To examine whether suppressive doses of thyroxine have any adverse effects on bone, we evaluated various bone metabolic markers (lectin-precipitated alkaline phosphatase, osteocalcin, carboxyl-terminal region of type I collagen propeptide, tartrate-resistant alkaline phosphatase, and urinary excretion of hydroxyproline and pyridinium crosslinks), incidence of vertebral deformity, total body and regional (lumbar spine and radius) bone mineral densities (BMDs), and rates of bone loss in 24 late postmenopausal (more than 5 years after menopause) women who were treated with levothyroxine (L-T4) after total thyroidectomy for differentiated carcinoma. Depending on the clinical records, including serum TSH levels measured by immunoradiometric assay, these patients were divided into two groups. One group of patients was given suppressive doses of L-T4 (TSH < 0.1 mU/L, n = 12) and the other group was given nonsuppressive doses of L-T4 (TSH > 0.1 mU/L, n = 12). There was no difference in bone metabolic markers and incidence of vertebral deformity between the groups. In patients with TSH suppression, Z-scores of BMDs calculated from age-matched healthy women (n = 179, aged 55 to 80) were nearly in the zero range of values (0.077 at total body, 0.228 at lumbar spine, and -0.117 at trabecular region of lumbar spine). The rate of bone loss in TSH-suppressed patients (-0.849 +/- 0.605%/year) was not significantly different from that of nonsuppressed patients (-0.669 +/- 0.659). These prospective and cross-sectional data suggest that long-term levothyroxine therapy using suppressive doses has no significant adverse effects on bone.