In order to determine objectively the efficacy of corticosteroids in relieving severe acute episodes of asthma, we administered infusions of hydrocortisone or placebo in a random, double-blind manner to 20 asthmatic subjects after they had been documented to be refractory to eight hours of conventional therapy. Eleven subjects received hydrocortisone (2 mg/kg bolus, then 0.5 mg/kg per hour for 24 hours) and nine received saline. All were given identical bronchodilator treatment during the study period, and all had multiple aspects of lung function serially recorded along with plasma cortisol levels. Although subjects in both groups had severe obstruction of similar magnitude at the beginning of treatment (one-second forced expiratory volume [FEV1] in placebo-treated group = 32 +/- 3 [SEM] percent of predicted, and 25 +/- 3 percent of predicted in steroid-treated group, p = NS), at the end of 24 hours, the subjects given corticosteroids had significantly greater resolution of airway obstruction (FEV1 in steroid-treated group increased 118 +/- 25 percent from control value, versus 35 +/- 22 percent with placebo). In five of nine subjects treated with placebo, pulmonary mechanics either were unchanged or deteriorated during the period of observation. There was no effect of the glucocorticoids on arterial blood gases, and no significant correlation could be found between plasma cortisol levels and the improvement in pulmonary mechanics and clinical status. These results provide objective documentation of the time course over which administration of parenteral corticosteroids speeds the recovery of asthmatic patients who are unresponsive to standard therapy.