A prospective, double-blind, randomized study of the role of corticosteroids in the treatment of tuberculous pleurisy was performed in 40 patients. All patients received adequate antituberculosis chemotherapy (isoniazid, 300 mg/day; rifampin, 450 mg/day; ethambutol, 20 mg/kg/day) for more than nine months. They were randomly assigned to take prednisolone 0.75 mg/kg/day orally or placebo for the initial treatment, which was tapered gradually for the next two to three months. Twenty-one were treated with steroids and 19 were given a placebo. The two groups were identical with regard to age, sex, duration from onset of symptoms to diagnosis, and initial amount of pleural effusion. The mean duration from symptoms (fever, chest pain, dyspnea) to relief was 2.4 days in the steroid-treated group, and 9.2 days in the placebo group (p less than 0.05). Complete reabsorption of pleural effusion occurred an average of 54.5 days in the steroid-treated group and 123.2 days in the placebo group (p less than 0.01). The development of residual pleural thickening was not influenced by the administration of corticosteroids. No serious side effects were noted during the treatment in either group. We conclude that the administration of corticosteroids, in conjunction with antituberculosis chemotherapy, will resolve the clinical symptoms more quickly and hasten the absorption of pleural effusion in patients with tuberculous pleurisy.