Background: Antibodies to thyroid-stimulating hormone (TSH) receptors that stimulate the thyroid gland cause hyperthyroidism in patients with Graves' disease, and their production during antithyroid drug treatment is an important determinant of the course of the disease. One factor that might contribute to the persistent production of antibodies to TSH receptors is stimulation of the release of thyroid antigens by TSH during antithyroid drug therapy. We therefore studied the effect of the suppression of TSH secretion by thyroxine on the levels of antibodies to TSH receptors after thyroid hormone secretion had been normalized by methimazole.
Methods and results: The levels of antibodies to TSH receptors were measured during treatment with methimazole, either alone or in combination with thyroxine, in 109 patients with hyperthyroidism due to Graves' disease. The patients first received 30 mg of methimazole daily for six months. All were euthyroid after six months, and their mean (+/- SD) level of antibodies to TSH receptors decreased from 64 +/- 9 percent to 25 +/- 15 percent (P less than 0.01; normal, 2.9 +/- 1.4 percent). Sixty patients then received 100 micrograms of thyroxine and 10 mg of methimazole and 49 received placebo and 10 mg of methimazole daily for one year. In the thyroxine-treated group, the mean serum thyroxine concentration increased from 108 +/- 16 nmol per liter to 145 +/- 11 nmol per liter (P less than 0.01), and the level of antibodies to TSH receptors decreased from 28 +/- 10 percent to 10 +/- 3 percent after one month of combination therapy. In the patients who received placebo and methimazole, the mean serum thyroxine concentration decreased and the level of antibodies to TSH receptors did not change. Methimazole, but not thyroxine or placebo, was discontinued in each group 1 1/2 years after the beginning of treatment. The level of antibodies to TSH receptors further decreased (from 6.6 +/- 3.2 percent at the time methimazole was discontinued to 2.1 +/- 1.2 percent one year later) in the patients who continued to receive thyroxine, but it increased (from 9.1 +/- 4.8 percent to 17.3 +/- 5.8 percent during the same period) in the patients who received placebo. One patient in the thyroxine-treated group (1.7 percent) and 17 patients in the placebo group (34.7 percent) had recurrences of hyperthyroidism within three years after the discontinuation of methimazole.
Conclusions: The administration of thyroxine during antithyroid drug treatment decreases both the production of antibodies to TSH receptors and the frequency of recurrence of hyperthyroidism.