Professional Documents
Culture Documents
Jurnal Ebm
Jurnal Ebm
CONTEXT: Inhaled corticosteroids (ICS) are associated with an increased risk of pneumonia in abstract
adult patients with chronic obstructive pulmonary disease.
OBJECTIVE: To assess the association between ICS use and risk of pneumonia and other
respiratory infections in children with asthma.
DATA SOURCES: We searched PubMed from inception until May 2015. We also searched
clinicaltrials.gov and databases of pharmaceutical manufacturers.
STUDY SELECTION: We selected randomized trials that compared ICS with placebo for at least 4
weeks in children with asthma.
DATA EXTRACTION: We included 39 trials, of which 31 trials with 11615 patients contributed data
to meta-analyses.
RESULTS: The incidence of pneumonia was 0.58% (44/7465) in the ICS group and 1.51%
(63/4150) in the placebo group. The meta-analysis of 9 trials that revealed at least 1 event
of pneumonia revealed a reduced risk of pneumonia in patients taking ICS (risk ratio [RR]:
0.65; 95% confidence interval [CI]: 0.44 to 0.94). Using risk difference as effect measure,
the meta-analysis including all 31 trials revealed no significant difference in the risk of
pneumonia between the ICS and placebo groups (risk difference: 0.1%; 95% CI: 0.3% to
0.2%). No significant association was found between ICS and risk of pharyngitis (RR: 1.01;
95% CI: 0.87 to 1.18), otitis media (RR: 1.07; 95% CI: 0.83 to 1.37), and sinusitis (RR: 0.89;
95% CI: 0.76 to 1.05).
LIMITATIONS: Lack of clearly defined criteria for respiratory infections and possible publication
bias.
CONCLUSIONS: Regular use of ICS may not increase the risk of pneumonia or other respiratory
infections in children with asthma.
aPostgraduate Program in Public Health, dPostgraduate Program in Health Science, and bFaculty of Medicine, Federal University of Rio Grande, Rio Grande, Brazil; and cDepartment of
Respiratory and Sleep Medicine, Lady Cilento Childrens Hospital and School of Medicine, The University of Queensland, Brisbane, Australia
Ms Cazeiro conceptualized and designed the study, participated in trial selection, quality assessment, data collection, data analysis and interpretation, and
drafted the protocol and the review article; Dr Silva, Ms Mayer, and Dr Mariany provided input for study conception and design, participated in trial selection,
quality assessment and data collection, and critically revised the manuscript; Dr Wainwright provided input for study conception and design and critically revised
the manuscript; Dr Zhang contributed to study conception and design, participated in trial selection, quality assessment, data collection, data analysis and
interpretation, and critically revised the review protocol and the manuscript; and all authors approved the nal manuscript as submitted.
DOI: 10.1542/peds.2016-3271
To cite: Cazeiro C, Silva C, Mayer S, et al. Inhaled Corticosteroids and Respiratory Infections in Children With Asthma: A Meta-analysis. Pediatrics. 2017;139(3):e20163271
5
6
TABLE 1 Continued
Study ID, Country, Sponsor Participants Study Design Intervention and Control Primary Outcome Adverse Event Outcomes
RTI Reported Systematic Data
Collection
Kerwin et al 2008,51 Patients 6 to 17 y with mild asthma 12-wk, multicenter, RDBPCT Budesonide 180, 200, 720, 800 g/d, FEV1 No Yes
Indonesia, Philippines, DPI (n = 410); Placebo, DPI
Thailand, Singapore, (n = 106)
USA, AstraZeneca
Kooi et al 2008,24 The Children aged 2 to 5 y with asthmalike 6-mo, multicenter, RDBPCT Fluticasone 200 g/d, MDI (n = 25); Symptom score Yes Yes
Netherlands symptoms Placebo, MDI (n = 20)
LaForce et al 2000,25 USA, Children aged 4 to 11 y with persistent asthma 12-wk, multicenter, RDBPCT Fluticasone 200 g/d, DPI (n = 164); FEV1, PEF Yes Yes
GSK Placebo, DPI (n = 78)
Levy et al 2006,26 USA, GSK Children aged 4 to 11 y with 6-mo history of 12-wk, multicenter, RDBPCT Fluticasone 200 g/d, MDI (n = PEF Yes Unclear
asthma requiring pharmacotherapy (FAP30010 trial) 160); Placebo, MDI (n = 81)
MacKenzie et al 1993,27 Children aged 6 to 12 y with asthma 4-wk, multicenter, RDBPCT Fluticasone 100 g/d, DPI (n = 128); Symptom score Yes Unclear
Austria, Belgium, (FLIP20 trial) Placebo, DPI (n = 130)
Finland, Germany,
Greece, Ireland, Israel,
Italy, the Netherlands,
Norway, South Africa,
UK, GSK
Martinez et al 2011,28 USA Patients aged 5 to 18 y with mild persistent 44-wk, multicenter, RDBPCT Beclomethasone 100 g/d, MDI (n = PEF Yes Unclear
asthma (US NAEPP criteria) (NCT00394329 trial) 72); Placebo, MDI (n = 74)
Nayak et al 2002,29 USA, 3M Children aged 5 to 12 y with moderate 12-wk, multicenter, RDBPCT Beclomethasone 100, 200 g/d, MDI FEV1 Yes Yes
Pharmaceuticals symptomatic asthma (n = 237); Placebo, MDI (n = 116)
NCT00163293 trial 2012,41 Children aged 4 to 11 y with mild persistent 12-mo, multicenter, RDBPCT Ciclesonide 200, 400 g/d, MDI (n = Asthma exacerbations Yes Yes
Canada, Hungary, South asthma. 155); Placebo, MDI (n = 85)
Africa, Takeda
NCT00392288 trial 2012,42 Children aged 4 to <12 y with persistent 12-wk, multicenter, RDBPCT Ciclesonide 100, 200 g/d, MDI (n = FEV1 Yes Yes
USA, Hungary, Mexico, asthma 349); Placebo, MDI (n = 172)
Poland, Russian
CAZEIRO et al
TABLE 1 Continued
Study ID, Country, Sponsor Participants Study Design Intervention and Control Primary Outcome Adverse Event Outcomes
RTI Reported Systematic Data
Collection
Pedersen et al 2010,31 Children aged 6 to 11 y with persistent asthma 12-wk, multicenter, RDBPCT Ciclesonide 50, 100, 200 g/d, MDI PEF Yes Yes
Bulgaria, Germany, (RAINBOW trial) (n = 930); Placebo, MDI (n = 150)
Hungary, Poland,
Romania, Russia, South
Africa, Spain, Ukraine,
Nycomed
Qaqundah et al 2006,32 Children aged 1 to <4 y with asthma 12-wk, multicenter, RDBPCT Fluticasone 200 g/d, MDI (n = Symptom score Yes Yes
Argentina, Chile, USA, 239); Placebo, MDI (n = 120)
GSK
7
TABLE 2 Risk of Pneumonia Associated With Use of ICS in Children With Asthma Using RD as effect estimate, we
Study ID Participants Affected/ Treatment Effect (95% CI) conducted the meta-analysis including
Participants Randomly all 31 trials that revealed at least 1
Assigned respiratory infection event. Twenty-
ICS Placebo RR Petos OR RD three trials1622,2426,28,29,3137,39,42,45,46
Allen et al 199816 0/219 0/106 UC UC 0 (0.01 to 0.01) did not explicitly reveal pneumonia as
Becker et al 200617 0/119 0/121 UC UC 0 (0.02 to 0.02) adverse events, and we considered no
Berger et al 200518 0/92 0/49 UC UC 0 (0.03 to 0.03) events of pneumonia in both treatment
Berger et al 200619 0/197 0/99 UC UC 0 (0.02 to 0.02) arms. Thus, 44 patients in the
Carlsen et al 200520 0/79 0/81 UC UC 0 (0.02 to 0.02)
de Blic et al 199621 0/20 0/18 UC UC 0 (0.09 to 0.09)
corticosteroid group (44/7465, 0.58%)
FLTA2007 200540 0/179 2/83 0.09 (0.01 to 0.04 (0.002 to 0.024 (0.06 to 0.01) and 63 patients in the placebo group
1.92) 0.83) (63/4150, 1.51%) had pneumonia
FMS30059 200546 2/142 0/69 2.45 (0.12 to 4.45 (0.23 to 0.01 (0.02 to 0.04) as adverse events. The summary RD
50.29) 85.99)
between the corticosteroid and placebo
Gelfand et al 200622 0/764 0/254 UC UC 0 (0.01 to 0.01)
Katz et al 199823 1/172 0/93 1.63 (0.07 to 4.67 (0.08 to 0.01 (0.01 to 0.03) groups was 0.1% (95% CI: 0.3% to
39.62) 283.50) 0.2%, P = .72, I2 = 0%; Table 2). The
Kooi et al 200824 0/25 0/20 UC UC 0 (0.08 to 0.08) posthoc sensitivity analysis excluding
LaForce et al 0/164 0/78 UC UC 0 (0.02 to 0.02) 8 trials17,19,21,2628,40,44 in which
200025
adverse events were not collected by
Levy et al 200626 0/160 0/81 UC UC 0 (0.02 to 0.02)
Mackenzie et al 0/128 3/130 0.15 (0.01 to 0.14 (0.01 to 0.02 (0.05 to 0.01) systematic assessment yielded the
199327 2.78) 1.31) similar results.
Martinez et al 0/143 0/74 UC UC 0 (0.02 to 0.02)
201128 Use of ICS and Risk of Other
NCT00163293 1/155 0/84 1.64 (0.07 to 4.67 (0.07 to 0.01 (0.02 to 0.03) Respiratory Infections
201241 39.69) 283.50)
Figure 2 reveals the results of meta-
NCT00392288 0/174 0/172 UC UC 0 (0.01 to 0.01)
201242 analyses of secondary outcomesother
NCT00569192 1/249 0/111 1.34 (0.06 to 4.25 (0.06 to 0.004 (0.01 to 0.02) respiratory infections. No significant
201343 32.74) 295.90) association was found between use of
NCT00920959 0/304 1/300 0.33 (0.01 to 0.13 (0.003 to 0.003 (0.01 to 0.01) ICS and risk of pharyngitis (23
201344 8.04) 6.73)
trials,16,1820,22,23,26,27,29,31,3342,4446
NCT01136382 0/152 0/152 UC UC 0 (0.01 to 0.01)
201445 RR: 1.01, 95% CI: 0.87 to
Nayak et al 200229 0/237 0116 UC UC 0 (0.01 to 0.01) 1.18, I2 = 19%), otitis media
Peden et al 199830 1/351 0/86 0.74 (0.03 to 3.48 (0.03 to 0.003 (0.01 to 0.02) (17 trials,16,1820,23,26,27,30,32,33,3741,44,46
18.05) 480.40) RR: 1.07, 95% CI: 0.83 to 1.37,
Pedersen et al 0/927 0/146 UC UC 0 (0.01 to 0.01)
201031
I2 = 34.2%), sinusitis (15
Qaqundah et al 0/239 0/120 UC UC 0 (0.01 to 0.01) trials,16,18,22,25,26,29,30,34,3741,44,46 RR:
200632 0.89, 95% CI: 0.76 to 1.05, I2 = 0%),
Schokker et al 0/48 0/48 UC UC 0 (0.04 to 0.04) bronchitis (9 trials,16,23,27,28,32,3739,46
200833 RR: 0.89, 95% CI: 0.75 to 1.05, I2 =
Shapiro et al 0/183 0/91 UC UC 0 (0.02 to 0.02)
200134
0%), and influenza (4 trials,16,23,41,45
Silverman et al 38/1004 57/977 0.65 (0.43 to 0.64 (0.43 to 0.02 (0.04 to RR: 1.12, 95% CI: 0.65 to 1.93, I2 =
200638 0.97) 0.96) 0.00) 0%). Posthoc subgroup analyses did
Skoner et al 200835 0/440 0/221 UC UC 0 (0.01 to 0.01) not reveal significant effects of drug
Skoner et al 201036 0/38 0/12 UC UC 0 (0.11 to 0.11) molecules, delivery devices, treatment
Skoner et al 201137 0/142 0/45 UC UC 0 (0.03 to 0.03)
Wasserman et al 0/219 0/113 UC UC 0 (0.01 to 0.01)
duration, patients age, and data source
200639 on the treatment effect size of ICS
All trials combined 44/7465 63/4150 0.65 (0.44 to 0.63 (0.43 to 0.001 (0.003 to (Table 3).
0.94) 0.93) 0.002)
Use of ICS and Risk of Respiratory
UC, unable to calculate using RR and Petos OR because no events in either group.
Infections: DoseResponse
Relationship
association between use of ICS and sensitivity analysis, excluding 3
risk of pneumonia (RR: 0.62, 95% CI: trials27,40,44 in which adverse events Seventeen trials16,18,19,22,23,2931,
0.21 to 1.86, P = .39, I2 = 0%; Petos were not collected by systematic 3437,39,4143,46 in which 2 or more
OR: 0.57, 95% CI: 0.17 to 1.89, P = assessment did not significantly daily doses of ICS were compared
.36, I2 = 31.6%). In another posthoc change the results. had contributed data to the
9
FIGURE 2
Use of ICS and risk of other respiratory infections. RR > 1 favors placebo, suggesting a higher risk of respiratory infections in the ICS group. Weights are
from random effects analysis. ES, effect size; RTI, respiratory tract infection. a Rate: cases/participants at risk.
meta-analyses. Five trials16,23,30,39,46 associated with lower risk of otitis of pneumonia. The meta-analysis of
used fluticasone (100 vs 200 g/ media (9 trials, RR: 0.73, 95% CI: 9 trials using RR as effect measure
day), 3 trials18,34,43 used budesonide 0.57 to 0.93, I2 = 0%). reveals that use of ICS could result
(200 to 500 vs 400 to 1000 g/ in a 35% reduction in the risk of
day), 1 trial29 used beclomethasone pneumonia when compared with
(100 vs 200 g/day), 3 trials19,36,37 DISCUSSION placebo. The meta-analysis using
used mometasone (100 vs 200 g/ Petos OR as effect measure yields
This systematic review and meta-
day), and 5 trials22,31,35,41,42 used a similar result. These results are
analysis of randomized trials suggests
ciclesonide (50 to 100 vs 100 to 200 consistent with those found in a
that regular use of ICS is unlikely
g/day). Three trials compared 3 meta-analysis9 of 10 trials in which
to increase the risk of pneumonia
daily doses, of which 2 trials22,31 the majority of participants were
or other respiratory infections in
used ciclesonide (50, 100, and adults with asthma. However, caution
children with asthma. Drug molecules,
200 g/day), and 1 trial36 used daily doses, delivery devices, should be taken when interpreting
mometasone (200, 400, and 800 g/ treatment duration, and patients age these results. Firstly, meta-analysis
day). In these 3 trials, we combined appear to have no significant impact may overestimate the incidence of
2 higher doses into a single group. on the size of risk of respiratory pneumonia if excluding the trials in
The meta-analyses did not reveal infections related to use of ICS. which no events of pneumonia were
a significant doseresponse reported in either arm. Secondly,
relationship between ICS and risk In this review, we used 2 data sets 1 trial38 contributed with 89% of
of respiratory infections (Fig 3). and 3 different effect size measures the weight to the meta-analysis
Moreover, an inverse relationship (RR, Petos OR, and RD) to estimate and the results were no longer
was observed between dose of ICS the risk of pneumonia due to ICS. statistically significant when this
and risk of otitis media, that is, The first data set includes only 9 trial was excluded in the sensitivity
higher dose of ICS was significantly trials that revealed at least 1 event analysis. However, exclusion
of this trial changed the upper arms if pneumonia had not been that the method used for dealing with
limit of 95% CI from 0.94 to 1.86, explicitly reported. The rationale possibly no events in both arms is
suggesting a potential increase of for this strategy is that no events of not yet validated, but no other better
risk of pneumonia on the basis of pneumonia in either treatment arm method is currently available.
the results of smaller trials. Thirdly, were likely not to be mentioned in
asthma exacerbations could be easily the trial reports because the majority ICS may affect local immunity
misclassified as pneumonia given of trials were designed to assess the and microbial community in the
the lack of clearly defined criteria for efficacy of ICS and pneumonia was respiratory tract.55 One cross-
pneumonia in the trials and similar rarely considered an a priori safety sectional study demonstrates that
clinical manifestations between end point. children with asthma regularly taking
pneumonia and asthma exacerbation. ICS are almost 4 times more likely
On the other hand, if the trial to have oropharyngeal colonization
The second data set consists of all revealed at least 1 respiratory by S pneumoniae than those not
31 trials that revealed at least 1 infection event, it is unlikely taking such drugs when adjusted
respiratory infection event, even that pneumonia, a more severe for potential confounders.10 It could
if pneumonia was not explicitly respiratory infection, could be be expected that children with an
mentioned. Using RD as effect ignored. Taking all together, we increased microbial burden in the
measure, the meta-analysis of believe that the meta-analysis on the oropharyngeal area may be at higher
31 trials reveals no significant basis of second data set may provide risk for respiratory infections, not
association between use of ICS more accurate estimates, suggesting only pneumonia but also other
and risk of pneumonia in children no significant association between respiratory infections. However,
with asthma. In this database, use of ICS and risk of pneumonia in this systematic review does not
we considered no events in both children with asthma. We recognize find a significantly increased risk
REFERENCES
1. Wilkie M, Finch S, Schembri S. 10. Zhang L, Prietsch SO, Mendes AP, 56-week randomized double-blind
Inhaled corticosteroids for chronic et al. Inhaled corticosteroids increase study. Ann Allergy Asthma Immunol.
obstructive pulmonary diseasethe the risk of oropharyngeal colonization 2006;96(6):800807
shifting treatment paradigm. COPD. by Streptococcus pneumoniae in
18. Berger WE, Qaqundah PY, Blake K,
2015;12(5):582590 children with asthma. Respirology.
et al. Safety of budesonide inhalation
2013;18(2):272277
2. Thomson NC, Spears M. Inhaled suspension in infants aged six to
corticosteroids for asthma: on-demand 11. Liberati A, Altman DG, Tetzlaff J, et al. twelve months with mild to moderate
or continuous use. Expert Rev Respir The PRISMA statement for reporting persistent asthma or recurrent
Med. 2013;7(6):687699 systematic reviews and meta-analyses wheeze. J Pediatr. 2005;146(1):9195
of studies that evaluate healthcare
3. Barnes PJ, Pedersen S, Busse 19. Berger WE, Milgrom H, Chervinsky P, et al.
interventions: explanation and
WW. Efcacy and safety of inhaled Effects of treatment with mometasone
elaboration. BMJ. 2009;339:b2700
corticosteroids. New developments. furoate dry powder inhaler in children
Am J Respir Crit Care Med. 1998; 12. Zhang L, Cazeiro C, Silva C, Mayer S, with persistent asthma. Ann Allergy
157(3 pt 2):S1S53 Mariany V. Inhaled corticosteroids Asthma Immunol. 2006;97(5):672680
and risk of pneumonia in children
4. Allen DB. Safety of inhaled 20. Carlsen KC, Stick S, Kamin W, Cirule I,
with asthma. PROSPERO 2015:
corticosteroids in children. Pediatr Hughes S, Wixon C. The efcacy and
CRD42015020656. Available at: www.
Pulmonol. 2002;33(3):208220 safety of uticasone propionate in
crd.york.ac.uk/PROSPERO/display_
5. Kew KM, Seniukovich A. Inhaled record.asp?ID=CRD42015020656 very young children with persistent
steroids and risk of pneumonia asthma symptoms. Respir Med.
13. Virtual Health Library of the Latin 2005;99(11):13931402
for chronic obstructive pulmonary
American and Caribbean Center on
disease. Cochrane Database Syst Rev. 21. de Blic J, Delacourt C, Le Bourgeois M,
Health Sciences Information (BIREME).
2014;3(3):CD010115 et al. Efcacy of nebulized budesonide
Available at: http://bvsalud.org.
6. Singh S, Amin AV, Loke YK. Long-term Accessed May 25, 2015 in treatment of severe infantile
use of inhaled corticosteroids and asthma: a double-blind study. J Allergy
14. Higgins JPT, Green S, eds. Cochrane Clin Immunol. 1996;98(1):1420
the risk of pneumonia in chronic Handbook for Systematic Reviews
obstructive pulmonary disease: a of Interventions. Version 5.1.0 22. Gelfand EW, Georgitis JW, Noonan M,
meta-analysis. Arch Intern Med. [updated March 2011]. The Cochrane Ruff ME. Once-daily ciclesonide in
2009;169(3):219229 Collaboration; 2011. Available at: www. children: efcacy and safety in asthma.
7. Drummond MB, Dasenbrook handbook.cochrane.org. Accessed May J Pediatr. 2006;148(3):377383
EC, Pitz MW, Murphy DJ, Fan E. 10, 2015
23. Katz Y, Lebas FX, Medley HV, Robson R;
Inhaled corticosteroids in patients 15. Higgins JPT, Thompson SG, Fluticasone Propionate Study Group.
with stable chronic obstructive Deeks JJ, Altman DG. Measuring Fluticasone propionate 50 micrograms
pulmonary disease: a systematic inconsistency in meta-analyses. BMJ. BID versus 100 micrograms BID in the
review and meta-analysis. JAMA. 2003;327(7414):557560 treatment of children with persistent
2008;300(20):24072416 asthma. Clin Ther. 1998;20(3):424437
16. Allen DB, Bronsky EA, LaForce CF,
8. OByrne PM, Pedersen S, Carlsson et al; Fluticasone Propionate Asthma 24. Kooi EM, Schokker S, Marike Boezen H,
LG, et al. Risks of pneumonia in Study Group. Growth in asthmatic et al. Fluticasone or montelukast for
patients with asthma taking inhaled children treated with uticasone preschool children with asthma-like
corticosteroids. Am J Respir Crit Care propionate. J Pediatr. 1998; symptoms: Randomized controlled
Med. 2011;183(5):589595 132(3 pt 1):472477 trial. Pulm Pharmacol Ther.
9. Bansal V, Mangi MA, Johnson MM, 17. Becker AB, Kuznetsova O, Vermeulen 2008;21(5):798804
Festic E. Inhaled corticosteroids J, et al; Pediatric Montelukast 25. LaForce CF, Pearlman DS, Ruff ME, et al.
and incident pneumonia in patients Linear Growth Study Group. Linear Efcacy and safety of dry powder
with asthma: Systematic review growth in prepubertal asthmatic uticasone propionate in children with
and meta-analysis. Acta Med Acad. children treated with montelukast, persistent asthma. Ann Allergy Asthma
2015;44(2):135158 beclomethasone, or placebo: a Immunol. 2000;85(5):407415
The online version of this article, along with updated information and services, is
located on the World Wide Web at:
/content/139/3/e20163271.full.html