Ventilator-dependent premature infants are often treated with dexamethasone. However, the optimal timing of therapy is unknown. We compared the benefits and hazards of initiating dexamethasone therapy at two weeks of age and at four weeks of age in 371 ventilator-dependent very-low-birth-weight infants (501 to 1500 grams) who had respiratory-index scores (mean airway pressure x the fraction of inspired oxygen) of greater than or equal 2.4 at two weeks of age. The primary outcome was the number of days from randomization to extubation not requiring reintubation (extubation score or death). The secondary outcomes were death before discharge from the hospital; the duration of assisted ventilation, supplementary oxygen therapy and hospital stay; the incidence of chronic lung disease (defined as the need for supplemental oxygen at 36 weeks postconceptional age by best obstetrical estimate) and rates of morbidity and mortality from respiratory causes during the first year. Additional secondary endpoints were hyperglycemia, hypertension, growth, bacteremia, nectrotizing enterocolitis and upper GI bleeding. The sample size of 370 was based on a 0.60 probability that the extubation score of late treatment was greater than early treatment, a 5% two-sided type

1 error, 85% power, and 10% treatment noncompliance.

Infants were randomized to either receive dexamethasone for two weeks followed by saline placebo for two weeks, or saline placebo for two weeks followed by either dexamethasone or additional placebo for two weeks (if they still met entry criteria). Dexamethasone was given at a dose of 0.25 mg per kilogram of body weight twice daily intravenously or orally for five days, and the dose then tapered.

The median time to ventilator independence was 36 days in the dexamethasone-placebo group and 37 days in the placebo-dexamethasone group. The incidences of chronic lung disease (defined as the need for oxygen supplementation at 36 weeks postconceptional age) were 66 percent and 67 percent, respectively.

Dexamethasone was associated with an increased incidence of nosocomial bacteremia (relative risk, 1.5; 95 percent confidence interval, 1.1 to 2.1) and hyperglycemia (relative risk, 1.9; 95 percent confidence interval, 1.2 to 3.0) in the dexamethasone-placebo group, elevated blood pressure (relative risk, 2.9; 95 percent confidence interval, 1.2 to 6.9) in the placebo-dexamethasone group, and diminished weight gain and head growth (P less than 0.001) in both groups. Treatment of ventilator-dependent premature infants with dexamethasone at two weeks of age is more hazardous and no more beneficial than treatment at four weeks of age.