The use of adrenocorticoids to reduce the morbidity associated with laryngotracheitis (croup) remains controversial despite ten published reports of randomized trials involving 1,286 patients. To determine whether, viewed in aggregate, these studies demonstrate a significant benefit of steroid treatment for this disorder, a meta-analysis of the nine methodologically satisfactory trials was performed. Clinical improvement 12 and 24 hours posttreatment and incidence of endotracheal intubation were evaluated. For each end point, an estimate of the overall effect was obtained by calculating a typical odds ratio and 95% confidence interval. This analysis indicates that the use of steroids in children hospitalized with croup is associated with a significantly increased proportion of patients showing clinical improvement 12 hours (odds ratio = 2.25, 95% confidence interval = 1.66, 3.06) and 24 hours (odds ratio = 3.19, 95% confidence interval = 1.70, 5.99) posttreatment and a significantly reduced incidence of endotracheal intubation (odds ratio = 0.21, 95% confidence interval = 0.05, 0.84). Higher initial doses of steroid (≥125 mg of cortisone or ≥100 mg of hydrocortisone) were associated with a larger proportion of patients improved 12 hours posttreatment than was seen with lower doses. These results support the use of steroids in the treatment of hospitalized children with croup and, in the absence of a randomized clinical trial of sufficient size, provide the most reliable estimate of the impact of steroid therapy on the morbidity associated with croup. In addition, the results of this meta-analysis may be used to estimate the number of subjects who would be required to conduct a randomized clinical trial of steroids for the treatment of croup.
Estimation of the odds ratio in a proportional odds model with censored time‐lagged outcome in a randomized clinical trial
