Steroids and bronchodilators for acute bronchiolitis in the first two years of life: systematic review and meta-analysis

Hartling L, Fernandes RM, Bialy L, et al.

Publication Details

CRD summary

The authors concluded that adrenaline with or without dexamethasone appeared beneficial and safe for short-term outcomes in young children outpatients with bronchiolitis, but further research is needed. There was no clear benefit from any intervention among inpatients. The low quality and short duration of included trials, and uncertainties surrounding statistical analyses, mean the authors' conclusions should be interpreted with caution.

Authors' objectives

To compare the safety and efficacy of bronchodilators and steroids, alone or in combination, for the acute management of bronchiolitis in children aged less than two years.

Searching

PubMed, EMBASE, Cochrane Central Register of Controlled Trials (CENTRAL), Scopus, LILACS, and IranMedEx were searched up to 2009 with no language restrictions. Search terms were reported in an online appendix (see URL for Additional Data). Six relevant conference proceedings and six clinical trial registers were searched for unpublished data. Reference lists of relevant studies and reviews were manually searched. Experts in the field were contacted.

Study selection

Randomised controlled trials (RCTs) that compared bronchodilators (salbutamol or terbutaline, adrenaline, ipratropium bromide) or steroids (inhaled or systemic) alone or in combination versus placebo or another intervention (including bronchodilators and steroids) in inpatients or outpatients (aged 24 months or less) with bronchiolitis were eligible for inclusion. Bronchiolitis was defined as the first episode of wheezing with respiratory distress associated with clinical evidence of viral infection. Trials in the intensive care setting or including intubated or ventilated patients were excluded, as were studies that assessed longer courses of steroids started during the acute phase of bronchiolitis for the prevention of post-bronchiolitic wheezing.

Primary outcomes were rate of admissions at day one and day seven for outpatient studies, and length of stay in hospital for inpatient studies. Secondary outcomes were also assessed.

Included trials were conducted worldwide from 1990 to 2009. Some trials included only infants aged less than one year. Methods of drug administration differed between interventions and trials. Findings were recorded at time points between one to 24 hours and three to 10 days.

Two reviewers independently screened studies for inclusion, with disagreements resolved through consensus or referral to a third reviewer.

Assessment of study quality

Two reviewers independently assessed trial quality according to Cochrane Collaboration criteria: randomisation, allocation concealment, blinding, incomplete outcome data, selective outcomes reporting, and other sources of bias. Trials were rated as being at low, unclear, or high risk of bias. Discrepancies were resolved by consensus among three reviewers.

Data extraction

One reviewer extracted outcome data to calculate mean differences for continuous data and risk ratios (RRs) for dichotomous data, with 95% confidence intervals (CIs). Inpatients and outpatients were treated separately. The extracted data were checked by a second reviewer, and quantitative data were checked by a statistician.

Methods of synthesis

A random-effects model was used to pool weighted or standardised mean differences and risk ratios, along with their 95% confidence intervals. Numbers needed to treat (NNT) were also calculated. Statistical heterogeneity was assessed using the I2 statistic. Sensitivity analyses were conducted based on risk of bias.

In the direct comparisons (for trials in which patients received combined interventions that followed a protocol), the common intervention was cancelled out. Groups from factorial trials (where two or more intervention comparisons are carried out simultaneously) could contribute to the same analysis. Subgroup analyses were undertaken to determine the influence of trials with co-interventions following a protocol and trials with co-interventions at the discretion of the physician.

A Bayesian mixed-treatment analysis was conducted on the primary outcomes to calculate odds ratios (ORs) and 95% credible intervals (CrIs). All trial group comparisons were considered separately in the mixed-treatment analysis. All groups from factorial trials were included. Interactions between groups were taken into consideration in the computations. Consistency was checked by comparing the results from the mixed treatment analysis with results from direct comparisons.

Results of the review

Forty eight RCTs (n=4,897 patients) were included in the review. Eight trials were at low risk of bias, risk was unclear for 25 trials, and high for 15 trials.

Direct comparisons

Outpatients: Four trials (n=920 patients) showed a statistically significant reduction in admission rates from the emergency department on day one using adrenaline compared with placebo (RR: 0.67, 95% CI 0.50 to 0.89; NNT=15; I2=0%). Removal of trials at risk of bias and the comparison of trials following a protocol and those not following a protocol resulted in non-significant differences. No other treatment comparisons were statistically significant (26 trials).

One trial (n=400 patients) showed that adrenaline plus dexamethasone significantly reduced admission rates up to seven days after the emergency department visit compared with placebo (RR 0.65, 95% CI 0.44 to 0.95; NNT=11). No other comparisons were statistically significant (10 trials).

Inpatients: Four trials (n=261 patients) showed that adrenaline significantly reduced the length of stay compared with salbutamol (mean difference -0.28 days, 95% CI -0.46 to -0.09; I2=0%). Exclusion of trials at risk of bias resulted in no significant difference, but comparison of steroids plus bronchodilators that followed a protocol did not significantly alter the results. No other comparisons showed statistically significant differences (21 trials), including trials comparing adrenaline with placebo (two trials).

Mixed treatment analyses

Outpatients: Adrenaline (probability 45%) and adrenaline plus dexamethasone (probability 38.9%) had the highest probability of being most effective in reducing admission rates at day one. However, none of the interventions were statistically more effective compared with placebo.

Steroids plus bronchodilators (adrenaline or salbutamol) had the highest probability of being most effective in reducing admission rates up to seven days after emergency department visit, but credible intervals were wide.

Inpatients: Adrenaline plus dexamethasone had the highest probability of being the best intervention (54%) in reducing length of stay, but credible intervals were wide and results were based on one small trial. No other comparisons were significantly different compared to placebo in terms of length of stay in hospital.

Adverse events: The incidence of adverse events was low and there were not significant differences reported across trials.

Other outcomes were reported in the review.

Authors' conclusions

Adrenaline appeared to be beneficial and safe for short-term outcomes among outpatients, with adrenaline plus dexamethasone also showing benefit, but further research is needed. There was no clear benefit from any one intervention on length of stay in hospital for inpatients.

CRD commentary

The review question and supporting inclusion criteria were clear. A comprehensive literature search was undertaken, with no language restrictions, and included a search for unpublished data, reducing the possibility of missed data. Each stage of the review process was conducted in duplicate, which reduced the potential for reviewer error and bias.

Trial quality was assessed using appropriate criteria, but risk of bias was unclear or high for most trials, which meant the robustness of the findings was unclear. The authors acknowledged the potential risk for bias in the included trials and the limited evidence available for some outcomes and comparisons. Trial publication dates ranged over 19 years, in which time medical practice may have changed. Few patient characteristics were reported, which made it difficult to assess whether pooling of the results was appropriate. There was some evidence of statistical heterogeneity among direct comparisons, but this was not assessed in the mixed-treatment comparisons. It was unclear whether methods used to combine trials for the direct comparisons were appropriate or how this may have impacted on the results. The limitations of mixed treatment analyses should also be borne in mind.

The review process was generally well conducted, but the quality of the included trials, evidence of heterogeneity, and uncertainties surrounding the statistical analyses suggest that the authors' conclusions should be interpreted with caution. The short term duration of the trials should also be borne in mind. Also, as the authors highlighted, the trials were not designed to assess the safety of interventions, so the findings on adverse events should be interpreted with caution.

Implications of the review for practice and research

Practice: The authors stated that adrenaline and adrenaline plus dexamethasone appear to be emerging as the preferred treatments for outpatients, while no clear advantages of steroids or bronchodilators were apparent among inpatients.

Research: The authors stated that the findings from the trials should guide the choice of comparators, including their dose and combinations in future large trials. They also state that future research should focus on areas where there is some suggestion of benefit but the strength of evidence is moderate or low, to compare findings with the current review.

Funding

Canadian Institutes of Health Research, knowledge synthesis grant programme. One author was supported by the Calouste Gulbenkian Foundation (Programme for Advanced Medical Education) and Fundacao Ciencia e Tecnologia (FCT, Portugal).

Bibliographic details

Hartling L, Fernandes RM, Bialy L, Milne A, Johnson D, Plint A, Klassen TP, Vandermeer B. Steroids and bronchodilators for acute bronchiolitis in the first two years of life: systematic review and meta-analysis. BMJ 2011; 342:d1714. [PMC free article: PMC3071611] [PubMed: 21471175]

Indexing Status

Subject indexing assigned by NLM

MeSH

Ambulatory Care; Bronchiolitis /drug therapy; Bronchodilator Agents /therapeutic use; Drug Therapy, Combination; Hospitalization; Humans; Infant; Randomized Controlled Trials as Topic; Steroids /therapeutic use

AccessionNumber

12011002228

Database entry date

13/04/2011

Record Status

This is a critical abstract of a systematic review that meets the criteria for inclusion on DARE. Each critical abstract contains a brief summary of the review methods, results and conclusions followed by a detailed critical assessment on the reliability of the review and the conclusions drawn.