It is uncertain what is the most appropriate medical therapy for patients with severe Graves' ophthalmopathy. Therefore, we carried out a single-blind, randomized clinical trial to compare the efficacy of prednisone with that of cyclosporine in 36 patients who had been euthyroid for at least two months. The two groups, each consisting of 18 patients, were similar in age, sex, and the duration and severity of ophthalmopathy. The initial dose of cyclosporine was 7.5 mg per kilogram of body weight per day, and that of prednisone was 60 mg per day, which was subsequently tapered to 20 mg per day. During the 12-week treatment period, 11 prednisone-treated and 4 cyclosporine-treated patients responded to therapy (61 percent vs. 22 percent; P = 0.018); response was manifested by decreases in eye-muscle enlargement and proptosis and improved visual acuity and total and subjective eye scores. There were no differences at base line between the patients who later responded and those who did not. Prednisone was tolerated less well than cyclosporine. After 12 weeks, patients who did not respond were treated for another 12 weeks with a combination of cyclosporine and a low dose of prednisone. Among the 9 patients who initially received prednisone, the addition of cyclosporine resulted in improvement in 5 (56 percent); among the 13 patients who received cyclosporine initially, 8 (62 percent) improved after the addition of prednisone. Combination therapy was better tolerated than prednisone treatment alone. We conclude that single-drug therapy with prednisone is more effective than cyclosporine in patients with severe Graves' ophthalmopathy. The combination can be effective in patients who do not respond to either drug alone.