Preparation with recombinant human thyroid-stimulating hormone (rhTSH) for thyroid remnant ablation results in lower extrathyroidal radiation than does hypothyroidism. The objective of this prospective study was to compare the damage caused by 131 I (3.7 GBq) when these 2 preparations are used. Methods: Ninety-four consecutive patients who underwent total thyroidectomy and remnant ablation with 3.7 GBq of 131 I were studied. Thirty patients (group A) received rhTSH, and 64 (group B) were prepared by levothyroxine withdrawal. Damage to salivary glands, ovaries, and testes; hematologic damage; and oxidative injury were evaluated by measurement of serum amylase, folliclestimulating hormone (FSH), complete blood count, and plasma 8-epi-PGF 2a before and after radioiodine. The 2 groups were similar in sex, age, and the results of baseline assessment. Results: The rate of successful ablation (stimulated thyroglobulin level , 1 ng/mL and negative findings on neck ultrasonography) was 90% in group A and 80% in group B. Considering only patients with a preablation thyroglobulin level greater than 1 ng/mL, these rates were 80% and 70.6%, respectively. Only 1 patient (3.3%) reported transient headaches with rhTSH. Elevated FSH levels after therapy were observed in 4 of 9 (44%) men in group A versus 16 of 18 (89%) in group B (P , 0.03), with a mean increase of 105% versus 236% (P , 0.001), respectively. In women, elevated FSH was observed in 1 of 13 (7.7%) patients in group A versus 6 of 30 (20%) in group B (P 5 0.4), with a mean increase of 65% versus 125% (P , 0.001). Thrombocytopenia or neutropenia occurred in 2 of 28 (7%) patients in group A versus 12 of 56 (21.4%) in group B (P 5 0.1), with a mean decrease of 20% versus 45% and 25% versus 52% (P , 0.01) for neutrophils and platelets, respectively. Hyperamylasemia and symptoms of acute sialoadenitis occurred in 11 of 30 (36.6%) versus 48 of 60 (80%) (P , 0.001) and in 9 of 30 (30%) versus 35 of 60 (58.3%) (P 5 0.01), respectively. 8-Epi-PGF 2a was found to be elevated after 131 I in 14 of 25 (56%) patients in group A versus 45 of 45 (100%) in group B (P , 0.001), with a mean increase of 60% versus 125% (P , 0.001). Conclusion: The lower radiotoxicity with rhTSH, suggested in dosimetry studies, was confirmed in the present prospective investigation, and this advantage occurred without compromising the efficacy of treatment.