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

Dexamethasone in bronchiolitis: a randomised controlled trial

Roosevelt G, Sheehan K, Grupp-Phelan J, Tanz RR, Listernick R.

Lancet 1996;348:292-5. [abstract]

Reviewed by Eric Stiles MD, New York Presbyterian Hospital Ð Weill Medical College of Cornell University

Review posted January 2, 2001

---

I. What is being studied?:

The study objective:

Determine if corticosteroids given to previously healthy infants admitted with a diagnosis of bronchiolitis were safe and efficacious.

The study design:

Prospective, randomized, placebo-controlled and double-blind controlled trial.

The patients included:

First-time wheezing infants between 4 weeks and 12 months of age seen in the emergency department of a tertiary care children's hospital, and admitted to the regular pediatric inpatient unit for further care.

The patients excluded:

Patients admitted to the pediatric intensive care unit; patients with a history of congenital heart disease; patients previously receiving ventilation; patients previously intubated; patients previously on supplemental oxygen.

The interventions compared:

The treatment group received dexamethasone 1 mg/kg intramuscular daily for a maximum of three doses; the placebo group received an equivalent volume of normal saline intramuscular daily for a maximum of three days.

The outcomes evaluated:

The primary outcome evaluated was time to resolution of symptoms for each patient. This was defined as the number of twelve hour periods needed for all of the following criteria to be met: (1) pulse blood oxygen saturation of more than 95% while receiving no supplemental oxygen; (2) accessory muscle score of 0 (no intercostal or other retractions); (3) wheeze score of 0 or 1 (no wheezes or end-expiratory wheezes only); and (4) resumption of normal feeding.

II. Are the results of the study valid?

Primary questions:

1. Was the assignment of patients to treatments randomized?

Yes. The hospital pharmacy prepared and coded drug and placebo randomly, although the method of randomization was not specified.

2. Were all patients who entered the trial properly accounted for and attributed at its conclusion?

Was followup complete?

Yes. No patients were lost in the analysis. Four were excluded. These included one found to have a protocol violation (intubation in the first 48 hours of life), one found to have cystic fibrosis two weeks after discharge, one who did not receive the first two doses of study treatment and one who received corticosteroids prescribed by the attending physician due to deteriorating clinical status. Given that these were only four out of 122 enrolled, the exclusions seem reasonable and not a gross violation of the principle of "intention to treat".

Were patients analyzed in the groups to which they were randomized?

Yes. There were no crossovers.

Secondary questions:

3. Were patients, health workers, and study personnel "blind" to treatment?

Yes. An equal volume of saline was administered to the placebo group, and those involved were unaware which treatment the patient received.

4. Were the groups similar at the start of the trial?

The dexamethasone group and the placebo group were similar at enrollment with respect to temperature, heart rate, blood pressure, respiratory rate, accessory muscle score, wheeze score, age weight, ethnic origin, atopic family history, exposure to tobacco and laboratory evidence of RSV infection. Chest radiographs were similar. The lone significant difference was that the dexamethasone group had more patients with initial oxygen saturations of less than 95% (p = 0.02). (See comments below).

5. Aside from the experimental intervention, were the groups treated equally?

Yes. There were no significant differences between the groups in terms of use of antibiotics, nebulized beta-agonist drugs or other interventions.

III. What were the results?

1. How large was the treatment effect?

In the 197 patients analyzed, there was no significant difference between the dexamethasone group and the placebo group in the time to symptom resolution, which was the primary outcome. When six subpopulations were compared (RSV positive, RSV negative, oxygen saturation greater than 95%, oxygen saturation less than or equal to 95%, family history of atopic disease, and family history of atopic disease with RSV positive), there was still no statistically significant difference between those receiving dexamethasone and those receiving placebo.

2. How precise was the estimate of the treatment effect?

Initial calculations determined that a sample size of 60 patients in each treatment group would be necessary to detect a 30% reduction in time to resolution of symptoms. (alpha = 0.05, power = 0.80). There were 65 patients in the treatment group and 53 in the placebo group. Therefore, we can state with 80% confidence that there was not a 30% or greater reduction in time to resolution of symptoms in the treatment group.

Survival analysis was used with a proportional hazards model. The hazard ratio was not considered significant if 1.0 was included in the 95% confidence intervals. So, for example, the hazard ratio of dexamethasone in the resolution of symptoms was 1.3, with 95% confidence intervals of 0.9 and 1.3, suggesting no effect of dexamethasone. The hazard ratio for oxygen saturation on admission, with respect to duration of oxygen therapy, was 0.5, with 95% confidence intervals of 0.3 and 0.7. This suggests that those patients whose oxygen saturation on admission was greater than 95% on room air had a significantly shorter duration of oxygen therapy. (A good explanation of the hazard ratio is available at http://www.graphpad.com/www/book/compsurv.htm).

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

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

Admission to the pediatric intensive care unit was an exclusion criteria for the study, so the results of this study will not be directly applicable to pediatric ICU patients. However, any study that adds to our understanding of bronchiolitis, a fairly common entity in the pediatric ICU, could prove helpful by suggesting new areas of research and new sub-populations to be studied.

2. Were all clinically important outcomes considered?

Yes. By making the time to resolution broad enough to include the patient's status with regard to oxygen requirement, wheezing, accessory muscle use and feeding, the investigators appear to have covered the relevant outcomes very well. A more clinically relevant endpoint might be duration of hospitalization, although length of stay can certainly be influenced by factors other than duration of symptoms.

3. Are the likely treatment benefits worth the potential harms and costs?

There was no treatment benefit.

Reviewer's Comments

There is an increasing body of evidence that corticosteroid treatment provides no improvement in outcome when used to treat bronchiolitis (1-3, 5). A recent meta-analysis by Garrison et al., however, associates corticosteroid therapy with a statistically significant reduction in length of stay and duration of symptoms (see PedsCCM EBJC Review by M. Bonnie Dutta)(6). Our understanding is becoming more finely-honed as different populations of bronchiolitics (RSV positive and negative, for instance), and even different delivery methods of steroids are employed (2,4). The study by Roosevelt, Sheehan et. al. provides further data to bolster the case against corticosteroids. The study was designed to detect an improvement in symptoms and was done in a randomized, prospective, double blind manner. Infants with their first episode of wheezing admitted for inpatient management of their bronchiolitis comprise a common pediatric problem, and the study therefore has potential applicability to a great number of patients.

The greatest limitation of the current study is the difference in oxygen saturation between the two groups upon enrollment. The dexamethasone group included 51 out of 65 patients with oxygen saturation of 95% or less upon admission, while the placebo group only included 31 out of 53 with a saturation of 95% or less (79% vs. 59%, respectively, giving p = 0.02). It is conceivable that if the dexamethasone group were somewhat more ill upon admission but experienced no clinically significant difference in time to resolution of symptoms when compared with the placebo group, that the dexamethasone group did receive some added benefit from dexamethasone. The authors analyzed the subgroup of all patients with saturations of 95% or less, however, and found that dexamethasone did not improve their outcome compared with placebo.

Interesting directions for further study are suggested: Are there subgroups and subpopulations who might benefit from steroids? Patients admitted to the ICU, those with RSV and atopic family history, and so forth? (5) If steroids are going to be found to have any benefit in bronchiolitis, it appears likely that it may only happen in the context of a very specific population.

References

1. Klassen TP, Sutcliffe T, Watters LK, Wells GA, Allen UD, Li MM. Dexamethasone in salbutamol-treated inpatients with acute bronchiolitis: A randomized, controlled trial. J Pediatr 1997; 130:191-6. [abstract] [PedsCCM EBJC Review]
2. Richter H, Seddon P. Early nebulized budesonide in the treatment of bronchiolitis and the prevention of postbronchiolitic wheezing. J Pediatr 1998; 132: 849-53. [abstract]
3. Springer C, Bar-Yishay E, Uwayyed K, Avital A, Vilozni D, Godfrey S. Corticosteroids do not affect the clinical or physiological status of infants with bronchiolitis. Pediatr Pulmonol 1990; 9:181-5. [abstract]
4. Wong JYW, Moon S, Beardsmore C, OÕCallaghan C, Simpson H. No objective benefit from steroids inhaled via a spacer in infants recovering from bronchiolitis. Eur Respir J 2000; 15:388-94. [abstract]
5. De Boeck K, van der Aa N, Van Lierde S, Corbeel L, Eeckels R. Respiratory syncytial virus bronchiolitis: A double-blind dexamethasone efficacy study. J Pediatr. 1997;131(6):919-21.[abstract]
6. Garrison MM, Christakis DA, Harvey E, Cummings P, Davis RL. Systemic Corticosteroids in Infant Bronchiolitis: A Meta-Analysis. Pediatrics 2000; 105:e44. [abstract] [full-text] [PedsCCM EBJC Review]
7. Dawson B, Trapp RG. Basic & Clinical Biostatistics, Third Edition. New York: 2001, McGraw-Hill Books.
8. Motulsky, H. Intuitive Biostatistics. 1995, Oxford University Press. Chapter 33: Comparing Two Survival Curves. Available at: http://www.graphpad.com/www/book/compsurv.htm.

Comments

Back to the EB Journal Club Index