Objective: To compare the rate of bone mineral loss in thyroxine-treated women with low thyrotropin (thyroid stimulating hormone, TSH) levels with that in women without known thyroid disease.
Design: Cases selected from a prospective calcium trial.
Setting: Subjects were recruited from the Boston area.
Measurements and main results: Of 361 women enrolled in a 2-year calcium supplement trial, 18 received thyroxine for hypothyroidism. Of these, 10 were considered overtreated, because they had low TSH levels. Rates of loss of bone mineral density from the radius, spine, and hip during 1.9 +/- 0.6 years were measured by single- and dual-photon absorptiometry. When compared with women with no known thyroid disease (236 controls for the spine, 246 for the radius, and 237 for the femoral neck), women with low TSH levels had greater annualized, adjusted mean rates of bone loss from the spine (-2.89% +/- 0.65% compared with -1.13% +/- 0.13%, P = 0.009) and similar but not significant trends at the radius (-1.18% +/- 0.75% compared with -0.13% +/- 0.17%) and femoral neck (-1.39% +/- 0.80% compared with -0.28% +/- 0.19%). These means were adjusted for variables that affected the rate of loss in the control group (baseline bone mineral density and body mass index, calcium intake, and years since menopause). There were no statistical differences between the low TSH and control groups for any laboratory variables measured, including serum calcium, phosphorus, parathyroid hormone or alkaline phosphatase, plasma 25-hydroxyvitamin D or 1,25-dihydroxyvitamin D, or 24-hour urine calcium-to-creatinine ratio.
Conclusions: Thyroxine-treated women with low TSH levels lose bone mineral from the spine more rapidly than do women without known thyroid disease. These patients are therefore at increased risk for osteoporosis. The absence of detectable biochemical changes in women with low TSH levels may result from their relatively modest degree of overtreatment.