Nebulized Budesonide, Intramuscular Dexamethasone, and Placebo for Moderately Severe Croup

Abstract & Commentary

Synopsis: In children with moderately severe croup, treatment with intramuscular dexamethasone or nebulized budesonide resulted in more rapid clinical improvement than did administration of placebo with dexamethasone offering the greatest improvement. Treatment with either glucocorticoid resulted in fewer hospitalizations.

Source: Johnson DW, et al. A comparison of nebulized budesonide, intramuscular dexamethasone, and placebo for moderately severe croup. N Engl J Med 1998;339:498-503.

Johnson and associates at the hospital for Sick Children in Toronto compared the effects of parenteral dexamethasone and nebulized budesonide on symptoms and need for hospitalization in children with moderately severe croup. They conducted a double-blind, randomized study involving 144 children with croup who were seen in their ED during an eight-month period of time in the winter and spring of 1995-1996 because of the acute onset of stridor with a "seal-like" barking cough. Ages ranged between 3 months to 9 years of age.

The children were randomized to receive either nebulized budesonide (4 mg), intramuscular dexamethasone, (0.6 mg/kg), or placebo. They were assessed prior to therapy and then hourly for five hours. The characteristics of the three groups were similar at base line, including the types of viruses identified, the types of croup, and the assessed clinical severity of illness. The overall rates of hospitalization were 71% (35 of 49) in the placebo group, 38% (18 of 48) in the bedesonide group, and 23% (11 of 47) in the dexamethasone group. Children treated with budesonide or dexamethasone had a statistically, significantly greater improvement in croup scores than those receiving placebo, and those treated with dexamethasone had a greater improvement than those treated with budesonide.

Comment by Thomas F. Dolan, MD, FAAP

Johnson et al have performed a careful study comparing the treatment of moderately severe croup with intramuscular dexamethasone, nebulized budesonide, or placebo. Patients in all three groups additionally received standard therapy including racemic epinephrine by inhalation and mist therapy.

Parainfluenza virus A, B, or C was isolated from approximately 30% of their patients. Another 10% grew influenza or RSV virus. Only six of 144 patients were diagnosed as spasmodic croup. Twenty-seven percent of patients had a history of previous croup, and 18% had a history of having prior treatment with racemic epinephrine. There was more rapid improvement clinically in patients treated with steroids than placebo and a marked decrease in hospitalization rate. The intramuscular dexamethasone (0.6 mg/kg) given once proved superior to inhaled budesonide. This is somewhat surprising to me, since, theoretically, more steroid should be delivered to the subglottic area more rapidly by inhalation. Perhaps in an ED setting, a frightened toddler might not do a good job inhaling steroids.

Historically, the management of croup has changed drastically over the past 30 years. Those of us with gray hair may remember the infamous "croup rooms" in many pediatric hospitals. One had to don a raincoat when entering these rooms and was barely able to see the patient through the fog. The mist was supposed to liquefy mucus. Cold mist rooms were replaced by mist tents, and it became fashionable to treat all children with respiratory tract disease with mist tents. Perhaps the good results people thought they were seeing were due to the fact that the mist was generated by oxygen rather than room air (this was before oximetry was available). Mist tent therapy was abandoned when Wolsdorf et al showed that mist therapy was only an expensive way to moisturize the nasal canal.1 A more serious problem with mist tents was that one cannot easily see or evaluate the infant. A study comparing the effects of humidified air vs. non-humidified air showed no statistical benefits.2

Antibiotics were commonly used to treat croup for fear the patient had epiglottitis due to Haemophilus Influenzae type b. I never thought the presentation was similar, but the problem has become moot since virtually all patients receive Hib vaccine, and Hib epiglottis has essentially disappeared in the United States.

There were great arguments about the value of racemic or l-epinephrine. Many clinicians believed that any child receiving these drugs should be admitted to the hospital because of a concern that a late rebound of symptoms might occur.

The value of steroids in croup was the topic of violent (and mostly anecdotal) arguments until the article by Super et al definitively proved that steroids were beneficial.3 Newer studies, such as the one conducted by Johnson et al, show that nebulized or parenteral steroids improve outcome.

I personally favor using intramuscular dexamethadose because oral steroids, both tablet and liquid preparations, are bitter and may be difficult to administer to an excited, frightened child with respiratory distress in the ED milleau. I suspect that a combination of an oral or IM steroid plus one or two doses of a nebulized steroid will be evaluated in the near future in an attempt to further decrease hospitalization rates. (Dr. Dolan is Professor of Pediatrics at the Yale-New Haven Children’s Hospital.)


1. Wolstorf J, et al. Mist therapy reconsidered: An evaluation of the respiratory deposition of labeled water aerosols produced by jet and ultrasonic nebulizers. Pediatrics 1969;43:799-794.

2. Bourchier D, et al. Humidification in viral croup in a controlled trial. Aust Paediatr J 1989;20:289-291.

3. Super PM. A prospective randomized double-blind study to evaluate the effect of dexamethasone in the out-patient management of acute laryngotracheitis. J Pediatr 1989;115:323-329.