Every year, croup affects up to 6 percent of children younger than six years of age. Approximately one third of these patients require hospitalization, and almost all require treatment in emergency departments and ambulatory clinics. The standard treatment of children with acute croup includes therapy with humidified air, although no evidence exists to show that this is effective, and epinephrine, which has been shown to provide temporary relief. Glucocorticoid therapy has shown some benefit, and interest in this agent is increasing. Ausejo and colleagues conducted a meta-analysis of all randomized controlled trials of glucocorticoid therapy in children with croup to determine its efficacy and clarify areas for research.

They identified studies conducted between January 1966 and August 1997 by searching databases, conducting follow-up on references of published trials and contacting researchers directly. Two researchers reviewed all studies for potential relevance. Two reviewers then independently assessed these studies. Only 24 of the 97 studies met the criteria for methodologic quality and suitability for inclusion in the meta-analysis. Each study included 36 to 60 children. Five trials compared active treatments; 19 were placebo-controlled. More than one half of the studies used the standardized Westley croup scale to assess outcome.

Glucocorticoid therapy was associated with a significant improvement in Westley croup score at six hours after treatment. The baseline rate of clinical improvement between the treated and control groups was 41 percent, with a number needed to treat of seven. At 12 hours after treatment, baseline clinical improvement was 68 percent, and the number needed to treat was five. At 24 hours after treatment, baseline clinical improvement was 83 percent, with a number needed to treat of eight. Children treated with glucocorticoids also showed a reduction in the number of adrenaline treatments, a decrease in time spent in emergency departments and a reduced length of hospital stay for those who were admitted. No significant difference was found in use of antibiotics between the treatment and control groups when expressed as the difference in risk.

The authors conclude that glucocorticoid (i.e., dexamethasone and budesonide) treatment is effective in relieving symptoms of croup in children as early as six hours and up to at least 12 hours after treatment and, in addition, reduces the need for other interventions. Although the meta-analysis consisted of mostly small studies and found evidence of publication bias, glucocorticoid treatment of children with croup who show signs of respiratory distress is recommended. Large trials using standardized outcome measures are needed to validate the published trial data and clarify the most effective dosage of glucocorticoids.