway collapse. It can also occur as a secondary
phenomenon in association with tracheo-oesphageal fistula, fol-
lowing prolonged intubation or infective tracheobronchiolitis, or
secondary to extrinsic compression. Due to its dynamic nature,
the diagnosis is usually made at traditional bronchography.
Although not commonly used in practice due to relatively high
radiation burden, in principle, dynamic (inspiratory and expira-
tory) CT can be used for diagnosis, the airways collapse becom-
ing apparent in the expiratory phase. CT can, however, be very
useful in identifying extrinsic causes of TBM.
Pulmonary masses: solitary masses and mediastinal lesions –
solitary pulmonary masses in children are more frequently benign
or developmental in origin than they are in adults. They may be
picked up incidentally on imaging, either antenatally or in child-
hood, and require further evaluation, or the child may present
with symptoms which require further investigation.
Masses identified at antenatal screening are most commonly
bronchopulmonary foregut malformations which include congen-
ital cystic adenomatoid/hamartomatous malformation (CCAM)
(Figure 5), pulmonary sequestration (PS), and bronchogenic cysts.
In some cases, the mass is seen to regress during the antenatal
period, with little or no abnormality identifiable in the neonate.
Those that persist will require further imaging, in the majority
of cases with CT. The depiction of the systemic (aortic) feeding
vessel is important for the definitive diagnosis of sequestration or
hybrid lesions (a mixture of sequestration and CCAM) and there-
fore a combiscan or CT angiogram is advised (discussed below).
No
Consider
• CXR
• Ultrasound
• VQ
• Interstitial lung disease
• Oesophogram
• Airways disease
• Bronchogram
• MRI
Initial scan: Combiscan
Follow-up: HRCT
© 2009 Elsevier Ltd. All rights reserved.
 Occasional review

Figure 3 Bronchogenic cyst

causing compression of the
left main bronchus. a Chest
radiograph of a left-sided
mediastinal mass with
associated hyperlucency of
the left hemithorax secondary
to air-trapping. b Coronal
maximum intensity projection
(MIP) demonstrating the
compression of the left main
bronchus.

Masses identified later in life may be found incidentally, pres-
ent with recurrent symptoms such as recurrent chest infections,
or present with persistent changes on imaging, such as lobar
collapse or persistent mass lesion. There is a wide spectrum
of pathologies that can present as a solitary mass ranging from
bronchopulmonary foregut malformations to sequelae of infec-
tion (lung abscess, loculated effusion) to the less common pri-
mary thoracic neoplasms (Figure 6) or solitary metastases.
The use of intravenous (IV) contrast material is highly recom-
mended in the imaging of solitary pulmonary and anterior or
middle mediastinal masses. Scanning during the time of peak
contrast enhancement permits optimal definition of anatomic

a
Apparent
narrowing
secondary to
endotracheal
tube

Long segment
tracheal
stenosis

Figure 4 a Computed
Narrow
trachea tomography (CT) scan volume
rendered for air, demonstrating
a long-segment tracheal
Pulmonary stenosis (between the two
sling arrows). b Axial post-contrast CT
maximum intensity projection
(MIP) demonstrating an
associated pulmonary sling.

PAEDIATRICS AND CHILD HEALTH 19:6 285 © 2009 Elsevier Ltd. All rights reserved.
 Occasional review

Neuroblastoma. a Chest radiograph demonstrates a well-defined,

right-sided, apical mass. b Axial computed tomography (CT) of
the mass.

Figure 6

Right lower lobe cystic adenomatoid malformation (CCAM)

identified on antenatal scanning seen on a chest radiograph,
and b axial computed tomography (CT) scan on lung window
settings.
Figure 5
features, mediastinal vasculature and lymphadenopathy, which
is of particular importance in children, who lack intrinsic ­contrast
differences due to their paucity of mediastinal fat. For posterior
mediastinal masses, magnetic resonance imaging (MRI) is the
modality of choice as it allows assessment of any intraspinal
involvement and does not involve radiation.
Multiple pulmonary masses – a number of pathologies can
lead to multiple pulmonary masses requiring further investiga-
tion. These include septic emboli, granulomatous processes,
­ ulmonary metastases, and multiple arteriovenous malforma-
p
tions. Helical CT is excellent at depicting pulmonary nodules and
has been shown to accurately detect the nodules even in children
unable to breath-hold who must be scanned during quiet respira-
tion. The sagittal and coronal reconstructions are invaluable in
differentiating vessels or vascular lesions from nodules, which
can de more difficult in the axial plane. It can also clarify the
spatial relationships of a nodule to the pleura or diaphragm and
depict cavitation not demonstrated on CXR. IV contrast is not
usually required. When screening for metastases as a staging
procedure, low-dose helical CT can be used as a highly sensitive
alternative to conventional-dose helical CT.
Other: a contrast-enhanced volume scan may also be of ben-
efit in investigation of pulmonary and pleural infections in the

Enlarged left main

pulmonary artery
Enlarged right main
pulmonary artery
Figure 7 Computed tomography
Aorta (CT) scan volume rendered
Small ’pruned’ for contrast in a patient with
peripheral vessels pulmonary hypertension
demonstrates classic enlargement
Heart Vertebral column of the main pulmonary arteries
with pruning peripherally.

PAEDIATRICS AND CHILD HEALTH 19:6 286 © 2009 Elsevier Ltd. All rights reserved.
 Occasional review
a Chest radiograph, and b minimum intensity projection (minIP)
of an unusual case of congenital lobar overinflation affecting
both the left upper and right middle lobes with compression of
the unaffected lobes.
Figure 8
Intralobar sequestration
PAEDIATRICS AND CHILD HEALTH 19:6
immunocompetent (if the clinical response is not as expected
with therapy) and immunocompromised child, and also in chest
trauma.
Indications for a combiscan or CT angiogram
Cardiovascular anomalies: MRI is the technique of choice for
imaging congenital large-vessel anomalies and other vascular
anomalies, but in patients in whom a long MRI study is inadvis-
able or who have combined vascular and airways abnormali-
ties, CT is invaluable with its high-speed acquisition time and
low sedation rates. High-quality vascular imaging can be used
to analyse vascular abnormalities of the pulmonary arteries and
veins, aortic arch and great vessels (double-aortic arch, pulmo-
nary sling etc) (Figures 1 and 7) as well as congenital lung mal-
formations (CCAM, PS, congenital lobar overinflation, scimitar
syndrome) in which depiction of the systemic and/or pulmonary
vasculature is important in the definitive diagnosis and manage-
ment (Figures 8 and 9). Post-processing techniques are of par-
ticular use in analysing the vascular anatomy and can provide
information that is superior to conventional angiograms due to
the three-dimensional nature of the images with clear demonstra-
tion of the adjacent anatomical structures. Helical CT angiocar-
diography with three-dimensional reconstructions is superior to
echocardiography for the non-invasive assessment of pulmonary
artery anatomy, and is equal to angiography in patients with
complex congenital heart disease, with the advantages of lower
patient morbidity and reductions in cost and time.
A contrast-enhanced MDCT scan can also be invaluable in
the definition of surgical shunts and postoperative vascular anat-
omy, as well as in suspected central pulmonary thromboemboli
(Figure 10).
Short focal tracheobronchial stenoses: the CT imaging of tra-
cheobronchial stenoses have been discussed above; however,
in the case of short focal stenoses, it is important to exclude
vascular compression as a cause of the stenosis and therefore
a combiscan or CT angiogram is advised. This is particularly
important in neonates and infants, in whom extrinsic compres-
sion is most commonly secondary to vascular anomalies. The
most common are double aortic arch, right-sided aortic arch with
left ligamentum arteriosus, innominate artery compression syn-
drome, and pulmonary artery sling. Chest x-ray or ECHO may
be suggestive of an abnormality, and an initial screening upper
Heart
Arterial supply
from abdominal
aorta
Figure 9 Computed tomography
Aorta
(CT) scan volume rendered
Vertebral column for contrast demonstrates a
right lower lobe intralobar
sequestration with arterial supply
from the abdominal aorta.
287 © 2009 Elsevier Ltd. All rights reserved.
 Occasional review

Figure 10 Axial maximum intensity

projection (MIP) computed
tomography (CT) demonstrating a
thrombosed left pulmonary artery
with metallic stent post Glenn
procedure (shunt from superior vena
cava to pulmonary arteries).

gastrointestinal (UGI) study or oesophagram is often performed

in suspected cases (Figure 11). In older children, an UGI study
performed for other reasons may alert the radiologist to a previ-
ously undiagnosed vascular anomaly. In the case of a suggestive
UGI study, a combiscan is advised which allows visualization of
the thoracic vasculature, tracheobronchial tree, and any more
distal sequelae.

Other congenital pulmonary anomalies: pulmonary underde-

velopment describes a spectrum of anomalies ranging from com-
plete absence of the bronchus and lung (agenesis) to bronchial
hypoplasia with a variable reduction of lung tissue (hypoplasia).
The exact aetiology of pulmonary agenesis is uncertain, but pul-
monary hypoplasia is caused by factors which either directly
or indirectly compromise the thoracic space available for lung
growth, or compromise the pulmonary vascular perfusion. Chest
x-ray depicts either complete opacification of a hemithorax (in
agenesis) or decreased aeration of the affected hemithorax (hypo-
plasia) with a small thoracic cage and compensatory hyperinfla-
tion and herniation of the contralateral normal lung across the
midline. Conventional MR with angiography is again the imaging
modality of choice, but in patients in whom a long MRI study is
inadvisable or in whom depiction of the airways abnormalities is
of importance, contrast enhanced CT is invaluable.

Diagnostic scan for interstitial lung disease and airways

­disease: both in airways disease (AD) and in diffuse interstitial
lung disease (DILD) detailed visualization of the lung paren-
chyma can be provided by non-contiguous acquisition HRCT
at a third of the radiation burden of a conventional volumetric
CT. However, in complex cases the initial diagnostic scan may
require a volumetric acquisition with contrast to exclude any
associated pathologies such as lymphadenopathy, mediastinal
lesions, or vascular anomalies.
The data obtained can be reconstructed with a high-spatial-
resolution algorithm and displayed with a wide window setting
so that the lungs can be visualized in a more ‘HRCT-like’ view.
Follow-up with HRCT is then advised to reduce radiation dose in
surveillance imaging (see below).

Indications for high-resolution CT scan

Figure 11 Lateral image from an upper gastrointestinal series The use of a low-dose (50 mA, 0.75 s) limited (1-mm slices every
demonstrating posterior indentation of the oesophagus secondary to a 15–20 mm) HRCT in inspiration with three supplementary expi-
right aortic arch with aberrant left subclavian artery. ratory scans allows accurate assessment of the presence and

PAEDIATRICS AND CHILD HEALTH 19:6 288 © 2009 Elsevier Ltd. All rights reserved.
 Occasional review
extent of lung disease at a dose equivalent to approximately 10
chest radiographs and at a third of the dose of a conventional
spiral CT. If the child is unable to breath-hold, the scans can be
performed during quiet breathing and decubitus scans used to
replace expiratory scans (the dependent lung behaving as the
‘expiratory lung’), and good-quality images of the non-­dependent
lung in inspiration can be useful.
High-resolution computed tomography (HRCT) scan
demonstrating a bronchial dilatation and bronchial wall
thickening in a child with cystic fibrosis. b Inspiratory and
c expiratory views in the same child show a mosaic pattern on
expiration indicating small-airways disease and air-trapping.
Figure 12
PAEDIATRICS AND CHILD HEALTH 19:6
These images are ideal for follow-up of diffuse interstitial lung
disease or airways disease.
Airways disease: small-airways disease (inflammatory dam-
age to the small airways) can occur secondary to a number of
conditions, and in children is most frequently seen following
a viral infection, particularly adenovirus types 7, 14 and 27. It
is also common in asthma, recurrent aspiration, cystic fibrosis
and graft-versus-host disease after organ transplantation. HRCT
signs of small-airways disease include patchy areas of decreased
parenchymal attenuation (which gives rise to a ‘mosaic’ attenu-
ation pattern), pulmonary vascular abnormalities, bronchial
abnormalities, and air-trapping on expiratory CT (Figure 12).
HRCT surveillance in paediatric cystic fibrosis patients is
established in some centres but remains controversial. A recent
study by Owens et al suggests that a lung clearance index from
multiple breath washouts, a more sensitive measure of pulmo-
nary function than spirometry, detects pulmonary abnormalities
with a similar frequency to CT and may be a safer non-radiation-
burdening alternative (or complementary) means of surveillance
in the future.
Interstitial lung disease: interstitial lung disease (ILD) is rare
in the paediatric population and represents a heterogeneous
group of disorders characterized by restrictive lung function and
impaired gas exchange. The chest x-ray is often non-specific, and
HRCT has been shown in both adults and children to increase
the accuracy at diagnosis of diffuse lung disease when compared
with chest radiographs.
Paediatric ILD occurs on a background of developing lungs
and a developing immune system, and often differs markedly
from adult ILD in terms of presentation, clinical features and
progress. At present, the role of HRCT in paediatric ILD is evolv-
ing and has a first-choice diagnostic accuracy of approximately
50–60%. Diseases that are correctly diagnosed on HRCT with a
high degree of confidence include alveolar proteinosis, pulmo-
nary lymphangiectasia, and idiopathic haemosiderosis. Differ-
entiation between non-specific interstitial pneumonitis (NSIP),
desquamative interstitial pneumonitis (DIP), and lymphocytic
interstitial pneumonitis (LIP) has proved more difficult.
The role of CT in empyema/pleural effusion
Chest ultrasound (US) is the initial investigation of choice in
suspected pleural effusion or empyema. There is no radiation
involved, and due to the high spatial and contrast resolution of
ultrasound, it is more sensitive than CT for depicting septations
within an effusion. US can also differentiate ‘complicated’ effu-
sions (those with loculations and/or echogenic fluid) from ‘simple’
effusions (those which are anechoic and unseptated), which may
impact on management. However, it is important to note that in
the setting of a parapneumonic effusion, no imaging modality can
accurately differentiate between infected and reactive effusions,
as even pus can appear anechoic on ultrasound. Empyema can-
not be excluded on the basis of imaging, and a clinical decision
regarding management – such as drainage – is required. Studies
suggest that CT is no better at differentiating ‘simple’ effusion
from empyema with pleural enhancement seen in almost all cases
of parapneumonic effusion, infected or not. CT should therefore
289 © 2009 Elsevier Ltd. All rights reserved.
 Occasional review
be reserved for the more complicated cases that do not respond to
conventional treatment, or in the immunocompromised host.
Conclusion
Careful teamwork between respiratory physicians, surgeons and
radiologists is paramount in choosing the correct imaging path-
ways for children. It is important always to consider the risk/ben-
efit ratio whenever a child undergoes any examination requiring
the use of ionizing radiation, but most especially CT. ◆
Further reading
Berrocal T, Madrid C, Novo S, et al. Congenital anomalies of the
tracheobronchial tree, lung, and mediastinum: embryology,
radiology, and pathology. Radiographics 2004; 24: e17.
Brink JA, Heiken JP, Semenkovich J, et al. Abnormalities of the
diaphragm and adjacent structures: findings on multiplanar spiral
CT scans. AJR Am J Roentgenol 1994; 163: 307–10.
Coakley FV, Cohen MD, Waters DJ, et al. Detection of pulmonary
metastases with pathological correlation: effect of breathing on the
accuracy of spiral CT. Pediatr Radiol 1997; 27: 576–9.
Copley SJ, Padley SP. High-resolution CT of paediatric lung disease. Eur
Radiol 2001; 11: 2564–75.
Gartenschlager M, Schweden F, Gast K, et al. Pulmonary nodules:
detection with low-dose vs conventional-dose spiral CT. Eur Radiol
1998; 8: 609–14.
Ghersin E, Khoury A, Litmanovich D, et al. Comprehensive multidetector
computed tomography assessment of severe cardiac contusion in a
pediatric patient: correlation with echocardiography. J Comput Assist
Tomogr 2005; 29: 739–41.
Jaffe A, Calder AD, Owens CM, et al. Role of routine computed
tomography in paediatric pleural empyema. Thorax 2008; 63:
897–902.
Katz M, Konen E, Rozenman J, et al. Spiral CT and 3D image
reconstruction of vascular rings and associated tracheobronchial
anomalies. J Comput Assist Tomogr 1995; 19: 564–8.
Owens CM. Radiology of diffuse interstital pulmonary disease in
children. Eur Radiol 2004; 14(Suppl 4): L2–12.
PAEDIATRICS AND CHILD HEALTH 19:6
Ramachandran N, Owens CM. Imaging of the airways with multidetector
row computed tomography. Paediatr Respir Rev 2008; 9: 9–76.
Remy-Jardin M, Remy J, Wattinne L, Giraud F. Central pulmonary
thromboembolism: diagnosis with spiral volumetric CT with
the single-breath-hold technique – comparison with pulmonary
angiography. Radiology 1992; 185: 381–7.
Riccabona M. Ultrasound of the chest in children (mediastinum
excluded). Eur Radiol 2008; 18: 390–9.
Turner A, Gavel G, Coutts J. Vascular rings – presentation, investigation
and outcome. Eur J Pediatr 2005; 164: 266–70.
Ventilation inhomogeneity, gas trapping and HRCT findings in young
children. ESPR, 2008.
Westra SJ, Hill JA, Alejos JC, et al. Three-dimensional helical CT of
pulmonary arteries in infants and children with congenital heart
disease. AJR Am J Roentgenol 1999; 173: 109–15.
Wiel E, Vilette B, Darras JA, et al. Laryngotracheal stenosis in children
after intubation. Report of five cases. Paediatr Anaesth 1997; 7:
415–9.
Practice points
• Computed tomography is the most sensitive way of
imaging the lungs and allows superb visualization of the
tracheobronchial tree and pulmonary and systemic circulation
• The radiation burden to the population, with the associated
increased lifetime cancer mortality, is the highest and most
significant contributor compared to any other imaging
modality
• Indications for use need to be restricted to those of real and
evidence-based diagnostic benefit, and optimal use of the
different types of CT protocol is essential
• Radiation dose must be kept to a minimum
• There should be careful teamwork between physicians,
surgeons and radiologists in choosing the optimal imaging
pathway for our children
290 © 2009 Elsevier Ltd. All rights reserved.
 Personal Practice
Non-specific isolated
persistent cough
Jenny Hughes
Michael D Shields
Abstract
Cough is a common problem in childhood and generates much anxiety
amongst parents. There are numerous different causes of cough in chil-
dren and this article provides the reader with a logical approach, using
a worked example, for its differential diagnosis. Non-specific isolated
persistent cough is discussed.
Keywords childhood; cough; diagnosis; management; persistent
Case history
Peter, a 4-year-old boy, has been referred to his local paediatri-
cian by his GP. His mum had requested his referral as she was
concerned about his cough. The letter reads as follows:
‘Please see this 4-year-old boy at your Outpatient Clinic. He
has had a persistent cough for 2 years. Asthma?
Key points in the history
How and when did the cough start?
An acute cough is defined as being of recent onset with a dura-
tion of less than 3 weeks. The majority of these children will
have a viral respiratory tract infection. In most cases this post-
viral cough will improve after 14 days; however, a minority will
take 3–4 weeks to resolve. A chronic cough is defined as one
lasting more than 8 weeks. A grey area thus exists between 3 and
8 weeks which can be termed a subacute cough. Many children
who present with a cough of duration within this period (i.e.
3–8 weeks) will be recovering from a viral infection or may be
suffering from a pertussive-like syndrome. Careful observation
with a wait-and-see approach is usually appropriate. However,
if during this time the cough becomes more severe in frequency
and intensity, further investigations are warranted.
Onset of cough in the neonatal period usually indicates signifi-
cant disease and must be investigated further.
A history of very sudden onset of cough after an episode of
choking may indicate an inhaled foreign body.
Jenny Hughes MB BCH BSc MRCPCH is a Paediatric Specialist Registrar at
the Department of Child Health, Queen’s University Belfast, Clinical
Institute, Grosvenor Road, Belfast BT12 6BJ, UK.
Michael D Shields MB ChB MD FRCP FRCPCH is a Professor of Child Health at
the Department of Child Health, Queen’s University of Belfast, Clinical
Institute, Belfast, UK.
PAEDIATRICS AND CHILD HEALTH 19:6
What is the nature of the cough?
It is important to determine whether the cough is productive or
dry. A wet- or moist-sounding cough, which can be accurately
reported by parents, needs further investigation and implies
either an increase in airway secretions or abnormalities in airway
clearance mechanisms. This can be difficult to assess in children
less than 5 years of age as they generally are unable to expecto-
rate and tend to swallow their sputum.
Paroxysmal spasmodic cough with or without an inspiratory
‘whoop’ is a major feature of the pertussis syndromes. Post-
­tussive vomiting often occurs. This should be considered in any
child coughing for more than 2 weeks, even if the child has been
immunized.
Is the cough an isolated symptom?
An impression of the general health of the child during the
coughing period is important. If the cough has been associated
with night sweats or weight loss, then tuberculosis must be
excluded.
Cough associated with wheezing or breathlessness may indi-
cate the onset of asthma, particularly if there is a history of
other atopic conditions (e.g. eczema, food allergies or allergic
­rhinitis).
If the child is otherwise well, a diagnosis of non-specific iso-
lated persistent cough should be considered.
What triggers the cough?
Trigger factors are important to identify. Throughout the consul-
tation the presence of the cough should be listened for; a sudden
spasm of cough occurring when ‘cough’ is mentioned can raise
suspicions of a psychogenic cough. This may also be suggested
in the history if there is an increased frequency with increased
attention or disappearance with sleep. Exacerbation with exer-
cise, cold air or early morning would suggest asthma, especially
if there is no current viral upper respiratory tract infection. A
postprandial cough suggests aspiration.
Is there a family history of respiratory symptoms, disorders
and atopy?
There is an increased likelihood of asthma if there is a strong
family history in a first-degree relative (parents or siblings) of
asthma, eczema or hay fever.
If there is a relative with tuberculosis then this must be
excluded.
What medications are the child on, what treatments have the
child had for the cough and what effect have they had on the
cough frequency and severity?
It is important to collate a complete list of treatment trials and
their duration. Frequently, an insufficient dosage of asthma
medication is prescribed to give a definitive answer of a good
response, or alternatively the outcome of the trial has not been
documented. Consideration must also be given to the level of
compliance with any trials of medication.
If the child has been prescribed an ACE inhibitor for an under-
lying cardiac condition then this may be causing the cough.
Does the cough disappear when asleep?
This is suggestive of a psychogenic cough.
291 © 2009 Elsevier Ltd. All rights reserved.
 Personal Practice
Does the child smoke cigarettes or are they exposed to envi-
ronmental smoke?
Smoking is well recognized as an irritant to the airways. Be aware
of the adolescent who may be smoking.
Findings from the history
Peters’ cough began 2 years ago. It was initially very severe
and harsh in nature, and was associated with a runny nose and
high temperature. It did not occur in spasms, but was associ-
ated with vomiting on occasions. Peters’ mum remembers some
coughing at night. Peter was given a course of amoxicillin which
he took for 4 days, but it did not make any difference to the
cough. This severe cough lasted for 3–4 weeks before resolv-
ing; however, since that time Peter has continued to have a mild
nocturnal cough, and during the day the cough is worse with
exercise and can limit his activity. Peters’ mum feels that he has
never wheezed. He has been given a salbutamol inhaler with
aerochamber over the last few months, but this has not helped
the cough.
Within the family, Peter has two siblings – 1 brother aged 9
years and a sister aged 4 years. Neither sibling has a history of
cough. His elder brother suffered from mild eczema as a baby,
but this is now resolved. Peter’s father suffers from hay fever and
mild asthma.
Peters’ mum is concerned that this may be asthma, and also
that his immune system is poor given this persistent cough.
Examination
All children with a cough should have a full clinical systematic
examination. In particular:
• growth and nutrition, including plotting growth percentiles,
should be assessed
• evidence of finger-clubbing and chest shape (Harrison’s sulci,
overinflation) should be specifically looked for
• evidence of clinical signs of allergy – such as eczema, hori-
zontal nasal crease, allergic shiners, nasal speech – should be
sought
• ENT examination, including the external auditory meatus for
the presence of a foreign body, should be performed
• throughout the consultation the character of the cough should
be observed; the child should be asked to cough or ‘huff’ as
part of the assessment.
Examination findings
Peter is appropriately grown with weight and height on the 50th
percentile. Auscultation of the chest is clear. There are no clinical
signs of atopy. ENT examination is normal.
Investigations
• Chest x-ray is indicated for most children with a history of
chronic cough (more than 8 weeks).
• To assess lung function, spirometry – with or without a test
of bronchial hyper-responsiveness – should be attempted in all
children over the age of 5 years.
• Specific IgE (or skin-prick tests) to inhalant allergens (not
food allergens) – e.g. house-dust mite, grass and tree pollen, cat
PAEDIATRICS AND CHILD HEALTH 19:6
and dog – can help identify the atopic individual (making cough-
variant asthma more likely).
• Observation of the cough and a sample of sputum should be
obtained. An experienced physiotherapist can help, particularly
with young children. The sputum sample should be sent for cul-
ture and sensitivity, virology and differential cytology.
Management
Given Peter’s history of nocturnal cough and exercise-related
symptoms, as well as the paternal history of asthma and the ma-
ternal concern, a trial of asthma medication would be appropri-
ate in this case. Peter is prescribed the following:
• budesonide 400 μg bd via metered dose inhaler (MDI) and
aerochamber
• monteleukast 4 mg once daily
• salbutamol up to 10 puffs prescribed as needed (PRN) via MDI
aerochamber.
A sweat test is requested and blood taken for immunoglobu-
lins and IgG subclasses and total IgE.
CXR is reported as normal.
Outpatient clinic review
Peter is reviewed at outpatient clinic 6 weeks later. His mum
feels that the inhalers may have helped a bit but she is not sure.
Action plan: given the lack of definite improvement with the
asthma medication it would be appropriate to stop them and
assess the response.
Second outpatient clinic review
Peter is again reviewed at outpatient clinic 8 weeks later. His
Mum feels that his cough has not deteriorated since withdrawing
the inhalers. If anything the cough might be a bit better.
At this time Peters’ blood test results are available and show
normal levels of Immunoglobulins and IgG subclasses, as well as
a normal total IgE. His sweat test is also reported as normal.
Peters’ mum is advised of a diagnosis of non-specific isolated
persistent cough and is reassured that in time this will improve.
No further treatment is prescribed.
Non-specific isolated persistent cough
This is typically found in an otherwise well child with no associ-
ated wheezing or sputum production or ‘alert signs’ of serious
pathology (Table 1). Evidence to date suggests that the ­majority
Red flag alert signs in the history
• Neonatal onset
• Cough with feeding
• Sudden onset cough
• Chronic moist cough with phlem production
• Associated night sweats/weight loss
• Continuous unremitting or worsening cough
• Signs of chronic lung disease
Table 1
292 © 2009 Elsevier Ltd. All rights reserved.
 Personal Practice
of these children do NOT have ‘cough-variant asthma’. They typ-
ically do not have underlying eosinophilic airways inflammation,
have different preceding risk factors, and don’t respond well
to anti-asthma therapy. However, it can be difficult to be sure
that they do not have a cough-predominant asthma, and thus an
effective trial of anti-asthma therapy may be warranted. Results
of the trial should be documented and treatment stopped if there
has been no response. ◆
Further reading
Shields MD, Bush A, Everard ML, et al. On behalf of the British
Thoracic Society Cough Guideline Group. Recommendations for the
assessment and management of cough in children. Thorax 2008;
63: 1–15.
Practice points
• Cough starting in the neonatal period usually indicates
significant disease and must be further investigated
• In a child where the diagnosis of asthma cannot be excluded,
a trial of anti-asthma therapy may be used. This treatment
should be effectively delivered in adequate doses, with
documentation of clear-cut outcomes. A definite period of
PAEDIATRICS AND CHILD HEALTH 19:6
time should be set for review and the medication stopped.
If significant response has not occurred, then asthma is
unlikely, and the cough is unresponsive to asthma therapy.
If the cough has improved, this could be due to natural
resolution and a further relapse that responds to therapy
would then be suggestive of cough-variant asthma
• Acute cough (less than 3 weeks’ duration) is caused by a
viral infection in the majority of children
• Many children who present with a prolonged acute cough
(between 3 and 8 weeks duration) will be recovering from
a viral infection or may be suffering from a pertussive-like
syndrome. Careful observation is usually adequate; however,
deterioration should prompt further investigation
• Management of chronic cough (more than 8 weeks’ duration) is
dependent on diagnosis. Initial assessment includes history, full
clinical examination, and basic investigations including chest
radiograph, spirometry, sputum analysis and allergy testing
• A chronic wet productive cough should be investigated for
underlying conditions associated with persistent bronchial
infections (e.g. cystic fibrosis, immune deficiencies, ciliary
dyskinesia, persistent bacterial bronchitis)
• Non-specific isolated persistent cough can be used to
describe children who typically have a persistent cough, no
other respiratory symptoms, are otherwise well with no signs
of chronic lung disease, and have a normal chest radiograph
293 © 2009 Elsevier Ltd. All rights reserved.
 self-assessment

Self-assessment
Questions (a) What is the most likely diagnosis?
Irritable hip (reactive arthropathy)
Case 1 Juvenile idiopathic arthritis
A 3-year-old girl presented to the Emergency Department with Leukaemia
a 6-week history of a reluctance to walk, and a limp more Multifocal osteomyelitis
pronounced in the morning. Prior to onset of these symptoms Trauma
she had had 2 days of fever with some associated rhinorhoea (b) What investigation would you recommend?
and cough. Repeat bone scan
A bone scan done at another hospital a week earlier had been Joint aspiration/biopsy
reported as normal. There was no recent history of night sweats, Magnetic resonance imaging (MRI) of hips/knee joints with
weight loss or gastrointestinal symptoms. Treatment had included contrast
non-steroidal anti-inflammatory drugs (NSAIDs), with minimal Bone-marrow aspirate
improvement. Antibiotics had not been given at any stage. (c) What treatment would you recommend?
On examination she was afebrile, had an antalgic gait, and Corticosteroid joint injections
on hip examination had limited internal rotation. There was Intravenous methylprednisolone
also mild swelling and a decreased range of movement of Intravenous antibiotics
the left knee. There was no bony tenderness and no obvious Surgery
swelling or limitation of movement of other joints. There was
no lymphadenopathy, and the eye, cardiovascular, respiratory Case 2
and abdominal examination were normal. A 9-week-old boy was well on the day of his routine 2-month
scheduled vaccinations (diphtheria/tetanus/pertussis (acellu-
Investigations
lar)/haemophilus influenzae type b/polio/pneumococcus). He
• Full blood count was a little unsettled immediately post-immunization and was
Value Reference range put down for a sleep on returning home, now 2 hours post-
Haemoglobin 113  g/l 110–140  g/l vaccination. When his mother went back into the bedroom
Haematocrit 0.32 (L) 0.34–0.42 l 5 minutes later he was staring upwards and ‘she thought he
RCC 4.31  ×  1012/l 3.9–5.3  ×  1012/l was dead’, as he appeared very pale, unresponsive and floppy.
MCH 26.2  pg 24–30  pg She immediately picked him up; he was breathing, and slowly
MCV 74 L* fl 75–90  fl
came around over the next 5–10 minutes. An ambulance was
called and he was taken to hospital and monitored for 4 hours,
RDW 12.7% 11–14%
where he remained well. Over the next 24 hours he was more
Platelets 602  ×  109/l (H) 150–400  ×  109/l
sleepy than normal, but there were no further events.
WCC 13.2  ×  109/l 6.0–17.0  ×  109/l
Neut seg 6.69  ×  109/l 1.5–8.5  ×  109/l
Past medical history: he was born via a normal vaginal
Lymphocytes 5.77  ×  109/l 3.0–9.5  ×  109/l
delivery. At 1 month of age he had some recurrent vomiting
Monocytes 0.63  ×  109/l 0.1–1.0  ×  109/l consistent with gastro-oesophageal reflux, and settled with
Eosinophils 0.05  ×  109/l 0–0.8  ×  10e9/l thickened feeds. Developmentally he was smiling, interactive,
Basophils 0.05  ×  109/l 0–0.1  ×  109/l and had normal newborn hearing screening. There is no sig-
RCC, red cell count; MCH, mean corpuscular haemoglobin; MCV, mean corpus- nificant family history.
cular volume; RDW, red cell distribution width; WCC, white cell count; L, low;
H, high.
• Erythrocyte sedimentation rate (ESR) 140 mm/h Examination
(range 0–6 mm/h)
• Growth: weight = 5 kg (50th centile)
• C-reactive protein (CRP) 63 mg/l (<8 mg/l) • Head circumference = 41 cm (75th centile)
• Hip x-ray: normal • Length = 58 cm (75th centile)
• Hip ultrasound: thickened synovium bilaterally, nil joint On examination he was alert, smiling, had a number of Mon-
effusion golian blue spots on the buttocks, but no other neurocutaneous
stigmata. He had normal tone, power, and reflexes. Cardiovas-
cular, respiratory, and abdominal examinations were normal.
Nigel Crawford BMBS MPH FRACP (Paediatrics) is a paediatrician for the (a) What is the most likely diagnosis?
NHMRC CCRE in Childhood & Adolescent Immunization, SAEFVIC- Episode related to gastro-oesophageal reflux (GORD)
Immunization Safety Surveillance Unit, Department of General Arrhythmia
Medicine, MCRI/Royal Children’s Hospital, Melbourne, Australia. Hypotonic hyporesponsive episode (HHE)

Paediatrics and child health 19:6 294 © 2009 Elsevier Ltd. All rights reserved.
 self-assessment

Atonic seizure Value Reference range

Acute life-threatening episode (ALTE)
Lymphocytes 12.62  ×  109/l (H) 4.0–10.0  ×  109/l
(b) What investigations would you recommend?
Electrocardiogram (ECG) Monocytes 2.85  ×  109/l (H) 0.1–1.0  ×  109/l
Electroencephalogram (EEG) Eosinophils 2.44  ×  109/l (H) 0–0.8  ×  109/l
pH study Bands 0.81  ×  109/l (H) 0–0.5  ×  109/l
None required
RCC, red cell count; MCH, mean corpuscular haemoglobin; MCV, mean corpus-
(c) What would you recommend for future vaccinations
cular volume; RDW, red cell distribution width; Nuc RBC, nucleated red blood
No further vaccinations cells; WCC, white cell count; I/T ratio, immature to total neutrophil ratio; Neut
Separate vaccinations (i.e. one vaccine at a time) seg, segmented neutrophils; H, high.
Vaccinate under supervision in clinic/hospital
(d) Vaccinate as per schedule with normal immunization
provider Blood film

Case 3 • Left shift

A 10-month-old boy presented with a 4-week history of vomit- • High neutrophil count and reactive lymphocytes
ing, increased peripheral oedema and weight loss. Five days • Normal morphology
prior to admission he developed diarrhoea Six weeks earlier • Nil blasts
ESR: 12 mm/h (range 0–6)
he had had an episode of bronchiolitis, but no other signifi-
CRP: 18 mg/l (less than 8)
cant past medical history. He was exclusively breast-fed for
6 months before starting solids, and at presentation was on an
extensive diet and three to four breast-feeds daily. There was
no recent history of overseas travel, he was not on any medi- Faeces
cations, and immunizations were up to date. Macroscopic Cream unformed
appearance:
On admission Microscopy White blood cells Not detected

• Weight = 8.65 kg (10–25th centile) Red blood cells Not detected

• Blood pressure = 90/60 Fat globules Not detected
• Pulse rate = 80/min regular Fatty acid crystals ++
• Capillary return less than 2 seconds Ethyl acetate conc. No growth
• Oedema of legs (non-pitting) to ankles and mild swelling performed (Y/N)
of hands
• Chest clear, normal percussion Parasites Ova, cysts or parasites
• Abdomen soft and non-tender, no masses, no organomegaly, not detected
no ascites Culture Salmonella, Shigella, Campylobacter,
• Cardiovascular examination normal Aeromonas, Plesiomonas, Yersinia
not isolated
Investigations
Electrolytes & liver function tests
• Full blood count 
Value Reference range Value Reference range
Haemoglobin 136  g/l (H) 105–135  g/L Sodium 130  mmol/l (L) 135–145  mmol/l
Haematocrit 0.40 0.33–0.41 Potassium 3.2  mmol/l (L) 3.5–5.1  mmol/l
RCC 5.56  ×  1012/l (H) 3.7–5.3  ×  1012/l Chloride 110  mmol/l 98–110  mmol/l
MCH 24.5  pg 23–31  pg Urea 3.1  mmol/l 1.3–6.6  mmol/l
MCV 72  fl 70–90  fl Creatinine 0.03  mmol/l 0.01–0.03  mmol/l
RDW 16.8% (H) 11–14% Total Bili 1  μmol/l
Nuc RBC 0.4  ×  109/l (H) 0–0 (Bu  +  Bc)
Platelets 524  ×  109/l (H) 150–400  ×  109/l Bu 1  μmol/l 0–10   μmol/l
WCC 40.7  ×  109/l (H) 6.0–18.0  ×  109/l Bc 0  μmol/l 0–5  μmol/l
I/T ratio 0.04 less than 0.12 ALP 85  IU/l (L) 100–350  IU/l
Neut seg 14.98  ×  109/l (H) 1.0–8.5  ×  109/l GGT less than 5  IU/l 0–40  IU/l

Paediatrics and child health 19:6 295 © 2009 Elsevier Ltd. All rights reserved.
 self-assessment

Value Reference range Chest x-ray: normal, no evidence of pleural or pericardial

effusion.
Total protein 30  g/l (L) 50–71  g/l
Albumin 14  g/l (L) 29–45  g/l Ultrasound:
ALT 54  IU/l less than 55  IU/l • Multiple peristalsing thick-walled fluid-filled loops of
Bili, bilirubin; Bu, unconjugated bilirubin; Bc, conjugated bilirubin; ALP, alka-
bowel seen throughout the abdomen
line phosphatase; GGT, γ-glutamyl transferase ALT, alanine aminotransferase; • No free intraperitoneal fluid
L, low. • No abnormal mass or collection
• Normal liver, spleen, kidneys

α1 antitrypsin
(a) What is the most likely diagnosis?
Value Reference range
Intestinal lymphangiectasia
Faecal α1 1615  mg/l (H) 2–324  mg/l
Leukaemia
antitrypsin
Coeliac disease
A1 antitrypsin 27.8  ml/day (H) 0.5–16.4  ml/day Eosinophilic gastroenteritis
clearance Inflammatory bowel disease
H, high.

(b) What investigations would you recommend?

Urine protein/creatinine ratio Computed tomography (CT) of abdomen
Bone-marrow aspirate
Value Reference range
Serum immunoglobulins
Urine protein (random) 0.11  g/l Gastroscopy/colonoscopy and biopsy
Urine Creatinine: 2.21  mmol/l All of the above
Protein/Creatinine Ratio: 0.05  g/mmol (less than 0.06)
creatinine (c) What would you recommend as initial treatment
whilst awaiting investigations?
Coeliac screen Amino-acid-based formula
Medium-chain triglyceride (MCT) formula
Value Reference range
20% albumin infusion
IgA (quantitative) 0.12 0.1–1.29 Gluten-free diet
Deamidated gliadin peptide Ab IgA 8  LU Negative (less
than 20)
Answers
Tissue transglutaminase Ab IgA 3  LU Negative (less
than 20) Case 1
Ab, antibody; IgA, immunoglobulin A.

(a) Juvenile idiopathic arthritis (JIA)

Serum allergy test results JIA comprises a group of diseases of unknown aetiology that
are manifested by chronic joint inflammation. Over the past
RAST
decade JIA has been treated with more intensive therapy
Egg white 0.02  kUA/L through a multidisciplinary approach to try and minimize the
Milk 0.05  kUA/L morbidity from chronic arthritis. Multiple classification sys-
Wheat 0.06  kUA/L
tems are used internationally, but the key features are onset at
less greater than 16 years of age, clinical arthritis, duration of
Interpretation Of RAST scores less than 0.35  kUA/l: negative
greater than 6 weeks, and type defined by presentation: oligo-
arthritis/pauci-articular disease, polyarthritis, or systemic-
0.35–0.70  kUA/l: low positive onset disease.
0.7–3.5  kAU/l: moderate positive A thorough rheumatological examination and baseline
3.5–17.5  kAU/l: high positive
investigations are very helpful in excluding differential diag-
noses, including infection (septic arthritis, osteomyelitis) or
greater than 17.5  kAU/l: very high
postinfectious causes such as reactive arthropathy. The rheu-
positive
matological differential includes other conditions that can be
RAST, radioallergosorbent test. present with arthritis, including systemic lupus erythematosus,

Paediatrics and child health 19:6 296 © 2009 Elsevier Ltd. All rights reserved.
 self-assessment
juvenile dermatomyositis sarcoidosis and ­vasculitic ­syndrome.
Arthritis can also be seen in leukaemia, although it is often
deeper bone pain, and oligoarthritis can be seen with other
chronic diseases such as inflammatory bowel ­disease.
Haematological investigations in JIA may show a raised white
cell count and inflammatory markers. These investigations are
often done in conjunction with radiological investigations such
as skeletal x-rays, bone scans and ultrasound, especially when
considering the differential diagnoses of infection or trauma.
Rheumatoid factor is positive in around 7–10% of older patients
with polyarticular JIA and is a indicator of poor prognosis. Other
features associated with poor prognsosis include: older age at
onset, the presence of rheumatoid nodules, and disease of the
cervical spine or hip. JIA rarely presents as isolated arthritis
of the hip, so an irritable hip or reactive arthropathy was the
most likely diagnosis early in the course of the illness in the
case outlined above. JIA became more likely as the symptoms
continued at 6 weeks, were worse in the morning, response to
NSAIDs was minimal, and other joints became involved.
(b) MRI joints with contrast
MRI is a helpful adjunct in investigating joint pathology,
as it is the only radiological modality that can differentiate ­
proliferating synovial tissue from an effusion and the other
joint components (e.g. cartilage, bone, ligaments and ten-
dons). Since proliferating tissue enhances, contrast (e.g.
gadolinium), is required to improve the sensitivity, but as it
dissipates quickly only three or four joints can be reviewed
at one time. MRI of the joints may also be used to monitor
disease progression and response to treatment.
(c) Corticosteroid joint injections
Treatment of JIA aims to minimize longer-term morbidity
and progression into adulthood. A multidisciplinary approach
includes paediatric rheumatologists, nurse specialists, physio-
therapists, occupational therapists and psychologists. Regular
ophthalmological review is also required to detect asymptom-
atic uveitis. Standard drug therapy initially involves the use
of NSAIDS and corticosteroid injections into affected joints.
Oral or systemic glucocorticoids are usually reserved for
overwhelming inflammatory or systemic illness (e.g. sys-
temic-onset disease), their longer-term role being limited by
side-effects. Methotrexate is the most commonly used sec-
ond-line agent which can be given weekly either orally or
subcutaneously. More intensive therapy is also now offered
to refractory patients and may include disease-modifying bio-
logic agents such as etanercept (a tumour necrosis factor α
(TNFα) inhibitor), anakinra (an interleukin 1 (IL-1) receptor
antagonist), infliximab (a TFNα inhibitor), and adalimumab
(an IgG1 humanized monoclonal anti-TNFα antibody). Newer
drugs are emerging with the ever-increasing understanding of
the role of cytokines and cytokine genotypes in subgroups of
JIA. This will allow for improved therapy based on individual
host factors.
Paediatrics and child health 19:6
Further reading
Gardner-Medwin JM, Irwin G, Johnson K. MRI in Juvenile Idiopathic
Arthritis and Juvenile Dermatomyositis. Annn y Acad Sci 2009;
1154: 52–83.
McCann LJ, Wedderburn LR, Hasson N. Juvenile idiopathic arthritis
Arch Dis Child Ed Pract 2006; 91: ep29–ep36.
Miller ML, Cassidy JT. Juvenile rheumatoid arthritis. pp: 1001–
1010. In: Kliegman RM, Behrman RE, Jenson HB, Stanton BF,
eds. Nelson textbook of pediatrics. 18th Edn. Philadelphia:
Saunders; 2007.
Case 2
(a) Hypotonic hyporesponsive episode (HHE)
This reaction post-vaccination is consistent with a significant
adverse event following immunization (AEFI) called a hypo-
tonic hyporesponsive episode (HHE). Other terms for these
events have included ‘collapse’ or ‘shock-like episodes’. They
have clinical features of the child appearing acutely pale,
floppy (hypotonic), and unresponsive. Usually patients do
not respond to direct stimulation and slowly come around
over the next 1–30 minutes. They are often more sleepy than
usual over the next 24 hours. The reactions can be a frighten-
ing experience for parents and physicians. It is an uncommon
AEFI which occurs in approximately one in 20–40,000 vaccine
doses. It is most commonly seen with the first infant vac-
cines at 2 months of age, and usually occurs around 3–4 hours
post-immunization, but can occur immediately or up to 24–48
hours post-vaccination. It was seen more commonly with
whole-cell pertussis vaccine, but does still occur with other
vaccines, including those containing acellular pertussis.
The differential includes an atonic seizure, but this is usu-
ally characterized by unconsciousness rather than hypore-
sponsiveness, and whilst tone is decreased there is usually no
pallor or cyanosis. An acute life-threatening episode (ALTE)
has similar features that overlap with an HHE, but classically
ALTE has associated apnoea and/or cyanosis. In addition, an
HHE, as outlined above, usually occurs within 48 hours of an
immunization. In this case there was no history of previous
events and no family history or examination findings to sug-
gest a cardiac anomaly.
The pathophysiology of HHE remains uncertain. It appears
that multiple factors are implicated, including host or vac-
cinee characteristics, neurological development, immunologi-
cal phenomena, and some component of the vaccine(s). The
constellation of symptoms described in an HHE is the same
as a vaso-vagal or fainting type of event, and this term is
preferred in children over the age of 2 years who more com-
monly have an immediate event post-immunization.
(b) None required
Assessment following an HHE includes a thorough history
and physical/neurological examination. Investigations are
297 © 2009 Elsevier Ltd. All rights reserved.
 self-assessment
often not indicated, but should be undertaken if the patient
has atypical features: e.g. ECG if the cardiac exam is abnor-
mal or there is a family history of arrhythmias, EEG if features
or history are suggestive of a possible seizures.
(c) Vaccinate under supervision
In the past an HHE was a contraindication to further whole-cell
pertussis immunization, which had an HHE rate of 1 in 3000
vaccine doses, but is not for acellular pertussis or other vac-
cines. The recurrence rate in the limited number of published
studies has been between 0 and 5%. As the recurrence rate
is low, most vaccinologists would not recommend separating
the future vaccinations required. The difficulty is that the HHE
episodes can occur up to 24 hours post-vaccination, and many
centres would recommend administration of future vaccina-
tion under medical supervision either in a day medical unit or
with overnight admission, especially if the HHE was delayed.
This will often help reassure the parents and ensure the child
is adequately protected from vaccine-preventable diseases.
Importantly, the studies following up children who have had
an HHE showed that these episodes have no long-term effect on
development when compared with school age-matched peers.
Further reading
Baraff LJ, Shields WD, Beckwith L, et al. Infants and children
with convulsions and hypotonic–hyporesponsive episodes
following diphtheria–tetanus–pertussis immunization: follow-up
evaluation. Pediatrics 1988; 81: 789–94.
Buettcher M, Heininger U, Braun M, et al. Hypotonic-hyporesponsive
episode (HHE) as an adverse event following immunization in
early childhood: case definition and guidelines for data collection,
analysis, and presentation. Vaccine 2007; 25: 5875–81.
Goodwin H, Nash M, Gold M, et al. Vaccination of children following
a previous hypotonic hyporesponsive episode. J Paediatr Child
Health 1999; 35: 549–52.
Case 3
(a) Intestinal lymphangiectasia
Children with a protein-losing enteropathy classically present
with ascites, peripheral oedema, and a low serum albumin.
Albumin loss from the gastrointestinal tract is usually less
than 15% of the total body albumin, but may be high as 60%
in an enteropathy as seen in this case. Faecal antitrypsin lev-
els are elevated as this protein is negligibly broken down in
the gastrointestinal tract and is therefore a good screening
test for protein loss.
There is a long list when It comes to the differential diag-
nosis of a protein-losing enteropathy. Diagnoses are usually
classified into groups depending on the aetiology:
Mucosal damage leading to increased permeability:
• Erosive: ulceration of stomach/duodenum, pseudomem­
branous colitis, ulcerative colitis
Paediatrics and child health 19:6
• Non-erosive: connective tissue disorders, coeliac diseases,
allergic (eosinophilic) gastroenteritis, parasitic infection
Mechanical lymphatic obstruction:
• For example, intestinal lymphangiectasia, tuberculosis,
sarcoidosis
(b) All of the above
Intestinal lympangiectasia presents with the symptoms of a pro-
tein-losing enteropathy, and the obstruction of lymphatic drain-
age may be due to congenital defects in lymphatic ducts or to
secondary causes such as constrictive pericarditis, malrotation,
lymphoma and post-cardiac surgery. Patients with congenital
lympangiectasia usually present in the first decade of life, with
the first signs often being peripheral oedema and diarrhoea. Fal-
tering growth may be a presenting feature in younger children.
Immunoglobulins are usually low due to protein loss, but oppor-
tunistic infections rarely occur. A bone-marrow aspirate should
be strongly considered to rule out occult malignancy, which
may have supportive examination and haematological findings.
Abdominal ultrasound and CT scans can help identify dilated
intestinal loops, thickened folds and nodular protrusions, and
can be very helpful when considering other differential diagno-
ses. Chylous pleural effusions may also be seen on chest x-ray
in some cases. Endoscopy and jejunal biopsy is the gold stan-
dard in histologic confirmation of intestinal lymphangiectasia.
Macroscopically, white nodules and xanthomatous plaques are
often visualized and submucosal elevations identified. Because
of the patchy nature of the disease in the small intestine, often
multiple biopsies need to be taken which show dilated lacteals
with distortion of mucosal and submucosal villi in the absence
of an inflammatory response. Capsule endoscopy is a new tech-
nique for visualizing characteristic features in the small intestine
which cannot be seen with standard endoscopy.
(c) Medium-chain triglyceride (MCT) formula
Treatment of intestinal lymphangiectasia includes dietary
management, minimizing long-chain fat intake by using a
medium-chain triglyceride (MCT) formula. Supplementation
with fat-soluble vitamins (e.g. A, D, E, and K) is important
because of losses due to malabsorption. If an isolated area of
bowel is involved, surgical resection may be a management
option. Treatment with albumin infusions is usually reserved
for haemodynamic instability, as replacement will not treat
the underlying cause. Octreotide has been reported as a
therapy in refractory cases. Treatment of secondary intestinal
lymphangiectasia involves dietary modulation as per above,
as well as management of the underlying cause.
Further reading
Sood MR. Disorders of malabsorption. In: Kliegman RM, Behrman
RE, Jenson HB, Stanton BF, eds. Nelson textbook of pediatrics.
18th Edn. Philadelphia: Saunders; 2007, pp. 1584–1593.
Radhakrishnan K, Rockson SG. The clinical spectrum of lymphatic
disease. Ann NY Acad Sci 2008; 1131: 155–84.
298 © 2009 Elsevier Ltd. All rights reserved.