We compared the effects of high and low dosages of antithyroid drugs in 113 patients with Graves' hyperthyroidism. The patients were randomly divided into 2 groups. In group A, 65 patients received either methimazole (MMI): 60 +/- 14.5 mg/day (mean +/- SD); range 40-100 mg/day, or propylthiouracil (PTU): 693 +/- 173 mg/day; range 500-1200 mg/day. These high doses were maintained throughout treatment with later addition of 50-75 micrograms T3 daily. Forty eight patients (group B) were treated with lower doses of MMI or PTU without thyroid hormone addition. The maintenance dose of MMI was 13.6 +/- 7 mg/day (range 5-25 mg/day) and that of PTU was 180 +/- 58 mg/day (range 100-300 mg/day). The treatment period was 15.1 +/- 4.2 (range 10-30) months for group A and 13.5 +/- 2.2 (range 12-20) months for group B. Remission occurred in 75.4% patients from group A and in 41.6% patients from group B (P less than 0.001). The mean follow-up was 42 +/- 14 months (17-81 months). The free T4 index (FT4I) in group A remained below the normal range during treatment. The mean FT4I, obtained during the course of treatment, of patients who went into remission from group A was significantly (P less than 0.001) lower than in relapsed patients (4.8 vs. 6.5). Moreover, there was an inverse correlation between mean FT4I and maintenance daily dose of either MMI (r = -0.567; P less than 0.001), or PTU (r = -0.379; P less than 0.01). A fall in microsomal antibody (MCHA) titer occurred mainly in remission patients, and was more significant (P less than 0.05) in group A patients. In contrast, 11 (7 from group B) of the 16 patients with an increase of microsomal antibody levels relapsed. The frequency of negative tests of thyroid-stimulating antibody was higher in group A patients (71%) than in group B (29%) at the end of therapy (P less than 0.01). No correlation was found between thyroid T3 suppressibility and either mean FT4I or thyroid-stimulatory antibody activity during treatment. Our findings show that patients treated with high doses of PTU or MMI throughout treatment have a higher remission rate when compared to those treated with a more conventional regimen. These results support the hypothesis that large antithyroid drug doses may have greater immunosuppressive effects than low dosage regimens. Furthermore, a high dosage regimen could permit the restoration of the immune surveillance mechanisms and, thus, lasting remission of Graves' disease.