Efficacy of oral corticosteroids for acute preschool wheeze: a systematic review and individual participant data meta-analysis of randomised clinical trials.

Abstract

Oral corticosteroids are commonly used for acute preschool wheeze, although there is conflicting evidence of their benefit. We assessed the clinical efficacy of oral corticosteroids by means of a systematic review and individual participant data (IPD) meta-analysis. In this systematic review with IPD meta-analysis, we systematically searched eight databases (PubMed, Ovid Embase, CINAHLplus, CENTRAL, ClinicalTrials.gov, EudraCT, EU Clinical Trials Register, WHO Clinical Trials Registry) for randomised clinical trials published from Jan 1, 1994, to June 30, 2020, comparing oral corticosteroids with placebo in children aged 12 to 71 months with acute preschool wheeze in any setting based on the Population, Intervention, Comparison, Outcomes framework. We contacted principal investigators of eligible studies to obtain deidentified individual patient data. The primary outcome was change in wheezing severity score (WSS). A key secondary outcome length of hospital stay. We also calculated a pooled estimate of six commonly reported adverse events in the follow-up period of IPD datasets. One-stage and two-stage meta-analyses employing a random-effects model were used. This study is registered with PROSPERO, CRD42020193958. We identified 16 102 studies published between Jan 1, 1994, and June 30, 2020, from which there were 12 eligible trials after deduplication and screening. We obtained individual data from seven trials comprising 2172 children, with 1728 children in the eligible IPD age range; 853 (49·4%) received oral corticosteroids (544 [63·8%] male and 309 [36·2%] female) and 875 (50·6%) received placebo (583 [66·6%] male and 292 [33·4%] female). Compared with placebo, a greater change in WSS at 4 h was seen in the oral corticosteroids group (mean difference -0·31 [95% CI -0·38 to -0·24]; p=0·011) but not 12 h (-0·02 [-0·17 to 0·14]; p=0·68), with low heterogeneity between studies (I2=0%; τ2<0·001). Length of hospital stay was significantly reduced in the oral corticosteroids group (-3·18 h [-4·43 to -1·93]; p=0·0021; I2=0%; τ2<0·001). Subgroup analyses showed that this reduction was greatest in those with a history of wheezing or asthma (-4·54 h [-5·57 to -3·52]; pinteraction=0·0007). Adverse events were infrequently reported (four of seven datasets), but oral corticosteroids were associated with an increased risk of vomiting (odds ratio 2·27 [95% CI 0·87 to 5·88]; τ2<0·001). Most datasets (six of seven) had a low risk of bias. Oral corticosteroids reduce WSS at 4 h and length of hospital stay in children with acute preschool wheeze. In those with a history of previous wheeze or asthma, oral corticosteroids provide a potentially clinically relevant effect on length of hospital stay. Asthma UK Centre for Applied Research.