Clinical Review 116: Bone mineral density, androgens, and the polycystic ovary: the complex and controversial issue of androgenic influence in female bone.

Polycystic ovary syndrome (PCOS), characterized by chronic anovulation, hyperandrogenemia, obesity, central adiposity, and insulin resistance (1–3), represents a unique, natural model for study of the influences of androgenic hormones on bone mass among women. PCOS usually manifests at puberty and is currently recognized as one of the most common endocrine disorders among women of reproductive age, affecting approximately 4% of this population (4). Women with PCOS typically have acyclic production of 17b-estradiol, with circulating concentrations similar to those seen in the follicular phase of cycling women, but considerably lower than the mean 17b-estradiol concentration observed across the normal menstrual cycle (1). This static level of 17b-estradiol, with the absence of the estradiol surge associated with ovulation, would intuitively be expected to negatively affect bone density. Testosterone and androstenedione, however, are produced in excess by the PCOS ovary and perhaps by the adrenals of some affected women. This chronic elevation in androgens may exert a positive influence on bone in PCOS women, either directly through androgen receptors on bone-related cells or indirectly after conversion to 17b-estradiol and estrone, respectively, in peripheral tissues. Moreover, elevated circulating insulin levels, also associated with PCOS, may offer some additional protection against a reduction in bone mass in these women. Hyperinsulinemia has been shown to positively affect bone density (5, 6) through direct stimulatory effects on osteoblastic activity (7) and by the suppressive influences exerted by insulin on the production of sex hormone-binding globulin (SHBG) (8) and insulin-like growth factor (IGF)-binding proteins (IGFBPs) (9). This insulininduced suppression of these binding proteins may result in increased exposure of target tissues, including bone, to the stimulatory effects of elevated concentrations of free sex steroids and IGFs. Both thin and obese women develop PCOS, a presentation that allows for evaluation of the effects of life-long obesity, alterations in body composition (e.g. increased central adiposity), and related metabolic abnormalities (e.g. hyperandrogenemia and hyperinsulinemia) on the skeleton. The relatively high prevalence of PCOS and its early-life presentation render this disorder of particular importance as a model for assessing the effects of androgen and potentially insulin on the attainment of maximal bone mass. The objectives of this review article are to 1) summarize the literature exploring the androgen-bone mass relationship in normal women; 2) assess the effects of PCOS, androgen excess, and androgen-estrogen balance on bone mass in women; 3) examine study design issues related to PCOS, androgen excess, and bone mass; and 4) suggest areas for future research related to the influence of PCOS, androgen excess, and insulin on bone in women.