BACKGROUND. Inhaled corticosteroids are not as effective as oral corticosteroids in school-aged children with severe acute asthma. It is uncertain how inhaled corticosteroids compare with oral corticosteroids in mild to moderate exacerbations.
PRIMARY OBJECTIVE. The purpose of this work was to determine whether there is a significant difference in the percentage of predicted forced expiratory volume in 1 second in children with mild to moderate acute asthma treated with either inhaled fluticasone or oral prednisolone.
METHODS. This was a randomized, double-blind controlled trial conducted between 2001 and 2004 in a tertiary care pediatric emergency department. We studied a convenience sample of 69 previously healthy children 5 to 17 years of age with acute asthma and forced expiratory volume in 1 second at 50% to 79% predicted value; 41 families refused participation. Albuterol was given in the emergency department and salmeterol was given after discharge to all patients, as well as either 2 mg of fluticasone via metered dose inhaler and valved holding chamber in the emergency department plus 500 μg twice daily via Diskus for 10 doses after discharge (fluticasone group, N = 35) or 2 mg/kg of oral prednisolone in the emergency department plus 5 daily doses of 1 mg/kg of prednisolone after discharge (prednisolone group, N = 34). We measured a priori defined absolute change in percent predicted forced expiratory volume in 1 second from baseline to 4 and 48 hours in the 2 groups.
RESULTS. At 240 minutes, the forced expiratory volume in 1 second increased by 19.1% ± 12.7% in the fluticasone group and 29.8% ± 15.5% in the prednisolone group. At 48 hours, this difference was no longer significant (estimated difference: 4.0 ± 3.4; P = .14). The relapse rates by 48 hours were 12.5% and 0% in the fluticasone group and prednisolone group, respectively.
CONCLUSION. Airway obstruction in children with mild to moderate acute asthma in the emergency department improves faster on oral than inhaled corticosteroids.
Systemic corticosteroid therapy has become a mainstay in the treatment of acute asthma. It has been shown to reduce hospitalizations within 4 hours,1 improve normalization of lung function, and decrease relapses.2,3 This early benefit has been attributed in part to its upregulation of β2 receptors, as well as mucosal vasoconstriction.4 However, numerous patients require repetitive treatment with systemic corticosteroids, which many perceive as potentially unsafe.5 Although corticosteroid-associated adverse effects of most concern are associated with long-term use, giving many short courses of systemic steroids in asthmatic patients with frequent relapses may not be without risks and may contribute to suboptimal physician asthma management plans and patient compliance.6,7 Therefore, the question arises about possible alternatives.
Inhaled corticosteroids are well established in the control of chronic asthma and, because of their high degree of anti-inflammatory activity,8 they may be good candidates for controlling mild acute asthma. Because pediatric asthma guidelines advocate an increase in the intensity of inhaled steroids if control is insufficient,9 and boosting their dose early in asthma exacerbations may be effective in adults,10 the physicians may opt for initiating inhaled corticosteroids for mild acute asthma. The advantages of inhaled fluticasone are its rapid metabolism by the liver, low oral bioavailability,11 and high topical anti-inflammatory potency with fewer systemic effects than those associated with beclomethasone.12–14 The upregulation of β2 receptors associated with potent steroids suggests an advantage for the concomitant use of long-acting β2 agonists (with a short-acting β2 agonist rescue if necessary) during the poststabilization period.
Although there is no doubt that inhaled corticosteroids are effective in chronic asthma management,12,15–19 their role in treating acute asthma is much less clear, and the severity threshold for the necessity of initiating oral corticosteroids has not been established. Although inhaled corticosteroids are less effective than oral corticosteroids in children with severe disease,20–21 their efficacy in more moderate exacerbations is inconclusive because of the conflicting results.22–26 The main limitations of past literature include both lack of outcome measurement within hours of the intervention and lack of an objective physiologic outcome evaluation, such as the forced expiratory volume in 1 second (FEV1). There is considerable practice variation in the use of inhaled corticosteroids in acute asthma,27–29 also evidenced by a recent meta-analysis,30 which identified too great a heterogeneity of the pediatric studies to reach a meaningful consensus.30 We have demonstrated previously that children with severe acute asthma derive greater benefit from oral prednisone than from inhaled fluticasone within 4 hours of commencing therapy.20 Children with mild to moderate acute asthma may have less airway inflammation and edema, and the delivery of the inhaled drug into the lungs may be correspondingly higher compared with their sicker counterparts. Therefore, because inhaled corticosteroids may still be of benefit in children with less severe disease, we conducted a study comparing the efficacy of inhaled fluticasone with that of oral prednisolone in children ≥5 years of age presenting to the emergency department (ED) with mild to moderate acute asthma. The primary objective was to compare the change in percent predicted (%pred) FEV1 from baseline to 4 and 48 hours in the 2 groups. Secondary objectives included comparisons of the changes in %pred FEV1 to day 6, as well as the return for care and use of additional corticosteroids by 48 hours and day 6.
Study Population and Study Design
The study nurses were notified of all children with acute asthma between 6 am and 9 pm in our ED between October 2001 and November 2004. Children were enrolled if they were 5 to 17 years of age, were diagnosed with acute asthma by the ED pediatrician, and had baseline FEV1 50% to 79% predicted.31
Excluded were children without previous wheezing, children with persistent vomiting, airway instability, treatment with oral corticosteroids within 7 days, coexistent cardiopulmonary/neuromuscular disease, varicella contact within 21 days of the study entry, previous treatment in the intensive care unit for asthma, and limited command of the English language. The study was approved by our research ethics board, and consent was obtained from each patient/parent. A log of patients who were missed, excluded, or refused participation was kept to assess the generalizability of the study.
We conducted a randomized, double-blind, double-dummy trial in which the eligible children were randomly assigned to receive either 2 mg of fluticasone propionate (8 inhalations, 250 μg each) via metered dose inhaler (MDI)/valved holding chamber (VHC) with a mouthpiece (AeroChamber, Trudell Medical) and prednisolone placebo syrup in the ED (fluticasone group [FG]) or 8 inhalations of fluticasone placebo and 2 mg/kg (maximum: 60 mg) of active prednisolone syrup (prednisolone group [PG]). In the ED, all of the children received nebulized albuterol (Ventolin 5% solution, GlaxoSmithKline, Canada) and ipratropium bromide (Boehringer Ingelheim, Canada), 250 μg per dose, via PariLC Star nebulizer with a mouthpiece before the experimental therapy (−20 minutes randomization baseline), immediately after the initial dose of the experimental therapy (0 minutes), and at 60, 120, 180, and 240 minutes. After enrollment of the first 10 patients, the dose of albuterol of 0.15 mg/kg was decreased to 0.075 mg/kg because of tremor and tachycardia noted in the initial cohort of patients with this high-efficiency nebulizer.32 After discharge, the FG received 10 doses of fluticasone, 500 μg per dose, with salmeterol, 50 μg per dose (Advair, GlaxoSmithKline, Canada), every 12 hours via a Diskus inhaler and 5 daily doses of prednisolone placebo, whereas the PG got 10 doses of fluticasone placebo with salmeterol, 50 μg per dose, every 12 hours and 5 daily doses of 1 mg/kg of active prednisolone syrup. The placebo MDIs and Diskus were provided by GlaxoSmithKline and the oral syrup placebo by our pharmacy. They were indistinguishable from the active drugs in appearance, taste, and smell. Both the patients/parents and the study nurses were blinded to treatment assignment. All of the discharged children were also provided with a rescue albuterol MDI to take every 4 hours for the initial 24 hours and then every 4 hours as needed in case of deterioration. At 48 hours and on day 6, the children were visited at home by a study nurse for clinical assessment, FEV1 measurements, and ascertainment of other outcomes. Children with persistent symptoms were referred either to our ED or to their physicians for assessment. The research pharmacist assessed compliance by weighing the inhalers and measuring the volume of syrup before and after the study. The families were also telephoned on day 28 regarding asthma relapses or residual symptoms.
We used a computer-generated randomization scheme with random block sizes of 10, produced by the research pharmacist. The research pharmacist supplied the corresponding MDI/syrup (for the ED phase) and Diskus/syrup (to be used after discharge) double-dummy setup to the study nurses who also enrolled participants and delivered the experimental therapy. The randomization codes were secured at our pharmacy until enrollment and analysis decisions had been terminated. The study nurses were unable to determine which intervention a given patient had received.
Delivery of Experimental Therapy and Other Treatments
To maximize the probability that the inhaled drugs would be uniformly distributed through the lungs, all of the patients received the first dose of bronchodilators immediately before the experimental therapy. The experimental therapy in the ED consisted of 8 inhalations of fluticasone or equivalent placebo via the MDI and VHC with the mouthpiece. The patients were taught a deep inhalation technique with 5 seconds of breath holding after each actuation33 by the study nurse. This technique was also used with the Diskus at home. At home, the children took a single inhalation containing 50 μg of salmeterol with either 500 μg of fluticasone or its placebo equivalent every 12 hours. Children who vomited the syrup within 30 minutes received the second dose; further vomiting necessitated withdrawal from the study.
All of the decisions regarding disposition and additional treatments during or after the study were made by the ED physicians and subsequently by the patients' pediatricians who were masked to the intervention and outcome measurements. These decisions were based on persistent respiratory distress and/or other asthma symptoms.
All of the outcomes were assessed by 4 study nurses trained to perform spirometry meeting American Thoracic Society guidelines34 in our pulmonary function laboratory. The FEV1 was measured at −20 minutes (randomization baseline) and at 0, 60, 120, 180, and 240 minutes in the ED and at 48 hours and on day 6, using a Spirolab II spirometer (Medical International Research), according to the recommendation of the American Thoracic Society.34 The FEV1 was measured in sets of 6 and expressed as percentages of predicted values for height and gender.31 The highest value was accepted for analysis. The primary outcome was the absolute change in %pred FEV1 at 4 and 48 hours in the groups.35 Also, an excellent response was arbitrarily defined as an increase in FEV1 of >30% from baseline to 240 minutes. Prespecified secondary outcomes composed of the change in %pred FEV1 on day 6 and those of a more exploratory nature also included a priori determined unscheduled return for care and administration of additional corticosteroids by 48 hours and day 6. We have also documented these outcomes on day 28, as well as hospitalizations by day 6.
The sample size was based on an estimated SD of 15 for the change in the %pred FEV1 in the PG. To allow detection of a 10% difference between the groups in the improvement in %pred FEV1 from baseline to 48 hours and to maintain α of.05 and β of .2, the required sample size was 74 children.
The change in the FEV1 from baseline to 4 and 48 hours was analyzed by mixed model regression, which incorporated repeated observations of individual patients. This method requires an assumption about the complex covariance associated with repeated measurements over time.36 Its advantage reflects its covariance structure suitable for unequal time spacing of the measurements. This covariance matrix assumes that observations closer together are more correlated that those further apart. A quadratic time variable was added to test for nonlinearity. Data on all of the patients, including those with missing values, were included in the model. The outcomes correlated over time were analyzed using the Proc Mixed program (SAS Institute Inc, Cary, NC).37 All of the statistical tests were 2-tailed, and all of the analyses were intent-to-treat.
Baseline differences between the 2 regimens were tested using the Student t test for continuous and normally distributed variables, as was the difference in the change of %pred FEV1 from baseline to day 6 between the 2 groups. The differences in proportions were analyzed with the χ2 and Fisher's exact tests.38
Characteristics of the Patients
Between October 2001 and November 2004, 2082 children 5 to 17 years of age were seen in our ED with asthma. Of these, 1502 were not screened, because the study nurse was not present for 1314 patients, 13 children were missed, 143 children had no distress and did not require therapy, and 32 were enrolled previously. Of the 580 patients screened for eligibility, 89 could not perform spirometry, 110 had FEV1 <50% predicted, and 119 had FEV1 ≥80%. Also excluded were 34 children with comorbidities, 55 children on oral corticosteroids, 27 children admitted previously to the intensive care unit, 5 patients for wheezing for the first time, 1 patient for pregnancy, 2 for contact with varicella, 22 families because of insufficient command of the English language, and 41 for refusal to participate. Six families lived outside commuting distance. Sixty-nine children were randomly assigned to the experimental therapy, 35 to the FG, and 34 to the PG. One parent changed her mind regarding participation at 120 minutes, 5 children vomited the syrup and were too nauseated for the FEV1 assessment, and 3 children were too sleepy to cooperate with this maneuver. Therefore, 60 children performed spirometry at 240 minutes, 30 in the FG and 30 in the PG. All of the randomly assigned patients received follow-up visits, with the exception of the aforementioned patient whose mother refused participation, 1 parent who was unavailable for the 48-hour visit, and 1 child who remained in the hospital for 7 days and his parents were not accessible. Sixty-six children were, thus, evaluated at 48 hours, 32 in the FG and 34 in the PG. The day-6 visit had 67 participants, 33 in FG and 34 in the PG. There were no significant baseline demographic or clinical differences between the groups (Table 1).
The PG achieved significantly faster relief of airway obstruction as measured by FEV1 than their fluticasone-treated counterparts (Fig 1; Table 2). The difference in the change in %pred FEV1 from baseline to 240 minutes was statistically significant (P = .001), favoring prednisolone (Table 2). Most of the overall improvement in the FEV1 in the PG had already occurred by 240 minutes, whereas the FEV1 of the FG improved more slowly (Fig 1). At 48 hours, this difference was no longer clinically or statistically significant (Table 2), and on day 6 it was negligible (increase of 33.6% ± 21.2% in the FG vs 37.0% ± 16.9% in the PG; P = .47). The mixed models for the FEV1 included a significant coefficient for a quadratic time effect because of nonlinearity (P = .0001). Adjustment for covariates such as age, atopy, and inhaled corticosteroids before the study did not significantly alter the results, and neither did the analysis of only the patients remaining on experimental therapy at 48 hours and day 6.
Although a total of 23 (76.7%) of 30 children in the PG had an excellent response at 240 minutes, only 14 (46.7%) of 30 children given fluticasone exhibited excellent response (P = .017). The “number needed to treat” with prednisone to obtain this benefit at 240 minutes is 3.3.
Secondary and Other Outcomes
At 240 minutes, the differences in the changes in the respiratory rate and oxygen saturation were negligible (0.30 breaths per minute; 95% confidence interval [CI]: −1.50 to 2.10; and 0.13%; 95% CI: 1.07 to 1.33, respectively). By 48 hours, the PG underwent unscheduled visits for asthma symptoms less often than the FG (0 of 34 vs 4 of 32; 95% CI: 1.04 to 23.96; P = .03). Likewise, the difference in frequency of administration of corticosteroids outside the experimental protocol during the study favored prednisolone but did not reach statistical significance (7 of 33 in FG vs 2 of 34 in PG; 95% CI: −0.01 to 31.3; P = .07). These outcomes were comparable between the groups between days 6 and 28, as was the frequency of the use of rescue albuterol MDI (data not shown). With the exception of 1 hospitalization in each group from the ED and 5 children vomiting (4 in FG and 1 in PG), neither intervention had experienced adverse events or adverse effects of the experimental therapy. One child in the PG group did not take all of the experimental syrup; other children were compliant with the assigned intervention.
Our study shows that children with mild to moderate acute asthma treated with prednisolone derive greater immediate relief of airway obstruction than those given inhaled fluticasone, with a clinically and statistically significant difference at 4 hours. This is also illustrated by the higher number of children showing clear benefit (“an excellent response”) on prednisolone. This benefit seems to be maintained, because children taking prednisolone also returned for care less frequently than those on fluticasone, although the differences in airway obstruction are no longer significant by 48 hours. The importance of these findings relates to a higher respiratory reserve at and immediately after discharge from the ED because of superior improvement with prednisolone within a short time period.
These findings are quite different from that of the previous studies comparing inhaled and oral corticosteroids in children with moderate acute asthma, 2 of which have favored inhaled corticosteroids24,26 and 2 that have found no difference.23,25 Their population may have been milder than the majority of children seen in most EDs, for example, no child in 1 study required hospitalization or return visits,23 and patients in another study had mean baseline FEV1 75% predicted.26 Two studies did not assess outcomes until several days after initiation of therapy25,26 and, thus, could not demonstrate any differences in short-term benefit. Two studies evaluated peak expiratory flow rate,24,25 which has a lower sensitivity than FEV1 for detecting small airway obstruction.39 In contrast, our design allowed us to compare serial objective FEV1 measurements within a well-defined asthma severity range. The FEV1 has been found to be the best objective clinical indicator of the degree of asthma severity in school-aged children with asthma,40 is responsive to the clinical status,41 and correlates with the inflammatory change in the airways.42,43 A recent publication investigating the influence of pulmonary function testing on the management of pediatric asthma concluded that spirometry results were frequently abnormal in patients without symptoms and with normal examination and that in the absence of FEV1 results, the physicians frequently overestimate how well a patient's asthma is controlled.44
The differences in the changes in FEV1 in the groups were paralleled by the differences in the unscheduled medical visits and a nonsignificant trend in the differences in the administration of additional corticosteroids. Although these results support the conclusion in the primary outcome, the number of children experiencing these outcomes was small. A very large sample size would be required to obtain reliable estimates of these infrequent events.
Interestingly, these children with mild to moderate acute asthma had better outcomes on prednisolone than on fluticasone, like the children with severe disease in our previous study.20 Similar to the children with severe asthma, our population likely had enough inflammatory edema and mucous plugging to prevent adequate delivery of inhaled fluticasone into the lungs, despite giving a large dose of the medication, pretreatment with albuterol via a highly efficient nebulizer with a mouthpiece,32,45 and the use of a proven inhalation technique33 and of the VHC, which optimizes pulmonary drug delivery and minimizes oropharyngeal deposition and systemic absorption.33,46 The lack of efficacy of inhaled fluticasone in our study, therefore, cannot be attributed to insufficient dosing. By 48 hours, enough of the initial airway obstruction had likely diminished to allow fluticasone to exert its full effect.
Our results may not be generalizable to children with very mild asthma. Many children with FEV1 >80% predicted were excluded, yet some of these were symptomatic enough to be given systemic corticosteroids. It is possible that this population may be more suitable for therapy with high-dose inhaled corticosteroids than that included in our study.
In conclusion, children with mild to moderate asthma improve faster when given oral prednisolone than with inhaled fluticasone. We recommend that fluticasone or other inhaled corticosteroids not be used in their ED management.
Supported by an investigator initiated grant from the Canadian Institutes for Health Research and GlaxoSmithKline Canada.
We are indebted to the medical and nursing staff for their collaboration and to Violeta Dukic for typing the article.
- Accepted March 20, 2006.
- Address correspondence to Suzanne Schuh, MD, FRCP(C), Division of Paediatric Emergency Medicine, Hospital for Sick Children, 555 University Ave, Toronto, Ontario, Canada M5G 1X8. E-mail:
The authors have indicated they have no financial relationships relevant to this article to disclose.
This work was presented in part at the annual meeting of the Pediatric Academic Societies; May 14–17, 2005; Washington, DC.
- ↵Scarfone RJ, Fuchs SM, Nager AL, Shane SA. Controlled trial of oral prednisone in the emergency department treatment of children with acute asthma. Pediatrics.1993;92 :513– 518
- ↵Shapiro GG, Furukawa CT, Pierson WE, Gardinier R, Bierman CW. Double-blind evaluation of methylprednisolone versus placebo for acute asthma episodes. Pediatrics.1983;71 :510– 514
- ↵Clinical Practice Guidelines. Expert Panel Report 2: Guidelines for the Diagnosis and Management of Asthma. NIH Publication No. 97–4051. Bethesda, MD: National Institutes of Health, National Heart, Lung and Blood Institute; 1997
- ↵Harding SM. The human pharmacology of fluticasone propionate. Respir Med.1990;84(suppl A) :25– 29
- ↵Barnes NC, Marone G, Di Maria GU, Visser S, Utama I, Payne SL. A comparison of fluticasone propionate, 1 mg daily, with beclomethasone dipropionate, 2 mg daily, in the treatment of severe asthma. International Study Group. Eur Respir J.1993;6 :877– 885
- Fabbri L, Burge PS, Croonenborgh L, et al. Comparison of fluticasone propionate with beclomethasone dipropionate in moderate to severe asthma treated for one year. International Study Group. Thorax.1993;48 :817– 823
- Adams RJ, Fuhlbrigge A, Finkelstein JA, et al. Impact of inhaled antiinflammatory therapy on hospitalization and emergency department visits for children with asthma. Pediatrics.2001;107 :706– 711
- ↵Volovitz B, Bentur L, Finkelstein Y, et al. Effectiveness and safety of inhaled corticosteroids in controlling acute asthma attacks in children who were treated in the emergency department: a controlled comparative study with oral prednisolone. J Allergy Clin Immunol.1998;102 :605– 609
- ↵American Thoracic Society. Standardization of spirometry. 1994 update. American Thoracic Society. Am J Respir Crit Care Med.1995;152 :1107– 1136
- ↵Diggle PJ, Liang K-Y, Zegerm SL. Analysis of longitudinal data. New York, NY: Oxford University Press; 1995
- ↵Little RC, Milliken GA, Walter W, Wolfinger RD. SAS system for mixed models: Cary, NC: SAS Institute; 1996
- ↵International Agency for Research of Cancer. Classical methods of analysis of grouped data. In: Statistical Methods in Cancer Research. Vol. 1. The Analysis of Case-Control Studies. Lyon, France: International Agency for Research of Cancer; 1980: IARC scientific publications no. 32
- ↵Ernst P, Suissa S. Measures of outcome. In: FitzGerald J, Ernst P, Boulet LP, O’Bryne P, eds. Evidence-Based Asthma Management. Hamilton, Ontario, Canada: BC Decker Inc; 2001
- ↵Kellie SE, Attfield MD, Hankinson JL, Castellan RM. Spirometry variability criteria - association with respiratory morbidity and mortality in a cohort of coal miners. Am J Epidemiol.1987;125 :437– 444
- ↵Newnham DM, Lipworth BJ. Nebuliser performance, pharmacokinetics, airways and systemic effects of salbutamol given via a novel nebuliser delivery system (“Ventstream”). Thorax.1994;49 :762– 770
- ↵Newman SP, Millar AB, Lennard-Jones TR. Improvement of pressurised aerosol deposition with Nebuhaler spacer device. Thorax.1984;39 :935– 941
- Copyright © 2006 by the American Academy of Pediatrics