Glucocorticoids in the treatment of croup: barking up the right tree.

Croup, a syndrome characterized by a barking cough, inspiratory stridor and respiratory distress, afflicts up to 3% of children less than 6 years of age each year. The syndrome is usually caused by infection with a virus, most often parainfluenza 1 virus, and this infection is believed to cause inflammation of the larynx and trachea, which results in the characteristic symptoms. Admission rates for patients with croup presenting to emergency departments have been documented at up to 30%. The optimal management of croup has been hotly debated for 40 years, with the dividing line at times falling between the “town” and “gown” groups (see box this page). However, the issue now appears to be resolved. Meta-analyses, which increase the power and precision of overall estimates, and recent randomized controlled trials (RCTs) (including our own) have shown that steroids reduce both the frequency of complications of croup and the rates of visits to emergency departments and admissions to hospital. Kairys and colleagues, in a meta-analysis on the effectiveness of glucocorticoids for inpatients with croup, found that there was a significant clinical improvement by 12 and 24 hours after administration of these agents, as well as a significant decrease in the probability of intubation. These results supported Coffin’s assertions from the early 1970s, an interesting paradox considering his abhorrence of double-blind studies. We performed a cumulative meta-analysis, which showed that there was evidence as early as 1966 for the effectiveness of glucocorticoids in producing a significant improvement in croup patients 12 hours after administration. This evidence had been gathered from experimental work involving inpatients with croup and hence was not immediately generalizable to outpatients. An intriguing question has been whether intervening at an earlier stage in the child’s illness, while he or she is an outpatient, would lead to a reduction in hospital admission rates. Furthermore, would direct administration of budesonide, a nebulizable steroid, to the affected area — the larynx and trachea — be an effective method of administering glucocorticoid? One of our trials demonstrated that, relative to placebo, nebulized budesonide did indeed lead to faster clinical improvement, shorter stay in the emergency department and reduced probability of admission to hospital. Earlier in the 1990s there was a dramatic increase in the number of croup trials, most involving outpatients. It was becoming evident that not only were glucocorticoids effective for inpatients, but they also offered significant benefits to children with croup who had Editorial