In their "Practical Guidelines for Clinicians Who Treat Patients With Amiodarone," Goldschlager et al1 missed some important points on the effect of amiodarone on thyroid function. First, it seems important to recall that amiodarone inhibits the conversion of thyroxine to triiodothyronine, resulting in slightly but abnormally increased free thyroxine levels contrasting with normal thyrotropin concentrations in the majority of patients treated with amiodarone despite clinical euthyroidism.2 Therefore, screening for thyroid dysfunction (reference 1, Table 2) requires only thyrotropin measurement in amiodarone-treated patients. Second, in Goldschlager and colleagues' Table 1, the incidence of hyperthyroidism is much lower than that of hypothyroidism in these patients. This is true in North America, not in Europe or other countries that have a suboptimal iodine supply. In Europe, amiodarone-induced hyperthyroidism (AIH) is more frequent than hypothyroidism, occurring in up to 10% of patients treated with amiodarone.3 In our experience, AIH accounts for 50% of severe hyperthyroidism in subjects 70 years of age and older.4 The diagnosis of hyperthyroidism may be difficult in patients taking amiodarone, who often lack the cardiac signs of thyrotoxicosis because of the inhibitory effect of desethylamiodarone on the binding of triiodothyronine on the cardiac thyroid hormone α1-receptor.5 Third, Goldschlager and colleagues stated that "hyperthyroidism due to amiodarone is usually inflammatory thyroiditis." Although the mechanisms of AIH are complex and have not been completely elucidated,2,6- 7 in our experience, as in that of other investigators, only about half of patients with AIH have biologic, ultrasonographic, or histologic signs of thyroiditis (referred to by some authors as type II AIH) and have a favorable outcome with steroid therapy. Iodide excess is the other major mechanism of AIH (type I); however, differentiation between the 2 mechanisms is difficult in most cases.6 Potassium perchlorate inhibits iodide accumulation in the thyroid gland by acting, among other mechanisms, on the sodium/iodide symporter of the thyroid follicle. Perchlorate is an effective treatment of iodine-induced hyperthyroidism in some patients, with few adverse effects when a low dosage (<1 g/d) is used.7 Thus, in our opinion, thyroidectomy must be considered only in patients whose AIH does not respond to a combination therapy consisting of carbimazole (or methimazole or 6-propyl-2-thiouracil), corticosteroids, and perchlorate. Clinicians have to remember that the discontinuation of amiodarone therapy is sufficient to allow recovery of AIH in about 50% of the patients.5 In some patients in whom amiodarone therapy can be discontinued but who may need it again, preventive radioiodine treatment (after normalization of iodide uptake) is probably the best solution.