Radioiodine (RAI) therapy is a safe and effective treatment for hyperthyroidism and individual doses are frequently administered. Initial thyroid volume (TV) is an important parameter for RAI therapy. Ultrasonography (US) is considered the most reliable method of determining TV. The aim of this study was to evaluate TV by means of US in a cohort of 75 hyperthyroid patients before and after RAI therapy. According to clinical examination, thyroid US and technetium-99m (99mTc)-pernechnetate scintiscan, the diagnosis of hyperthyroidism was multinodular goiter (MNG) in 27, diffuse goiter (DG) in 32 and uninodular goiter (UNG) in 16 patients. The RAI dose to be administered was calculated according to TV and RAI uptake, up to a maximum of 600 MBq. TV was further evaluated 1, 3 and 6-12 months after RAI therapy. The initial TV was 42.3+/-4.0 ml for MNG, 29.7+/-2.8 ml for DG and 34.5+/-3.7 ml for UNG. After 6-12 months a non-significant TV reduction was observed in the MNG group even though the fraction of initial TV was 53.3+/-6.5%. Moreover, a significant TV reduction was noticed in the DG group (8.8+/-2.3 ml; p<0.001). In this group the fraction of initial TV was 28.6+/-3.2% at 6-12 month evaluation. A less marked, though still significant (p=0.04) TV reduction (19.6+/-3.2 ml) was also observed in the UNG group, the fraction of initial TV being 57.8+/-5.3% 6-12 months after RAI. In the whole patient population there was no significant correlation between TV reduction or TV at the last examination and initial TV, RAI dosage, baseline free T4 and TSH levels. No correlation was found between clinical condition at the last examination and TV reduction. In conclusion, these data justify TV estimation by means of US in the protocol of individual RAI dose for the therapy of hyperthyroidism. Our follow-up documents a poorly predictable TV reduction in all clinical conditions, but this is more pronounced and predictable in patients with diffuse toxic goiter.