Efficacy of oral corticosteroids in the treatment of acute wheezing episodes in asthmatic preschoolers: systematic review with meta-analysis

Abstract

Rationale

Systemic corticosteroids (SCS) are used for treat preschoolers with acute asthma or wheezing exacerbations, with conflicting results.

Objective

To evaluate the effectiveness of oral corticosteroids (OCS) compared to placebo in preschoolers presenting with acute asthma/wheezing exacerbations.

Methods

Five electronic databases were searched for all placebo-controlled, randomized clinical trials of OCS in children <6 years of age presenting with recurrent wheezing/asthma exacerbations of any severity. Primary outcomes were hospitalizations, unscheduled emergency department (ED) visits in following month, need of additional OCS courses, and length of stay (ED or hospital).

Results

Eleven studies met inclusion criteria (n=1733); four were conducted on an outpatient basis, five in inpatients, and two in the ED. Significant heterogeneity was found when pooling all studies, and thus analysis was stratified by trial setting. Among the outpatient studies, children who received OCS had a higher hospitalization rate (RR: 2.15 [95%CI=1.08-4.29], I²=0%) compared to those to received placebo. Among the ED studies, children who received OCS had a lower risk of hospitalization (RR: 0.58 [0.37-0.92], I²=0%). Among the inpatient studies, children who received OCS needed fewer additional OCS courses than those on placebo (RR: 0.57 [0.40 to 0.81], I²=0%).

Conclusions

Treatment with OCS in the ED or hospital may be beneficial in toddlers and preschoolers with frequent asthma/wheezing exacerbations. However, more studies are needed before OCS can be broadly recommended for this age group. Future trials should be carefully designed to avoid bias and according to our findings regarding administration setting.

INTRODUCTION

The management of asthma/wheezing in infants, toddlers and preschoolers (less than 6 years of age) is crucial due to several factors: an increase of asthma incidence in early ages, principally explained by variation in preschoolers¹; the greatest decline in lung function in asthmatics may occur during the preschool period²; preschool age is the age bracket with the poorest asthma control in childhood,³ with 48% of preschoolers with asthma reporting an exacerbation in the preceding year⁴. The annual rate of emergency department (ED) visits is 23-42 per 1000 for preschoolers vs. less than 15 per 1000 for those aged 6-70 yrs⁵, with the same pattern hospitalizations^5,6. Furthermore, the age at first hospitalization for asthma has decreased over time¹.

For decades, it has been standard practice among community pediatricians, allergists and pulmonary specialists to treat infants, toddlers, and preschoolers with acute episodes of wheeze with systemic corticosteroids (SCS), based on strong evidence of their efficacy in school-aged children and adolescents with asthma⁷. International guidelines recommend the use of SCS in children under 6 years old with severe wheeze/asthma exacerbations that do not respond to short active beta-2 agonists (SABA)^8,9, and thus up to 38% of wheezing infants receive OCS during their first year of life¹⁰. This practice is based on the general belief that episodes of wheeze in young children are likely early manifestations of asthma caused by the same pathophysiologic process of airway inflammation and narrowing⁸. However, recent trials evaluating the efficacy of OCS in preschoolers presenting to the ED or outpatient care for acute wheezing/asthma exacerbations have shown conflicting results, and a recent update on the efficacy of OCS in preschoolers does not support their efficacy, suggesting that the heterogeneity of early childhood wheezing might be at least partly responsible¹¹. At present, there is no systematic review with meta-analysis published for this specific age group.

The objective of this systematic review is to evaluate the efficacy of SCS (mainly OCS) use in children up to six years of age (including infants, toddlers and preschoolers) presenting with acute asthma or recurrent wheezing exacerbations. We hypothesized that SCS efficacy may vary according to severity of exacerbation and timing of administration, and thus may be more be effective in some settings than others.

METHODS

Search and Selection Criteria

This study was registered with the International Prospective Register of Systematic Reviews (PROSPERO, http://www.crd.york.ac.uk/PROSPERO) as CRD42015024332. We followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines to perform this review¹². The authors identified studies published in MEDLINE, EMBASE, CINAHL, SCOPUS and the Cochrane Controlled Trials Register (CENTRAL) databases and ClinicalTrials.gov until May 2015, using the terms “(oral corticosteroids OR steroids) AND (asthma OR wheeze) AND (infant OR toddler OR preschool).” Additionally, a search of relevant files from the drug manufacturer’s databases (published and unpublished) was performed. Finally, the references of all selected studies were also reviewed, and backward and forward references searches were performed. Language restrictions were not applied. To be included, studies had to meet all the following criteria: 1) children under 6 years of age with recurrent wheezing/asthma exacerbations of any severity presenting to the ED, receiving treatment at home (“outpatient studies”) or hospitalized for an asthma/wheezing exacerbation (“inpatient studies”); 2) randomized clinical trials (RCTs; parallel group or cross-over design) of any duration; 3) comparison of OCS (any type) vs. placebo; and 4) report at least one of the following primary outcomes: need of hospitalization, unscheduled visits to the ED in the four weeks following the trial intervention, length of hospital stay, or need for additional courses of SCS; or the following secondary outcomes: improvement of lung function measured by forced expiratory volume in 1 second (FEV₁) or peak expiratory flow (PEF), length of stay during the first ED visit, symptom scores, withdrawals (total and due to adverse effects [AEs]), and safety (AEs and serious AEs [SAEs]). An SAE was defined as any untoward medical occurrence that results in death, is life-threatening, requires inpatient hospitalization, or results in persistent or significant disability or incapacity¹³. We used study setting (outpatient, ED, or inpatient) as a proxy for exacerbation severity and timing of OCS administration.

Data Extraction and Assessment of Risk of Bias

Titles, abstracts, and citations were independently analyzed by the two authors (JCR and AB). From the full text, all studies were independently assessed for inclusion. Both authors were independently involved in all stages of study selection, data extraction, and risk of bias assessment. The latter was assessed according to recommendations outlined in the Cochrane Handbook¹⁴ for the following items: 1) adequacy of sequence generation; 2) allocation concealment; 3) blinding of participants and investigators; 4) blinding of outcome assessment; 5) incomplete outcome data; 6) selective outcome reporting, and other bias. Disagreements were discussed and resolved by the third investigator (EF).

Data Analysis

Analysis was performed by intention to treat and included all participants to minimize bias. Outcomes were pooled using mean differences (MD) (inverse variance method) or Mantel-Haenszel risk ratios (RR). Estimate precision was quantified by 95% confidence intervals (CI). Heterogeneity was measured by the I² test¹⁵ (≤ 25% absence of bias; 26 to 39% unimportant; 40% to 60% moderate; and 60% to 100% substantial bias). A fixed-effects model was used when there was no evidence of significant heterogeneity in the analysis (I² < 40%); if significant heterogeneity was found a random-effects model was used^16,17. A priori subgroup analyses included: type of OCS, age (<2 versus >2 years old), severity of exacerbation (mild versus moderate to severe), and trials sponsored by pharmaceutical industry versus independent trials. The meta-analysis was performed with the Review Manager 5.3.5 software (Cochrane IMS, 2014).

RESULTS

A total of 122 studies were initially identified in the database search and from other sources (Figure 1). Of these, 109 studies were excluded (reasons for exclusion: abstracts, letters, reviews, pooled analysis, no RCTs), leaving thirteen studies for which eligibility was assessed. Two of these studies were excluded because they did not meet inclusion criteria: one included the same population as another included study and added children with no prior wheezing episodes; the other included only participants with none or just one prior wheezing episodes.

FIGURE 1.

Process of study selection.

Included studies

Eleven RCTs published from 1986 to 2013 were included^18-28 (Table 1). Six were conducted in the United Kingdom^{18,20,24-26,28}, three in the U.S.^19,21,23, one in Israel²² and one in Finland²⁷. A total of 1,733 patients were randomized. Four of the studies were conducted on an outpatient basis^18-21, two were realized in the ED ^22,23 and five were inpatient studies^24-28.

Table 1. Characteristics of Included Studies.

Study	Design	Patients, n (% Male)	Mean Age Steroid group (SD or range)	Mean Age Placebo group (SD or range)	Selected Comparisons
Webb 198618	R,DB, CO, SC	38 (74%)	10.4 mo (4.1)	9.3 mo (3.7)	Prednisolone 1 mg/kg BID 5d, vs. placebo
Grant 199519	R,DB, CO, SC	86 (65%)	NR	NR	Prednisone 2 mg/kg 5d, vs. placebo
Oommen 200320	R,DB, PG, SC	225 (65%)	25 mo (17-37)	27 mo (19-38)	Prednisolone 20 mg daily 5d, vs. placebo
Beigelman 201321	R,DB, PG, MC	215 (NR) 278 (NR)	29 mo (0.84) 34.5 mo (0.77)	30 mo (0.84) 34.5 mo (0.77)	Prednisolone 2 mg/kg 2 d, then 1 mg/kg 2d, prescribed as rescue treatment according to predefined protocol
Tal 199022	R,DB, PG, SC	74 (62%)	23.1 mo (7-54)		Methylprednisolone 4 mg/kg IM once, vs. placebo
Scarfone 199323	R,DB, PG, SC	75 (72%)	59 mo (47)	63 mo (49)	ED: Prednisone 2 mg/kg once, vs. placebo After discharge: all prednisone 1 mg/kg BID 5d
Storr 198724	R,DB, PG, SC	140 (69%)	5.2 y (NR)	5.4 y (NR)	Prednisolone 30 mg once
Gleeson 199025	R,DB, PG, SC	39 (74%)	4.7 y (0.5)	5.1 y (0.7)	Hydrocortisone 6 mg/kg IV once then 2 mg/kg IV 4 hourly 1d then prednisolone 1 mg/kg BID 5d, vs. placebo
Fox 199626	R,DB, PG, SC	62 (69%)	7 mo (4-13)	7 mo (3-14)	Prednisolone 2 mg/kg 5d, vs. placebo
Jartti 200727	R,DB, PG, SC	58 (66%)	2.1 y (1.1)	2.9 y (1.4)	Prednisolone 2 mg/kg 3d, vs. placebo
Panickar 200928	R,DB, PG, MC	443 (65%)	25.8 mo (13.3)	26.2 mo (14.7)	Prednisolone 10 mg daily for children 10-24 mo and 20 mg daily for older children 5d, vs. placebo

AD=after discharge; BID=twice daily; CO=crossover; d=days; DB=double-blind; ED=emergency department; IM=intramuscular; MC=multicenter; mo= months; NR=not reported; PG=parallel group; R=randomized; SC=single center; y=years.

Seven studies used prednisolone^{18,20,21,24,26-28}, two studies used prednisone^19,23, one used methylprednisolone²² and one used hydrocortisone followed by prednisolone²⁵. Most studies maintained the treatment for three to five days^{18-21,23,25-28} and two studies used a single dose of OCS^22,24. All studies compared the treatment with OCS to placebo, with the exception of one post hoc analysis study²¹. One study compared OCS and placebo in the ED²³, but both groups received OCS after discharge, so only the first phase of the study was used for analysis. In one ED study²² the OCS was administered within 30 minutes of arrival at the ED, and in the other study this subject was not specified²³.

Two studies included only infants and toddlers^18,26 and four studies also randomized preschoolers^20-22,28. Five studies also included older children by protocol, but finally enrolled only a small proportion of older children or analyzed their results by different age range; therefore they were also included in the present analysis^19,23-25,27. Most studies enrolled children with a moderate acute asthma or recurrent wheezing exacerbation^18,21-28 and two studies gave no specification on episode severity^19,20. Four studies included some children being treated with inhaled corticosteroids (ICS)^20,21,27,28, three studies specifically excluded such population^22,23,25 and four studies did not mention ICS use^18,19,24,26. Among all of the studies, two were sponsored by the pharmaceutical industry^24,27, seven received other financial support^{18-21,25,26,28} and two did not specify funding source^22,23. Two studies were post-hoc analyses of two prior protocols^21,27: one of them was a subgroup analysis of a larger study comparing OCS with placebo for any wheezing episode, and included only recurrent wheezing episodes²⁷; the other one was a multicenter study, including two different trials, and compared the use of high dose ICS with montelukast and with albuterol alone during acute asthma²¹. In this latter study²¹, participants were not randomly assigned to treatment with OCS, and therefore it was included in the review but excluded from the meta-analysis. Methodological quality of included studies is described in Table 2; most studies had low methodological quality.

TABLE 2.

Risk of Bias of the Included Studies

Study	Random Sequence Generation	Allocation Concealment	Blinding of Participants & Personnel	Blinding of Outcome Assesment	Incomplete Outcome Data	Selective Reporting	Funded by pharmaceutical industry	Other bias
Webb 198618	U	U	Y	U	N	U	N	-
Storr 198724	U	U	Y	U	N	U	Y	-
Tal 199022	U	U	Y	Y	N	U	U	-
Gleeson 199025	U	U	Y	U	N	U	N	-
Scarfone 199323	U	U	Y	Y	N	U	U	-
Grant 199519	U	U	Y	U	N	U	N	-
Fox 199626	U	U	Y	U	N	U	N	-
Oommen 200320	Y	Y	Y	Y	Y	U	N	-
Jartti 200727	Y	Y	Y	Y	N	U	Y	Subgroup analysis of study designed for comparison of interest
Panickar 200928	Y	Y	Y	Y	N	U	N	-
Beigelman 201321	Y	Y	Y	Y	N	Y	N	Post hoc analysis of study not designed for comparison of interest

N=No; U=Unknown; Y=Yes.

Primary Outcomes

1. Hospital admission

Five studies reported hospital admission rates^18-20,22,23 (Figure 2a). There was no significant difference between OCS and placebo (RR: 1.00; 95% CI: 0.49-2.05), and there was significant heterogeneity between studies (I²=63%, p=0.03). Analyzing only outpatient studies^18-20, OCS treatment was associated with a higher hospital admission rate (Figure 2b; RR: 2.15; 95% CI: 1.08-4.29) with no heterogeneity (I²=0%, p=0.57). Considering only the two studies conducted in the ED^22,23, OCS treatment had a lower risk of hospital admissions (Figure 2c; RR: 0.58; 95% CI: 0.37-0.92) with no heterogeneity (I²=0%, p=0.73). In our subgroup analysis, we found no significant difference between OCS and placebo for children less than 2 years of age vs. older (RR: 0.41; 95% CI: 0.17 to 1.03; I²=0%, p=0.52)^18,22. When analyzing studies not sponsored by the pharmaceutical industry, those who received OCS had significantly lower risk of hospital admission than those on placebo (Figure 2b); however, opposite results were found when analyzing studies that did not specify funding (Figure 2c). When excluding the ED study by Scaropone et al.²³, in which 37% of the study population was over 5 years of age, the effect of OCS on hospital admission became non-significant. No differences were found by type of OCS. There were insufficient data to perform subgroup analyses by episode severity. No difference was observed after adjusting for methodological quality.

FIGURE 2.

2. Additional course of systemic corticosteroids

One outpatient study²⁰ and two inpatient studies^24,28 reported the need for additional courses of SCS (Figure 3a), showing no significant difference between OCS and placebo (RR: 0.74; 95% CI: 0.40 to 1.34). There was significant heterogeneity between studies (I²=62%, p=0.07). Analyzing only inpatient studies^24,28, the difference became significant favoring the OCS group (RR: 0.57; 95% CI: 0.40 to 0.81; I²=0%, p=0.34), (Figure 3b). However, when excluding the inpatient study by Storr et al.²⁴, in which 50% of the study population was over 4 years of age, the effect of OCS became non-significant. Subgroup analysis was not performed due to insufficient data.

FIGURE 3.

3. Unscheduled visits

Three inpatient studies reported unscheduled visits for asthma symptoms in the month following^25,27,28. There was no significant statistical difference between OCS and placebo (RR: 0.73; 95% CI: 0.35 to 1.52; I²=54%, p=0.11) (Figure 4). When excluding the study by Gleeson et al.²⁵, which included some children older than 5 years, the results did not change. No differences were observed when excluding data obtained from Jartti²⁷ (RR: 0.72; 95% CI: 0.25 to 2.07, I²=77%, p=0.04), considering it was performed on a post-hoc basis. On the other hand, one outpatient study reported more ED consults in the OCS group than placebo during the treatment period (0.87 ± 1.5 vs. 0.41 ± 0.81, respectively)¹⁹ in children under 6 years of age. Subgroup analysis was not performed due to insufficient data.

FIGURE 4.

Unscheduled visits for asthma symptoms

4. Hospital length of stay

Finally, four studies^25-28 reported no differences in hospital length of stay were found between OCS and placebo, while one study reported the OCS group had a shorter stay²⁴; however length of stay was reported differently (means vs medians) and thus we were not able to perform a pooled analysis.

Secondary Outcomes

Outpatient and ED studies did not give any data on pulmonary function. Two inpatient studies found an improvement in PEF measurements in the OCS group^24,25, while a third study showed no differences between both groups in impulse oscillometry values²⁷. There were insufficient data to perform a meta-analysis.

Outpatient studies found no differences between OCS and placebo on symptom scores,^18,20,21 but ED studies found better symptom improvement in the OCS group^22,23. No differences were shown in four inpatient studies^24,26-28, while a single one showed a bigger fall in heart rate²⁵. There were insufficient data to perform a meta-analysis of symptom scores due to different scores used.

None of the ED studies reported length of stay during the first consult in the ED. Some studies reported no relevant AEs^18,27,28, while one found minor AEs similar in both groups¹⁹. There were insufficient data to perform meta-analysis of AEs.

DISCUSSION

The use of OCS in the treatment of recurrent wheezing in infants, toddlers and preschoolers remains controversial. In this specific population, considering the overall group, we did not find evidence that OCS administration is effective compared to placebo in any of our primary outcomes: hospital admissions, additional systemic corticosteroid courses, hospital length of stay, and unscheduled ED visits for asthma symptoms.

Our analysis revealed very significant heterogeneity when pooling all identified studies. Strikingly, this heterogeneity disappeared when the analysis was stratified by the clinical setting in which the trials were performed (i.e. outpatient setting, ED, or inpatient setting). When analyzing studies performed in the ED, OCS treatment was associated with a lower hospitalization rate. Similarly, when analyzing studies performed in the inpatient setting, OCS treatment was associated with a lower need for additional courses of SCS. In the outpatient studies, on the other hand, OCS administration was associated with more hospital admissions, suggesting that OCS may not be beneficial in all clinical settings for this age group. In addition, it must be taken under consideration that the largest study in this group had a high rate of treatment noncompliance²⁰. Also, behavioral changes have been reported in children during OCS therapy, and this factor might affect the clinical decision to admit to the hospital²⁰.

These stratified findings support our a priori hypothesis that OCS may be more effective in certain subgroups depending on exacerbation severity or timing of administration: in this age group, OCS may be more beneficial among children who present with more severe exacerbations that require urgent care or hospitalization. However, asthma in young children is a heterogeneous condition with different underlying pathophysiological pathways, which could explain why some patients may benefit from OCS prescription while others show no response. In a recent study on preschoolers with severe recurrent wheeze, biopsies of children up to 36 months were found to contain more inflammatory cells with fewer eosinophils than those on older preschoolers²⁹. Most asthma/wheezing exacerbations in preschoolers are triggered by episodic viral infections^5,30. Moreover, the particular virus causing an asthma/wheezing exacerbation should be considered in this age group, since they produce a different immune response, which could affect the effectiveness of OCS;³¹ while prednisolone reduces wheezing relapse in children with acute rhinovirus infection, this effect was not observed in children with acute respiratory syncytial virus infection³². Other factors that may affect OCS response include the timing of administration^33,34, vitamin D deficiency³⁵, or genetic predisposition³⁶.

In regards to methodology, all included studies were randomized, double-blinded and placebo-controlled studies; however, in most of them allocation concealment was unclear, and therefore their overall reporting quality was suboptimal. While the dose (single or multiple) of OCS used in these eleven RCTs was adequate, as was stated recently³⁷, the different protocols used could also have contributed to the observed heterogeneity. Additionally, one study²⁷ was a post hoc analysis, although results did not change when it was excluded from the meta-analysis. Moreover, although the mean age of participants was within preschool age range, some of the analyzed studies included older children. Three studies^23-25 included some proportion of children 6 years of age or older; when excluding those studies the protective effect for hospital admission of OCS among the ED studies and for need of additional course of systemic corticosteroids among the inpatient studies became non-statistically significant. All these differences make it difficult to reach uniform conclusions on the use of OCS for asthma/wheeze in this age group.

There were not enough data to analyze results in terms of lung function or symptom scores. Similarly, available data on adverse effects was not enough to perform a meta-analysis. It is important to consider that OCS may have multiple systemic adverse effects, including alterations in adrenal function and bone metabolism^38,39,40. Therefore, it is critically important to accurately identify those preschoolers with asthma/wheezing exacerbations who would benefit from OCS administration.

In summary, current evidence is inadequate to formulate any broad clinical recommendations regarding the use of OCS in infants, toddlers and preschoolers with recurrent episodes of acute wheezing. OCS might potentially be beneficial (lower hospital admission rates and less need for additional courses of systemic steroids) in children with more severe asthma/wheezing exacerbations that present to the ED or require hospitalization; however, in order to answer this question more definitively, it will be important that future studies utilize a standardized case definition, larger sample size, and more homogeneous methodological quality are needed to give a more definitive answer. It will also be imperative to take into account our findings in terms of the setting of OCS administration.

LIST OF ABBREVITATIONS

AEs

adverse effects

CI

Confidence intervals

ED

Emergency department

FEV₁

Forced expiratory volume in 1 second

ICS

Inhaled corticosteroids

MD

Mean differences

OCS

Oral corticosteroids

PEF

Peak expiratory force

RCTS

Randomized clinical trials

RR

Risk ratio

SAEs

Serious adverse effects

SCS

Systemic corticosteroids