The Cochrane Library and safety of systemic corticosteroids for acute respiratory conditions in children: an overview of reviews

Background

Acute respiratory conditions are a leading cause of childhood morbidity and mortality. Corticosteroids are effective and established treatments in some acute respiratory infections (e.g. croup) and asthma exacerbations; however, their role is controversial in other conditions owing to inconsistent effectiveness or safety concerns (e.g. bronchiolitis, acute wheeze).

Objectives

To examine clinically relevant short-term safety outcomes related to acute single or recurrent systemic short-term (<2 weeks) corticosteroid use based on systematic reviews of acute respiratory conditions.

Methods

We searched the Cochrane Database of Systematic Reviews in February 2013 for systematic reviews comparing systemic corticosteroids with placebo for children (aged 0–18 years) with acute asthma, preschool wheezing, bronchiolitis, croup, pharyngitis/tonsillitis or pneumonia. We selected the following outcomes a priori: gastrointestinal (GI) bleeding and abdominal pain; behavioural effects (tremor or hyperactivity, jitteriness, irritability or emotional distress); hypertension; serious adverse events, including death, length of stay in hospital; and relapse leading to hospitalization. One reviewer extracted data and another reviewer independently verified data. Results were combined using Peto odds ratios and risk differences (RD) for dichotomous outcomes and mean differences for continuous outcomes.

Main results

Seven reviews containing 44 relevant randomized controlled trials were included. Three reviews were on asthma and one each on bronchiolitis, croup, wheeze and pharyngitis/tonsillitis. Six trials (2114 patients) assessed GI bleeding and/or abdominal pain and showed no significant differences between corticosteroids and placebo (1.5% vs. 1.8%, respectively). Various behavioural effects and hypertension/blood pressure were measured in four trials each (838 and 1617 patients, respectively), with no significant differences reported. None of the trials reported deaths in any of the treatment groups. Based on 17 trials (2056 patients), there were significantly fewer admissions at day 1 with corticosteroids (risk differences = −0.11, 95% confidence interval −0.18 to −0.05; Peto odds ratios = 0.63, 95% confidence interval 0.52 to 0.78). Based on 16 trials (1502 patients) corticosteroids resulted in over 8 fewer hours in hospital compared with placebo (mean differences = −8.49 hours, 95% confidence interval −1.76 to −3.23). There were significantly fewer relapses leading to hospitalization (13 trials, 1099 patients) with corticosteroids (Peto odds ratios 0.42, 95% confidence interval 0.23 to 0.76). While differences favouring corticosteroids in hospital-related outcomes were restricted to asthma and/or croup, we did not find any increase in hospital admission at day 1, length of stay or re-hospitalization in the other acute respiratory conditions.

Authors' conclusions

Practitioners may prescribe systemic corticosteroids in otherwise healthy children when indicated for the management of acute respiratory conditions (i.e. infections or asthma exacerbations) with minimal concern about short-term adverse effects.