The Effect of Inhaled Budesonide on Symptoms,
Lung Function, and Cold Air and Methacholine
Responsiveness in 2- to 5-year–old Asthmatic Children
KIM GJERUM NIELSEN and HANS BISGAARD
Department of Pediatrics, Copenhagen University Hospital, Rigshospitalet, Copenhagen, Denmark
We hypothesized that measurement of lung function (LF) and
bronchial hyperresponsiveness (BHR) could serve as supplemental
tools in evaluating the efficacy of treatment with inhaled cortico-
steroids in asthmatic children aged 2 to 5 yr. We studied 38 chil-
dren (mean age: 53 mo; range: 35 to 71 mo) with moderately se-
vere asthma in a single-center, randomized, double-blind, parallel-
group, placebo-controlled study involving 8 wk of treatment.
Budesonide (BUD) 400 ␮g twice daily was administered via a pres-
surized metered-dose inhaler and metal spacer device. Symptom
scores (SSc) and use of short-acting ␤2-agonist were monitored
with diary cards. LF in awake children was measured as the specific
airway resistance (sRaw), using whole-body plethysmography; as
resistance by the interrupter technique (Rint); and as resistance
and reactance at 5 Hz (Rrs5, Xrs5) by the impulse oscillation tech-
nique. Cold air challenge (CACh) and methacholine challenge
(MCh) were used to assess BHR. Children in the BUD group experi-
enced significantly fewer night- and daytime symptoms (p ⬍ 0.05)
and more symptom-free days (p ⬍ 0.05), but not nights (p ⫽
0.07), than children in the placebo group. Daytime (p ⬍ 0.05) but
not nighttime (p ⫽ 0.09) use of rescue medication and asthma ex-
acerbation rates (3.7 versus 9.3 exacerbations/yr) (p ⫽ 0.006)
were both in favor of BUD. LF measured with the Rint technique,
Rrs5, and Xrs5 were significantly improved by BUD. BHR as mea-
sured by CACh improved significantly with BUD, whereas no im-
provement was found on MCh. In conclusion, inhaled BUD at a
total dose of 800 ␮g daily significantly improved SSc, asthma exac-
erbation rates, lung function, and BHR as assessed by CACh in
asthmatic children aged 2 to 5 yr.
Asthma is the most common chronic disease of young children
(1), and presents a considerable burden to the child, the fam-
ily, and society because of its high prevalence and lack of good
control with treatment. Diagnosis of asthma is ambiguous, ow-
ing to the lack of objective measurements. The poor control of
asthma in young children is partly due to a lack of validated
objective methods for studying lung function and bronchial re-
activity in young children, which hampers both disease moni-
toring and studies of pharmacotherapy.
The efficacy of inhaled corticosteroids (ICS) in young pre-
school children was first documented in our earlier study mon-
itoring symptom score (SSc) and use of rescue treatment (2).
Subsequent supportive trials have also been limited to such
health outcomes (3–8), since objective measurements have not
been available.
Young children under 6 yr of age can rarely perform the
maneuvers needed for lung function measurements, such as
forced expiration, which require active cooperation. Although
(Received in original form on February 7, 2000 and in revised form May 31, 2000)
Supported by AstraZeneca.
Correspondence and requests for reprints should be addressed to Hans Bisgaard,
Professor of Paediatrics, Copenhagen University Hospital, Rigshospitalet, DK-
2100 Copenhagen, Denmark. E-mail: Bisgaard@RH.DK
Am J Respir Crit Care Med Vol 162. pp 1500–1506, 2000
Internet address: www.atsjournals.org
single readings may occasionally be obtained, poor repeatabil-
ity prevents the use of conventional methods for measuring
lung function (9, 10). Recently, we successfully adapted the
measurement of specific airway resistance (sRaw) by whole-
body plethysmography; resistance measured with the inter-
rupter technique (Rint); and resistance and reactance at 5 Hz
(Rrs5, Xrs5) as measured with the impulse oscillation (IOS)
technique for use in awake children aged 2 yr and older (11–
13). These methods require no active cooperation by the
young child, are well accepted by the child, and are more sen-
sitive than spirometry for detecting the response to methacho-
line (11).
Bronchial hyperresponsiveness (BHR) is an essential fea-
ture of the pathophysiology and clinical manifestation of
childhood asthma (14), and is related to disease control. Bron-
choconstriction induced by hyperventilation of cold, dry air is
commonly found in asthmatic adults and children, and can be
used to assess BHR (15–21). The challenge with cold, dry air
has two main attractive features: the reaction probably reflects
the pathophysiology of asthma better than does pharmaco-
logic bronchoprovocation with histamine or methacholine
(22), and the method is simple to perform and standardize
even in children as young as 2 yr of age (23). We recently re-
ported a sensitivity of 68% and a specificity of 93% using
sRaw and cold air challenge (CACh) in asthmatic children
from 2- to 5-yr old (23). Furthermore, we recently used this
model to document clinically relevant bronchoprotection by
the leukotriene receptor antagonist montelukast (24).
The aim of the present study was to evaluate the effect of
budesonide (BUD) on symptoms, lung function (as measured
through SRaw, Rint, Rrs5, and Xrs5), and BHR to CACh and
methacholine challenge (MCh), with a view to evaluating
whether these objective measures could serve as additional
tools in the study of symptomatic asthmatic children aged 2 to
5 yr.
METHODS
Patients
Asthmatic children aged 2 to 5 yr from our outpatient clinic were eli-
gible for the study. The diagnosis of asthma was made empirically on
the basis of recurrent asthma symptoms, clinical improvement with
regular ICS therapy, and relapse during interruption of treatment.
Only patients fully cooperating in all lung function test procedures
and with CACh were considered for inclusion in the run-in period of
the study. Asthma symptoms, use of rescue medication, and distur-
bance of parents’ sleep were registered in a diary during the run-in pe-
riod. Patients with a daily SSc of at least 4 (not including scoring of
parents’ sleep disturbance) on at least seven of 14 consecutive days
were randomized.
Study Design
The study was a single-center, double-blind, parallel-group, random-
ized, placebo-controlled trial. Patients who fulfilled the inclusion cri-
teria at Visit 1 entered an observational run-in period in which regular
treatment was stopped and terbutaline as required was the only medi-
Nielsen and Bisgaard: Budesonide in Young Asthmatic Children
cation allowed. The run-in period lasted from 2 wk to a maximum of
8 wk, in which asthma symptoms, use of rescue medication, and distur-
bance of parents’ sleep were registered in a diary. Scoring during peri-
ods with respiratory infections was suspended. Patients were excluded
if they received inhaled steroids or were hospitalized for asthma during
the run-in period. At randomization, both a CACh challenge test and
an MCh test were performed within 72 h. Treatment medication and a
new diary card, identical to the one used during the run-in period,
were delivered after the last of the two challenge tests. Subjects were
scheduled for a visit after 4 wk of treatment. At this visit, diary cards
were checked, old medication was returned and new medication was
delivered, and baseline lung function was measured. After 8 wk of
treatment, CACh and MCh were repeated.
The local ethics committee (KF-02-179/96) and the national health
authorities of Denmark approved the study. Written informed con-
sent was obtained from parents or guardians of the subjects.
Randomization and Treatment
Subjects were randomly assigned to receive either two puffs of BUD
200 ␮g twice daily (total daily dose of 800 ␮g budesonide) or two puffs
twice daily of placebo from a pressurized metered-dose inhaler
(pMDI) via a metal spacer (Nebuchamber, AstraZeneca, Lund, Swe-
den). The randomization was computer generated in balanced blocks
of four treatment regimens.
Inhaled ␤2-agonist, delivered via the pMDI and spacer, was used
as rescue medication throughout the study. In the case of moderate
asthma attacks, one puff of formoterol 12 ␮g twice daily, delivered
from a pMDI and spacer, and/or a 3-d oral course of prednisolone at
1 mg/kg body weight twice daily, was allowed at the discretion of the
investigator. No other antiasthma medication was allowed. All short-
and long-acting ␤2-agonists were stopped at 6 and 24 h, respectively,
before baseline lung function measurement and bronchial provocation.
Assessments
Diary cards. Diary cards were constructed according to the method
described in our previous study (7). Symptoms were scored on a scale
of 0 to 3 (with a score of 0 indicating no symptoms and 1, 2, and 3 indi-
cating mild, moderate, and severe symptoms, respectively) by one of
the subject’s parents in three symptom domains (wheezing, cough,
and shortness of breath), in addition to scoring for use of relief medi-
cation for daytime symptoms and for nighttime symptoms individu-
ally. Furthermore, exercise-induced asthma symptoms during the day-
time and parents’ sleep disturbance caused by subjects’ asthma during
the night were scored on a scale of 0 to 3.
Pulmonary function testing. Pulmonary function testing was done
with a Master Screen unit, version 4.22 (Erich Jaeger GmbH, Würzburg,
Germany). Flow and volume were measured with a heated pressure-
screen–type pneumotachograph with a resistance of 0.036 kPa*s*L⫺1.
The equipment was calibrated daily. The methods and equipment
used for pulmonary function testing have previously been described
in detail (11, 12).
The children used a face mask (Astratech No. 2; ASTRA, Denmark)
fitted with a flexible, noncompressible mouthpiece that supported the
cheeks and provided stable access to the airways via the mouth (11).
Measurements were made during tidal breathing. Readings from
duplicate measurements made with each method were used as base-
line values before bronchial provocation tests. The sequence of mea-
surements was Rint, Xrs5, Rrs5, and sRaw. The same observer made
measurements at baseline and after bronchial provocation tests.
The subjects did not have clinical respiratory symptoms on the
days of CACh and MCh.
sRaw. sRaw was measured in a constant-volume, whole-body
plethysmograph as the relationship between simultaneous variations
in respiratory flow and maximum changes in plethysmographic pres-
sure during inspiration and expiration (11). sRaw was calculated from
the S-shaped resistance loops presented graphically by the computer
connected with the plethysmograph. Compensation for body temper-
ature, barometric pressure, and water vapor-saturated (BTPS) condi-
tions was done electronically (13). The respiratory rate was 30 to 45
breaths/min. When needed, an adult accompanied the subject during
testing (11, 13). The median value of five sequential measurements of
SRaw was used as the result.
1501
Rint. Rint is based on the assumption of a simple relationship be-
tween mouth pressure at the end of interruption of airflow and the
airflow after reopening of a shutter valve mounted on the pneumo-
tachograph. At every second inspiratory phase, inspiration of 50 ml of
air activated the shutter for 80 ms. Mouth pressure was measured dur-
ing the last 5 ms of the interruption. Flow was measured over a 70-ms
period after reopening of the shutter. The mean value of five consecu-
tive measurements was used as the result.
Xrs5 and Rrs5. Xrs5 and Rrs5 were measured with the IOS tech-
nique. Rectangular impulses were generated mechanically by a loud-
speaker and were applied to the respiratory system through the
mouthpiece of the pneumotachograph. The resulting pressure and
volume signals were analyzed for amplitude and phase difference to
determine the Xrs and Rrs of the respiratory system. Thirty seconds
of undisturbed measurements were used as the results (11, 12).
CACh. Cold, dry air was generated by a respiratory heat exchange
system (RHES; Erich Jaeger). Methods and equipment used for the
CACh have previously been described in detail (23). The CACh was
done as a single-step, 4-min, isocapnic hyperventilation test. We used
cold, dry air at ⫺15⬚ C, mixed with 5% CO2. The subject breathed
through a face mask fitted with a mouthpiece, which effectively se-
cured mouth breathing and prevented inhalation of room air. The rate
of ventilation was 1 L/min/kg body weight. The subject was motivated
to hyperventilate by competing with a computer-animated balloon,
which reflected the ventilation rate (23).
The response to hyperventilation of cold, dry air was recorded at
4 min after the end of the challenge.
Methacholine challenge test. The MCh test was done as a multistep
challenge with a dosimetric method. Isotonic methacholine chloride so-
lution was nebulized with a Wright nebulizer (Clement Clarke, Essex,
UK) that delivered aerosol into an aerosol box that prevented entrain-
ment of air (25). Methacholine was inhaled by tidal breathing from the
aerosol box through a face mask and mouthpiece. At each step of the
challenge the subject inhaled 200 ml of the aerosol per kilogram of
body weight. Doubling concentrations of methacholine, from 0.0625 to
a maximum of 64 mg/ml, were used. Increasing doses were inhaled until
sRaw increased by 50%, the maximum methacholine dose of 64 mg/ml
was reached, clinical airway obstruction was apparent, the subject com-
plained of discomfort, or transcutaneous oxygen pressure (TcO2), which
was monitored continuously during provocation, exhibited a decrease
of ⭓ 3 kPa. The heated (44⬚ C) transcutaneous pressure electrode was
placed on the middle part of the flexor side of the lower arm. Duplicate
measurements of Rint were made 2 min after provocation at each step
until Rint had increased by 10% or more from baseline. At that point,
sRaw, Rrs5, or Xrs5 were measured in that sequence.
The response to a challenge was measured at 3 to 5 min after the
end of the challenge.
Data Analysis
Study variables. Primary outcome measures were SSc and need for
rescue medication. Secondary outcome measures were changes in
lung function and change in response to CACh and MCh during the
treatment period.
All variables from the diary cards were summarized for each pa-
tient as a run-in mean and a treatment-period mean, after excluding
the first 14 d of the treatment period. Total nighttime and daytime
asthma symptoms were summarized by first adding the scores for
each subsymptom and then averaging the scores over the treatment
periods. Total 24-h asthma symptoms were summarized by adding the
nighttime and daytime mean scores. Identical periods and data were
used for calculation of symptom-free days.
An asthma exacerbation was defined as at least two consecutive
24-h periods with symptoms of wheezing and the need for at least 3
puffs of rescue treatment. The exacerbation was considered ongoing
as long as the two conditions were fulfilled, although for computa-
tional use, exacerbations experienced by a particular subject were
separated by periods of 5 d from the last previous day of exacerbation.
The means of duplicate measurements of sRaw, Rint, Xrs5, and
Rrs5 defined the baseline at each visit. Effect on lung function after
treatment was calculated as the average lung function after 4 wk and 8 wk
as a percent of lung function at baseline.
Predicted values based on height (in centimeters) for the lung
1502 AMERICAN JOURNAL OF RESPIRATORY AND CRITICAL CARE MEDICINE VOL 162 2000

function variables were based on a previous study of healthy controls
(26). Response to CACh was quantified as the percent change from
baseline. Thus, the formula used in calculating the response to CACh
was 100 ⴢ (post-CACh value ⫺ baseline value)/(baseline value). The
result of the MCh challenge was expressed as PCx, the provocative
concentration producing a change of 50% in SRaw, 30% in Rint and
Rrs5, 80% in Xrs5, and 15% in TcO2 (25). PCx was found through lin-
ear interpolation on a logarithmic scale. If x% change was not found
after the last dose step, PCx was estimated by one-step linear extrapo-
lation, using the last two previous values. If x% change was found af-
ter the first dose step, PCx was estimated by linear interpolation on
the linear scale between 0 and the first MCh concentration used.
Statistical considerations. An analysis of variance model was ap-
plied to compare BUD and placebo, using run-in mean SSc and base-
line lung function values as covariates. Ninety-five percent confidence
intervals (CIs) were constructed for the differences with and without
treatment. When analyzing asthma SSc, sleep disturbance, use of res-
cue medication, response to CA challenge, and change in lung func-
tion, we used additive models that gave arithmetic means. When ana-
lyzing lung function tests in measured units or PCx from the MCh
challenge, we used multiplicative models giving geometric means,
which were used to calculate treatment differences as ratios. The dif-
ferent subsymptoms were analyzed with a multivariate analysis of vari-
ance model, using each of the variables and baseline values as covari-
ates.
Asthma exacerbation rates were compared for treatment versus
placebo with Poisson’s distribution, assuming equal exacerbation rates
for all patients within a treatment group. The numbers of patients in
each treatment group experiencing at least one asthma exacerbation
during the treatment period were compared through Fisher’s exact
test.
From experience with a previous study using the same diary cards
(7), we estimated that a difference of 0.6 between the treatment and
placebo groups in total asthma SSc and score for use of rescue medi-
cation would be detected with 80% power with 22 patients in each
treatment group.
All statistical tests were two sided, and values of p ⬍ 5% were con-
sidered statistically significant.
RESULTS
Patients
A total of 91 patients were recruited for the study, and 39 were
randomized to treatment or placebo. Fifty-two patients were
withdrawn prior to randomization, 50 because their SSc did
not fulfil the criteria and two because of withdrawal of consent
prompted by increasing symptoms after stopping of regular
treatment. One patient randomized into the study had asthma
exacerbations during most of the treatment period due to lack
of compliance and was therefore withdrawn before analysis of
the study data. Thirty-eight asthmatic children (22 boys and 16
girls; mean age 53 mo [range: 35 to 71 mo]), were analyzed as
an intention-to-treat population. Baseline characteristics (Ta-
ble 1), including height, atopic disposition, relevant allergy, ex-
posure to passive smoking, SSc, use of rescue medication, and
lung function were comparable in the treatment and placebo
groups. The mean (range) duration of a history of recurrent
asthma symptoms was 18 mo (range: 4 to 56 mo). Twenty-one
(55%) subjects had a first-degree relative with atopic disease,
and 18 (47%) were exposed to passive smoking at home.
Twenty-two (58%) subjects had other manifestations of atopy
(atopic dermatitis and/or rhinitis). All 39 randomized subjects
were given skin prick tests or had assays of specific serum-IgE
antibody to the 10 inhalant allergens most common in Scandi-
navia, and 15 of the 39 (39%) had at least one positive test rel-
evant to a history of asthma.
Until the start of the run-in period, 33 (87%) of the subjects
were undergoing treatment with ICS, consisting of BUD at 100
to 1,200 ␮g/d (mean: 440 ␮g/d) delivered via a metered-dose
inhaler with a metal spacer (Nebuchamber), or fluticasone at
50 to 250 ␮g/d (mean: 133 ␮g/d) delivered via a metered dose
inhaler with a plastic spacer (Babyhaler, GlaxoWellcome,
Brøndby, Denmark). Short acting ␤2-agonists delivered via a
metered dose inhaler and spacer device were given as rescue
medication. Four subjects were not treated with ICS: two were
treated only with a short-acting ␤2-agonist as required, one
with a long-acting ␤2-agonist as required, and one with sodium
cromoglycate. The mean (range) duration of the run-in period,
between the cessation of ICS therapy and qualification for ran-
domization into the study, was 27 d (range: 14 to 56 d).
SSc and Rescue Medication Use
A statistically significant difference between BUD and pla-
cebo was found for nighttime asthma symptoms (p ⫽ 0.01),
and a tendency toward a difference was found for daytime

TABLE 1
BASELINE CHARACTERISTICS, USE OF RESCUE MEDICATION* AND TOTAL
SYMPTOM SCORE DURING RUN-IN, AND BASELINE LUNG FUNCTION IN
BUDESONIDE AND PLACEBO GROUP AT RANDOMIZATION

Budesonide Placebo p Value

n 19 19
Age, mo (range) 53 ⫾ 11 (35–70) 54 ⫾ 10 (37–71) ns
Sex, M/F 10/9 12/7 ns
Height, cm 105.1 ⫾ 7.1 106.7 ⫾ 6.7 ns
Atopic disposition 63% 47% ns
Other atopic diseases 47% 63% ns
Positive allergy test 39% 33% ns
Passive smoking 37% 58% ns
Total symptom score, night (range) 2.37 (0–5.9) 1.69 (0–4.3) ns
Total symptom score, day (range) 4.25 (1.1–7.2) 3.45 (1.9–6.9) ns
Rescue, night (range) 0.59 puffs (0–2.86 puffs) 0.90 puffs (0–3.86 puffs) ns
Rescue, day (range) 1.65 puffs (0–7.08 puffs) 1.54 puffs (0–5.08 puffs) ns
SRaw kPa*s (% pred) 1.48 ⫾ 0.29 (113 ⫾ 20) 1.57 ⫾ 0.48 (120 ⫾ 38) ns
Rint kPa*s*L⫺1 (% pred) 1.19 ⫾ 0.38 (118 ⫾ 32) 1.22 ⫾ 0.22 (120 ⫾ 22) ns
Xrs5 kPa*s*L⫺1 (% pred) ⫺0.41 ⫾ 0.21 (84 ⫾ 39) ⫺0.40 ⫾ 0.14 (84 ⫾ 39) ns
Rrs5 kPa*s*L⫺1 (% pred) 1.20 ⫾ 0.28 (90 ⫾ 20) 1.21 ⫾ 0.25 (89 ⫾ 18) ns

Definition of abbreviations: F ⫽ female; M ⫽ male; Rint ⫽ resistance by the interrupter technique; Rrs5 ⫽ airway resistance at 5 Hz by the
impulse oscillation technique; SRaw ⫽ specific airway resistance; Xrs5 ⫽ airway reactance at 5 Hz by the impulse oscillation technique.
* Mean number of puffs.
Nielsen and Bisgaard: Budesonide in Young Asthmatic Children
Figure 1. Effect on subsymptom scores, total symptom score, and day-
time and nighttime use of rescue medication of treatment with budes-
onide in contrast to placebo. Adjusted mean (95% CI) differences
(BUD–PLAC) between changes from run in period to completion of
budesonide treatment and placebo periods are shown.
asthma symptoms (p ⫽ 0.07) (Figure 1). The combined 24-h
SSc was significantly in favor of BUD (p ⫽ 0.03).
All subsymptom scores were reduced during treatment,
more so in the BUD than in the placebo group. Nighttime
cough was significantly reduced with active treatment (p ⫽
0.04). Borderline significance was found for a reduction in
daytime cough (p ⫽ 0.09) and disturbance of parents’ sleep (p ⫽
0.06) with treatment, but no other subsymptoms were signifi-
cantly reduced (Figure 1).
BUD significantly reduced daytime use of rescue medica-
tion (p ⫽ 0.01), but not its nighttime use (p ⫽ 0.09) (Figure 1).
The change with BUD and the difference between BUD
and placebo in asthma-symptom–free days and nights are
shown in Figure 2. There was a significant difference in the
change in percentage of symptom-free days (p ⫽ 0.03) and to-
tal 24-h symptom-free periods (p ⫽ 0.01), and a tendency to-
ward a difference in symptom-free nights (p ⫽ 0.07).
The percentage of days with asthma exacerbations was
4.9% in the BUD group versus 19.4% in the placebo group,
1503
which was significant (p ⫽ 0.01). We found a total of 12 exacer-
bations in the BUD group and 29 in the placebo group, giving
exacerbation rates of 3.7/yr versus 9.3/yr for BUD and placebo,
respectively (p ⫽ 0.006). Nine of 19 (47%) subjects in the BUD
group and 14 of 19 (74%) subjects in the placebo group (p ⫽
0.18) experienced at least one asthma exacerbation.
Lung Function Measurements
Baseline lung function measured as sRaw and Rint at the day
of randomization was 116% (95% CI: 106 to 126) and 119%
(95% CI: 110 to 128), respectively, of the predicted lung func-
tion values in relation to the height of the subjects (26) (i.e.,
the subjects showed significantly increased airway resistance
at baseline) (Table 1). Baseline measurements of Xrs5 and
Rrs5 did not differ from the reference values.
Lung function was significantly improved with BUD (Table
2). The difference between BUD and placebo was reflected by
Rint (p ⫽ 0.01), Rrs5 (p ⫽ 0.01), and Xrs5 (p ⫽ 0.001) at the
endpoint after 8 wk of treatment, and by the average lung
function through Visits 3 and 4. As shown in Table 2, Rint,
Xrs5, and Rrs5 improved during active treatment and deterio-
rated during administration of placebo. Xrs5 and Rrs5 re-
flected this improvement by Visit 3. No treatment effect was
reflected by SRaw (p ⫽ 0.35) (Table 2).
CACh
Responsiveness to CA challenge measured with each study
method at different visits is shown as the mean percentage
change from baseline (95% CI) in Figure 3. BUD significantly
reduced responsiveness to CA challenge after 8 wk of treat-
ment. This was reflected in measurements of SRaw (p ⬍ 0.001),
Rint (p ⫽ 0.01), Rrs5 (p ⫽ 0.01), and Xrs5 (p ⫽ 0.07). Figure 3
shows the decrease in responsiveness in all parameters during
BUD treatment and the reverse effect during placebo adminis-
tration. At the final CA challenge, 11 of 19 patients in the BUD
group, versus three of 19 subjects in the placebo group, reached
normal responsiveness (defined as a change of ⬍ 3 SD within
subject [w]) (23) measured in terms of sRaw (p ⫽ 0.02).
Methacholine Challenge Response
There was no statistically significant difference between BUD
and placebo in any of the parameters used in the study, al-
though a trend in favor of BUD treatment was seen in SRaw
measurements (p ⫽ 0.11). The provocative concentration of
MCh eliciting a 50% increase in SRaw could be computed by
interpolation for all curves except two. Considerably more of
the PCx values for Rint, Xrs5, Rrs5, and TcO2 had to be esti-
mated as extrapolated values.
DISCUSSION
This is the first report of the efficacy of ICS in asthmatic chil-
dren aged 2 to 5 yr to be based on objective measurements of
Figure 2. Increase in percentage of symptom-free
nights and days after treatment with budesonide and
placebo administration.
1504 AMERICAN JOURNAL OF RESPIRATORY AND CRITICAL CARE MEDICINE VOL 162 2000

TABLE 2
EFFECT ON LUNG FUNCTION OF TREATMENT WITH BUDESONIDE AND PLACEBO
p Value
Budesonide Placebo
Budesonide
Total Total versus
Variable 0 wk 8 wk Period 0 wk 8 wk Period Placebo

sRaw, kPa*s (95% CI) 1.49 (1.36–1.63) 1.44 (1.33–1.55) 1.41 (1.31–1.51) 1.58 (1.34–1.82) 1.54 (1.31–1.76) 1.48 (1.37–1.59) ns
Rint, kPa*s*L⫺1 (95% CI) 1.20 (1.03–1.38) 1.10 (0.98–1.22) 1.10 (1.03–1.18) 1.20 (1.09–1.32) 1.26 (1.14–1.38) 1.25 (1.17–1.34) 0.01
Xrs5, kPa*s*L⫺1 (95% CI) ⫺0.41 (⫺0.31–⫺0.50) ⫺0.34 (⫺0.27–⫺0.42) ⫺0.33 (⫺0.29–⫺0.37) ⫺0.44 (⫺0.36–⫺0.51) ⫺0.49 (⫺0.39–⫺0.60) ⫺0.46 (⫺0.40–⫺0.53) 0.001
Rrs5, kPa*s*L⫺1 (95% CI) 1.22 (1.07–1.37) 1.09 (0.96–1.21) 1.09 (1.02–1.17) 1.18 (1.07–1.30) 1.29 (1.19–1.39) 1.27 (1.18–1.36) 0.01

Definition of abbreviations: CI ⫽ confidence interval; Rint ⫽ resistance by the interrupter technique; Rrs ⫽ airway resistance at 5 Hz by the impulse oscillation technique; sRaw ⫽
specific airway resistance; Xrs5 ⫽ airway reactance at 5 Hz by the impulse oscillation technique.
Mean (95% CI) baseline values at randomization (0 wk), after 8 wk of treatment, and average (geometric mean, 95% CI) baseline lung function during treatment period (Total Pe-
riod) are given.

lung function and bronchial responsiveness. The study was ex-
plorative, evaluating objective measures together with con-
ventional measures of symptoms as primary outcomes of
asthma disease activity. The effect of 400 ␮g BUD twice daily,
delivered via a pMDI with a metal spacer device, was docu-
mented by improvement in SSc, reduced consumption of res-
cue medication, reduced asthma exacerbation rates, and in-
creased numbers of days and nights without symptoms, as
found in previous studies (2–8). In addition to this expected
treatment effect, the disease control with BUD was reflected
in measures of lung function and responsiveness to CACh.
The design of the present study was biased by regression
toward the mean, since randomization took place immediately
after a period of significant symptom scoring. This was re-
flected by the pronounced placebo effect, and the study results
should therefore be interpreted in the light of these unfavor-
able odds, strengthening the conclusions of the study. Also,
the small sample size of only 38 patients hampered the power
to uncover treatment differences in some subsymptoms, SRaw,
and MCh challenge.
Previous studies of effect of ICS on asthma in young chil-
dren have focused solely on subjective parameters such as SSc,
exacerbation rate, and use of rescue medication (2–8), and all
have shown an effect of ICS against moderate to severe persis-
tent asthma symptoms. However, none of these studies used
any objective parameters in children under 6 yr old. In a study
of infants aged 5 to 18 mo with recurrent wheezing (27), lung
function as measured with the rapid thoracoabdominal com-
pression technique, and bronchial responsiveness assessed
with histamine challenge, were used as objective measures in
parallel with subjective measures. ICS was found to signifi-
cantly improve bronchial responsiveness, but failed to show an
effect on baseline lung function or SSc, possibly because of a
lack of statistical power.
Ninety-one patients, of whom 90% were receiving regular
inhaled steroid treatment, were screened for our study and
stopped their steroid treatment during the run-in period.
Thirty-nine patients qualified for randomization. One reason
for noneligibility may have been that some of the patients did
not have asthma. Alternatively, some would still have been
protected at 8 wk after treatment interruption, and some had
only mild asthma. Therefore, the final study group in the
present study consisted of asthmatic children carefully se-
lected on the basis of a history of recurrent asthmatic epi-

Figure 3. Change (%) in measures of lung function

(sRaw, Rint, Xrs5, and Rrs5) in response to CA chal-
lenge at randomization (0 W) and after 8 wk of treat-
ment with budesonide or placebo (8 W). Mean (95%
CI).
Nielsen and Bisgaard: Budesonide in Young Asthmatic Children
sodes, need for current antiasthmatic therapy, and relapse of
asthma symptoms during interruption of treatment. From the
very high percentages of days (95%) and nights (78%) with
symptoms during the run-in period, it may be concluded that
these patients represented a group of children with moderate
to severe asthma. Their baseline lung function showed signifi-
cantly increased airway resistance as compared with that of
healthy controls (26), which is in agreement with our previous
findings in a random group of young asthmatic children (28)
and in selected groups of young asthmatic children (23, 24).
The dose of BUD given in the study was 400 ␮g twice daily,
which is above the pediatric dose range recommended as effi-
cacious in young children with moderate to severe persistent
asthma symptoms, and is probably on the flat part of the dose–
response curve. The present study was explorative, and the
high dose of BUD was chosen with a view toward not over-
looking any possible effect on lung function and BHR. Impor-
tantly, despite the high dosage of BUD, it was not possible to
eliminate asthma symptoms in all of the children in the study.
Whole-body plethysmography (sRaw), the interrupter
technique (Rint), and impulse oscillometry (Rrs5 and Xrs5)
are convenient methods for measuring lung function in young
children (11). Furthermore, the CACh test as a provocative
stimulus and measurement of sRaw to quantify the bronchial
response can be used to disclose BHR in awake, young chil-
dren, and to distinguish asthmatic and healthy children (23). In
the present study we demonstrated long-term improvement in
baseline lung function, attributed to treatment, as measured
by the interrupter technique and impulse oscillation tech-
nique. Whole-body plethysmography seemed less sensitive in
detecting long-term changes in baseline lung function.
Hyperventilation with cold, dry air has been shown to be a
potent stimulus to bronchoconstriction in asthma, and has been
applied in several studies both in adults and school children
(15–21). We used the single-step method for CA challenge,
since multistep protocols are more time consuming and iden-
tify the same subjects as hyperresponsive (21, 22). The CA
challenge test is simple to perform and standardize even in
children as young as 2 yr of age, and imposes no discomfort on
the child (23). We recently reported a sensitivity of 68% and a
specificity of 93% for measurement of SRaw and CA chal-
lenge testing in a group of 2- to 5-yr–old asthmatic children
and a group of healthy children (23).
The present study is the first to demonstrate an effect of
BUD on the bronchial responsiveness to CA challenge associ-
ated with improvement in symptom control in young children.
Hyperventilation of cold, dry air is believed to cause airway
narrowing through the release of leukotrienes (24, 29) and
other mediators, and the level of responsiveness is hypothe-
sized to reflect the degree of airway inflammation (22). We re-
cently reported clinically relevant bronchoprotection against
BHR provided by the leukotriene receptor antagonist mon-
telukast as quantified by CA challenge and SRaw measure-
ments in young asthmatic children (24). Previous studies of
adults (30, 31) have shown an effect of ICS on bronchocon-
striction induced by cold, dry air. In the present study, respon-
siveness improved in the group receiving active medication
and deteriorated in the placebo group, probably reflecting an
effect of BUD on the underlying inflammation during treat-
ment and loss of disease control when treatment was stopped.
Four children showed little or no change in responsiveness
with BUD. All other children in the BUD group improved in
responsiveness to CA challenge.
The importance of tailoring the dose of inhaled steroids to
the degree of BHR was recently emphasized by Sont and co-
workers (32). A treatment strategy in which BHR as reflected
1505
by CA challenge is used to titrate the steroid dose may pro-
vide an improvement in the long-term management of asthma
in young children.
MCh has been extensively studied in children, and with
success in children as young as 2 to 5 yr of age, showing satis-
factory repeatability (11, 12). However, this method is time
consuming and probably does not reflect the pathophysiology
of asthma to the same extent as does CACh (22). In the
present study we found no effect of BUD on responsiveness to
MCh, in contrast to the effect found on CACh. Inadequate du-
ration of treatment may explain the lack of effect on MCh re-
sponsiveness. Furthermore, it has been suggested that ICS
may provide greater protection against constrictor stimuli that
act indirectly, such as CA, than those that act directly, such as
MCh. It has also been found that in asthmatic children, the air-
way reactivities induced by cold, dry air and methacholine
challenge have no significant relationship (33).
The present study found a concordant treatment response
to BUD in both subjective parameters and objective parame-
ters such as lung function and bronchial responsiveness in
young asthmatic children. This suggests that measurement of
these objective parameters could be implemented and applied
in future studies of the clinical management of asthma in
young children. Furthermore, a dose–response study of ICS
on symptoms, lung function, and BHR in a larger number of
young asthma patients is needed.
In conclusion, inhaled BUD at a total dose of 800 ␮g daily
significantly improved SSc, asthma exacerbation rates, lung
function, and BHR as assessed by CA challenge in young asth-
matic children aged 2 to 5 yr.
References
1. Martinez, F. D., A. L. Wright, L. M. Taussig, C. J. Holberg, and M. Ha-
lonen, W. J. Morgan. 1995. Asthma and wheezing in the first six years
of life. N. Engl. J. Med. 332:133–138.
2. Bisgaard, H., S. L. Munck, J. P. Nielsen, W. Petersen, and S. V. Ohlsson.
1990. Inhaled budesonide for treatment of recurrent wheezing in early
childhood. Lancet 336:649–651.
3. Ilangovan, P., S. Pedersen, S. Godfrey, K. Nikander, N. Noviski, and
J. O. Warner. 1993. Treatment of severe steroid dependent preschool
asthma with nebulised budesonide suspension. Arch. Dis. Child. 68:
356–359.
4. Connett, G. J., C. Warde, E. Wooler, and W. Lenney. 1993. Use of bu-
desonide in severe asthmatics aged 1–3 years. Arch. Dis. Child. 69:
351–355.
5. de Blic, J., C. Delacourt, M. Le Bourgeois, B. Mahut, J. Ostinelli, C.
Caswell, and P. Scheinmann. 1996. Efficacy of nebulized budesonide
in treatment of severe infantile asthma: a double-blind study. J. Al-
lergy Clin. Immunol. 98:14–20.
6. Baker, J. W., M. Mellon, J. Wald, M. Welch, M. Cruz-Rivera, and K.
Walton-Bowen. 1999. A multiple-dosing, placebo-controlled study of
budesonide inhalation suspension given once or twice daily for treat-
ment of persistent asthma in young children and infants. Pediatrics
103:414–421.
7. Bisgaard, H., J. Gillies, M. Groenewald, and C. Maden. 1999. The effect
of inhaled fluticasone propionate in the treatment of young asthmatic
children: a dose comparison study. Am. J. Respir. Crit. Care Med. 160:
126–131.
8. Kemp, J. P., D. P. Skoner, S. J. Szefler, K. Walton-Bowen, M. Cruz-Riv-
era , and J. A. Smith. 1999. Once-daily budesonide inhalation suspen-
sion for the treatment of persistent asthma in infants and young chil-
dren. Ann. Allergy Asthma Immunol. 83:231–239.
9. Sly, P., and C. F. Robertson. 1990. A review of pulmonary function test-
ing in children. J. Asthma 27:137–147.
10. Kanengiser, S., and A. J. Dozor. 1994. Forced expiratory manoeuvres in
children aged 3 to 5 years. Pediatr. Pulmonol. 18:144–149.
11. Bisgaard, H., and B. Klug. 1995. Lung function measurement in awake
young children. Eur. Respir. J. 8:2067–2075.
12. Klug, B., and H. Bisgaard. 1996. Measurement of lung function in awake
2–4-year-old asthmatic children during methacholine challenge and acute
asthma. Pediatr. Pulmonol. 21:290–300.
1506 AMERICAN JOURNAL OF RESPIRATORY AND CRITICAL CARE MEDICINE
13. Klug, B., and H. Bisgaard. 1996. Measurement of the specific airway re-
sistance by plethysmography in young children accompanied by an
adult. Eur. Respir. J. 10:1599–1605.
14. Warner, J. O., and C. K. Naspitz. 1998. Third international pediatric con-
sensus statement on the management of childhood asthma. Interna-
tional Pediatric Asthma Consensus Group. Pediatr. Pulmonol. 25:1–17.
15. Deal, C. E., E. R. McFadden, R. H. Ingram, F. J. Breslin, and J. J. Jae-
ger. 1980. Airway responsiveness to cold air and hyperpnea in normal
subjects and in those with hay fever and asthma. Am. Rev. Respir. Dis.
121:621–628.
16. McLaughlin, J., and A. J. Dozor. 1983. Cold air inhalation challenge in
the diagnosis of asthma in children. Pediatrics 72:503–509.
17. Tal, A., H. Pasterkamp, C. Serrette, F. Leahy, and V. Chernick. 1984.
Response to cold air hyperventilation in normal and in asthmatic chil-
dren. J. Pediatr. 104:516–521.
18. Zach, M. S., G. Polgar, H. Kump, and P. Kroisel. 1984. Cold air chal-
lenge of airway hyperreactivity in children: practical application and
theoretical aspects. Pediatr. Res. 18:469–478.
19. Zach, M. S., and G. Polgar. 1987. Cold air challenge of airway hyperreac-
tivity in children: dose-response interrelation with a reaction plateau.
J. Allergy Clin. Immunol. 80:9–17.
20. Reisman, J., L. Mappa, F. de Benedictis, J. McLaughlin, and H. Levison.
1987. Cold air challenge in children with asthma. Pediatr. Pulmonol. 3:
251–254.
21. Modl, M., E. Eber, B. Steinbrugger, E. Weinhandl, and M. S. Zach. 1995.
Comparing methods for assessing bronchial responsiveness in chil-
dren: single step cold air challenge, multiple step cold air challenge,
and histamine provocation. Eur. Respir. J. 8:1742–1747.
22. Zach, M. S. 1995. Cold dry air challenge for measuring bronchial respon-
siveness—where do we stand? Pediatr. Pulmonol. 19:323–325.
23. Nielsen, K. G., and H. Bisgaard. 2000. Lung function response to cold air
challenge in asthmatic and healthy children aged 2 to 5 years. Am. J.
Respir. Crit. Care Med. 161:1805–1809.
24. Bisgaard, H., and K. G. Nielsen. 2000. Bronchoprotection from leuko-
VOL 162 2000
triene receptor antagonist in 3–5 year old children. Am. J. Respir. Crit.
Care Med. 162:187–190.
25. Klug, B., and H. Bisgaard. 1997. Repeatability of methacholine chal-
lenges in 2- to 4-year-old children with asthma, using a new technique
for quantitative delivery of aerosol. Pediatr. Pulmonol. 23:278–286.
26. Klug, B., and H. Bisgaard. 1998. Specific airway resistance, interrupter
resistance and respiratory impedance in healthy children aged 2–7
years. Pediatr. Pulmonol. 25:322–331.
27. Stick, S. M., P. R. Burton, J. B. Clough, M. Cox, P. N. LeSouëf, and P. D.
Sly. 1995. The effects of inhaled beclomethasone dipropionate on lung
function and histamine responsiveness in recurrently wheezy infants.
Arch. Dis. Child. 73:327–332.
28. Klug, B., and H. Bisgaard. 1999. Lung function and short-term outcome
in young asthmatic children. Eur. Respir. J. 14:1185–1189.
29. Fischer, A. R., M. A. Rosenberg, M. Roth, M. Loper, S. Jungerwirth,
and E. Israel. 1997. Effect of a novel 5-lipoxygenase activating protein
inhibitor, BAYx 1005, on asthma induced by cold dry air. Thorax
52:1074–1077.
30. Vathenen, A. S., A. J. Knox, A. Wiesniewski, and A. E. Tattersfield.
1991. Effect of inhaled budesonide on bronchial reactivity to hista-
mine, exercise, and eucapnic dry air hyperventilation in patients with
asthma. Thorax 46:811–816.
31. Claussen, M., and V. Sill. 1993. Influence on non-specific bronchial hy-
perreactivity to cold air hyperventilation and carbachol by nedocromil
and budesonide. Pneumologie. 47:209–214.
32. Sont, J. K., L. N. A. Willems, E. H. Bel, J. H. J. M. van Krieken, J. P.
Vandenbroucke, P. J. Sterk, and the AMPUL Study Group. 1999.
Clinical control and histopathological outcome of asthma when using
airway hyperresponsiveness as additional guide to long-term treat-
ment. Am. J. Respir. Crit. Care Med. 159:1043–1051.
33. de Benedictis, F. M., J. C. Canny, B. MacLusky, and H. Levison. 1995.
Comparison of airway reactivity induced by cold air and methacholine
challenges in asthmatic children. Pediatr. Pulmonol. 19:326–329.