Efficacy of Inhaled Corticosteroids in Infants and Preschoolers With Recurrent Wheezing and Asthma: A Systematic Review With Meta-analysis

METHODS

RESULTS

Eighty-nine abstracts were identified in the initial search. Of them, 29 randomized, controlled trials (3592 children) met the inclusion criteria and were selected for analysis.^9,10,18–44 Details of the selection process are shown in Fig 1. Twelve trials included fluticasone MDI^{9,10,31,35,36,38–44}, 7 nebulized budesonide suspension,^{23,26,27,29,30,32,33} 7 budesonide MDI,^{19–22,24,25,34} 2 beclomethasone MDI,^28,37 and 1 nebulized beclomethasone suspension¹⁸ like ICS treatment. The main characteristics of these studies are described in Table 1.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 1

Flowchart for identification of usable studies.

Wheezing/Asthma Exacerbations

Data from 16 studies*with 2805 patients showed a significant reduction in the incidence of WAEs between the ICS and placebo groups (18.0% and 32.1%, respectively), with a RR of 0.59 (95% CI: 0.52–0.67; P = .0001; I² = 10%) (Fig 2). Treatment of 7 subjects with ICS therapy prevented 1 person from experiencing a WAE compared with placebo (NNT: 7 [95% CI: 6–9]). Although patients with wheeze who received ICSs showed a significant reduction of WAEs compared with those taking placebo, they presented a higher rate of WAEs compared with asthmatic patients (Table 2). Otherwise, there were no significant differences according to age (infants versus preschoolers), atopic status, Jadad score (<4 vs ≥4), method of ICS delivery (MDI versus nebulizer), ICS choice (budesonide MDI versus fluticasone MDI), or duration of study (<12 vs ≥12 weeks).

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 2

Pooled RRs for WAEs (with 95% CIs) of eligible studies comparing ICSs with placebo.

View this table:

• View inline
• View popup

TABLE 2

Sensitivity Analysis of WAEs

DISCUSSION

To our knowledge, this is the first meta-analysis performed exclusively to explore the efficacy of ICSs in infants/preschoolers with wheeze or asthma. Regarding the primary outcome, we found that ICS therapy was associated with significant reduction in WAEs (in nearly 40% and with a NNT of 7). This is a very relevant finding, because WAEs account for the largest part of direct health costs for asthma. Although this WAE reduction appeared in both wheeze and asthmatic groups, posthoc subgroup analysis suggests that this effect was more relevant in children with a diagnosis of asthma. On the other hand, this beneficial effect was independent of age, atopic condition, type of ICS, mode of delivery, and study quality and duration.

In addition, children treated with ICSs had significantly fewer withdrawals because of WAEs (in nearly 50%), less albuterol use, and more clinical (change in symptoms score) and functional (change in PEF and FEV₁ from baseline) improvement than those children on placebo treatment.

There is no doubt that chronic asthma in school-aged children responds well to ICSs, accounting for their adoption by most of the latest guidelines worldwide.^6–8 However, the efficacy of ICSs in infant/preschooler wheezing has been controversial, with the different response at this age hypothesized to be a consequence of dissimilar underlying pathology.⁵ From epidemiologic studies, we know that before the age of 6, mainly 2 phenotypes of wheezing can coexist: “transients” (those who have symptoms in the first 3 years of life, and after that 70%–80% become asymptomatic and without risk for later development of asthma) and “persistents” (those who continue to be symptomatic into childhood). This latter group can be further divided into 2 groups: “immunoglobulin E–associated wheezers,” who almost certainly have eosinophilic airway inflammation, the counterpart of adult atopic asthma, and “nonatopic wheezers,” who also have persistent symptoms and in whom viral infection will be implicated.¹

Obviously, these divisions are an oversimplification, because some atopic children may stop wheezing by 5 years of age, whereas others (nonatopic children) could wheeze with viral infection into midchildhood. Moreover, an apparent phenotype may change with time (eg, an infant may initially not have any atopic manifestations but may wheeze with viral colds and only later in life develop skin-prick test sensitivity and interval symptoms). However, a recent study⁴⁵ revealed similar pathologic features (epithelial reticular basement membrane thickness and eosinophilic inflammation on the bronchial biopsies) from preschoolers with wheezing with and without atopic characteristics as those seen in adults and school-aged children with asthma. Another division that can also be helpful to clinicians is the pattern of symptoms: “intermittent” (symptoms mainly associated with viral infections or virus-induced wheeze and without interval symptoms) and “persistent” (those associated with wheezing episodes both associated with viral infections and between viral infections [eg, other triggers such as allergens, cold, pollutions, etc] and with continued symptoms over time).⁵

However, our results show that the ICSs were clinically superior to placebo in children with recurrent wheezing, either among infants (the range of age in which we can speculate that more transients wheezers can exist, although no studies on children with intermittent wheeze and intermittent treatment were included) and preschoolers (in whom there is likely to be a predominance of persistent wheezing) or among children with and without atopic predisposition. However, the effect of ICSs was more efficacious in reducing WAEs in children with a diagnosis of asthma than in those with a diagnosis of wheeze. However, it is important to note that many of these children are too young to have a confirmed diagnosis of asthma and to have completely manifested atopy. Finally, the widespread guideline recommendations^6–8 and recent task force report¹² to use ICSs in infants/preschoolers with recurrent episodes of wheeze (of appropriate severity) are supported by the present meta-analysis, with the advantage that in this one the NNT were calculated. Nevertheless, it is important to stress that, ideally, the children who are likely to outgrow symptoms (that could be those with the diagnosis of wheeze) should be treated with minimal therapy, in comparison with the child who will progress to having chronic asthma that might merit more aggressive or longer therapy.

Ten years ago, Calpin et al⁴⁶ published a meta-analysis on 24 randomized placebo-controlled trials in childhood asthma (only 10 studies were performed in infants/preschoolers [n = 377]) and showed that those with persistent wheezing who received ICSs (beclomethasone or budesonide) for 4 to 24 weeks had better clinical (lower symptom score and β₂-agonist and oral steroid uses) and functional (increased PEF rate) improvement than those who used a placebo. However, there was not a significant difference in the rate of hospital admissions (only 3 of 10 trials used this outcome variable). A recent systematic review without meta-analysis⁴⁷ that included 9 additional randomized, double-blind, controlled trials of ICSs in preschool-aged children (n = 1196) with multiple-trigger or persistent wheeze confirmed those results, and the authors suggested that only children with troublesome symptoms and frequent use of systemic corticosteroids should receive ICSs as a maintenance treatment for reducing their symptoms.

As was stated in the recent guidelines,^6–8 the goal of asthma therapy is to maintain the longest duration of control asthma possible with the least amount of medication and, hence, with the lowest risk for adverse effect. Therefore, it is crucial to consider the importance of monitoring the response to ICSs, and if a clear and beneficial response is not obvious in 4 to 6 weeks with satisfactory medication technique and adherence, the treatment should be stopped and alternative therapies or an alternative diagnosis should be considered. On the other hand, if a clear and beneficial response is sustained for at least 3 months, it is appropriate to consider a step down in therapy to evaluate the need for continued, daily long-term therapy, because some children have high rates of spontaneous remission of symptoms.⁷

For selecting those preschoolers with a greater risk of asthma, several clinical scores have been developed, the most popular of which is the asthma predictive index. This score, for use in children with 3 episodes of wheezing before the age of 3 years, includes 2 major (parental history of asthma or personal history of eczema) and 3 minor (blood eosinophilia, wheezing without colds, and allergic rhinitis) criteria.⁴⁸ The presence of 1 major or 2 minor criteria is associated with an increased risk of continued wheezing at 5 years of age. However, it is one thing to identify a child who is more prone to be an asthmatic in the future and quite another to be able to modify the natural history of the disease. The latter was tested by Guilbert et al,⁹ who after studying 285 preschoolers with a positive modified asthma predictive index found that those children on fluticasone improved their clinical and functional parameters during the 2 years of treatment in comparison to those in the placebo group, but at the end of the 1-year observation study (free of ICSs) the symptoms and pulmonary function were not different from those treated with placebo for 2 years, confirming that ICSs are helpful in decreasing symptoms but do not modify the natural history of asthma.

CONCLUSIONS

This meta-analysis shows that ICSs are useful in infants and preschoolers with persistent wheeze/asthma in reducing exacerbations (nearly in 40%) and withdrawals caused by exacerbations (nearly in 50%) as compared with placebo independent of age, diagnosis, atopy, mode of delivered, and ICSs used. Also, infants/preschoolers on ICSs had less albuterol use and greater clinical (change in symptoms score) and functional (change PEF and FEV₁ from baseline) improvement than those on placebo. These results support the current recommendations of most international asthma guidelines.