Treatment with androgens such as testosterone has a number of salutary effects on postmenopausal women, including greater energy, improved sexual function, preservation of bone mass, and possibly a protective effect on the breast. There is concern, however, that excessive androgen may lead to an insulin-resistant state. This study addressed the issue of insulin sensitivity using testosterone undecenoate (TU), an orally active preparation with only minor liver-related side effects. Participants were 63 naturally postmenopausal women in good general health, 2 of whom were obese. The women were randomly assigned to receive 40 mg testosterone (T) in the form of TU every other day, 2 mg estradiol (E) valerate daily, or both treatments (T+E), all for 3 months. The euglycemic hyperinsulinemic clamp technique served to estimate insulin sensitivity. The 3 groups were similar in age, body weight, bone mass index, blood pressure, serum hormone levels, and a history of using conventional hormone therapy. Diastolic but not systolic blood pressure fell significantly in the T group but, in general, treatment was well tolerated. No hyperandrogenic signs such as acne or hirsutism developed. Total serum testosterone rose to median levels of 35 to 46 ng/dL in both the T and T+E groups-within the normal range for premenopausal women. Fasting blood glucose levels did not change significantly, but plasma insulin levels fell significantly in the E and T+E groups. Basal insulin sensitivity increased significantly in E-treated women only. Steady-state plasma insulin and blood glucose levels were similar in all groups before and after treatment. Whole-body glucose uptake fell significantly by approximately 20% in the T group and also after combined treatment. No significant correlation was found between changes in free T levels and altered anthropometric or metabolic variables. Body weight increased by approximately 1 kg, a small but significant change, in all treatment groups. Total body fat mass and fat distribution remained stable, but total lean body mass increased significantly in the T+E group. Total cholesterol fell significantly by approximately 20 mg/dL in the same group. High-density lipoprotein cholesterol decreased significantly in the T group but increased in the E group. Low-density lipoprotein cholesterol decreased significantly in both E-treated groups. It appears that short-term treatment with TU, alone or combined with estradiol, induces insulin resistance in postmenopausal women and also worsens the serum lipid profile. Treatment may, however, increase lean body mass. More work is needed to appreciate the clinical significance of these findings.
Read full abstract