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

A comparison of inhaled fluticasone and oral prednisone for children with severe acute asthma

Schuh S, Reisman J, Alshehri M, et al.

N Engl J Med. 2000;343(10):689-94. [abstract]

Reviewed by Al Torres, MD , University of Illinois College of Medicine at Peoria

Review posted February 16, 2001

---

I. What is being studied?:

The study objective:

To compare the efficacy of 2 mg of inhaled fluticasone delivered by a metered-dose inhaler with a valved holding chamber (i.e., spacer) with that of 2 mg of prednisone per kg as evaluated within 4 hours after administration in children five years or more of age who were treated in the emergency department for severe acute asthma.

The study design:

A randomized, double-blind, double dummy controlled trial.

The patients included:

Children were eligible if they were at least five years old, if the baseline forced expiratory volume in one second (FEV1), expressed as a percentage of the predicted value, was < 60, and if they were able to use an inhaler and undergo pulmonary function tests reliably. The investigators enrolled a total of 103 children.

The patients excluded:

Of the 8001 children seen in the emergency department with acute severe asthma, 2084 were excluded because the research nurses were not present when they were evaluated, 5196 because they were too young, 264 because they could not perform spirometry reliably, and 73 because they had already been in this study. Also excluded were 160 children who were taking prednisone, 76 who were taking high-dose inhaled corticosteroids, and 45 because the parents refused permission for them to participate. Three additional children were eventually excluded: one child in the prednisone group repeatedly vomited after the experimental syrup, one child in the fluticasone group was subsequently excluded because his FEV1 reached 92% of the predicted value after the first nebulized treatment, and, one child became acutely ill within one hour after treatment with prednisone, necessitating admission to the intensive care unit. Of the 100 children who were able to proceed with spirometry, 51 were in the fluticasone group and 49 in the prednisone group.

The interventions compared:

Eligible children received either a single 2 mg dose of fluticasone through an inhaler and spacer with a mouthpiece or oral prednisone syrup (2 mg per kg; maximum 60 mg) All patients received 0.15 mg of nebulized albuterol per kg through a jet nebulizer with a tight fitting mask within 20 minutes before the experimental therapy, at baseline (i.e., immediately after the experimental therapy), and 20, 40, 60, 80, and 140 minutes after experimental therapy. Ipratropium bromide at a dose of 250 micrograms was added to the initial 3 nebulized doses of albuterol.

For the home phase of the study, those in the fluticasone group received 500 micrograms of fluticasone twice daily by inhaler with a spacer and placebo syrup for seven days, whereas the prednisone group received placebo inhaler and 1 mg of oral prednisone per kg per day (maximum 40 mg per day) for seven days.

The outcomes evaluated:

The change in FEV1 as a percentage of the predicted value from baseline (time 0) to 240 minutes was the primary outcome measure. Secondary outcome measures were the changes in the forced vital capacity (FVC) and the predicted peak expiratory flow rate as percentages of the predicted values, the respiratory rate, transcutaneous oxygen saturation while breathing room air, and the rate of hospitalization.

II. Are the results of the study valid?

Primary questions:

1. Was the assignment of patients to treatments randomized?

Yes. A blocked randomization code was prepared by the pharmacists from a computer-generated list of random numbers.

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

Was followup complete?

Yes. All patients were accounted for at the end of the 4 hours of spirometry measurements. The authors mentioned 2 pts in the fluticasone group who got worse and got IV steroids, and one in each groups who, "despite respiratory improvment" were unable to complete spirometry at 240 minutes. It is not clear if these patients were excluded or not. All the tables of results list the same 100 patients that were enrolled.

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

No. Three of the 103 patients who were randomized (2 in the predisone group and 1 in the fluticasone group) were not evaluated at 4 hours. With such small numbers it is unlikely an intention-to-treat analysis would have resulted in a significant change in the outcome measures. No such analysis was performed.

Secondary questions:

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

Yes. The pharmacy prepared sequential sealed packets containing the study drugs. The randomization code was revelaed after all the patients had completed the study. The placebo inhalers and oral syrup placebo were indistinguishable from the active drugs with respect to appearance, taste, and smell.

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

Almost. There was no significant demographic or clinical differences between the two groups, except for gender - the prednisone group had twice as many males as females whereas the fluticasone group had a near equal gender distribution. The investigators stated both groups were similar in severity since they had equivalent mean hospitalizations in the previous year, number using inhaled corticosteroids, baseline spirometry results, respiratory rates, and oxygen saturations.

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

No. While 9/51 (18%) of the fluticasone group received intravenous hydration, none of the patients in the prednisone group did. This difference in therapy and the greater prevalence of vomiting in the fluticasone group (47% vs. 16%) may have accounted for the greater rate of hospitalization in the fluticasone group (31% vs. 10%). It is possible that the increased rate of hospitalization in the fluticasone group might have been for asthma plus vomiting and needing IV hydration, whereas the predisone patients who were admitted might only have been admitted for their asthma. The high dose of fluticasone may have been responsible for the greater rate of vomiting (see VI. 3. below). It's not clear exactly why the fluticasone group had a higher hospitalization rate.

III. What were the results?

1. How large was the treatment effect?

The FEV1 increased by a mean of 9.4 +/- 12.5 percentage points in the fluticasone group and by 18.9 +/- 9.8 percentage points in the prednisone group four hours after therapy (p < 0.001). The FVC and PEFR also improved significantly greater from baseline to 4 hours in the prednisone group than in the fluticasone group. The following table provides the percentage of patients in each group with an excellent response (increase in FEV1 of at least 25 percentage points of the predicted value from baseline to 4 hours), moderate response, and poor response (increase in FEV1 of < 5 percentage points):

Groups	Excellent	Moderate	Poor
Prednisone	27	65	8
Fluticasone	10	59	31

(p = 0.002)

None of the patients taking the prednisone had a reduction in FEV1 from baseline to 4 hours, whereas 25% of patients taking fluticasone did (p < 0.001).

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

The absolute risk risk reduction (ARR) for a poor response if the patient receives prednisone instead of fluticasone is 0.23 (95% CI, 0.08 to 0.38). The relative risk reduction (RRR) of a poor response if the patient receives prednisone instead of fluticasone is 74% (95% CI, 28 to 91%). The odds ratio (OR) of a poor response after receiving prednisone instead of fluticasone is 0.19 (95% CI, 0.06 to 0.63). The number needed to treat (NNT) with prednisone instead of fluticasone to avoid a poor response is 4 (95% CI, 3 to 12).

The ARR, RRR, and OR for hospitalization in patients receiving prednisone instead of fluticasone is 0.21 (95% CI, 0.06 to 0.36), 68% (95% CI, 19 to 87%), and 0.25 (95% CI, 0.08 to 0.74), respectively. The NNT with prednisone instead of fluticasone to avoid hospitalization is 5 (95% CI, 3 to 18).

[HINT: Use the clinical significance calculator at this URL to calculate relative risk reduction, absolute risk reduction - and their confidence intervals! http://www.healthcare.ubc.ca/calc/clinsig.html ]

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

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

Yes. The results of the study are consistent with current ED treatment standards for acute severe asthma - oral steroids plus nebulized beta agonists (+/- anticholinergics). If there was poor delivery of the inhaled steroid to the distal airways for whatever reason (e.g., mucus plugging, failure to prime the spacer with several puffs of drug to eliminate electrostatic interference [1]), the study confirms that oral steroids is superior to placebo in improving pulmonary function and reducing the rate of hospitalization. Further investigation is needed with some differences in study design (e.g., nebulized fluticasone instead of fluticasone delivered by metered-dose inhaler, follow-up greater than 4 hours) is needed prior to concluding with any certainty that inhaled steroids are not as efficacious as oral steroids in the treatment of acute asthma. Although the investigators state they kept a log of the patients who were missed (e.g., acute asthma patients treated in the ED from 9 p.m. to 8 a.m.) or excluded, they do not state whether these patients were similar to the ones enrolled in the study making generalizability less likely. Only five percent of the eligible subjects were enrolled making internal generalizability suspect as well.

2. Were all clinically important outcomes considered?

No. Differences in cost and patient satisfaction were not addressed.

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

No. Besides a significant decrease in FEV1 within 4 hours of drug administration in 25% of the fluticasone group, 47% of the fluticasone group also experienced vomiting and 18% required intravenous hydration. A 16-week study of 660 mcg twice a day (n = 32) and 880 mcg twice a day (n = 32) in adolescent and adult asthmatics, resulted in a rate of vomiting of 16%.(2) The younger age and higher dose (2 mg) may have contributed to this high rate of vomiting in this study. Other potential explanations for the high rate of vomiting were the oral placebo administered to the fluticasone group induced vomiting or the fluticasone group had a greater rate of gastrointestinal inflammation as well as pulmonary inflammation. The authors did not address this adverse event.

References

1. Zar HJ, Brown G, Donson H, et al. Home-made spacers for bronchodilator therapy in children with acute asthma: a randomized trial. Lancet 1999;354:979-982. [abstract]
2. Physician Desk Reference (electronic version), 55 ed. Montvale: Medical Economics, 2001. http://physician.pdr.net

Comments

Back to the EB Journal Club Index