Relapse of Graves' disease after successful outcome of antithyroid drug therapy: results of a prospective randomized study on the use of levothyroxine

Antithyroid drugs are effective in restoring euthyroidism in Graves' disease, but many patients experience relapse after withdrawal. Prevention of recurrence would therefore be a desirable goal. In a prospective study, patients with successful outcome of 12 to 15 months antithyroid drug therapy were stratified for risk factors and randomly assigned to receive levothyroxine in a variable thyrotropin (TSH)-suppressive dose for 2 years or no treatment. The levothyroxine group was randomized to continue or discontinue levothyroxine after 1 year. End points included relapse of overt hyperthyroidism. Of 346 patients with Graves' disease enrolled 225 were euthyroid 4 weeks after antithyroid drug withdrawal and were randomly assigned to receive levothyroxine (114 patients) or no treatment (controls, 111 patients). Of those not randomized, 39 patients showed early relapse within 4 weeks, 61 endogenous TSH suppression, 7 TSH elevation, and 14 had to be excluded. Dropout rate during the study were 13.3%. Kaplan-Meier analyses showed relapse rates to be similar in the levothyroxine group (20% after 1 year, 32% after 2 years) and the randomized controls (18%, 24%), whereas relapses were significantly more frequent in the follow-up group of patients with endogenously suppressed TSH (33%, 49%). Levothyroxine therapy did not influence TSH-receptor antibody, nor did it reduce goiter size. The best prognostic marker available was basal TSH determined 4 weeks after withdrawal of antithyroid drugs (posttreatment TSH). The study demonstrates that levothyroxine does not prevent relapse of hyperthyroidism after successful restoration of euthyroid function by antithyroid drugs and characterizes posttreatment TSH as a main prognostic marker.