Context. Graves' orbitopathy (GO) is probably caused by autoimmune reactions against autoantigen(s) shared by thyroid and orbital tissues sustained by intrathyroidal autoreactive T-lymphocytes infiltrating the orbit. Total thyroid ablation may be beneficial for GO through removal of shared antigen(s) and autoreactive T-lymphocytes, but randomized studies are lacking. Objective. To evaluate the effects of total thyroid ablation in patients with GO treated with iv glucocorticoids (ivGC). Design. A prospective, single-blind, randomized study. Setting. Referral Center. Patients. Sixty patients with mild to moderate GO. Interventions. Patients were randomized into: i) TX (near-total thyroidectomy) or ii) TTA (near-total thyroidectomy plus ¹³¹I), and then treated with ivGC. Patients were evaluated 3 and 9 months after ivGC. Main outcome measure. Overall improvement of GO at 9 months. Results. The distribution of GO outcome at 9 months was significantly more favorable in TTA than in TX patients (P = 0.0014 by Chi square test). A cumulative significant (P = 0.0054) difference between the two groups at 3 and 9 months was found using a generalized linear model. Radioiodine uptake test and thyroglobulin assay in a patient sample showed complete ablation in the majority of TTA, but not of TX patients. Conclusions. Compared with thyroidectomy alone, total thyroid ablation is followed by a better outcome of GO in patients given ivGC. Whether total thyroid ablation maintains this advantage in the long-term remains to be established.