In routine use for more than 50 years, radioiodine ((131)I) is generally considered safe and devoid of major side effects. Therefore, it is surprising that relatively many aspects of radioiodine therapy are controversial, as illustrated by recent international questionnaire studies. Our review aims at highlighting three of these areas - namely, the influence of (131)I on the course of Graves' ophthalmopathy, the possible radioprotective effects of antithyroid drugs, and the use of (131)I in large goitres. (131)I therapy carries a small (but definite) risk of causing progression of Graves' ophthalmopathy. Identification of risk factors (thyroid dysfunction, high level of thyroid-stimulating hormone (TSH) receptor antibodies, cigarette smoking) allows the identification of patients at risk and the institution of concomitant glucocorticoid treatment, thereby hindering progression of eye disease. On the basis, largely, of retrospective data, it appears that carbimazole (or methimazole), if stopped 3-5 days before treatment, does not influence the outcome of (131)I therapy. Simultaneous thyrostatic medication most probably reduces the efficacy of (131)I, as does restarting it within 7 days. Propylthiouracil seems to have a more prolonged radioprotective effect than carbimazole. Surgery is the treatment of first choice in patients with a large goitre. However, in the case of patient ineligibility or preference, (131)I therapy may be an option. The treatment has a favourable effect on tracheal compression and inspiratory capacity, but the reduction in thyroid volume is only 30-40%. Inpatient treatment, necessitated by the large doses, makes the treatment cumbersome. Controversy related to radioiodine therapy is mainly based on the lack of adequate prospective randomised studies comparing efficacy, side effects, cost and patient satisfaction.