Anda di halaman 1dari 10

A Systematic Review of Long-Acting β2-Agonists Versus Higher Doses of

Inhaled Corticosteroids in Asthma

Jose A. Castro-Rodriguez and Gustavo J. Rodrigo
Pediatrics 2012;130;e650; originally published online August 27, 2012;
DOI: 10.1542/peds.2012-0162

The online version of this article, along with updated information and services, is
located on the World Wide Web at:

PEDIATRICS is the official journal of the American Academy of Pediatrics. A monthly

publication, it has been published continuously since 1948. PEDIATRICS is owned,
published, and trademarked by the American Academy of Pediatrics, 141 Northwest Point
Boulevard, Elk Grove Village, Illinois, 60007. Copyright © 2012 by the American Academy
of Pediatrics. All rights reserved. Print ISSN: 0031-4005. Online ISSN: 1098-4275.

Downloaded from at Indonesia:AAP Sponsored on June 30, 2014

A Systematic Review of Long-Acting b2-Agonists Versus
Higher Doses of Inhaled Corticosteroids in Asthma
AUTHORS: Jose A. Castro-Rodriguez, MD, PhD,a and
abstract Gustavo J. Rodrigo, MDb
aDepartments of Family Medicine and Pediatrics, School of
OBJECTIVE: To compare the efficacy of inhaled corticosteroids (ICS)
Medicine, Pontificia Universidad Católica de Chile, Santiago,
plus long-acting b2 agonist (LABA) versus higher doses of ICS in Chile; and bDepartamento de Emergencia, Hospital Central de las
children/adolescents with uncontrolled persistent asthma. Fuerzas Armadas, Montevideo, Uruguay
METHODS: Randomized, prospective, controlled trials published KEY WORDS
asthma, children, adolescents, LABA, inhaled corticosteroids,
January 1996 to January 2012 with a minimum of 4 weeks of LABA
+ICS versus higher doses of ICS were retrieved through Medline,
Embase, Central, and manufacturer’s databases. The primary outcome AEs—adverse events
was asthma exacerbations requiring systemic corticosteroids; sec- BDP—beclomethasone dipropionate
ondary outcomes were the pulmonary function test (PEF), with- CI—confidence interval
FDA—Food and Drug Administration
drawals during the treatment period, days without symptoms, use FEV1—flow expiratory volume in the first second
of rescue medication, and adverse events. ICS—inhaled corticosteroids
LABA—long-acting b2 agonist
RESULTS: Nine studies (n = 1641 patients) met criteria for inclusion
OR—odds ratio
(7 compared LABA+ICS versus double ICS doses and 2 LABA+ICS ver- PEF—peak expiratory flow
sus higher than double ICS doses). There was no statistically signif- RCT—randomized controlled trial
icant difference in the number of patients with asthma exacerbations RR—relative risk
WMD—weighted mean differences
requiring systemic corticosteroids between children receiving LABA
Dr Castro-Rodriguez has made substantial contributions to the
+ICS and those receiving higher doses of ICS (odds ratio = 0.76; 95% conception, design, and interpretation of data; has revised the
confidence interval: 0.48–1.22, P = .25, I2 = 16%). In the subgroup article critically for important intellectual content; and has
analysis, patients receiving LABA+ICS showed a decreased risk of provided final approval of the version to be published.
Dr Rodrigo has made substantial contributions to the
asthma exacerbations compared with higher than twice ICS doses
conception and design, acquisition of data, analysis and
(odds ratio = 0.48; 95% confidence interval: 0.28–0.82, P = .007, I2= 0). interpretation of data; has drafted the submitted article and
Children treated with LABA+ICS had significantly higher PEF, less use revised it critically for important intellectual content; and has
of rescue medication, and higher short-term growth than those on provided final approval of the version to be published.

higher ICS doses. There were no other significant differences in This trial has been registered with the International Prospective
Register of Systematic Reviews (
adverse events. PROSPERO/) (CRD42011001435).
CONCLUSIONS: There were no statistically significant group differences
between ICS+LABA and double doses of ICS in reducing the incidence doi:10.1542/peds.2012-0162
of asthma exacerbations but it did decrease the risk comparing to Accepted for publication May 17, 2012
higher than double doses of ICS. Pediatrics 2012;130:e650–e657
Address correspondence to José A. Castro-Rodríguez, MD, PhD,
Departments of Family Medicine and Pediatrics, School of
Medicine, Pontificia Universidad Católica de Chile, Lira 44, 1er.
Piso, casilla 114-D, Santiago, Chile. E-mail:

(Continued on last page)


Downloaded from at Indonesia:AAP Sponsored on June 30, 2014

According to the most commonly used the literature. Therefore, it is important subjects with asthma exacerbations re-
international asthma guidelines,1–3 chil- to know which option (increased doses quiring the use of systemic cortico-
dren with persistent asthma should be of ICS or the addition of LABA) is better steroids. Secondary outcome measures
started on controller therapy with in- for step 3 of the guidelines for children were the following: withdrawals during
haled corticosteroids (ICS) as the pre- when low doses of ICS do not control treatment period, pulmonary function
ferred drug, with leukotriene modifiers their asthma. tests (FEV1 or PEF), days without asthma
(eg, montelukast) as an alternative for The objective of this systematic review symptoms, use of rescue medication,
patients who are unable or unwilling was to assess the safety and efficacy of adverse events (AEs), and severe AEs. A
to use ICS. A recent meta-analysis con- the LABA/ICS combination compared serious AE was defined as any untoward
cludes that children receiving ICS with an increased dose of ICS (double or medical occurrence that sometimes
showed a significantly decreased risk greater) in children and adolescents results in death, is life-threatening, re-
of asthma exacerbation requiring sys- with uncontrolled persistent asthma. quires inpatient hospitalization, or results
temic corticosteroids than children re- in persistent or significant disability/
ceiving montelukast.4 As well, children METHODS incapacity.7
treated with ICS had significantly higher
Search and Selection Criteria
pulmonary function and better clini- Data Abstraction and Assessment
cal parameters compared with those We identified studies from Medline, of Risk of Bias
receiving montelukast.4 Moreover, Embase (search January 2012), and the
This systematic review was performed
the latest study comparing ICS and Cochrane Controlled Trials Register
according to Preferred Reporting Items
montelukast showed that fluticasone (CENTRAL) (search January 2012 data-
for Systematic Reviews and Meta-
(100 mg twice daily) was the most ef- bases using the following medical sub-
analyses guidelines.8 Titles, abstracts,
fective therapy; however, uncontrolled ject headings, full text, and keywords:
and citations were independently ana-
asthma occurred in more than 50% of long-acting b-2 agonists OR salmeterol
lyzed by all reviewers. From the full
the children, and 39% of the children OR formoterol OR indacaterol AND cor-
texts, the reviewers independently as-
had at least 1 asthma exacerbation that ticosteroids OR fluticasone OR budeso-
sessed all studies for inclusion based
was treated with oral corticosteroids nide OR ciclesonide OR mometasone OR
on the criteria for population inter-
during a 48-week period.5 beclomethasone OR flunisolide OR tri-
vention, study design, and outcomes.
amcinolone). The search was then lim-
In cases where ICS is not sufficient to After obtaining full reports about po-
ited with the terms children OR child OR
control the disease in children, in- tentially relevant trials, they assessed
pediatric OR adolescents OR infants OR
ternational guidelines recommend in- eligibility. The authors were indepen-
preschoolers. As well, we performed a
creasing the dose of ICS or adding dently involved in all stages of study
search of relevant unpublished files
leukotriene modifiers or long-acting b selection, data extraction, and risk of
from drug manufacturer databases
agonists (LABAs).1–3 A previous system- bias assessment. The latter was asses-
atic review6 showed that in children, but sed according to recommendations
result_compounds.jsp; http://www.
not in adults, LABA added to ICS had not outlined in the Cochrane Handbook.9; and
significantly reduced the risk of exac- Disagreements were resolved by group
erbations requiring a short course of consensus. In the case of multiple pub-
Trials published solely in abstract form
systemic corticosteroids (relative risk lished or unpublished reports, data
were excluded because the methods
[RR] = 1.28, 95% confidence interval [CI] from the most recent version were
and results could not be fully analyzed.
0.58–2.66) compared with the use of extracted.
The specific inclusion criteria were as
higher doses of ICS. Moreover, children follows: (1) children and adolescents
could be almost 3 times more likely than aged 4 to 18 years with persistent Data Analysis
adults to require oral steroids when asthma and having received ICS daily; The present analysis was done by in-
they were treated with a LABA than with (2) the addition of LABA to ICS com- tention to treat with all participants,
ICS; however, some children included in pared with a higher doses of ICS; (3) including withdrawals, to minimize bias
the meta-analysis came from trials studies with at least 4 weeks’ duration; owing to differences among groups. We
performed in mixed population (chil- (4) randomized (parallel group or cross- calculated the Mantel-Haenszel odds
dren and adults together). over) controlled trials (RCTs) without ratios (ORs) and 95% CIs for binary
In recent years, more studies enrolling language restriction. The primary out- outcomes. When effect estimates were
children exclusively have appeared in come of the study was proportion of significantly different between groups,

PEDIATRICS Volume 130, Number 3, September 2012 e651

Downloaded from at Indonesia:AAP Sponsored on June 30, 2014
the number needed to treat to benefit
or to harm was obtained. Continuous
outcomes were pooled using weighted
mean differences (WMDs) and 95% CIs.
Heterogeneity was measured by the I2
test (,40% could be unimportant, 40%
to 60% could be moderate, and 60% to
100% could be substantial).10 Because
selected studies differed in the mixes
of participants and interventions, a
random-effects meta-analysis was per-
formed to address this variation
across studies in all outcomes.11 We
used a priori subgroup analysis to ex-
plore the influence of the ICS dose
(double versus more than double), type
of LABA (salmeterol versus formoterol),
length of treatment (,24 weeks versus
$24 weeks), age range (4–11 vs 11–17
years), and severity of airway ob-
struction (prebronchodilator FEV1; and
morning and evening PEF from base-
line) Subgroups were compared by
using the interaction test.12 Additional
predefined sensitivity analyses were FIGURE 1
Process of study selection.
done to explore the influence on effect
size of risk of bias (low-risk trials ver-
sus high-risk trials), and the statistical asthma, low doses of ICS (200–500 µg/ (successfully complied with at least 5
model (fixed versus random effects). d beclomethasone dipropionate [BDP] of the 6 domains of bias assessment)
A low-risk bias was defined as a minimum or equivalent). The remaining study (Table 2).
of 5 of 6 domains filled in an acceptable recruited children with mild asthma.15
way. Publication bias of primary out- Almost all studies tested the com- Primary Outcome
comes was evaluated by funnel plots.13 monly recommended doses of LABAs The analysis of 8 studies (n = l616
A P , .05 using a 2-tailed test was (ie, salmeterol 50 mg twice daily, or subjects)14,16–22 showed no statistically
considered to indicate significance. formoterol 9–12 mg twice daily). One significant differences in the number of
Meta-analysis was performed with Re- study used the combination formoterol/ patients with asthma exacerbations
view Manager 5.1.2 software (The Nordic budesonide as maintenance, plus addi- requiring systemic corticosteroids be-
Cochrane Centre, The Cochrane Collab- tional doses as needed.21 Intervention tween children receiving LABA+ICS and
oration, 2011, Copenhagen, Denmark). groups received BDP equivalent doses those receiving higher doses of ICS (OR =
of 400 mg/d in 7 studies14,16–20,22 and 200 0.76; 95% CI: 0.48–1.22, P = .25) (Fig 2).
mg/d in 2 studies.15,21 The dose of ICS There was no evidence of publication
that the control group received was bias (Egger’s test, 0.35; 95% CI: –0.4
Nine RCTs,14–22 involving a total of 1641 twice,14–20 or more than twice, the amount to 0.74) or significant heterogeneity
children and adolescents, fulfilled the received by the LABA/ICS group.21,22 among studies (I2 = 16%). However,
inclusion criteria (Fig 1). One trial was Rescue medications, such as inhaled among the subgroup studies that com-
unpublished.20 All studies examined short-acting b2-agonists and sys- pared LABA+ICS versus higher than a
the combination LABA/ICS in 1 device temic steroids, were permitted in all double dose of ICS, combination therapy
(Table 1). The mean age of participants the trials. Most of the studies14,16–21 significantly reduced the risk of exac-
was 9 years (range 4–17), with 59% were funded by the pharmaceutical erbations (OR = 0.48; 95% CI: 0.28–0.82,
being male. Eight trials14,16–22 included industry. Six studies16–18,20–22 were P = .007, I2 = 0%) (Fig 3B). This difference
subjects with inadequately controlled judged to have a low risk of bias was compatible with a number needed


Downloaded from at Indonesia:AAP Sponsored on June 30, 2014

TABLE 1 Characteristics of Included Studies

Study Design Location and Patients, MeanAge, Atopy Mean Baseline Selected Comparisons
Duration n(% Male) y (Range) status (%) FEV1 (% Predicted)
Verberne14 R,DB,PG Multicenter 120 (63) 11.1 89 88.5 SALM/BDP 50/200 mg BID
54 wk (6–16) versus BDP 400 mg BID
Heuck15 R,DB,CO SC 27 (52) 9.6 NR 88.5 FORM/BUD 12/100 mg BID
6 wk (6–13) versus BUD 200 mg BID
Vaessen- R,DB,PG Multicenter 158 (58) 9.3 75 100 SALM/FLUT 50/100 mg BID
Verberne16 versus FLUT 200 mg BID
(SAM 101667) 26 wk (6–16)
de Blic17 R,DB,PG Multicenter 303 (64) 8.1 88 1.7 liters SALM/FLUT 50/100 mg BID
(SAM 104926) 12 wk (4–11) versus FLUT 200 mg BID
Gappa18 R,DB,PG Multicenter 283 (68) 9.5 NR 91 SALM/FLUT 50/100 mg BID
(VIAPAED 102318) 8 wk (4–16) versus FLUT 200 mg BID
Murray19 R,DB,PG Multicenter 24 (50) 7.3 75 82 SALM/FLUT 50/100 mg BID
(SAM 40100) 6 wk (4–11) versus FLUT 200 mg BID
GSK SAM 4001220 R,DB,PG Multicenter 367 (69) 7.7 75 NR SALM/FLUT 50/100 mg BID
24 wk (4–11) versus FLUT 200 mg BID
Bisgaard21 R,DB,PG Multicenter 224 (69) 8 NR 76 FORM/BUD 4.5/80 mg BID plus
(SD-039-0673) 54 wk (4–11) additional doses as needed
versus BUD 320 mg BID
Lemanske22 R,DB,CO Multicenter 120 (40) 10.9 NR 96 SALM/FLUT 50/100 mg BID
16 wk (6–17) versus FLUT 250 mg BID
BID, twice daily; BUD, budesonide; CO, cross over; DB, double-blind; FEV1= forced respiratory volume in the first second; FLUT, fluticasone; FORM, formoterol; NR, not reported; PG, parallel group;
R, randomized; SALM, salmeterol; SC, single center.

TABLE 2 Risk of Bias of the Eligible Studies

Study Random Sequence Allocation Blinding of Participants Blinding of Outcome Incomplete Outcome Selective
Generation Concealment & Personnel Assessment Data Addressed Reporting
Verberne 199814 Y Y Y Y U N
Heuck 200015 Y U Y Y U N
Vaessen-Verberne 201016 Y Y Y Y U Y
DeBlic 200917 Y Y Y Y Y Y
Gappa 200918 Y Y Y Y Y N
Murray 201019 U U Y Y U N
GSK SAM4001220 Y Y Y Y Y N
Bisgaard 200621 Y Y Y Y Y N
Lemanske 201022 Y Y Y Y Y Y
N, No; U, Unknown; Y, Yes.

to treat of 9 (95% CI: 5–45). Post hoc (OR = 0.53; 95% CI: 0.53–1.40, P = .20). possibility of comparing trials spon-
subgroup analysis showed that sub- Because the number of studies was sored by the pharmaceutical industry
jects in studies testing higher than low, the impact of the baseline severity and independent studies, as only 1 of
twice ICS doses had a significantly of airway obstruction by lung function the 2 independent studies had data on
lower risk of asthma exacerbations and type of LABA on size effect could not exacerbations.
than subjects in studies using a dou- be examined. In the same way, the ef-
ble ICS dose (OR = 0.38;95% CI: 0.37– fect size obtained using random or Secondary Outcomes
0.84, P = .01). fixed effects models did not differ (OR = The addition of LABA to ICS provided
A sensitivity analysis comparing age 0.92; 95% CI: 0.42–2.19, P = .9). Sensi- significantly greater improvements in
range groups (4–11 vs 11–17 years) tivity analysis based on the risk of bias morning PEF from baseline (Fig 4A)
was not possible to do, because the showed different results; trials with (WMD = 8.74; 95% CI: 4.87–12.51 L/min,
studies were not divided into these low risk of bias16–18,20–22 were not as- I2 = 0%) and evening PEF from baseline
2 age categories; in contrast, they had sociated with a significantly low risk of (WMD = 4.41; 95% CI: 1.77–7.05 L/min, I2 =
an age range not mutually exclusive exacerbation (OR = 0.68; 95% CI: 0.42– 0%) at the end point (Fig 4B), compared
(4–11 and 6–17 years). The duration 1.10, I2 = 8%) compared with trials with with higher ICS doses. The duration of
of treatment ($24 weeks versus ,24 high risk of bias14,15,19 (OR = 0.84; 95% interventions did not affect the
weeks) did not influence this effect size CI: 0.12–5.75, I2 = 42%). There was no magnitude of this improvement over

PEDIATRICS Volume 130, Number 3, September 2012 e653

Downloaded from at Indonesia:AAP Sponsored on June 30, 2014
To our knowledge, this is the first meta-
analysis performed of trials exclusively
about child and adolescent populations
to explore the efficacy of ICS+LABA
compared with higher doses of ICS for
uncontrolled persistent asthma. Over-
all, there were no statistically sig-
nificant group differences between
FIGURE 2 ICS+LABA and double or higher doses of
Pooled ORs and 95% CIs for the number of patients with at least 1 asthma exacerbation (with 95% CI) ICS in reducing the incidence of asthma
requiring systemic corticosteroids comparing LABA+ICS versus higher doses of ICS.
exacerbations requiring systemic cor-
time. There were no statistically sig- [–10.99 to 0.93]); (5) AEs (54.6% vs Curiously, comparing 2 trials by using
nificant group differences in pre- 55.6%); and (6) severe AEs (2.0% vs LABA+ICS versus higher than double
bronchodilator FEV1 between LABA+ICS 2.6%) (Table 3). On the other hand, the doses of ICS, significant effects were
versus higher ICS doses (WMD = 0.46; combination of LABA+ICS is associated observed that favor the combination
95% CI: 0.18–1.34 L/s; I2 = 74%, P = .68); with significantly lower, but modest, use therapy in reducing the risk of asthma
however, this information came from of rescue medication (–0.11 puffs/d, exacerbation (number needed to treat
only 3 studies.14,16,17 95% CI: –0.20 to –0.01) (Table 3). Fi- of 9); however, the effect on asthma
There were no significant differences nally, data from 3 trials15,16,21 showed exacerbations was not observed when
between the LABA+ICS and ICS groups in that short-term growth was signifi- trials comparing LABA+ICS versus
the following outcomes: (1) number of cantly greater in children treated with double doses of ICS were analyzed. The
prematurely discontinued patients combination therapy compared with paradoxical effect is biologically diffi-
(4.4% vs 4.1%); (2) withdrawals due to children treated with higher ICS doses cult to explain. Potential explanation
AEs (1.1% vs 1.1%); (3) withdrawals be- (WMD = 0.66 cm/y [95% CI: 0.08–1.25) could be attributable to the inclusion of
cause of asthma exacerbations (0.3% vs (Table 3). In almost all of the variables, 2 particular studies. In the Bisgaard
1.0%); (4) percentage of days free of the degree of heterogeneity was un- et al study,21 1 of the 2 groups with
asthma symptoms (WMD = –5.03% important or null. combination therapy used an adjustable

Pooled ORs and 95% CIs for the number of patients with at least 1 asthma exacerbation (with 95% CI) requiring systemic corticosteroids comparing LABA+ICS
versus double (A) or more than double dose of ICS (B).


Downloaded from at Indonesia:AAP Sponsored on June 30, 2014

Pooled WMD and 95% CIs for the mean change in morning (A) and evening (B) PEF (L/min) from baseline.

TABLE 3 Effect of LABA plus ICS Versus Higher ICS Doses on Secondary Outcomes health care system and for the com-
Outcome Studies n Measure (95% CI) P 2
I munity in general.23 Also, exacerbations
Prematurely discontinued 14–21 1543 OR = 1.0 (0.57 to 1.74) .99 46 are the most important cause of lost
patients school days for asthmatic children.24
Withdrawals owing to 14–15, 17,21 713 OR = 1.01 (0.26 to 3.99) .98 0 Asthma control has 2 aspects: cur-
adverse events
Withdrawals owing to asthma 14,17,21 665 OR = 0.26 (0.04 to 1.63) .15 0 rent control in response to day-to-day
exacerbations symptoms through the use of rescue
Percent of days without 14–16,18–21 1222 WMD = –5.03 (–10.99 to 0.93) .10 0 medications; and the burden imposed
asthma symptoms
Use of rescue medication, 14–15,18–19,21 697 WMD = –0.11 (–0.20 to –0.01) .02 0
by these symptoms, and the risk of
puffs/d asthma exacerbations, irreversible
AEs 14,16–21 1495 OR = 0.95 (0.73 to 1.25) .23 25 decrease in lung function, and side
Serious AEs 14,16–18,20–22 1593 OR = 0.76 (0.39 to 1.49) .43 0
effects from asthma medications.2–25
Linear growth rate, cm/y 15–16,21 430 WMD = 0.66 (0.08 to 1.25) .02 0
Therefore, the prevention of asthma
n, number of subjects.
exacerbations is an important compo-
nent of establishing ideal asthma con-
rather than fixed dose of LABA+ICS meta-analysis, no statistically significant trol. A control trial5 showed that in step
during exacerbations (and probably be- group difference on asthma exac- 2 of asthma management (low ICS
tween exacerbations) or step-up ther- erbation was found between LABA+ICS doses or leukotriene modifiers), more
apy during exacerbation, versus those versus higher doses of ICS. It is im- than 50% of children still have un-
in the group of ICS who received fixed portant to consider that a crossover controlled asthma and 39% have had at
ICS doses, given the possibility that study is probably the best design to least 1 asthma exacerbation that was
children in the latter group received explore individual response to drugs, treated with oral corticosteroids dur-
a lower total ICS dose. And in the however, and that trial22 showed the ing a 48-week period; for that reason it
Lemanske et al study,22 the design was superiority of adding LABA to ICS ver- is very important to prevent exacer-
cross sectional (child received for 16 sus higher doses of ICS in reducing bations. A previous meta-analysis6
weeks LABA+ 200 mg/d of fluticasone asthma exacerbation requiring sys- that included only 3 studies done in
and for 16 weeks 500 mg/d of flutica- temic corticosteroids. children showed a trend toward in-
sone or vice versa, with 4 weeks for Asthma exacerbations are common creased risk of rescue oral steroids
wash-out) given the possibility that the events in asthmatic patients and repre- (RR 1.24, 95% CI: 0.58–2.66) and hos-
wash-out period used was not enough. sent the greatest risk, and the highest pital admission (RR 2.21, 95% CI: 0.74–
When we exclude these 2 studies in our asthma-related treatment cost for the 6.64) associated with combination

PEDIATRICS Volume 130, Number 3, September 2012 e655

Downloaded from at Indonesia:AAP Sponsored on June 30, 2014
therapy versus higher ICS doses. How- by increasing gene transcription. These therapy” showed no presence of LABA
ever, the current study included 8 trials findings could explain the higher perfor- risk.33 However, the FDA called on
done exclusively in children showing mance of the combination of LABA+ICS manufactures of LABA to conduct large
a trend of decreased risk of asthma versus higher doses of ICS. clinical trials to definitively determine
exacerbations requiring systemic cor- We were not able to perform a sub- whether the addition of LABAs to ICS
ticosteroids in the group of LABA+ICS analysis of main outcome comparing increases the risk of serious asthma
versus higher doses of ICS (OR = 0.76; age groups (4–11 vs 11–17 years) be- outcomes.34 Conversely, a recent study32
95% CI: 0.48–1.22, P = .25). The differ- cause trials included in the meta- has summarized nearly 20 systemic
ence may be attributable to the number analysis had overlap in age (4–11 and reviews and databases on LABA safety
and type of studies included. More trials 6–17 years). This is relevant because and showed that there is no risk of
need to be done to definitively lay down an international asthma guideline1 serious asthma-related events when
the best treatment in children with recommends increasing ICS doses first using LABA associated with ICS, par-
persistent asthma. instead of adding LABA in children ticularly when concomitant use of
Another important direct effect of older than 5 years. Moreover, if we LABAs+ICS can be reasonably ensured
asthma exacerbations is the use of found that short-term growth was sig- (combined in a single inhaler). Evi-
rescue medication and lung function nificantly greater in children with dence from RCTs, meta-analysis of
deterioration.26 In the current study, we combination therapy (370 mg/d of BDP RCTs, and observational studies, al-
found a significant modest reduction in or equivalent) compared with children though limited by low statistical
the use of rescue medication among with higher ICS doses (770 mg/d of power, indicate that the use of com-
children on LABA+ICS than those on BDP or equivalent), this difference of bination therapy (LABAs+ICS) in chil-
higher doses of ICS. Also, we found 0.66 cm/y could be important, espe- dren and adults is associated with
a statistically significant but uncertain cially for children in their early years. a decreased risk of serious asthma-
clinically significant improvement in However, long-term growth studies related events.32
lung function (morning and evening need to confirm this finding. Also, it is
PEF) among children/adolescents us- important to consider the strong evi- CONCLUSIONS
ing LABA+ICS compared with those dence of the ICS molecule-dependant This meta-analysis showed no statisti-
using higher doses of ICS. ICS treatment effect on growth.30 cally significant group differences be-
has a plateau, such that increasing the Even though the studies included in the tween ICS+LABA and double doses of
dose does not necessarily improve the present meta-analysis have a wide ICS in reducing the incidence of asth-
clinical response, and systemic effects range of duration (6 to 54 weeks), no ma exacerbations requiring systemic
can start.27 In contrast, the synergistic statistically significant group differ- corticosteroids but it did decrease the
effect of adding LABA to ICS has been ences in AEs and serious AEs were found risk comparing to higher than double
reported,28,29 where LABA, along with between children on LABA+ICS versus doses of ICS. As well, children on com-
its bronchodilator effect, increases the higher doses of ICS. These findings are bination therapy had significantly im-
nuclear translocation of the glucocor- in accordance with the latest Food and proved lung function (morning and
ticoid receptor. At the same time, ICS is Drug Administration (FDA) recommen- evening PEF), reduced use of rescue
delivered in the same device and along dation,31,32 including 1 exclusively done medication and showed less effect on
with its anti-inflammatory effect, it in- in children by the FDA where those short-term linear grow rate than chil-
creases the expression of b2-receptors trials with LABA plus “assigned ICS dren on higher doses of ICS.

1. Global Initiative for Asthma. Global Strategy 2007. Available at: Accessed in children with mild-moderate asthma:
for Asthma Management and Prevention. August 3, 2011 results of a systematic review with meta-
Update 2010. Available at: www.ginasthma. 3. British Thoracic Society Scottish Inter- analysis. Arch Dis Child. 2010;95(5):365–370
com. Accessed August 3, 2011 collegiate Guidelines Network. British guide- 5. Sorkness CA, Lemanske RF Jr, Mauger DT,
2. National Heart, Lung, and Blood Institute. line on the management of asthma. Thorax. et al; Childhood Asthma Research and Ed-
Expert Panel Report 3: Guidelines for the 2008;63(suppl 4):iv1–121 ucation Network of the National Heart, Lung,
Diagnosis and Management of Asthma. 4. Castro-Rodriguez JA, Rodrigo GJ. The role and Blood Institute. Long-term comparison
Bethesda, MD: National Institutes of Health, of inhaled corticosteroids and montelukast of 3 controller regimens for mild-moderate


Downloaded from at Indonesia:AAP Sponsored on June 30, 2014

persistent childhood asthma: the Pediatric growth and collagen turnover in asth- impairment of health related quality of life
Asthma Controller Trial. J Allergy Clin matics treated with inhaled formoterol and in asthma: development of a questionnaire
Immunol. 2007;119(1):64–72 budesonide. Arch Dis Child. 2000;83(4):334– for use in clinical trials. Thorax. 1992;47(2):
6. Ducharme FM, Ni Chroinin M, Greenstone I, 339 76–83
Lasserson TJ. Addition of long-acting beta2- 16. Vaessen-Verberne AAPH, van den Berg NJ, 25. Bateman ED, Hurd SS, Barnes PJ, et al.
agonists to inhaled steroids versus higher van Nierop JC, et al; COMBO Study Group. Global strategy for asthma management
dose inhaled steroids in adults and chil- Combination therapy salmeterol/fluticasone and prevention: GINA executive summary.
dren with persistent asthma. Cochrane versus doubling dose of fluticasone in Eur Respir J. 2008;31(1):143–178
Database Syst Rev. 2010;(4):CD005533 children with asthma. Am J Respir Crit 26. Covar RA, Cool C, Szefler SJ. Progression of
7. International Conference on Harmonisation Care Med. 2010;182(10):1221–1227 asthma in childhood. J Allergy Clin Immu-
of Technical Requirements for Registration 17. de Blic J, Ogorodova L, Klink R, et al. nol. 2005;115(4):700–707
of Pharmaceuticals for Human Use. Clinical Salmeterol/fluticasone propionate vs. dou- 27. Stoloff SW, Kelly HW. Updates on the use of
safety data management: definitions and ble dose fluticasone propionate on lung inhaled corticosteroids in asthma. Curr
standards for expedited reporting. Avail- function and asthma control in children. Opin Allergy Clin Immunol. 2011;11(4):337–
able at: Pediatr Allergy Immunol. 2009;20(8):763– 344
pdf. Accessed July 21, 2011 771 28. Johnson M. Molecular mechanisms of beta
8. Liberati A, Altman DG, Tetzlaff J, et al. The 18. Gappa M, Zachgo W, von Berg A, Kamin W, (2)-adrenergic receptor function, response,
PRISMA statement for reporting systematic Stern-Sträter C, Steinkamp G; VIAPAED and regulation. J Allergy Clin Immunol. 2006;
reviews and meta-analyses of studies that Study Group. Add-on salmeterol compared 117:18–24
evaluate health care interventions: expla- to double dose fluticasone in pediatric 29. Barnes PJ. Scientific rationale for inhaled
nation and elaboration. Ann Intern Med. asthma: a double-blind, randomized trial combination therapy with long-acting beta2-
2009;151(4):W65-94 (VIAPAED). Pediatr Pulmonol. 2009;44(11): agonists and corticosteroids. Eur Respir J.
9. Higgins JPT, Green S. Cochrane handbook 1132–1142 2002;19(1):182–191
for systematic reviews of interventions. 19. Murray CS, Custovic A, Lowe LA, et al. Effect 30. Sharek PJ, Bergman DA. The effect of in-
Version 5.1.0 [updated March 2011]: The of addition of salmeterol versus doubling haled steroids on the linear growth of chil-
Cochrane Collaboration; 2011. Available at: the dose of fluticasone propionate on dren with asthma: a meta-analysis. Pediatrics. Accessed March specific airway resistance in children with 2000;106(1). Available at:
31, 2012 asthma. Allergy Asthma Proc. 2010;31(5): cgi/content/full/106/1/e8
10. Higgins JPT, Thompson SG, Deeks JJ, Altman 415–421 31. FDA Drug Safety Communication. New
DG. Measuring inconsistency in meta-analyses. 20. GlaxoSmithKline Clinical Trial Register. SAM safety requirements for long-acting in-
BMJ. 2003;327(7414):557–560 40012 trial. Available at: http://gsk-clinical- haled asthma medications called Long-
11. Borenstein M, Hedges LV, Higgins JPT, et al. Accessed July 7, 2011 Acting Beta-Agonists (LABAs). Available at:
Introduction to meta-analysis. Chichester 21. Bisgaard H, Le Roux P, Bjåmer D, Dymek A,
(West Sussex), United Kingdom: John Wiley Vermeulen JH, Hultquist C. Budesonide/ DrugSafetyInformationforPatientsandProviders/
& Sons, Ltd.; 2009 formoterol maintenance plus reliever ther- ucm200776.htm#_Ref252304350. Accessed
12. Altman DG, Bland JM. Interaction revisited: apy: a new strategy in pediatric asthma. October 12, 2010
the difference between two estimates. BMJ. Chest. 2006;130(6):1733–1743 32. Chowdhury BA, Dal Pan G. The FDA and safe
2003;326(7382):219 22. Lemanske RF Jr, Mauger DT, Sorkness CA, use of long-acting beta-agonists in the
13. Egger M, Davey Smith G, Schneider M, et al; Childhood Asthma Research and Ed- treatment of asthma. N Engl J Med. 2010;
Minder C. Bias in meta-analysis detected by ucation (CARE) Network of the National 362(13):1169–1171
a simple, graphical test. BMJ. 1997;315 Heart, Lung, and Blood Institute. Step-up 33. McMahon AW, Levenson MS, McEvoy BW,
(7109):629–634 therapy for children with uncontrolled Mosholder AD, Murphy D. Age and risks of
14. Verberne AAPH, Frost C, Duiverman EJ, Grol asthma receiving inhaled corticosteroids. FDA-approved long-acting b₂-adrenergic
MH, Kerrebijn KF; The Dutch Asthma Study N Engl J Med. 2010;362(11):975–985 receptor agonists. Pediatrics. 2011;128(5).
Group. Addition of salmeterol versus dou- 23. Barnett SB, Nurmagambetov TA. Costs of Available at:
bling the dose of beclomethasone in chil- asthma in the United States: 2002-2007. tent/full/128/5/e1147
dren with asthma. Am J Respir Crit Care J Allergy Clin Immunol. 2011;127(1):145– 34. Rodrigo GJ, Castro-Rodríguez JA. Safety of
Med. 1998;158(1):213–219 152 long-acting b agonists for the treatment of
15. Heuck C, Heickendorff L, Wolthers OD. A 24. Juniper EF, Guyatt GH, Epstein RS, Ferrie asthma: clearing the air. Thorax. 2012;67
randomised controlled trial of short term PJ, Jaeschke R, Hiller TK. Evaluation of (4):342–349

(Continued from first page)

PEDIATRICS (ISSN Numbers: Print, 0031-4005; Online, 1098-4275).
Copyright © 2012 by the American Academy of Pediatrics
FINANCIAL DISCLOSURE: Dr Castro-Rodriguez has participated as a lecturer and speaker in scientific meetings and courses under the sponsorship of
AstraZeneca, GlaxoSmithKline, Merck Sharp & Dohme, and Novartis; and Dr Rodrigo has participated as a lecturer and speaker in scientific meetings and courses
under the sponsorship of Admiral, AstraZeneca, Boehringer Ingelheim, Dr Esteve SA, GlaxoSmithKline, and Merck Sharp & Dome.
FUNDING: The funding for this study came from salary support for Drs Castro-Rodriguez and Rodrigo. No sponsorship from institutions or the pharmaceutical
industry was provided to conduct this study.

PEDIATRICS Volume 130, Number 3, September 2012 e657

Downloaded from at Indonesia:AAP Sponsored on June 30, 2014
A Systematic Review of Long-Acting β2-Agonists Versus Higher Doses of
Inhaled Corticosteroids in Asthma
Jose A. Castro-Rodriguez and Gustavo J. Rodrigo
Pediatrics 2012;130;e650; originally published online August 27, 2012;
DOI: 10.1542/peds.2012-0162
Updated Information & including high resolution figures, can be found at:
References This article cites 26 articles, 11 of which can be accessed free
Subspecialty Collections This article, along with others on similar topics, appears in
the following collection(s):
Permissions & Licensing Information about reproducing this article in parts (figures,
tables) or in its entirety can be found online at:
Reprints Information about ordering reprints can be found online:

PEDIATRICS is the official journal of the American Academy of Pediatrics. A monthly

publication, it has been published continuously since 1948. PEDIATRICS is owned, published,
and trademarked by the American Academy of Pediatrics, 141 Northwest Point Boulevard, Elk
Grove Village, Illinois, 60007. Copyright © 2012 by the American Academy of Pediatrics. All
rights reserved. Print ISSN: 0031-4005. Online ISSN: 1098-4275.

Downloaded from at Indonesia:AAP Sponsored on June 30, 2014