Patients with very large multinodular goitres, frequently found among elderly people, often suffering from cardiovascular or other disabling disorders, may be considered as unsuitable for surgery. We have evaluated the feasibility of relatively high-dose 131I therapy in such patients. As subclinical or clinical hyperthyroidism is commonly found in these patients, associated with a low radioiodine (RAI) uptake at 24 h, we pretreated a group of patients with a single intramuscular injection of recombinant human TSH (rhTSH 0.45 mg) in order to increase the uptake of the therapeutic dose of RAI. Forty-one patients with large, long-standing multinodular goitres, were recruited for this study. After a careful and detailed clinical and laboratory evaluation, 34 patients (28 women, six men) were included and randomly assigned to group 1 (control, n = 17, 15 women, two men, age 63.1 +/- 11.2 years) receiving only RAI. Patients in group 2 (n = 17, 13 women, four men, age 63.6 +/- 12.3 years) received the therapeutic dose of RAI, having been pretreated (24 h) with 0.45 mg of rhTSH. Both groups of patients were submitted to a low iodine diet, 4-6 months before RAI treatment, while seven thyrotoxic patients also received methimazole (40 mg/day) until they reached euthyroidism. Ultrasonographic studies were performed for all patients and fine-needle aspiration biopsy (FNAB) were performed on one or two nodules before RAI treatment. RAI was given as a single oral dose to the hospitalized isolated patients. Blood samples for thyroid function tests and serum thyroglobulin (Tg) were collected daily during the first week following RAI treatment, and at 1, 3, 6, 9 and 12 months thereafter. Goitre volume was estimated by computed tomography scan. Thyroid function tests (total T3, free T4, TSH and serum Tg) were assayed with commercial kits. Urinary excretion of iodine was assayed by the Sandell-Kolthoff method. After the RAI therapeutic dose, serum thyroid function tests were carried out daily for the first week and then on a monthly basis (1, 3, 6, 9 and 12 months). Serum TSH levels rose to a peak level of 45.9 +/- 19.1 mU/l (24 h) in group 2 returning to normal at 72 h. Free T4 serum concentrations rose significantly to 59.35 +/- 21.61 pmol/l at 48 h (in group 2) returning to normal at 7 days. Similarly, serum TT3 also peaked above normal levels only in group 2 (6.12 +/- 1.89 nmol/l) at 24 h. Serum Tg increased in both groups (significantly higher in group 2) and remained elevated during the following 12 months. Both groups had a significant reduction in goitre volume at 12 months (group 2: 57.8%vs. group 1: 39.7%, P < 0.05). Hypothyroidism was detected after RAI treatment, respectively, in 21.4% (group 1) and 64.7% (group 2), of the patients at 12 months. Our results indicate that the use of hTSH increased the efficacy of the RAI therapeutic dose. This was basically due to an increased uptake of the radionuclide (as a consequence of the higher serum TSH levels) and a more extensive distribution of 131I within the nodules of the multinodular goitre. A more intense radiation effect was reflected in there being a higher output of serum Tg and thyroid hormones (group 2). As a consequence this group had a significantly higher reduction of the goitre volume. Also incidence of hypothyroidism post-RAI was significantly higher in group 2. We concluded that pretreatment with rhTSH in elderly patients with large multinodular goitres increases the uptake of the RAI therapeutic dose, thereby significantly reducing the multinodular goitre volume and relieving the compressive symptoms with relatively few side-effects.