Reprinted from Eating Disorders Review
January/February 2005 Volume 16, Number 1
©2005 Gürze Books
Severe bone loss, mood disorders, and cognitive impairment are only a few of the many problems that women with anorexia nervosa experience. Low-dose androgen therapy has led to marked increases in bone density in hypogonadal men (J Endocrinol Metab 1990; 81:4358) and studies suggest that low-dose androgen therapy may increase bone formation (J Clin Endocrinol Metab 1990; 81:37) and bone density among postmenopausal women.
Harvard researchers Drs. Karen K. Miller, K.A. Griego, and A. Klibanski recently conducted a short-term controlled study of testosterone therapy in a group of 33 women with anorexia nervosa and relative testosterone deficiency (J Clin Endocrinol Metab 2004; 1181). According to the authors, this is the first study to examine markers of bone metabolism in women receiving testosterone without concurrent estrogen administration.
The anorexic women were selected after screening with the Beck Depression Inventory, and measurement of osteocalcin and a propeptide of type 1 collagen (PICP). The women were severely to moderately depressed. Dr. Miller and colleagues hypothesized that the short-term low doses of testosterone would increase bone formation, improve depression, and improve spatial ability in the women with anorexia nervosa.
Two study groups
The women were randomized to receive 150 mcg or 300 mcg of transdermal testosterone (Intrinsa) or placebo for three weeks. At baseline, free testosterone levels correlated with bone density (measured at L4), body mass index (BMI), depressive symptoms, and spatial cognition. Visual-spatial ability was assessed with Shepard and Metzler’s three-dimensional rotations test. Quality of life was evaluated with the Psychological General Well-being Index (PGWI).
After 3 weeks: improved PICP levels
Serum total and free testosterone levels increased significantly among women treated with testosterone. About 50% of those who received a dosage of 150 mcg and 67% of those who received 300 mcg had an increase in free testosterone levels after 3 weeks; those levels exceeded the upper limits of the normal premenopausal range. In contrast, estradiol, sex-hormone-binding globulin, dehydroepiandrosterone sulfate (an androgen), and insulin-like growth factor 1 levels did not change in the treatment group. The increase in free testosterone was greater in patients with higher BMIs. PICP levels were also higher during testosterone administration than placebo. Osteocalcin and bone-specific alkaline phosphatase levels did not change.
Depressed patients receiving testosterone therapy improved from severely depressed to moderately depressed; the placebo group was unchanged. Spatial cognition also improved in the testosterone group compared with placebo. There was no change in well-being, as measured at 10 days or three weeks.
Low doses of testosterone were well tolerated in this population, according to the authors. None of the women developed hirsutism, acne, decreases in high-density lipids or increases in transaminase levels. The most common side effect was local irritation at the patch sites (3 women in the testosterone group and 1 in the placebo group reported this effect). One patient in the placebo group was treated by her primary care physician for fatigue and vertigo. Five study participants were either discontinued or withdrew from the study, including one who became pregnant and another who developed multiple fractures in an automobile accident. Two subjects (one in the placebo group and one in the treatment group were discontinued from the study due to life-threatening weight loss.
An experimental treatment that needs more study
According to the authors, low-dose testosterone therapy may prevent decreased bone formation in patients with anorexia nervosa, but since testosterone did not affect all markers of bone formation studied, further studies will be needed to show its effects on bone formation.
Since the long-term risks for ongoing testosterone therapy in this population aren’t known, this treatment should be considered entirely experimental at this point, and cannot be recommended for general treatment plans.