Large Trial Shows Corticosteroids Don't Help Against Infant Wheeze

SALT LAKE CITY, July 25 -- Corticosteroids appear to be ineffective against first episode bronchiolitis in infants, despite frequent emergency room use, researchers here said.

In a large, carefully designed trial, hospitalization rates in infants with moderate-to-severe wheeze were no different for those given oral dexamethasone than for those given placebo (39.7% versus 41.0%, P=0.74).

Nor did respiratory status improve significantly with corticosteroids compared with placebo (P=0.21), Howard M. Corneli, M.D., of the University of Utah and colleagues reported in the July 26 issue of the New England Journal of Medicine.

But, "withholding therapy is much more difficult than giving it," commented Caroline Breese Hall, M.D., of the University of Rochester in Rochester, N.Y., in an accompanying editorial.

"Despite the value of this study," she wrote, "the long history of therapies and recommendations attending bronchiolitis suggest that the study's results will not appreciably change the nature of the care provided by the primary physician faced with a young, distressed infant and anxious parents."

Effective treatment has been elusive and controversial, Dr. Hall noted.

Corticosteroids are commonly used to treat bronchiolitis despite lack of overall benefit in reviews by the American Academy of Pediatrics, the Cochrane Collaboration, and the Agency for Healthcare Research and Quality, which were based predominantly on results of small studies.

Antiviral drugs, bronchodilators, and antibiotics are not routinely recommended either, yet 50% to 80% of infants hospitalized with bronchiolitis receive one of these medications or corticosteroids or a combination thereof, the researchers noted.

So to determine the efficacy of a single dose of dexamethasone, they conducted a trial among 600 infants ages two to 12 months treated in 20 emergency departments for a first episode of moderate-to-severe bronchiolitis and no history of asthma.

The infants were randomized to receive 1 mg dexamethasone per kilogram of body weight or an identical oral placebo solution.

The trial ran from January 2004 through April 2006 and baseline characteristics were similar between groups.

Other bronchiolitis treatments, diagnostic testing, and viral assays were left to the clinician's discretion. The majority of infants in both groups got inhaled albuterol (77.0% in the dexamethasone group and 80.3% in the placebo group) or epinephrine (15.5% and 16.7%, respectively) with a similar number of treatments in each group.

However, none of the endpoints favored dexamethasone treatment to a clinically significant degree.

For the primary endpoint of hospitalization or discharge at four hours after medication administration, there was no difference in admission rates (absolute difference -1.3%, 95% confidence interval -9.2 to 6.5, P=0.74).

This was corroborated by the secondary outcome of change in respiratory status after four hours as measured by the Respiratory Assessment Change Score. Mean scores were similar between groups (-5.3 dexamethasone versus -4.8 placebo, P=0.21).

The same was true for the prespecified subgroup of children with eczema or a family history of asthma (absolute difference -1.3% for hospitalization, 95% CI -11.1 to 8.5, and -0.4 for respiratory score change, 95% CI -1.3 to 0.6).

Some of the other endpoints significantly differed between groups but the differences were small, the researchers noted. These findings were:

• A 0.5-point better improvement in Respiratory Distress Assessment Instrument score for the dexamethasone group than the placebo group (P=0.03).
• A one-breath per minute lower respiratory rate change with placebo compared with dexamethasone (P=0.39).
• Better improvement in oxygen saturation with placebo than dexamethasone (0.9% versus 0.3% change, P=0.02).
• An eight-beat per minute better improvement in heart rate with dexamethasone than placebo (P<0.001).
• A 0.4? C better reduction in temperature with dexamethasone than placebo (P<0.001).
• Mean length of hospitalization was 2.55 days in the dexamethasone group versus 2.27 days in the placebo group (P=0.10).
• Subsequent hospital admissions in the week after intervention were more common with dexamethasone (4.2% versus 3.8%).

A subanalysis showed no difference in outcomes whether children tested positive or negative in tests for respiratory syncytial virus.

Because the study looked only at short-term outcomes, though, it could not determine whether corticosteroid treatment would impact development of later pulmonary problems, such as the recurrent wheezing that develops in 20% to 50% of infants who have had bronchiolitis, Dr. Hall said.

Furthermore, the study included only a small proportion of screened patients.

Among 8,686 infants screened for the study, 91% did not meet inclusion criteria. Two-thirds had either had prior wheezing (41%) or had too mild a case of bronchiolitis (a respiratory distress score less than six).

"Still, these infants were judged by their parents or primary physician to be ill enough to require evaluation in an emergency department, and they would make up an even larger proportion of the bronchiolitis cases seen in office practices than the infants who were enrolled in the study," Dr. Hall noted.

These limitations do not diminish the soundness of the results and their "clinical pertinence," she said. "The results, although negative, are important."

One researcher reported receiving grant support from Sepracor; another reported receiving consulting fees from St. Jude Medical, Tyco Health Care, and Ucyclyd Pharma and grant support from St. Jude Medical and GlaxoSmithKline. Dr. Hall reported no potential conflict of interest.