BACKGROUND. Patients with active Crohn's disease are often treated with corticosteroids, but the treatment has many side effects. Budesonide is a potent, well-absorbed corticosteroid, but because of a high rate of first-pass metabolism in the liver, its systemic bioavailability is low. METHODS. We conducted a randomized, double-blind, 10-week trial comparing the efficacy and safety of an oral controlled-release form of budesonide with the efficacy and safety of prednisolone in 176 patients with active ileal or ileocecal Crohn's disease (88 patients in each treatment group). The dose of budesonide was 9 mg per day for eight weeks and then 6 mg per day for two weeks. The dose of prednisolone was 40 mg per day for two weeks, after which it was gradually reduced to 5 mg per day during the last week. RESULTS. At 10 weeks, 53 percent of the patients treated with budesonide were in remission (defined as a score < or = 150 on the Crohn's disease activity index), as compared with 66 percent of those treated with prednisolone (P = 0.12). The mean score on the Crohn's disease activity index decreased from 275 to 175 in the budesonide group and from 279 to 136 in the prednisolone group (P = 0.001). Corticosteroid-associated side effects were significantly less common in the budesonide group (29 vs. 48 patients, P = 0.003). Two patients in the prednisolone group had serious complications (one had intestinal perforation and one an abdominal-wall fistula). The mean morning plasma cortisol concentration was significantly lower in the prednisolone group than in the budesonide group after 4 weeks (P < 0.001) and 8 weeks (P = 0.02) of therapy, but not after 10 weeks. CONCLUSIONS. Among patients with active Crohn's disease, both controlled-release budesonide and prednisolone are effective in inducing remission. In this trial, prednisolone reduced scores on the Crohn's disease activity index more, whereas with budesonide there were fewer glucocorticoid-associated side effects and less suppression of pituitary-adrenal function.