Introduction Acute bronchiolitis is the most common cause of hospitalisation in infants worldwide. A disease characterised by inflammation of the lower respiratory tract, corticosteroids and dexamethasone in particular have been investigated as possible therapeutic agents, with varying results. We set out to systematically review the efficacy of dexamethasone in treating bronchiolitis.

Material and methods Data sources were MEDLINE, EMBASE, Cochrane Central Register of Controlled trials, and reference lists of retrieved articles. We included randomised controlled trials comparing dexamethasone with placebo or other interventions and investigating infants, aged under 24 months that had been diagnosed with bronchiolitis. Two review authors independently extracted data from the included studies and assessed the trials' quality. Where appropriate, meta-analyses were conducted.

Results 12 studies met the inclusion criteria. Five of the studies investigated oral dexamethasone, three intravenous, three intramuscular and one nebulised preparations. Due to significant heterogeneity in results, there was limited scope for meta-analysis of primary and secondary outcomes. The oral dexamethasone trials demonstrated no significant difference in change-to-baseline clinical scores, with a pooled weighted mean difference of −0.29 (p=0.06), while the intravenous trials showed no significant difference in hospital stay, with a weighted mean difference of 15.25 h (p=0.39). Meta-analysis found no statistically significant difference in adverse events between those receiving dexamethasone and placebo.

Conclusions The evidence suggests that dexamethasone, although commonly used in bronchiolitis, has no significant benefit as compared to placebo in this disease. Despite much research having been conducted in this area, many studies have been of low quality and small statistical power. This review has found significant clinical and methodological heterogeneity between studies. Recommendations of the authors therefore are that future conducted studies should be of adequate power with robust methodology.