| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Editorial |
University of Virginia Health System Charlottesville, Virginia 22908
Address all correspondence and requests for reprints to: Dr. John C. Marshall, , University of Virginia Health System, Jefferson Park Avenue, West Complex OMS 5826, PO Box 800612, Charlottesville, Virginia 22908. E-mail: jcm9h{at}virginia.edu.
Data from the National Health and Nutritional Examination Survey (NHANES) provide compelling evidence that there has been a marked increase in obesity over the past 30 yr. The prevalence of being overweight (>95th percentile body mass index for age) has increased 4- to 5-fold in children and adolescents between the ages of 6 and 19 yr, so that 16% of both 6–11 and 12–19 yr olds met these criteria in 2002 (1). Coincident with the increase in obesity, there has been a marked increase in recognition of the prevalence of both type II diabetes and the metabolic syndrome (MBS) in this age group (2). The criteria commonly used to define MBS in adults are those proposed by the National Cholesterol Education Program Adult Treatment Panel (3) and consist of the presence of three or more of the following: waist circumference in women greater than 88 cm; blood pressure greater than 130/85 mm Hg; serum high-density lipoprotein (HDL)-cholesterol less than 50 mg/dl; fasting serum triglycerides greater than 150 mg/dl; fasting glucose of 110 mg/dl or greater. Some studies have shown associated increased risk for developing type II diabetes and cardiovascular disease (4, 5), but the precise consequences and future risks posed by the presence of MBS in adults remain controversial. Not surprisingly, the increase in cardiovascular risk factors in adolescents has led to increasing concerns related to the potential long-term effects of increased prevalence or the earlier incidence of vascular disease in adults (6). Hyperandrogenemia, predominantly of ovarian origin, is a common consequence of obesity, and increased androgens have been previously associated with an increased incidence of MBS in adults (7). Over the same time course, there has been enhanced recognition that the polycystic ovarian syndrome (PCOS) in adult women is a disorder of pre- or peripubertal onset and PCOS is commonly associated with obesity.
In this issue, Coviello et al. (8) present data showing that, in adolescent girls with PCOS, as many as 63% of obese girls had MBS, compared with 32% of obese girls in the NHANES-III data. After adjusting for BMI, girls with PCOS were four times more likely to show features of MBS. Of particular import, the odds of MBS characteristics being present were four times higher for each quartile increase in plasma unbound testosterone (8). These data provide strong evidence that adolescents with PCOS have a higher prevalence of MBS and that hyperandrogenemia is a risk factor for MBS, independent of obesity and insulin resistance.
In premenopausal women, approximately 50% of plasma testosterone is derived in equal proportions from ovarian and adrenal secretion. The remaining 50% derives from conversion of androstenedione in peripheral tissues, including adipose tissue (9). In obese individuals with insulin resistance, the associated hyperinsulinemia acts as a co-gonadotropin with LH to increase androgen production by ovarian theca cells, as the ovary remains sensitive to the actions of insulin (10). Additionally, insulin suppresses hepatic production of SHBG, leading to marked elevation of free or unbound plasma testosterone (11). Plasma LH is elevated or in the upper normal range in some 70–90% of women with PCOS (12). Thus, as PCOS is commonly associated with obesity, a double stimulus of elevated LH and hyperinsulinemia leads to increased ovarian testosterone production, which may be further enhanced as ovarian theca cells from PCOS ovaries hypersecrete androgens (13).
Elevated plasma androgens exert actions on several physiological systems, with generally negative implications for the normal regulation of reproduction and metabolism. In the ovary, androgens exert double actions that are incompletely understood. On the positive side, testosterone from the ovarian theca cell is an essential precursor for estrogen production, as it is aromatized to estradiol after FSH induction of aromatase activity in the granulosa cells. However, excess androgens may be a factor in inducing follicular atresia, as dihydrotestosterone inhibits aromatase activity and dihydrotestosterone is markedly elevated in atretic follicles (14). The effects of elevated androgens on the hypothalamus are evidenced in the impaired regulation of GnRH secretion. Both pre- and postmenarchal adolescent girls with hyperandrogenism have elevated LH levels and a rapid frequency of GnRH pulse secretion that is persistent throughout 24 h (15). The abnormal GnRH/LH secretion does not appear to be related to hyperinsulinemia or insulin resistance, as LH pulsatile secretion was unchanged after prolonged insulin infusion or after insulin resistance was reduced by pioglitazone treatment (16). Hyperandrogenemia has been shown to interfere with the normal inhibitory feedback actions of estradiol and progesterone on GnRH secretion in women. Estradiol induces hypothalamic progesterone receptors, and the normal action of progesterone is to slow GnRH pulses in the luteal phase of ovulatory cycles. The ability of progesterone to slow the frequency of GnRH pulse release is impaired in hyperandrogenic women (17, 18). This appears to reflect a direct effect of androgen, as sensitivity to progesterone feedback can be restored after blockade of androgen action by treatment with flutamide (19). The effect of testosterone in impairing progesterone inhibition of GnRH secretion is also present in adolescence (20). However, not all individuals appear to be susceptible, and, in a proportion of individuals, progesterone feedback is preserved despite the presence of elevated androgen levels. The mechanisms of this varied susceptibility to androgen action are unclear but may reflect genetic differences, as in initial studies individuals with preserved hypothalamic sensitivity to progesterone were of Hispanic descent. Similarly, it is unclear at what stage(s) of development the hypothalamic actions of androgen are exerted. In both sheep and monkeys, exposure to high levels of androgen in utero led to abnormalities of LH secretion during subsequent puberty, with elevated plasma LH, persistently rapid GnRH frequency, and impaired ovarian steroid feedback (21, 22). Together, these data from fetal primates and peripubertal humans suggest that excess androgen at both these stages of development can modify normal pubertal changes and impair feedback interactions between the ovary and the hypothalamus.
Elevated androgens may also exert actions to modify adipose cell function. Adipose tissue contains a wide spectrum of enzymes involved in steroid metabolism and, as noted, contributes up to 50% of circulating testosterone in premenopausal women (23). Human preadipocytes contain androgen receptors, which are up-regulated by androgens, and higher concentrations of androgen receptors are present in abdominal as opposed to sc adipocytes (24). Hyperandrogenemia is more pronounced in obese adolescent girls with predominant visceral adiposity, which is also associated with an increased risk of MBS (25). Adipose tissue is known to be a source of cytokines such as TNF 
, IL-6, and plasminogen activator inhibitor, and plasma cytokines are increased in individuals with MBS. Thus, it is possible that hyperandrogenemia exerts negative actions on adipocyte metabolism by modifying the production or actions of cytokines.
Further evidence supporting a role for excess androgens in the genesis of the MBS can be found in the results of treatment of adolescent girls. Successful weight loss programs lead to reduced plasma testosterone, increased SHBG, and consequently reduced circulating free testosterone (25). Blockade of androgen action by flutamide produced similar results and in addition markedly reduced low-density lipoprotein and triglycerides, although HDL and hyperinsulinemia were not improved (26). The addition of metformin to flutamide in hyperinsulinemic adolescents results in additional improvements in HDL and reduced abdominal fat (27).
The evidence briefly reviewed above emphasizes the marked elevation of testosterone that is commonly associated with obesity in adolescent girls. Excess androgens exert deleterious actions at several levels in the reproductive system, with the hypothalamic effects leading to increased LH secretion, which synergizes with elevated plasma insulin to augment and maintain androgen production by the ovary. The data from Coviello et al. (8) indicate the importance of hyperandrogenemia in exerting negative effects on metabolism, enhancing the likelihood that the features collectively recognized as MBS are present in obese adolescents, with potential long-term negative impacts on cardiovascular health. The marked increase in obesity in adolescent girls during the past 30 yr is associated with increased hyperandrogenemia, which in turn appears to be a genetic marker for the propensity to develop PCOS (28). This in turn emphasizes the importance of recognizing the presence of excess androgens in obese adolescents. In that regard, a note of caution is appropriate related to measurement of androgens, in particular to the estimation of free or unbound testosterone. This issue was recently reviewed by Matsumoto and Bremner (29), who cautioned on the use of direct measurement of free testosterone by analog RIA in women, where values are 10- to 20-fold lower than those in men. In view of variable correlations with measurements made using the gold standard of equilibrium dialysis, they recommend that free testosterone is derived by calculation from total testosterone and SHBG, the latter measured by immunoradiometric assay. Expanded recognition that hyperandrogenemia is a common consequence of obesity will lead to increased emphasis on efforts to reduce adolescent obesity and body weight. As these actions reduce androgen excess, implementation of therapeutic weight loss programs, together with pharmacological approaches to reduce androgens and insulin resistance, appear to be a critical step in efforts to reduce the likelihood of increased cardiovascular disease occurring at a younger age over the next 30-yr period.
Footnotes
Abbreviations: HDL, High-density lipoprotein; MBS, metabolic syndrome; PCOS, polycystic ovarian syndrome.
Received December 7, 2005.
Accepted December 19, 2005.
References
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
