Criteria abstracted from The Users' Guide to Medical Literature, from the Health Information Research Unit and Clinical Epidemiology and Biostatistics, McMaster University

Highlighted lines and questions below provide links to the pertinent description of criteria in The EBM User's Guide, now available at the Canadian Centres for Health Evidence

Article Reviewed:

Please visit the new Evidence Based Journal Club Reviews

Efficacy of Bronchodilator Therapy in Bronchiolitis: A Meta-analysis

Kellner JD, Ohlsson A, Gadomski AM, Wang EE

Arch Pediatr Adolesc Med 1996; 150: 1166-1172. [abstract]

Reviewed by Margarita Reloza-Macias, MD, Hope Childrens Hospital, Oak Lawn, Illinois

Review posted June 18, 1999

---

I. Are the results of the study valid?

A. Primary questions:

1. Did the overview address a focused clinical question?

Probably. "Bronchodilators" is a rather broad category of intervention, encompassing several kinds of therapies. The authors do, however, analyze the effect of beta-agonists specifically in the secondary analysis.

2. Were the criteria used to select articles for inclusion appropriate?

Yes. The study included articles that were prospective, randomized clinical trials comparing a bronchodilator with placebo in patients with bronchiolitis. The term bronchiolitis was similarly defined in all articles and only those published as complete manuscripts and that assessed signs and symptoms were included. To measure the effectiveness of bronchodilators in these cases, clinical scores, oxygen saturation and admission to hospital were selected outcomes of interest.

It is of interest that three of the fifteen articles selected were performed by two of the authors of this meta-analysis. However, from the methods described, there does not seem to be any bias in article selection.

B. Secondary questions:

3. Is it unlikely that important, relevant studies were missed?

Yes. A thorough search with no language restrictions was done using MEDLINE 1966-1994, Excerpta Medica 1974-1995, and Reference Update Files of two of the authors and bibliographies of all selected articles were examined. Likewise, unpublished data were requested from authors when necessary. There was, however, no personal communication with experts in the area.

4. Was the validity of the included studies appraised?

Yes. Selected outcome measures were established to determine the efficacy of bronchodilators versus placebo in the treatment of bronchiolitis. The quality of each study was evaluated with emphasis on double blinding and concealment of allocation to treatment groups. Two of the fifteen articles, however, were not double blinded and unfortunately, only two articles had adequate (central randomization and/or coded medication and/or drugs prepared by pharmacy) concealment of treatment allocation, with the rest labeled as either unclear or inadequate.

A more quantitative scale for grading methodologic quality in the included studies may have been useful. Other criteria, in addition to blinding and concealment of treatment group, could have been included in the methodologic quality assessment.

5. Were assessments of studies reproducible?

Not quite. Although there was complete concordance in the selection of articles for review by two of the authors, there was an interplay of so many variables to consider which weaken the confidence in the result of the overview.

6. Were the results similar from study to study?

Yes. Analysis of the fifteen included studies generally favored treatment with bronchodilators as shown by significant results for score improved and average score variables. However, at least five different bronchodilator agents in varying doses used alone or in combination were considered. Significant heterogeneity in some data hindered conclusive results for other variables, especially in the secondary analysis. Explanation for the heterogeneity included pooling of children with recurrent wheezing and first time wheezing, differences in clinical scoring systems, variations in the timing of evaluations, and state variation (asleep vs. awake).

II. What are the results?

1. What are the overall results of the review?

This overview concluded that bronchodilators produce a modest short term improvement in clinical features of mild or moderately severe bronchiolitis. Clinical scores were analyzed in two ways. First, it was looked at as a dichotomous variable: was the score improved or not. From this arose the binary measure to which the relative risk could be applied. Second, they looked at the scores in terms of effect size, which normalizes the magnitude of changes between scales for comparative purposes; and is treated as a continuous variable. It is probably not completely proper to treat a four point scale as a continuous variable, as it is really an ordinal variable.

Results were as follows. For primary analysis, which included first time wheezing only, the unbiased estimate of effect size (ES) with its 95% confidence interval was significant for average score (-0.32), favoring treatment. The relative risk for score improved (0.76) was also significant and favored treatment. There was no significant effect of treatment on either hospital admission rate or oximetry, however data for the latter showed statistically significant heterogeneity.

Secondary analysis looked at all studies, beta-agonists only, and first time wheezing outpatients. The results were similar to the primary analysis however there was more heterogeneity on some results (oximetry and score improved).

2. How precise were the results?

In the primary analysis, the effect size for average score (-0.32) had a 95% confidence interval of -0.54 to -0.11. An effect size of less than zero indicates a benefit (i.e. reduction of severity scores), which was significant in this case. The relative risk for score improved was 0.76 (95% CI 0.60 to 0.95). Thus the treatment group are .76 times as likely as the controls to not improve, with this estimate being as "strong" as 0.60 or as "weak" as 0.95 (with 95% certainty).

III. Will the results help me in caring for my patients?

1. Can the results be applied to my patient care?

It depends. There were so many factors to consider in these studies, such as severity of illness, whether the patients were first-time wheezers or not, duration of treatment as well as clinical evaluation, etc. All this heterogeneity make it difficult to interpret the true efficacy of bronchodilators in the treatment of bronchiolitis, in general. Furthermore, in some studies, a combination of bronchodilators were used. One could therefore question whether one of those drugs had a larger effect than others, or whether a difference in dose in single drug studies influenced results significantly.

The subjects in the study did not include PICU patients (or at least it wasnt mentioned as such), therefore we dont really know if the study results are applicable to such patients. One could speculate, though, that if clinical improvement is seen in general with bronchodilators and bronchiolitis, it could be reasonable to assume that there will likely be a clinical benefit in the most severely affected patients that we see in the PICU.

2. Were all clinically important outcomes considered?

Not really. Adverse effects were only briefly enumerated and not further discussed. It is interesting that one of the most clinically important outcomes admission to the hospital was not altered by therapy. Even if there is some clinical improvement, if it does not translate into reduction in hospital admissions (or earlier discharge of inpatients), it many not be very effective after all.

3. Are the benefits worth the harms and costs?

No. As the study mentions, not all children with bronchiolitis likely have the same propensity to have bronchospasm and bronchial hyperactivity for which they could benefit from bronchodilators. The authors state that there is a subgroup of children with bronchiolitis who respond to bronchodilator therapy. However the characteristics identifying this subgroup have not been described.

Furthermore, the cost of bronchodilator therapy in these cases is significant, with an estimated annual cost of $37.5 million and adverse effects have not been adequately addressed.

This remains a controversial issue that would benefit from further double blind placebo-controlled trials with more homogeneous subjects.

Related Reviews

Dobson JV, Stephens-Groff SM, McMahon SR, Stemmler MH, Brallier SL, Bay C. The Use of Albuterol in Hospitalized Infants With Bronchiolitis. Pediatrics 1998; 101: 361-368. [abstract]

Review, by K. Daphtary, K. Meert

Comments

Back to the EB Journal Club Index