Bone mineral density and metabolism in premenopausal women taking L-thyroxine replacement therapy

Excess endogenous thyroxine causes bone loss, but the effects of exogenous thyroxine are disputed. We report on bone mass and metabolism in women taking L-thyroxine therapy.

Cross-sectional and longitudinal studies.

Cross-sectional study: 40 healthy premenopausal women with autoimmune thyroiditis taking either physiological (serum TSH usually normal, 0.35-3.3 mU/l) or suppressive (serum TSH usually < 0.35 mU/l) doses of L-thyroxine; patients were also compared with previously acquired age and weight matched premenopausal volunteers with no history of thyroid dysfunction. Longitudinal study: 28 patients were followed-up > or = 1 year later.

In all subjects bone mineral density (BMD) was measured at the anteroposterior lumbar spine and left hip (femoral neck, greater trochanter and Ward's area) using dual-energy X-ray absorptiometry, and physical activity assessed using the Framingham physical activity index. Serum osteocalcin (OC), PTH and vitamin D, and urinary pyridinoline and deoxypyridinoline excretion were measured in patients.

Cross-sectional study: The patient groups were well matched for disease duration and physical activity although the suppressed group were slightly younger (mean 38.1 (SD 7.5) vs 43.3 (3.9) years, P < 0.05). BMD, serum OC, PTH and vitamin D, and urinary cross-link excretion did not differ significantly between the two groups. Multivariate analysis of the whole group suggested that BMD at the femoral neck and greater trochanter was related positively to weight and physical activity and negatively to thyroxine dose (microgram/kg/day) and patient age. At the lumbar spine BMD was reduced non-specifically in the presence of thyroid disease and treatment, but this effect appeared to decline with increasing duration of therapy. A similar analysis of the patient group suggested that urinary cross-link excretion correlated positively with thyroxine dose, and negatively with duration of treatment. Longitudinal study: Annualized changes in BMD were inversely related to thyroxine dose (microgram/kg/day) at all sites but achieved statistical significance only at the femoral neck and Ward's area.

We did not find any effect of persistent historical TSH suppression on current bone mass, and this might relate to the relative insensitivity of older TSH assays. However, the cross-sectional and longitudinal data suggest that high daily doses of thyroxine in relation to patient body weight might adversely affect bone mass, particularly at the hip. These findings support the contention that excess exogenous thyroxine might predominantly deplete skeletal sites, such as the femoral neck, rich in cortical bone.