Preventing Exacerbations in Preschoolers With Recurrent Wheeze: A Meta-analysis
CONTEXT: Half of children experience wheezing by age 6 years, and optimal strategies for preventing severe exacerbations are not well defined.
OBJECTIVE: Synthesize the evidence of the effects of daily inhaled corticosteroids (ICS), intermittent ICS, and montelukast in preventing severe exacerbations among preschool children with recurrent wheeze.
DATA SOURCES: Medline (1946, 2/25/15), Embase (1947, 2/25/15), CENTRAL.
STUDY SELECTION: Studies were included based on design (randomized controlled trials), population (children ≤6 years with asthma or recurrent wheeze), intervention and comparison (daily ICS vs placebo, intermittent ICS vs placebo, daily ICS vs intermittent ICS, ICS vs montelukast), and outcome (exacerbations necessitating systemic steroids).
DATA EXTRACTION: Completed by 2 independent reviewers.
RESULTS: Twenty-two studies (N = 4550) were included. Fifteen studies (N = 3278) compared daily ICS with placebo and showed reduced exacerbations with daily medium-dose ICS (risk ratio [RR] 0.70; 95% confidence interval [CI], 0.61–0.79; NNT = 9). Subgroup analysis of children with persistent asthma showed reduced exacerbations with daily ICS compared with placebo (8 studies, N = 2505; RR 0.56; 95% CI, 0.46–0.70; NNT = 11) and daily ICS compared with montelukast (1 study, N = 202; RR 0.59; 95% CI, 0.38–0.92). Subgroup analysis of children with intermittent asthma or viral-triggered wheezing showed reduced exacerbations with preemptive high-dose intermittent ICS compared with placebo (5 studies, N = 422; RR 0.65; 95% CI, 0.51–0.81; NNT = 6).
LIMITATIONS: More studies are needed that directly compare these strategies.
CONCLUSIONS: There is strong evidence to support daily ICS for preventing exacerbations in preschool children with recurrent wheeze, specifically in children with persistent asthma. For preschool children with intermittent asthma or viral-triggered wheezing, there is strong evidence to support intermittent ICS for preventing exacerbations.
- BID —
- twice daily
- CI —
- confidence interval
- EVW —
- episodic viral wheeze
- ICS —
- inhaled corticosteroids
- NNT —
- number needed to treat
- RR —
- risk ratio
- RCT —
- randomized controlled trial
- SFDs —
- symptom-free days
- URTI —
- upper respiratory tract infection
Half of all children experience one or more episodes of wheezing by 6 years of age,1 and these wheezing episodes lead to substantial morbidity, caregiver burden, and health care costs.2 In the United States, annual direct health care costs due to asthma in children exceed $50 billion,3 and rates of asthma-related emergency department visits and hospitalizations are highest among preschool children.4
Optimal strategies for preventing severe asthma exacerbations in this population are not well defined. The 2007 National Asthma Education and Prevention Program guidelines recommend that preschool children be classified in terms of asthma severity, and for those with persistent asthma, daily inhaled corticosteroids (ICS) be initiated to prevent severe exacerbations.5 Daily ICS have been shown to significantly reduce exacerbations in preschool children, especially those with persistent symptoms.6 However, there are concerns about effects on linear growth with prolonged treatment,5 and ICS do not modify the development of asthma or improve lung function after discontinuation.7
The majority of preschool children with recurrent wheezing have intermittent, but sometimes severe, exacerbations triggered by viral upper respiratory tract infections (URTIs) and minimal symptoms between exacerbations.1 This pattern of illness has been called episodic viral wheeze (EVW)8 or severe intermittent wheezing.9 Although wheezing patterns and phenotypes in young children can change over a short time,10 recent studies have examined phenotype-directed strategies for preventing severe exacerbations, as well as alternative strategies to daily ICS. Two alternative strategies include intermittent (started at the onset of URTI) ICS and the leukotriene inhibitor montelukast. Both of these strategies offer potential advantages to health care providers and caregivers by mitigating the burden and risks of daily ICS. The more recent 2015 Global Initiative for Asthma guideline, which integrates these recent studies, recommends considering intermittent ICS for preschool children with EVW.11
Because of the complexity of managing preschool children with recurrent wheeze, substantial practice variation exists regarding choice of therapy for preventing severe exacerbations.12 Given the magnitude of disease burden and health care costs of recurrent wheezing in preschoolers, it is paramount that we determine the optimal therapeutic strategy for preventing severe exacerbations in these children. The primary objective of this systemic review and meta-analysis is to synthesize the evidence of the effects of daily ICS, intermittent ICS, and montelukast as strategies for preventing severe exacerbations in preschool children with recurrent wheeze. Our secondary objective is to synthesize the evidence of the effects of these preventive strategies in specific phenotypes of preschool children with recurrent wheeze. Our work is intended to update and build on a recent review of the diagnosis, management, and prognosis of preschool wheeze by Ducharme et al.13 These data should assist all practitioners who provide primary care to young children, provide subspecialty care to children with recurrent wheezing, and provide care for children during acute exacerbations.
We conducted and reported this systematic review in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses statement.14 We did not register a protocol.
In consultation with a medical librarian, we created search strategies for 3 databases (Medline, Embase, and CENTRAL) from inception to February 2015. The detailed search strategy for Medline and Embase is outlined in the Supplemental Appendix. Briefly, the search used terms for glucocorticoids (glucocorticoid/ and detailed listing of all indexed drugs within the glucocorticoid category), montelukast, asthma (asthma/ or status asthmaticus/ or asthma* or reactive adj2 airway* or wheez*), and inhaled (inhalational or nebulizer/ or vaporizer/ or inhaler/ and related terms), limited to studies of humans and children. We also searched abstracts of the Pediatric Academic Societies (2002–2014) and American Academy of Allergy, Asthma, and Immunology conference proceedings (1996–2015), reference lists of all included papers and relevant reviews identified, and the top 200 citations from Google Scholar (using terms “asthma,” “wheeze,” “child,” “steroid,” and “montelukast”).
Study Inclusion Criteria
Studies were considered eligible for inclusion if they met criteria regarding population, intervention and comparator, outcomes, and study design. Participants were children ages ≤6 years with asthma or recurrent wheezing (≥2 episodes in last year). Studies that included only children <2 years were excluded because of the potential overlap with bronchiolitis in this age group.15,16 We included studies comparing the following interventions: daily ICS versus placebo, intermittent ICS versus placebo, daily ICS versus intermittent ICS, or any regimen of ICS versus any regimen of montelukast. We included any studies that reported on our outcome: severe wheezing exacerbations necessitating systemic (oral or intravenous) corticosteroid. Severe exacerbations were chosen as our primary outcome because they are a patient-important outcome that have significant consequences for children, caregivers, and the health care system.17 Only randomized controlled trials (RCTs) were included. Guidelines, reviews, commentaries, abstracts, and letters to editors were reviewed to identify any primary data; however, these publication types were not included because of lack of peer review and inability to judge bias.
Study Selection Process
All titles and abstracts were pooled in EndNote (Thompson Reuters, Philadelphia, PA), and duplicates were deleted. Two review authors independently screened all titles and abstracts to assess which studies met the inclusion criteria. We retrieved full-text copies of all potentially relevant articles for review. Unresolved disagreements on inclusion were referred to a third review author. The Supplemental Appendix includes studies that were excluded, and Fig 1 outlines the study selection process.
Data Extraction and Management
Data were extracted via a standardized data extraction form. Study design, patient characteristics (age, gender, atopy, family history), intervention (dose, frequency, duration), intervention and comparator groups, methodological quality, and key outcomes were noted. Corresponding authors were contacted for information not available in the journal article. This process was not blinded to manuscript origin (journal, authors, institution).
Assessment of Risk of Bias in Included Studies
Methodological quality of all studies was assessed with the Cochrane Risk of Bias tool, which assesses studies for risks of bias in selection, performance, detection, attrition, or reporting.18 This process was not blinded to manuscript origin. Study quality assessments were incorporated into a sensitivity analysis and the final conclusions.
Measures of Treatment Effect
For rates of severe wheezing exacerbations necessitating systemic steroids, we collected numbers of participants in each group with and without the outcome and determined pooled risk ratios (RRs) with 95% confidence intervals (CIs). To determine whether to pool results, we assessed for clinical heterogeneity by detailed consideration of each study (design, patient characteristics, intervention, comparison, outcomes, and conduct of study) and assessed statistical heterogeneity by visual inspection of forest plots and calculation of Cochran’s χ2 test of homogeneity and I2 test statistic. A fixed-effects model was used for the meta-analysis unless statistical heterogeneity was identified (Cochran’s χ2 test P ≤ .05 or I2 ≥50%). Analyses were done in Review Manager 5.3 (Copenhagen, Denmark). Publication bias was assessed with funnel plots.
Subgroup and Sensitivity Analyses
To determine the efficacy of daily ICS, intermittent ICS, and montelukast for specific phenotypes of preschool wheeze, we performed subgroup analyses. Descriptions of each study population’s baseline symptoms were carefully reviewed to determine phenotypic classification. We performed 1 subgroup analysis restricted to studies that described inclusion only of children with persistent asthma (symptoms >2 days/week, nighttime awakenings 1–2/month, short acting β-agonist use >2 days/week, or minor limitation with normal activity).5 We performed another subgroup analysis that described inclusion only of children with intermittent asthma (symptoms ≤2 days/week, no nighttime awakenings, short acting β-agonist use ≤2 days/week, and no limitation with normal activity)5 or viral-triggered wheezing and minimal symptoms between exacerbations (EVW or severe intermittent wheezing). Studies in which the phenotypes of included children were mixed or unclear were analyzed as a separate subgroup. We also performed a sensitivity analyses in which we excluded studies with high risk of bias in ≥1 domain or crossover design to determine whether effect size and direction was consistent with our primary analysis.
Description of Studies
Results of Search
Our search identified 4290 references. After removing duplicates and screening abstracts, we selected 123 for full-text review. Of these, 101 were excluded upon full-text review. Reasons for exclusion are described in the Supplemental Appendix. Twenty-two studies met inclusion criteria; all were included in the quantitative synthesis and meta-analysis (Fig 1).7,19–39
Characteristics of included studies are described in Table 1.
Fifteen studies compared daily ICS with placebo.7,20–22, 24–26, 28–32,35–37 All were double-blind RCTs. Thirteen studies had a parallel design, and 2 had crossover designs: Gleeson and Price28 and Webb et al36 (3-week washout period and no washout, respectively). The studies used several different delivery systems and types of ICS, with most studies using medium daily doses (budesonide 0.4 mg/day, fluticasone 0.2 mg/day, beclomethasone 0.15 mg/day, ciclesonide 0.16 mg/day, flucinolide 40 μg/kg/day). Study durations ranged from 6 weeks to 5 years, with most 12 weeks. Eight of 15 studies focused on preschool children with persistent asthma.20–22,24,28, 31,32,35 Only 1 study examined daily ICS for children with intermittent asthma or viral-triggered wheeze.37 A funnel plot of these studies was symmetric, suggesting no evidence of publication bias.
Six studies compared intermittent ICS with placebo.19,23,27, 30,33,38 All of these studies were double-blind RCTs. Two had a crossover design: Connett and Lenney23 (participants switched treatment arms after each URTI) and Wilson and Silverman38 (participants switched treatment arms after 2 URTIs). The studies used several different delivery systems and types of ICS at high dosages (budesonide 1.6–2 mg/day, fluticasone 1.5 mg/day, beclomethasone 2.3 mg/day). Study durations ranged from 12 to 52 weeks. Five studies focused on preschool children with intermittent asthma or viral-triggered wheeze.19,23, 27,33,38 A funnel plot of these studies was symmetric, suggesting no evidence of publication bias.
Two studies directly compared daily ICS with intermittent ICS.30,39 Both were double-blind RCTs with parallel designs. Papi et al30 compared high-dose daily (0.4 mg twice daily [BID]) or intermittent beclomethasone (0.8 mg as needed) to placebo for 12 weeks. Zeiger et al39 compared high-dose daily (0.5 mg daily) with intermittent budesonide (1 mg BID) for 52 weeks in children with viral-triggered asthma and positive modified asthma predictive index.40
Two studies compared ICS with montelukast.19,34 Both were RCTs with parallel design. Bacharier et al19 compared intermittent ICS (budesonide 1 mg BID) with intermittent montelukast (4 mg daily) over 52 weeks in children with intermittent wheezing. Szefler et al34 compared daily ICS (budesonide 0.5 mg BID) with daily montelukast (4 mg daily) over 52 weeks in children with persistent asthma.
Risk of Bias and Quality
The risk of bias in all included studies is illustrated in Fig 2. Twelve of 22 studies had low risk of bias. The most common area of concern was attrition bias due to incomplete outcome data. Studies with >20% loss to follow-up were rated as high risk.19,25,27–29, 34,36,38,39 Most studies did not describe how they handled missing data. Szefler et al34 was rated high risk of performance and detection bias because of open-label design. Wilson and Silverman38 was rated high risk because of potential confounding bias, because 13/24 participants were on daily controller therapy during the study, which could include ICS.
Risk of Severe Wheezing Exacerbations Necessitating Systemic Steroids
Our meta-analyses of strategies for preventing severe exacerbations in preschool children with recurrent wheeze are described in Fig 3. Data from 15 studies (N = 3278) showed a significant reduction in rates of exacerbations with daily ICS compared with placebo (12.9% and 24.0%, respectively; RR 0.70; 95% CI, 0.61–0.79; P < .001; I2 = 42%). Treatment of 9 children with daily ICS prevented 1 child from experiencing an exacerbation (number needed to treat [NNT] = 9; 95% CI, 7–12).
Data from 6 studies (N = 588) showed a significant reduction in rates of severe exacerbations with intermittent ICS compared with placebo (24.8% and 41.6%, respectively; RR 0.64; 95% CI, 0.51–0.81; P < .001; I2 = 0%). Treatment of 6 children with intermittent ICS therapy prevented 1 child from experiencing an exacerbation (NNT = 6; 95% CI, 4–11).
Data from 2 studies (N = 498) directly comparing daily with intermittent ICS showed no differences in rates of severe exacerbations (25.7% and 28.1%, respectively; RR 0.91; 95% CI, 0.71–1.18; P = .49, I2 = 43%).
Bacharier et al19 (n = 190) showed no significant differences in rates of severe exacerbations comparing intermittent ICS to intermittent montelukast (38.5% and 46.8%, respectively; RR 0.82; 95% CI, 0.59–1.15; P = .25). Szefler et al34 (n = 202) showed a significant reduction in rates of severe exacerbations with daily ICS versus daily montelukast (21.9% and 37.1%, respectively; RR 0.59; 95% CI, 0.38–0.92; P = .02).
We performed sensitivity analyses excluding studies with high risk of bias in ≥1 domain. Findings were similar to our primary analysis for 2 comparisons, with daily ICS better than placebo (5/15 studies excluded, RR 0.67; 95% CI, 0.58–0.77) and no differences comparing daily ICS versus intermittent ICS (1/2 studies excluded, RR 0.33; 95% CI, 0.07–1.62). With the exclusion of 3 out of 6 studies comparing intermittent ICS with placebo, the benefit of intermittent ICS was no longer statistically significant (RR 0.61; 95% CI, 0.35–1.07). Both studies comparing ICS with montelukast had high risk of bias in ≥1 domains. We also performed a sensitivity analysis excluding only the 4 studies with crossover design and found very similar results to our primary analysis.
Subgroup Analyses: Intermittent Asthma or Viral-Triggered Wheeze
Our subgroup analyses of preschool children with intermittent asthma or viral-triggered wheeze are described in Fig 4. There was only 1 study (n = 41) examining daily ICS versus placebo, which found no significant benefit (RR 1.05; 95% CI, 0.16–6.76). Most studies comparing intermittent ICS with placebo were focused on children with intermittent asthma or viral-triggered wheeze (5/6). Data from these 5 studies (N = 422) showed significant reduction in rates of severe exacerbations with intermittent ICS (33.9% vs 51.3%, respectively; RR 0.65; 95% CI, 0.51–0.81; P = .0002; I2 = 0%). Treatment of 6 children prevented 1 child from experiencing an exacerbation (NNT = 6; 95% CI, 4–12). There was no difference in rates of severe exacerbations with daily compared with intermittent ICS (1 study, n = 278; RR 0.97; 95% CI, 0.75–1.25) or intermittent ICS compared with intermittent montelukast (1 study, n = 190; RR 0.82; 95% CI, 0.59–1.15).
Subgroup Analyses: Persistent Asthma
Our subgroup analyses of preschool children with persistent asthma are described in Fig 5. Eight studies comparing daily ICS with placebo were focused on children with persistent asthma (N = 2505), and showed a reduction in rates of severe exacerbations with daily ICS (8.7% vs 18%, respectively; RR 0.56; 95% CI, 0.46–0.70; P < .001; I2 = 0%). Treatment of 11 children prevented 1 child from experiencing an exacerbation (NNT = 11; 95% CI, 8–15). Data from Szefler et al34 (n = 202) showed that daily ICS reduced rates of severe exacerbations compared with daily montelukast (RR 0.59; 95% CI, 0.38–0.92; P = .02). There were no studies of intermittent ICS for children with persistent asthma.
Subgroup Analyses: Unclear or Mixed Wheezing Phenotypes
Our subgroup analyses of preschool children with unclear or mixed phenotypes are described in Fig 6. Six studies compared daily ICS with placebo (N = 732) and showed no significant difference in rates of severe exacerbations (30.8% vs 40.1%, respectively; RR 0.86; 95% CI, 0.73–1.02; P = .08; I2 = 42%). Data from Papi et al30 showed no significant difference comparing intermittent ICS with placebo (RR 0.61; 95% CI, 0.19–1.91; P = .40) or daily ICS with intermittent ICS (RR 0.33; 95% CI, 0.07–1.62; P = .17).
Other Outcomes: Symptom-Free Days and Linear Growth
Seven studies comparing daily ICS with placebo (N = 1336) reported on symptom-free days (SFDs);7,22,24, 30–32,37 however, few provided adequate data for meta-analysis. Six of these studies31 found a benefit with daily ICS, with mean differences in percentage of SFDs ranging from 5% to 23%. Two studies that compared daily and intermittent ICS30,39 (N = 498) found no difference in SFDs. Bacharier et al19 found no differences in SFDs comparing intermittent ICS with intermittent montelukast or placebo.
We also reviewed linear growth effects, because this is the major concerning side effect with ICS.5 We were unable to meta-analyze these data given the small number of studies reporting growth data and the varied growth metrics reported. Six studies reported on linear growth outcomes7,19,27, 29,35,39 (N = 1461). Three studies compared daily ICS with placebo.7,29,35 Wasserman et al35 found no differences in growth velocity during their 12-week study. Guilbert et al7 found that children treated with daily ICS had a 1.1 cm lower mean increase in height at 2 years (12.6 ± 1.9 cm vs 13.7 ± 1.9 cm, P < .001), but 1 year after discontinuation of ICS, the difference in height increase was reduced to 0.7 cm (19.2 ± 2.2 cm vs 19.9 ± 2.2 cm, P = .008). Murray et al29 found a significantly smaller change in mean height z score after 6 months of daily ICS but no differences at 1, 2, or 5 years of follow-up. In studies comparing intermittent ICS with placebo,19,27,39 Bacharier et al19 found no significant differences in mean change in height comparing intermittent ICS with montelukast or placebo over 1 year. Ducharme et al27 found that intermittent ICS compared with placebo led to smaller mean change in height (6.23 ± 2.62 cm vs 6.56 ± 2.90 cm) and height z score (−0.19 ± 0.42 vs 0.00 ± 0.48) over 1 year. Zeiger et al39 found no significant differences in mean change in height, height percentile, or z score comparing daily with intermittent ICS over 1 year.
With this analysis, we aimed to synthesize the evidence of the effects of daily ICS, intermittent ICS, and montelukast in preventing severe exacerbations among preschool children with recurrent wheeze. In our primary analysis, we found that both daily and intermittent ICS were effective in preventing severe exacerbations. Daily ICS reduced the risk of exacerbations by 30%, intermittent ICS reduced risk by 36%, and there were no significant differences when these strategies were compared directly. Given the varying patterns of recurrent wheezing in preschool children, we performed subgroup analyses by wheezing phenotype. In line with the 2007 National Asthma Education and Prevention Program guideline, we found strong evidence to support daily ICS for preschool children with persistent asthma. For preschool children with intermittent asthma or viral-triggered wheeze, we found strong evidence to support intermittent ICS.
In our primary analysis of preschool children with recurrent wheeze, we found that daily ICS was effective in reducing the risk of severe wheezing exacerbations (NNT = 9), in line with a meta-analysis done in 2009.6 Daily ICS also led to an increase in SFDs. These findings are in line with studies in older children and adults that have established ICS as the most potent and consistently effective long-term control medication for asthma.5 The broad action of ICS on the inflammatory process probably accounts for their efficacy as preventive therapy.5 Overall, the growth-suppressive effects of ICS in preschool children improved over time in most children.7,29,35 A follow-up study by Guilbert et al41 found that children started on daily ICS at a younger age (<2 years) or lower weight (<15 kg) may experience greater effects on linear growth. A Cochrane meta-analysis found dose–response effects of ICS on growth.42 Consequently, children on ICS should have regular monitoring of growth, and health care providers should titrate ICS dosing to the lowest dose that is effective.
Our subgroup analyses by wheezing phenotype showed that most studies of daily ICS in preschool children have focused on children with persistent asthma. For these children, we found strong evidence to support daily ICS, with data from >1600 children demonstrating 44% reduced risk of severe exacerbations (NNT = 11). In addition, most studies that reported on symptom-free days found significant improvements with daily ICS compared with placebo.22,24,32 We also found that daily ICS reduced risk of exacerbations more than montelukast, but these data were limited to a single study. These findings support current national and international guidelines,5,8,11 which recommend daily ICS as first-line therapy for preschool children with persistent asthma.
We also performed a subgroup analysis of preschool children with intermittent asthma or viral-triggered wheeze, because this is the most common wheezing pattern in this age group.1 Most studies evaluated intermittent ICS. We found strong evidence to support intermittent ICS, with a 35% risk reduction in severe exacerbations (NNT = 6). In these studies, children generally received high-dose ICS started at the first sign of a URTI for 7 to 10 days. The children studied had minimal wheezing between URTIs, but the majority had a history of moderate to severe wheezing exacerbations with URTI necessitating systemic steroids, emergency department visits, and hospitalizations (severe intermittent wheezing).19,27,33,38 There were limited data for daily ICS in this population, with only 1 small study comparing daily ICS with placebo (N = 41) that found no difference. Zeiger et al39 directly compared daily ICS with intermittent ICS and found no differences; they also found that intermittent ICS led to a lower cumulative dose than daily ICS. Ducharme et al27 found slower linear growth in children treated with intermittent ICS compared with placebo. However, Bacharier et al19 (intermittent ICS versus placebo) and Zeiger et al39 (intermittent versus daily ICS) found no differences in linear growth. Overall, there is strong evidence to support the safety and efficacy of intermittent ICS for preschool children with intermittent asthma or viral-triggered wheeze, including those with severe intermittent wheezing, in line with the 2015 Global Initiative for Asthma guideline.11 We found limited data directly comparing montelukast with ICS, and a recent Cochrane meta-analysis comparing montelukast with placebo for preschool children with viral-triggered wheezing found no benefit with montelukast.43 More studies are needed that directly compare the efficacy of intermittent ICS, daily ICS, and montelukast for this population.
Previous systematic reviews of these therapies have either not focused on preschool children or not compiled data on multiple therapeutic strategies (daily ICS, intermittent ICS, and montelukast). Our findings are in line with previous studies that combined pediatric and adult data or examined a single therapy. A 2009 meta-analysis compared daily ICS with placebo in preschool children with recurrent wheeze and found a similar reduction in wheezing exacerbations (RR 0.59; 95% CI, 0.52–0.67; P = .0001; I2 = 10%).6 A 2015 Cochrane meta-analysis comparing intermittent ICS with placebo found a reduction in wheezing exacerbations with intermittent ICS in a subgroup analysis of preschool children (odds ratio 0.48; 95% CI, 0.31–0.73; P < .001).44 In addition, a 2013 Cochrane meta-analysis comparing intermittent and daily ICS found no significant differences in a subgroup analysis of preschool children (RR 1.09; 95% CI, 0.85–1.41; P = .49).45 In 2015, Ducharme et al13 published a nonsystematic review of preschool wheeze with meta-analyses of newer studies; they reported similar results comparing daily ICS with placebo (relative risk 0.57; 95% CI, 0.40–0.80) and daily and intermittent ICS (relative risk 0.91; 95% CI, 0.71–1.18). The conclusions of a 2016 nonsystematic review by Castro-Rodriguez et al46 were also in line with our findings.
One limitation to our study is heterogeneity among the included studies. We found moderate heterogeneity in our primary analyses of daily ICS versus placebo and daily versus intermittent ICS. Sources of heterogeneity likely include variations in clinical factors (population, study duration, cointerventions) and study design (parallel vs crossover). As expected, when we narrowed to more homogenous studies in our subgroup analyses, heterogeneity improved. Another limitation was the inclusion of studies that had a high risk of bias in ≥1 domain, usually because of incomplete outcome data. To address the influence of this potential bias, we ran sensitivity analyses that excluded these studies, which were in line with our primary findings. Additionally, the majority of studies included children <2 years, so they may include some children with bronchiolitis. However, all studies required children to have recurrent wheezing, and many additionally required other criteria that should have minimized recruitment of children with bronchiolitis (bronchodilator response, risk factors for asthma).
Our subgroup analyses highlighted that most studies of daily ICS focused on children with persistent asthma, and most studies of intermittent ICS focused on children with intermittent asthma or viral-triggered wheezing. Studies of intermittent ICS may have also preferentially recruited children with higher baseline risk, because rates of exacerbations in placebo groups were higher in studies comparing intermittent ICS with placebo (41.6%) than in studies comparing daily ICS with placebo (24.0%). The differences in study groups recruited for testing these strategies may correlate with treatment response, given that we found treatment benefits in phenotypically homogenous groups and did not find benefits in a group with mixed or unclear phenotypes. However, phenotypic classification of recurrent wheezing in preschool children has limitations. Although the pattern of episodic viral wheeze has been well described in the literature and advocated as a management tool by a European Respiratory Society Task Force,8 recent studies have demonstrated that most preschool children quickly change from 1 phenotype to another.47 Given these limitations, therapeutic decisions remain challenging until more studies are conducted that clearly describe the disease pattern and baseline risk of enrolled children and directly compare daily ICS, intermittent ICS, and montelukast. Our findings show significant reductions in risk of moderate to severe exacerbations with ICS, and they support initiation of ICS therapy in preschool children with symptoms of persistent asthma or those with high risk of severe exacerbations (>1 course of systemic steroids per year).5 Reasonable therapeutic strategies include initiation of daily ICS5 or intermittent ICS11 and should be based on symptom pattern, risk of severe exacerbations,5 and risk of developing chronic asthma.40 Therapy should be reevaluated frequently and adjusted based on symptom pattern.
This is the first study to our knowledge to systematically review and meta-analyze the effects of daily ICS, intermittent ICS, and montelukast in preventing severe exacerbations among preschool children with recurrent wheeze. We performed a thorough and extensive search of the literature. Our overall study population was large, including 4756 children from centers across the world. We found strong evidence to support daily ICS for preventing severe exacerbations in preschool children with recurrent wheeze, specifically in children with persistent asthma. For preschool children with intermittent asthma or viral-triggered wheeze, we found strong evidence to support intermittent ICS for preventing exacerbations. With either treatment strategy, we recommend frequent reassessment of wheezing symptoms and pattern, close monitoring of growth, and active titration to the lowest ICS dose that is effective. More studies are needed that directly compare these therapies.
We thank Elizabeth M. Uleryk for helping develop our search strategy and Dr Prakesh S. Shah for technical guidance with the design and analysis of this study.
- Accepted March 16, 2016.
- Address correspondence to Sunitha V. Kaiser, MD, MSc, 550 16th St, Box 3214, San Francisco, CA 94158. E-mail:
FINANCIAL DISCLOSURE: The authors have indicated they have no financial relationships relevant to this article to disclose.
FUNDING: No external funding.
POTENTIAL CONFLICT OF INTEREST: Dr Bacharier has received consulting fees from Merck and Teva and payment for lectures from Astra Zeneca and Teva; Dr Cabana has served as a consultant for Genentech (Data Registry Safety Board) and Merck (Speaker’s Bureau); and the other authors have indicated they have no potential conflicts of interest to disclose.
- American Lung Association
- Centers for Disease Control and Prevention
- National Asthma Education and Prevention Program
- Brand PL,
- Baraldi E,
- Bisgaard H, et al
- Bacharier LB,
- Phillips BR,
- Bloomberg GR, et al; Childhood Asthma Research and Education Network, National Heart, Lung, and Blood Institute
- Global Initiative for Asthma
- Sawicki GS,
- Smith L,
- Bokhour B, et al
- Moher D,
- Liberati A,
- Tetzlaff J,
- Altman DG,
- Group P; PRISMA Group
- Davies BR,
- Carroll WD
- Higgins JPAD,
- Gøtzsche PC,
- Jüni P, et al; Cochrane Bias Methods Group
- Bacharier LB,
- Phillips BR,
- Zeiger RS, et al; CARE Network
- Connett G,
- Lenney W
- Connett GJ,
- Warde C,
- Wooler E,
- Lenney W
- Gleeson JG,
- Price JF
- Murray CS,
- Woodcock A,
- Langley SJ,
- Morris J,
- Custovic A; IFWIN Study Team
- Wasserman RL,
- Baker JW,
- Kim KT, et al
- Webb MS,
- Milner AD,
- Hiller EJ,
- Henry RL
- Wilson N,
- Sloper K,
- Silverman M
- Wilson NM,
- Silverman M
- Copyright © 2016 by the American Academy of Pediatrics