Since January 2020 Elsevier has created a COVID-19 resource centre with

free information in English and Mandarin on the novel coronavirus COVID-

19. The COVID-19 resource centre is hosted on Elsevier Connect, the
company's public news and information website.

Elsevier hereby grants permission to make all its COVID-19-related

research that is available on the COVID-19 resource centre - including this
research content - immediately available in PubMed Central and other
publicly funded repositories, such as the WHO COVID database with rights
for unrestricted research re-use and analyses in any form or by any means
with acknowledgement of the original source. These permissions are
granted for free by Elsevier for as long as the COVID-19 resource centre
remains active.
 Articles

Azithromycin for episodes with asthma-like symptoms in

young children aged 1–3 years: a randomised, double-blind,
placebo-controlled trial
Jakob Stokholm, Bo L Chawes, Nadja H Vissing, Elín Bjarnadóttir, Tine M Pedersen, Rebecca K Vinding, Ann-Marie M Schoos, Helene M Wolsk,
Sunna Thorsteinsdóttir, Henrik W Hallas, Lambang Arianto, Susanne Schjørring, Karen A Krogfelt, Thea K Fischer, Christian B Pipper,
Klaus Bønnelykke, Hans Bisgaard

Summary
Background Bacteria and viruses are equally associated with the risk of acute episodes of asthma-like symptoms in Lancet Respir Med 2016;
young children, suggesting antibiotics as a potential treatment for such episodes. We aimed to assess the effect of 4: 19–26

azithromycin on the duration of respiratory episodes in young children with recurrent asthma-like symptoms, Published Online
December 15, 2015
hypothesising that it reduces the duration of the symptomatic period.
http://dx.doi.org/10.1016/
S2213-2600(15)00500-7
Methods In this randomised, double-blind, placebo-controlled trial, we recruited children aged 1–3 years, who were See Comment page 2
diagnosed with recurrent asthma-like symptoms from the Copenhagen Prospective Studies on Asthma in Childhood Copenhagen Prospective
2010 cohort; a birth cohort consisting of the general Danish population of Zealand, including Copenhagen. Exclusion Studies on Asthma in
criteria were macrolide allergy, heart, liver, neurological, and kidney disease, and, before each treatment, one or more Childhood, Herlev and
clinical signs of pneumonia (respiratory frequency of ≥50 breaths per min; fever of ≥39°C; C-reactive protein Gentofte Hospital
(J Stokholm PhD,
concentration of ≥476·20 nmol/L [≥50 mg/L]). Each episode of asthma-like symptoms lasting at least 3 days was B L Chawes PhD, N H Vissing PhD,
randomly allocated to a 3-day course of azithromycin oral solution of 10 mg/kg per day or placebo after thorough E Bjarnadóttir MD,
examination by a study physician at the Copenhagen Prospective Studies on Asthma research unit. Each episode was T M Pedersen MD,
randomly allocated independently of previous treatment from a computer-generated list of random numbers in blocks R K Vinding MD,
A-M M Schoos PhD,
of ten (generated at the Pharmacy of Glostrup). Investigators and children were masked until the youngest child turned H M Wolsk MD,
3 years of age and throughout the data validation and analysis phases. The primary outcome was duration of the S Thorsteinsdóttir MD,
respiratory episode after treatment, verified by prospective daily diaries and analysed with Poisson regression. Analyses H W Hallas MD, L Arianto MD,
K Bønnelykke PhD,
were per protocol (excluding those without a primary outcome measure or who did not receive treatment). This trial is
Prof H Bisgaard DMSc), and
registered with ClinicalTrials.gov, number NCT01233297. Section of Biostatistics,
Department of Public Health
Findings Between Nov 17, 2010, and Jan 28, 2014, we randomly allocated 158 asthma-like episodes in 72 children (79 (C B Pipper PhD), University of
Copenhagen, Copenhagen,
[50%] to azithromycin and 79 [50%] to placebo). The mean duration of the episode after treatment was 3·4 days for
Denmark; Department of
children receiving azithromycin compared with 7·7 days for children receiving placebo. Azithromycin caused a Pediatrics, Naestved Hospital,
significant shortening of the episode of 63·3% (95% CI 56·0–69·3; p<0·0001). The effect size increased with early Naestved, Denmark
initiation of treatment, showing a reduction in episode duration of 83% if treatment was initiated before day 6 of the (J Stokholm, E Bjarnadóttir,
T M Pedersen, R K Vinding); and
episode compared with 36% if initiated on or after day 6 (p<0·0001). We noted no differences in clinical adverse
Department of Microbiology
events between the azithromycin (18 [23%] of 78 episodes included in final analysis) and placebo (24 [30%] of 79) and Infection Control
groups (p=0·30), but we did not investigate bacterial resistance patterns after treatment. (S Schjørring PhD,
Prof K A Krogfelt PhD) and
Department of Microbiological
Interpretation Azithromycin reduced the duration of episodes of asthma-like symptoms in young children, suggesting
Diagnostics and Virology
that this drug could have a role in acute management of exacerbations. Further research is needed to disentangle the (Prof T K Fischer DMSc), Statens
inflammatory versus antimicrobial aspects of this relation. Serum Institut, Copenhagen,
Denmark

Funding Lundbeck Foundation, Danish Ministry of Health, Danish Council for Strategic Research, Capital Region Correspondence to:
Prof Hans Bisgaard, Copenhagen
Research Foundation.
Prospective Studies on Asthma
in Childhood, Herlev and
Introduction (COPSAC2000; a previous birth cohort of children born to Gentofte Hospital, University of
Childhood asthma is often preceded by recurrent episodes mothers with asthma), that airway bacteria (Haemophilus Copenhagen, Copenhagen,
DK-2820, Denmark
of troublesome lung symptoms in relation to airway influenzae, Streptococcus pneumoniae, and Moraxella
[email protected]
infections in the first years of life.1,2 Treatment of such catarrhalis) and respiratory viruses (at least one of
episodes represents a major unmet clinical need; they are picornavirus, respiratory syncytial virus, coronavirus,
the most common cause of admission to hospital in young parainfluenzavirus, influenza virus, human meta-
children, are a reason for stress and anxiety for families, pneumovirus, adenovirus, or bocavirus) are equally
and cause a major draw on health-care resources.3–5 closely associated with episodes of asthma-like symptoms
We discovered in our birth cohort, the Copenhagen in the first 3 years of life.6 Bacteria and viruses occurred
Prospective Studies on Asthma in Childhood 2000 together in most cases, challenging the previous

www.thelancet.com/respiratory Vol 4 January 2016 19

 Articles

Research in context
Evidence before this study Added value of this study
Findings from a publication from the Copenhagen This study is, to our knowledge, the first randomised controlled
Prospective Studies on Asthma in Childhood showed that trial of azithromycin treatment of acute episodes of asthma-like
both bacteria and viruses are equally associated with the risk symptoms in children aged 1–3 years with a history of recurrent
of acute episodes of asthma-like symptoms in young children, episodes and its findings show a clinically significant shortening
suggesting that antibiotics such as azithromycin could help in of symptom duration by 63% after intervention.
management of such episodes. We searched PubMed up to
Implications of all the available evidence
Oct 15, 2010, with no language limits, for various
Present guidelines do not recommend antibiotics for
combinations of the search terms “RCT”, “childhood”,
treatment of episodes of asthma-like symptoms in young
“asthma”, “wheeze”, and “antibiotics”. We identified all
children, yet antibiotics remain among the most commonly
previous randomised controlled trials of treatment with
prescribed drugs in these episodes. Our findings suggest that
antibiotics for asthma and wheezy exacerbations, but focused
azithromycin might be beneficial after medical assessment of
mainly on childhood asthma. The search led us to new
an acute asthma-like episode in young children with a known
articles, but also identified other relevant old publications for
history of such symptoms and without clinical signs of
background material. Investigators of two randomised trials
pneumonia. How the effect of azithromycin is compared with
concluded no beneficial effect of β-lactam treatment for
narrow-spectrum antibiotics and whether long-term effects
acute exacerbations, which has led to present guidelines not
are associated with recurrent use of azithromycin need to be
recommending antibiotic treatment for episodes of acute
investigated.
asthma-like symptoms.

hypothesis that episodes with asthma-like symptoms in concentration of 476·20 nmol/L [50 mg/L] or higher).
this age group are largely virally induced.7–9 This finding Most mothers from the COPSAC2010 cohort also
suggested that bacteria might play an equal part in the participated in other medical trials while pregnant and
pathogenesis of such episodes and that treatment with may have received dietary supplements or an influenza
antibiotics might ameliorate symptoms. At present, vaccination (NCT00856947, NCT00798226, and
guidelines do not recommend antibiotics for treatment NCT01012557).11,12 Additional details of baseline charac-
of early asthma-like episodes,10 yet they are widely used.4 teristics of the cohort are outlined in the COPSAC2010
We did a randomised controlled trial (RCT) of azithro- cohort design report.11
mycin for treatment of episodes of troublesome lung This trial was approved by the ethics committee for
symptoms in young children who were followed up Copenhagen (H-3-2010-065), Danish Data Protection
prospectively in our new unselected Copenhagen Agency (2010-41-5023), and Danish Health and Medicines
Prospective Studies on Asthma in Childhood 2010 Authority (2612-4329). Parents of children gave written
(COPSAC2010) birth cohort.11 and oral informed consent before enrolment of
participants. The complete biobank is publicly available
For the Danish National Methods at the Danish National Biobank. The entire COPSAC
Biobank see https://www. Study design and participants dataset, including the RCT-specific data, are currently
biobankdenmark.dk
In this randomised, double-blind, placebo-controlled being transferred to a publicly available database (the
For the Danish Data Archive see trial, we recruited children from the COPSAC2010 cohort, Danish Data Archive).
https://www.sa.dk which is a single-centre, population-based birth cohort of
700 children recruited from the general Danish Randomisation and masking
population of Zealand, including Copenhagen, at 1 week Each episode of troublesome lung symptoms that occurred
of age and followed up prospectively at the Copenhagen up to the age of 3 years or up to a maximum of seven
Prospective Studies on Asthma in Childhood (COPSAC) treatments per child was randomised individually to either
research unit (Copenhagen and Naestved, Denmark) with azithromycin or placebo. Treatments were randomly
deep clinical phenotyping.11 Children aged 1–3 years allocated at the Pharmacy of Glostrup (Copenhagen,
diagnosed with recurrent asthma-like symptoms Denmark) with a computer-generated list of random
(troublesome lung symptoms) as defined in the numbers in blocks of ten. Copies of the randomisation
Procedures section were eligible each time they had an code were kept in sealed envelopes at the research site and
episode of troublesome lung symptoms lasting at least the pharmacy. Investigators and participating families
3 days. Exclusion criteria were macrolide allergy, heart, were masked to treatment assignment until the youngest
liver, neurological, and kidney disease, and, before each child turned 3 years of age and throughout the data
treatment, one or more clinical signs of pneumonia validation and analysis phases. Those assessing the
(respiratory frequency of 50 breaths per min or higher, primary outcome were masked; those doing subanalyses
fever of 39°C or higher, or C-reactive protein [CRP] were not.

20 www.thelancet.com/respiratory Vol 4 January 2016

 Articles

Procedures Children diagnosed with recurrent troublesome lung

Troublesome lung symptoms, consisting of cough, symptoms and participating in this trial were prescribed
wheeze, or dyspnoea, severely affecting the wellbeing of azithromycin or placebo at the COPSAC research unit
the child, were monitored using daily diary cards filled when subsequent acute episodes of troublesome lung
out by the parents from birth.11 We defined an episode as symptoms occurred after diagnosis. We gave azithro-
at least 3 consecutive days of troublesome lung symptoms, mycin as an oral solution of 10 mg/kg per day in closed
at which point we requested that the parents brought the bottles (Teva) for 3 consecutive days or a matching
child to the COPSAC research unit for an acute visit. We placebo of similar look and taste (Pharmacy of
used the composite score of troublesome lung symptoms Glostrup).
to describe asthma-like symptoms in the children, a score
previously validated13,14 and used in our clinical Outcomes
observational cohort studies of young children13,15–17 and a The primary outcome was diary-verified duration of
randomised controlled trial.19 episodes of troublesome lung symptoms after initiation
At each acute visit, the diary cards were reviewed with of treatment. Secondary outcomes were the time from
the family by trained COPSAC paediatricians to validate treatment to the next episode of troublesome lung
symptom definitions and severity. Additionally, the symptoms, the number of episodes that turned into
research paediatrician did a thorough physical exam- severe exacerbations (need for oral steroids or admission
ination, consisting of assessment of fever, tachypnoea, to hospital), and the duration of β2 agonist use after
chest recessions, wheezing, and lung and heart treatment. Serious adverse events, any adverse events,
auscultation, and examination of the skin, ears, nose, and gastrointestinal symptoms, or other infections were
throat. Furthermore, we measured CRP concentration documented with daily diary cards and hospital records.
(detection limit of 76·19–1523·84 nmol/L [8–160 mg/L]) in
the peripheral blood with the QuickRead 101 instrument Statistical analysis
(Orion Diagnostica, Espoo, Finland). We collected a We based our power calculations on the duration of
hypopharyngeal aspirate using a soft suction catheter episodes of troublesome lung symptoms at age 1–3 years
passed through the nose into the hypopharynx, as in the children of the previous COPSAC2000 cohort.18,19,21
previously described.18 We cultured the samples and 86 independent episodes were needed to detect a
isolated the airway bacterial pathogens H influenzae, difference of 1 day duration of episodes with a power of
S pneumoniae, and M catarrhalis. We obtained a 90%, a p value of 0·05, and an SD of the duration of an
nasopharyngeal aspirate for viral identification with PCR. episode of 1·4 days.18,19,21
The viral airway pathogens analysed were rhinoviruses, We analysed the duration of an episode of troublesome
respiratory syncytial virus (RSV), and enteroviruses. lung symptoms and β2 agonist use after treatment with
We treated episodes of troublesome lung symptoms Poisson regression with a log link. This type of regression
with the β2 agonist salbutamol (Airomir; Teva, Kongens is ideal for modelling counts because it captures both
Lyngby, Denmark) inhaled from a pressurised metered skewness and variance heterogeneity and provides an
dose inhaler delivered via a spacer (AeroChamber; easy-to-interpret quantification of effects as relative
Trudell Medical International, London, ON, Canada) as change in mean counts. The model includes fixed effects
needed. We added 4 mg of montelukast in the evening of the categorical variable episode number and an effect
for 2 weeks in children who had previously benefited of treatment. We included a random effect of child to
from this treatment. We added prednisolone 1–2 mg/kg account for heterogeneity between children. We analysed
per day for 3 days for severe episodes at the discretion of factors potentially modifying the treatment effect with
the attending paediatricians in the COPSAC research robust Poisson regression to account for within-child
unit. correlation. We obtained inference by means of the
Recurrent troublesome lung symptoms were diagnosed generalised estimating equations procedure with a
if a child had: daily diary recordings of five episodes of working independence assumption.
troublesome lung symptoms within 6 months; 4 weeks We analysed gap times between episodes with
of continuous symptoms; or a severe acute episode Cox regression, including γ-distributed frailties shared by
needing oral prednisolone or hospital admission. This gap times within each child to account for between-child
diagnosis algorithm had previously been validated in our heterogeneity. We included fixed effects of treatment at
at-risk COPSAC2000 birth cohort.19,20 At diagnosis of the preceding episode in the model and stratified baseline
recurrent troublesome lung symptoms, we gave children hazards by episode number. We obtained estimates with
a 3-month course of 2 × 50 μg fluticasone (Flixotide; maximum likelihood estimation with Wald 95% CIs and
GlaxoSmithKline, UK) inhaled from a pressurised a 0·05 p value cutoff. We did analyses using R version
metered dose inhaler delivered via a spacer twice daily. If 3.2.2 and the add-on package lme4.
a second relapse of troublesome lung symptoms occurred Primary analyses were per protocol (excluding those
after cessation of inhaled corticosteroids, we initiated a without a primary outcome measure or who did not
6-month course of inhaled corticosteroids. receive treatment). Safety analyses included those

www.thelancet.com/respiratory Vol 4 January 2016 21

 Articles

without a primary outcome measure but who did receive 2·2 treatments (SD 1·5). In 121 (82%) of the 148 episodes
the study treatment. analysed (62 [84%] in the azithromycin-treated episodes and
This trial was monitored by the Good Clinical Practice 59 [80%] in the placebo-treated episodes), the child received
unit at Copenhagen University Hospital (Copenhagen, concurrent treatment with inhaled corticosteroids. In 89
Denmark). This trial is registered with ClinicalTrials.gov, (60%) episodes, the child received concurrent treatment
number NCT01233297. with montelukast (47 [64%] in the azithromycin-treated
episodes and 42 [57%] in the placebo-treated episodes).
Role of the funding source Treatment was complied with in 154 (97%) of 158 episodes:
The funders of the study had no role in study design, one (1%) azithromycin treatment was never given and three
data collection, data analysis, data interpretation, or (2%) treatments were discontinued after initiation; one (1%)
writing of the report. JS, CBP, and HB had access to the in the azithromycin group and two (1%) in the placebo
raw data. The corresponding author had full access to all group. 71 (99%) of 72 children had complete clinical follow-
the data in the study and had final responsibility for the up from inclusion in the study until age 3 years.
decision to submit for publication. Baseline characteristics did not differ significantly
between participants in the trial and other children
Results diagnosed with recurrent troublesome lung symptoms
207 (30%) of the 700 children enrolled in the main cohort from the COPSAC2010 cohort who did not participate in the
were diagnosed with recurrent troublesome lung
symptoms during the first 3 years of life; between Nov 17, RCT Non-RCT
2010, and Jan 28, 2014, we randomly allocated 158 episodes participants participants
for trial treatment (79 [50%] to azithromycin and 79 [50%] (n=72) (n=135)

to placebo; figure 1) from 72 (35%) of these children. Child

Before analyses, we excluded ten (6%) episodes from the Male sex 47 (65%) 74 (55%)
analysis (five [6%] in each group), nine (6%) because of White 70 (97%) 130 (96%)
missing diary information (azithromycin four [5%]; Older children in the home at birth 39 (54%) 70 (52%)
placebo five [6%]) and one (1%; in azithromycin group) Sensitisation (SPT or specific IgE) 8 (11%) 20 (15%)*
because the treatment was never given to the child. Atopic dermatitis 21 (30%)* 39 (30%)†
Mean age at randomisation was 2·0 years (SD 0·6). The 17q21 risk variant (RS2305480) 26 (41%)‡ 46 (39%)§
mean number of randomisations for each child was Smoking in pregnancy 9 (13%) 16 (12%)
Cat or dog at birth 26 (36%) 48 (36%)
Antibiotics in pregnancy 31 (43%) 50 (37%)
207 children assessed for eligibility Term birth >37 weeks 67 (93%) 127 (94%)
Caesarean section 18 (25%) 31 (23%)
Season of birth
135 children ineligible
Winter 25 (35%) 38 (28%)
Spring 17 (24%) 37 (27%)
72 children enrolled Summer 12 (17%) 29 (21%)
Autumn 18 (25%) 31 (23%)
Mother
79 episodes randomly assigned 79 episodes randomly assigned Maternal age at birth (years) 31·9 (4·7) 32·2 (4·5)
azithromycin placebo Maternal asthma¶ 31 (44%)* 38 (28%)
Maternal educational level

1 episode excluded because drug not

Low|| 9 (13%) 15 (11%)
ingested Medium** 53 (74%) 83 (61%)
High†† 10 (14%) 37 (27%)
Household annual income
4 episodes excluded because of 5 episodes excluded because of
missing primary outcome measure missing primary outcome measure Low‡‡ 4 (6%) 17 (13%)
Medium§§ 45 (63%) 73 (54%)
High¶¶ 23 (32%) 45 (33%)
74 episodes included in 74 episodes included in
per-protocol analysis per-protocol analysis Data are n (%) or mean (SD). RCT=randomised controlled trial. SPT=skin prick test.
*One child has missing information. †Seven children have data missing. ‡Nine
children have missing information. §17 children have data missing. ¶History of
78 episodes included in 79 episodes included in doctor-diagnosed asthma. ||Primary school, secondary school, or college
safety analysis safety analysis graduate. **Tradesman or Bachelor degree. ††Masters degree. ‡‡Less than
€50 000. §§€50 000–110 000. ¶¶More than €110 000.

Figure 1: Trial profile Table 1: Baseline characteristics

Each episode after inclusion was randomly allocated individually.

22 www.thelancet.com/respiratory Vol 4 January 2016

 Articles

trial, except for a higher proportion of mothers of RCT modify the treatment effect of azithromycin. None of
participants being asthmatic (31 [44%] of 71) than of mothers the detected viruses significantly modified treatment
of non-participants (38 [28%] of 135; p=0·03; table 1). The effects (table 2). Concurrent treatment with inhaled
mean total duration of respiratory episodes was 13·7 days. steroids (p value for interaction=0·57) or montelukast
The episode duration after randomisation was unrelated to (p value for interaction=0·69) did not significantly
sex, mother’s smoking status, allergic sensitisation to modify the treatment effect, and responses were equal
inhalant or food allergens at 6 months or 18 months, atopic (appendix). See Online for appendix
dermatitis, or 17q21 genetic risk variant (appendix).
Figure 2 shows the proportion of symptomatic children n (%) Mean Mean % reduction p value Modification
day-by-day during the 30 days after treatment, showing a azithromycin placebo (95% CI) p value
shortening of the symptomatic period after treatment episode episode
duration duration
with azithromycin. The average number of symptom days (days) (days)
after azithromycin treatment was 3·4 days versus 7·7 days
All 148 3·4 7·7 63·3% <0·0001
after placebo, corresponding to a calculated reduction in (56·0 to 69·3)
episode length of 63·3% (95% CI 56·0–69·3; p<0·0001) Clinical measures
within a child due to azithromycin treatment (table 2).
C-reactive protein 133 (100%) 0·6350
Restriction of the analysis to the first randomised concentration
treatment in each child substantiated a significant (nmol/L)
reduction of symptom duration by 44·4% (30·9–55·2; ≥76·19 nmol/L 23 (17%) 3·6 6·3 45·6% 0·2510
p<0·0001) after treatment with azithromycin, corre- (≥8 mg/L) (–53·9 to 80·8)
sponding to a mean duration of 4·0 days after azithromycin <76·19 nmol/L 110 (83%) 3·5 8·4 59·4% 0·0158
(<8 mg/L; (15·6 to 80·5)
versus 7·1 days after placebo. lowest
The effect of azithromycin was increased when the detection)
duration of symptoms before treatment was shorter, Fever (°C) 136 (100%) 0·4809
showing a reduction in episode duration of 83% if ≥38 23 (17%) 3·8 4·9 21·4% 0·5122
treatment was initiated before day 6 of the episode (–61·6 to 61·8)
compared with 36% if initiated on or after day 6 (p<0·0001). <38 113 (83%) 3·6 7·2 47·3% 0·0401
Figure 3 shows the reduction of episode duration after (2·9 to 71·4)
azithromycin treatment as a function of episode duration Objective wheeze 144 (100%) 0·8140
before treatment (less than or more than the median value Yes 26 (18%) 3·4 8·8 55·0% 0·0330
(6·3 to 78·4)
of 6 days). We noted no differential effect for episodes
No 118 (82%) 3·6 13·0 60·1% 0·0120
presenting with fever of ≥38°C or with increased CRP (18·3 to 80·5)
concentration of ≥76·19 nmol/L (≥8 mg/L), and treatment Bacterial infection
was equally effective in episodes with and without objective
Any pathogenic 135 (100%) 0·2864
wheeze during examination in the research clinic (table 2). bacteria
Presence of any pathogenic bacteria did not sig- Present 90 (67%) 4·2 7·9 41·6% 0·0881
nificantly modify the treatment effect compared with (–8·3 to 68·5)
episodes without detection of bacteria, but Not present 45 (33%) 2·0 5·5 64·7% 0·0007
azithromycin was more effective in episodes positive (35·6 to 80·7)
for H influenzae (table 2). Presence of any virus did not Haemophilus 135 (100%) 0·0323
influenzae
Present 32 (24%) 2·7 12·1 77·0% <0·0001
(58·0 to 87·4)
100 Placebo
Azithromycin Not present 103 (76%) 3·8 5·9 33·4% 0·2264
(–28·7 to 65·6)
80 Moraxella 135 (100%) 0·9062
catarrhalis
Children affected (%)

60 Present 64 (47% ) 4·4 8·7 40·5% 0·3163

(–64·3 to 78·5)
Not present 71 (53%) 2·8 5·2 45·0% 0·0436
40 (1·7 to 69·3)
Streptococcus 135 (100%) 0·8576
20
pneumoniae
Present 43 (32%) 3·3 6·2 44·4% 0·1436
(–22·1 to 74·7)
0 Not present 92 (68%) 3·6 7·5 49·6% 0·0377
0 5 10 15 20 25 30
(3·8 to 73·5)
Episode duration after treatment (days)
(Table 2 continues on next page)
Figure 2: Duration of episodes of troublesome lung symptoms after treatment

www.thelancet.com/respiratory Vol 4 January 2016 23

 Articles

secondary outcome to be analysed statistically (azithromycin

n (%) Mean Mean % reduction p value Modification
azithromycin placebo (95% CI) p value three [4%] episodes; placebo two [3%] episodes).
episode episode We noted no differences between treatment groups
duration duration during the 30 days after treatment with respect to serious
(days) (days)
or any adverse events, gastrointestinal symptoms, or
(Continued from previous page) other infections, as documented by daily diary cards and
Viral infection hospital records (appendix).
Any pathogenic 135 (100%) 0·7999
virus
Discussion
Present 58 (43%) 3·8 6·8 44·0% 0·1866
Azithromycin significantly reduced the duration of
(–32·4 to 76·4)
physician-verified episodes of asthma-like symptoms in
Not present 77 (57%) 2·7 7·4 50·7% 0·0004
(27·3 to 66·6) children aged 1–3 years with a history of recurrent
Rhinovirus 135 (100%) 0·5125 asthma-like symptoms. The duration was decreased by
Present 26 (19%) 4·6 6·9 26·7% 0·6430 about 63% after azithromycin treatment, with more
(–172·6 to 80·3) improvement if treatment was started early in the
Not present 109 (81%) 3·1 7·1 54·1% <0·0001 episode. Azithromycin had no long-term effect on risk of
(34·3 to 67·9) subsequent episodes.
RSV 135 (100%) 0·8886 This study is substantially strengthened by the
Present 22 (16%) 3·3 5·9 42·1% 0·3242 prospective, longitudinal, daily diary recordings of lung
(–71·7 to 80·5) symptoms before development of attacks, validated by
Not present 113 (84%) 3·6 7·2 46·9% 0·0289 study paediatricians at 6-monthly and acute visits. COPSAC
(6·3 to 69·9)
served as the primary health-care centre for the birth
Enteroviruses 135 (100%) 0·1997
cohort, ensuring a standardised approach to diagnosis and
Present 27 (20%) 2·1 6·8 66·3% 0·0026
(31·5 to 83·4)
treatment, which improves reliability of diagnoses
Not present 108 (80%) 4·0 7·1 41·1% 0·0619
compared with reporting from community doctors and
(–2·7 to 66·2) retrospective information from parents.18–20,22 Diagnosis was
based on an algorithm of symptom quantity, which has
p values correspond to associations and test for effect modification by covariants. RSV=respiratory syncytial virus.
been analysed and validated in detail13 and applied in our
Table 2: Duration of episodes of troublesome lung symptoms after intervention and effect modification previous RCT19 of inhaled corticosteroids in young children
from paraclinical measures in the at-risk COPSAC2000 birth cohort. Such strict diagnostic
procedure is paramount to clinical assessments, diagnoses,
100 and treatments, which are otherwise poorly standardised in
the community and more difficult in young children than
Reduction of episode duration

80 later in life.13 The in-depth clinical assessment of each

after treatment (%)

60 respiratory episode by the study paediatrician before

randomisation, including a thorough objective examination
40
and CRP concentration measurement in the research
20 clinic, ensured validity and homogeneity of the primary
outcome and exclusion of children with clinical signs of
0
<6 days ≥6 days pneumonia. This assessment minimised the possibility of
Episode duration before treatment the azithromycin effect being driven by treatment of
bacterial pneumonia misclassified as an episode of asthma-
Figure 3: Reduction of duration of episodes of troublesome lung symptoms
like symptoms.
after azithromycin treatment as a function of episode duration before
treatment Another strength is the centralised longitudinal
Circles correspond to estimates and solid lines to 95% pointwise CIs. clinical follow-up of the cohort by a research team with a
well established routine of doing clinical cohort studies.
Treatment with azithromycin did not significantly affect This routine ensured a high follow-up of the cohort, with
the time to next episode of troublesome lung symptoms in 99% of the randomised children completing full follow-
individual children (hazard ratio 0·95 [95% CI 0·65–1·40]; up until 3 years of age and only 6% of randomised
p=0·82). Treatment with azithromycin significantly treatments being unavailable. We obtained airway
reduced the duration of treatment with β2 agonists after samples before treatment in 91% of episodes, with only
intervention. The mean number of β2 agonist days after 13 episodes missed. Our results are generalisable to a
azithromycin treatment was 8·9 days versus 10·1 days after similar group of children with a known history of
placebo, corresponding to a calculated reduction in recurrent asthma-like symptoms, with or without
duration of β2 agonist treatment of 22·0% (95% CI concomitant treatment with inhaled steroids and with
7·0–34·6; p=0·006). Too few episodes requiring oral an episode duration of at least 3 days. An important
steroids or admission to hospital occurred for this limitation is that our data pertain to children with a

24 www.thelancet.com/respiratory Vol 4 January 2016

 Articles

detailed clinical history and acute worsening judged by in particular, might be present for episodes triggered by
the study clinician. They cannot be generalised to a more H influenzae.34 This corresponds with our finding of a
liberal setting, such as initiation at home by parents or stronger azithromycin effect in episodes triggered by
children with mild disease. H influenzae. Viral infection in general did not predict
This study is the first, to our knowledge, to investigate an altered effect of azithromycin treatment. Low
and show an effect of azithromycin for treatment of acute numbers in some viral species groups limit the
respiratory episodes in young children with a history of conclusions drawn about effect modification by specific
recurrent asthma-like symptoms, in a cohort designed viruses. Thus, the antibacterial, anti-inflammatory, and
and powered to explore such effects. A third of all children antiviral pathways of azithromycin could have
will experience an episode of asthma-like symptoms in contributed to the shortening of episode length observed
relation to airway infections before 3 years of age.23 in this study. We cannot rule out that the effects noted
Shortening of such episodes by 63% is therefore clinically are mainly anti-inflammatory because episodes
significant to the child, families, health-care user, and triggered by H influenzae induce neutrophilic
society. We noted that azithromycin had a marked effect in inflammation34 and azithromycin also reduced episode
relation to any given type of episode independently of the duration in children without any evidence of bacterial
trigger, clinical presentation, or symptom duration before infection. An RCT37 of treatment with azithromycin in
the intervention, which suggests a broad application. The RSV-positive children showed a reduction of the
sensitivity analysis restricting episodes to the first neutrophilic marker interleukin 8 at day 14 after
randomisation validated the primary finding. azithromycin treatment. This finding could point
Asthma-like episodes in young children probably towards an anti-inflammatory effect as the primary
represent a heterogeneous clinical syndrome. Importantly, mediator of our findings and also explain why findings
we excluded typical pneumonia based on predefined from previous RCTs38,39 have failed to show treatment
clinical criteria. Furthermore, most children (more than effects of non-anti-inflammatory antibiotics in acute
80%) had undetectable concentrations of CRP and no fever exacerbations of childhood asthma.
at randomisation, and the treatment effect was similar in Present guidelines do not recommend antibiotics for
these children, strongly suggesting that the effect of treatment of episodes of asthma-like symptoms in young
azithromycin is not due to misclassification of pneumonia. children10 and yet they are among the most commonly
Azithromycin treatment was particularly effective in prescribed drugs for such episodes in both the USA and
children who were treated shortly after symptom debut. Europe.4 Our data suggest an effect of azithromycin on
This finding makes us speculate that azithromycin is acute asthma-like episodes and thereby identify a
mainly acting on the acute inflammatory24 or infectious6,25 potential future treatment, but do not provide sufficient
processes related to exacerbations, rather than a persistent evidence to recommend this treatment in clinical practice.
underlying inflammation. This hypothesis is supported by How the effect of azithromycin compares with narrow-
the absence of effect on time to next episode. Alternatively, spectrum antibiotics and whether any long-term effects
azithromycin acts by clearing bacterial pathogens indirectly are associated with recurrent use need to be investigated.
responsible for the respiratory episode through subsequent We are keenly aware of the potential ecological issues
co-infection by a viral trigger, and therefore the treatment relating to use of antibiotics in terms of bacterial ecology
was more effective when initiated early in the episode than and resistance. Macrolide resistance in organisms
when initiated late.26 causing respiratory illness in children is already an
This study is based on our previous birth cohort study6 issue.40 Future research should establish the choice of
in which we discovered that pathogenic airway bacteria antimicrobial treatment and criteria for treatment, taking
and respiratory viruses are equally closely associated societal aspects into consideration. We did not identify
with acute episodes of asthma-like symptoms in young strong effect modifiers from the objective clinical
children and mostly occur together. Our finding in this measures, concomitant treatments, or the microbiological
study that the treatment effect was strong, even in profile other than the presence of H influenzae. Our data
episodes for which no bacterial pathogen was detected, suggest that the effect increases by starting early in the
suggests that the effect of azithromycin is not only episode. Future studies might help to identify specific
antibacterial. Macrolide antibiotics are active against disease phenotypes or biomarkers directing the treatment
both common airway pathogenic bacteria and atypical to specific groups of young children.
bacteria,27–29 but also have anti-inflammatory activity24 The results of this study identify a potential treatment
and, possibly, antiviral effects.30 Studies of macrolides in for a common childhood disease for which better
adults with severe asthma episodes are ambiguous,31,32 treatment options are needed than are currently available.
and macrolides reportedly reduce exacerbations in Better treatments might help to alleviate a substantial
adults with an asthma type characterised by chronic disease burden for children, families, and society.
neutrophilic inflammation.33,34 Indeed, recurrent Contributors
asthma-like symptoms in young children are also HB conceived, designed, and carried out the study, acquired, analysed,
characterised by neutrophilic inflammation,25,35,36 which, and interpreted data, and wrote the report. JS, BLC, and KB contributed

www.thelancet.com/respiratory Vol 4 January 2016 25

 Articles

to design of the study. JS acquired, analysed, and interpreted data and 18 Bisgaard H, Hermansen MN, Buchvald F, et al. Childhood asthma
drafted the report. CBP did the statistical analyses. SS and KAK cultured after bacterial colonization of the airway in neonates.
and identified the bacteria. TKF identified the viruses. BLC, NHV, EB, N Engl J Med 2007; 357: 1487–95.
TMP, RKV, A-MMS, HMW, ST, HWH, LA, and KB collected and 19 Bisgaard H, Hermansen MN, Loland L, Halkjaer LB, Buchvald F.
interpreted data and wrote the report. All coauthors have contributed Intermittent inhaled corticosteroids in infants with episodic
substantially to the analyses or interpretation of the data and have wheezing. N Engl J Med 2006; 354: 1998–2005.
provided important intellectual input and approval of the final version of 20 Bisgaard H, Bønnelykke K, Sleiman PM, et al. Chromosome 17q21
the manuscript. gene variants are associated with asthma and exacerbations but not
atopy in early childhood. Am J Respir Crit Care Med 2009; 179: 179–85.
Declaration of interests 21 Bisgaard H. The Copenhagen Prospective Study on Asthma in
HB has received funds for research and for members of research staff Childhood (COPSAC): design, rationale, and baseline data from a
and been paid as a consultant for Chiesi. All other authors declare no longitudinal birth cohort study. Ann Allergy Asthma Immunol 2004;
competing interests. 93: 381–89.
22 Chawes BL, Buchvald F, Bischoff AL, et al. Elevated exhaled nitric
Acknowledgments
oxide in high-risk neonates precedes transient early but not
Copenhagen Prospective Studies on Asthma in Childhood (COPSAC) is persistent wheeze. Am J Respir Crit Care Med 2010; 182: 138–42.
funded by private and public research funds all listed on www.copsac.com.
23 Martinez FD, Wright AL, Taussig LM, Holberg CJ, Halonen M,
The Lundbeck Foundation, Danish Ministry of Health, Danish Council Morgan WJ. Asthma and wheezing in the first six years of life. The
for Strategic Research, and Capital Region Research Foundation have Group Health Medical Associates. N Engl J Med 1995; 332: 133–38.
provided core support for COPSAC. We express our gratitude to the 24 Fonseca-Aten M, Okada PJ, Bowlware KL, et al. Effect of
children and families of the COPSAC2010 cohort for all their support and clarithromycin on cytokines and chemokines in children with an
commitment. We likewise acknowledge and appreciate the unique efforts acute exacerbation of recurrent wheezing: a double-blind,
and teamwork of the COPSAC research team. Steffen Lynge Jørgensen is randomized, placebo-controlled trial.
thanked for microbiological technical assistance. Ann Allergy Asthma Immunol 2006; 97: 457–63.
References 25 Schwerk N, Brinkmann F, Soudah B, Kabesch M, Hansen G. Wheeze
1 Masoli M, Fabian D, Holt S, Beasley R. The global burden of in preschool age is associated with pulmonary bacterial infection and
asthma: executive summary of the GINA Dissemination Committee resolves after antibiotic therapy. PLoS One 2011; 6: e27913.
report. Allergy 2004; 59: 469–78. 26 Jartti T, Kuneinen S, Lehtinen P, et al. Nasopharyngeal bacterial
2 Papadopoulos NG, Christodoulou I, Rohde G, et al. Viruses and colonization during the first wheezing episode is associated with
bacteria in acute asthma exacerbations—a GA² LEN-DARE longer duration of hospitalization and higher risk of relapse in
systematic review. Allergy 2011; 66: 458–68. young children. Eur J Clin Microbiol Infect Dis 2011; 30: 233–41.
3 Kocevar VS, Bisgaard H, Jönsson L, et al. Variations in pediatric 27 Freymuth F, Vabret A, Brouard J, et al. Detection of viral,
asthma hospitalization rates and costs between and within Chlamydia pneumoniae and Mycoplasma pneumoniae infections in
Nordic countries. Chest 2004; 125: 1680–84. exacerbations of asthma in children. J Clin Virol 1999; 13: 131–39.
4 Bisgaard H, Szefler S. Prevalence of asthma-like symptoms in 28 Biscardi S, Lorrot M, Marc E, et al. Mycoplasma pneumoniae and
young children. Pediatr Pulmonol 2007; 42: 723–28. asthma in children. Clin Infect Dis 2004; 38: 1341–46.
5 Bush A. Practice imperfect—treatment for wheezing in 29 Esposito S, Blasi F, Arosio C, et al. Importance of acute
preschoolers. N Engl J Med 2009; 360: 409–10. Mycoplasma pneumoniae and Chlamydia pneumoniae infections in
children with wheezing. Eur Respir J 2000; 16: 1142–46.
6 Bisgaard H, Hermansen MN, Bønnelykke K, et al. Association of
bacteria and viruses with wheezy episodes in young children: 30 Gielen V, Johnston SL, Edwards MR. Azithromycin induces
prospective birth cohort study. BMJ 2010; 341: c4978. anti-viral responses in bronchial epithelial cells. Eur Respir J 2010;
36: 646–54.
7 Minor TE, Dick EC, DeMeo AN, Ouellette JJ, Cohen M, Reed CE.
Viruses as precipitants of asthmatic attacks in children. JAMA 1974; 31 Koutsoubari I, Papaevangelou V, Konstantinou GN, et al. Effect of
227: 292–98. clarithromycin on acute asthma exacerbations in children: an open
randomized study. Pediatr Allergy Immunol 2012; 23: 385–90.
8 Kusel MM, de Klerk NH, Holt PG, Kebadze T, Johnston SL, Sly PD.
Role of respiratory viruses in acute upper and lower respiratory tract 32 Johnston SL, Blasi F, Black PN, Martin RJ, Farrell DJ, Nieman RB.
illness in the first year of life. Pediatr Infect Dis J 2006; 25: 680–86. The effect of telithromycin in acute exacerbations of asthma.
N Engl J Med 2006; 354: 1589–600.
9 Khetsuriani N, Kazerouni NN, Erdman DD, et al. Prevalence of
viral respiratory tract infections in children with asthma. 33 Brusselle GG, Vanderstichele C, Jordens P, et al. Azithromycin for
J Allergy Clin Immunol 2007; 119: 314–21. prevention of exacerbations in severe asthma (AZISAST):
a multicentre randomised double-blind placebo-controlled trial.
10 Reddel HK, Bateman ED, Becker A, et al. A summary of the new Thorax 2013; 68: 322–29.
GINA strategy: a roadmap to asthma control. Eur Respir J 2015;
46: 622–39. 34 Brusselle GG, Joos G. Is there a role for macrolides in severe
asthma? Curr Opin Pulm Med 2014; 20: 95–102.
11 Bisgaard H, Vissing NH, Carson CG, et al. Deep phenotyping of the
unselected COPSAC2010 birth cohort study. Clin Exp Allergy 2013; 35 Le Bourgeois M, Goncalves M, Le Clainche L, et al. Bronchoalveolar
43: 1384–94. cells in children <3 years old with severe recurrent wheezing.
Chest 2002; 122: 791–97.
12 Bischoff AL, Følsgaard NV, Carson CG, et al. Altered response to
A(H1N1)pnd09 vaccination in pregnant women: a single blinded 36 Saglani S, Malmström K, Pelkonen AS, et al. Airway remodeling
randomized controlled trial. PLoS One 2013; 8: e56700. and inflammation in symptomatic infants with reversible airflow
obstruction. Am J Respir Crit Care Med 2005; 171: 722–27.
13 Bisgaard H, Pipper CB, Bønnelykke K. Endotyping early childhood
asthma by quantitative symptom assessment. 37 Beigelman A, Isaacson-Schmid M, Sajol G, et al. Randomized trial
J Allergy Clin Immunol 2011; 127: 1155–64.e2. to evaluate azithromycin’s effects on serum and upper airway IL-8
levels and recurrent wheezing in infants with respiratory syncytial
14 Skytt N, Bønnelykke K, Bisgaard H. ‘To wheeze or not to wheeze’: virus bronchiolitis. J Allergy Clin Immunol 2015; 135: 1171–1178.e1.
that is not the question. J Allergy Clin Immunol 2012; 130: 403–07.e5.
38 Shapiro GG, Eggleston PA, Pierson WE, Ray CG, Bierman CW.
15 Chawes BL, Stokholm J, Bønnelykke K, Brix S, Bisgaard H. Double-blind study of the effectiveness of a broad spectrum
Neonates with reduced neonatal lung function have systemic low- antibiotic in status asthmaticus. Pediatrics 1974; 53: 867–72.
grade inflammation. J Allergy Clin Immunol 2015; 135: 1450–56.e1.
39 Graham VA, Milton AF, Knowles GK, Davies RJ. Routine antibiotics
16 Vissing NH, Chawes BL, Bisgaard H. Increased risk of pneumonia in hospital management of acute asthma. Lancet 1982; 1: 418–20.
and bronchiolitis after bacterial colonization of the airways as
neonates. Am J Respir Crit Care Med 2013; 188: 1246–52. 40 Spreading US macrolide resistance. JAMA 2015; 314: 1218.
17 Kreiner-Møller E, Chawes BL, Vissing NH, et al. VEGFA variants
are associated with pre-school lung function, but not neonatal lung
function. Clin Exp Allergy 2013; 43: 1236–45.

26 www.thelancet.com/respiratory Vol 4 January 2016