The effect of long-term thyroxine on bone mineral density and serum cholesterol

The effect of thyrotrophin suppression on bone mineral density (BMD) and serum cholesterol concentration was assessed in 31 treated hypothyroid women. Measurements of the BMD of the lumbar spine and femoral neck were repeated in seven of those with the lowest value after an average period of 22.7 months. Final cholesterol concentrations were compared with values before thyroxine was started. The dose of thyroxine was based on clinical assessment, serum triiodothyronine concentrations kept within the normal range, and thyrotrophin values within the normal range or suppressed. The patients had taken thyroxine replacement for a mean of 12.7 years. Two-thirds (21 subjects) had suppressed thyrotrophin concentrations, and it was normal in one-third (10). Fifteen subjects had a past history of thyrotoxicosis. BMD and cholesterol concentrations were compared between those with suppressed and normal thyrotrophin concentrations and between those with and without a past history of thyrotoxicosis. No patient had a pathological fracture. One had a Z value for the femoral neck of -1.6, denoting early but definite osteoporosis, and five had borderline osteoporosis with Z values for one or other site between -1.1 and -1.5. None of the seven with the lowest BMDs had any significant change when measurements were repeated. The difference in Z values between subjects with suppressed and normal thyrotrophin concentrations was not significant for either the lumbar spine (p = 0.68) or the femoral neck (p = 0.28). A past history of thyrotoxicosis had a greater effect on BMD for both sites than thyrotrophin suppression, but again the difference between those with and without a past history of thyrotoxicosis was significant neither for the lumbar spine (p = 0.18) nor for the femoral neck (p = 0.34). The combination of thyrotrophin suppression and a past history of thyrotoxicosis also failed significantly to reduce the BMD of the lumbar spine (p = 0.38) or femoral neck (p = 0.30) in comparison with those who had neither thyrotrophin suppression nor a past history of thyrotoxicosis. The mean fall in serum cholesterol concentration was 2.1 mmol/l (SD 1.78) (p = 0.001) in those with a suppressed thyrotrophin concentration taking a mean daily dose of thyroxine of 171 micrograms (SD: 34.7), compared with a fall of 0.89 mmol/l (SD: 1.04) (p = 0.065) in those whose thyrotrophin concentration was not suppressed on a mean daily thyroxine dose of 140 micrograms (SD: 50). No patient had atrial fibrillation or cardiographic evidence of coronary artery disease (CAD). The serum cholesterol concentration should play at least as important a part in influencing the dose of thyroxine as a fear of osteoporosis. Fractures are not a feature in the natural history of treated hypothyroidism, whereas CAD is a common cause of death in these patients.