Hyperandrogenism in Women with Polycystic Ovary Syndrome ...

ORIGINAL

ARTICLE

E n d o c r i n e

C a r e

Hyperandrogenism in Women with Polycystic Ovary Syndrome Persists after Menopause Marios C. Markopoulos, Demetrios Rizos, George Valsamakis, Efthimios Deligeoroglou, Odysseas Grigoriou, George P. Chrousos, George Creatsas, and George Mastorakos Second Department of Obstetrics and Gynecology (M.C.M., E.D., O.G., G.C.), Hormonal Laboratory (D.R.), Endocrine Unit (G.V., G.M.), Athens University Medical School, Aretaieion Hospital, Athens 11528, Greece; First Department of Pediatrics (G.P.C.), Athens University Medical School, Aghia Sophia Children’s Hospital, Athens 10674, Greece

Context: Ovarian and adrenal hyperandrogenism characterize premenopausal women with polycystic ovary syndrome (PCOS). Androgens decline with age in healthy and PCOS women. Objective: The objective of the study was to investigate hyperandrogenism in PCOS after menopause. Design: This was a case-control, cross-sectional study. Setting: The study was conducted at a university hospital endocrinology unit. Patients: Twenty postmenopausal women with PCOS and 20 age- and body mass index-matched controls participated in the study. Interventions: Serum cortisol, 17-hydroxyprogesterone (17-OHP), ⌬4-androstenedione (⌬4A), dehydroepiandrosterone sulfate (DHEAS), total testosterone (T), and free androgen index (FAI) levels were measured at baseline, after ACTH stimulation, and after 3-d dexamethasone suppression. The ACTH and cortisol levels were measured during the CRH test. Main Outcome Measures: Androgen profile at baseline, after ACTH stimulation, and 3-d dexamethasone suppression tests were the main outcome measures. Results: Postmenopausal PCOS women had higher 17-OHP, ⌬4A, DHEAS, total T, FAI (P ⬍ 0.05) and lower SHBG (P ⬍ 0.05) baseline levels than control women. ACTH and cortisol responses during the CRH test were similar in the two groups. After ACTH stimulation, ⌬4〈, DHEAS, and total T levels were equally increased in both groups. After dexamethasone suppression, LH levels did not change in either group; 17-OHP-, ⌬4A-, and FAI-suppressed levels remained higher in PCOS than in control women (P ⬍ 0.05), whereas total T and DHEAS levels were suppressed to similar values in both groups. Conclusions: In postmenopausal PCOS women, ACTH and cortisol responses to CRH are normal. Androgen levels at baseline are higher in PCOS than control women and remain increased after ACTH stimulation. The dexamethasone suppression results in postmenopausal PCOS women suggest that DHEAS and total T are partially of adrenal origin. Although the ovarian contribution was not fully assessed, increased ⌬4A production suggests that the ovary also contributes to hyperandrogenism in postmenopausal PCOS women. In conclusion, postmenopausal PCOS women are exposed to higher adrenal and ovarian androgen levels than non-PCOS women. (J Clin Endocrinol Metab 96: 623– 631, 2011)

ISSN Print 0021-972X ISSN Online 1945-7197 Printed in U.S.A. Copyright © 2011 by The Endocrine Society doi: 10.1210/jc.2010-0130 Received January 19, 2010. Accepted November 15, 2010. First Published Online December 22, 2010

Abbreviations: ⌬4A, ⌬4-Androstenedione; AUC, area under the curve; BMI, body mass index; CV, coefficient of variation; DHEAS, dehydroepiandrosterone sulfate; FAI, free androgen index; 17-OHCS, 17-hydroxycorticosteroid; 17-OHP, 17-hydroxyprogesterone; PCOS, polycystic ovary syndrome; T, testosterone.

J Clin Endocrinol Metab, March 2011, 96(3):623– 631

jcem.endojournals.org

623

624

Markopoulos et al.

Hyperandrogenism in PCOS after Menopause

yperandrogenism is a major pathophysiological feature of polycystic ovary syndrome (PCOS) with 60 – 80% prevalence (1). Increased ovarian androgen production by the theca cells is the main contributor to androgen excess in women of reproductive age with PCOS (2), whereas about 20 – 60% of women with classic anovulatory PCOS have adrenal androgen excess, as determined by elevated dehydroepiandrosterone sulfate (DHEAS) levels (3). The latter originates mainly from the zona reticularis of the adrenal cortex (4). Adrenal androgen hypersecretion in PCOS women of reproductive age is also observed after adrenal stimulation (5–7). Menopausal transition involves many changes regarding women’s androgen status. Both the ovaries and the adrenals seem to contribute to androgen production in healthy women after menopause. Certain authors reported that postmenopausal ovary is associated with approximately 50% of total testosterone (T) and about 30% of ⌬4-androstenedione (⌬4A) production (8 –10), with the remaining steroids originating from the adrenals and from peripheral conversion of androgen precursors in adipose tissue. Interestingly, in postmenopausal women with intact ovaries and adrenal insufficiency under glucocorticoid therapy, T levels were undetectable (11). Natural menopause coincides with the age-related decline of all androgens (12). Little is known about the clinical and biochemical features of PCOS after menopause. Enhanced androgen production, assessed by total T concentrations, persists in postmenopausal women with ultrasonographic evidence of polycystic ovaries alone (13). However, there is evidence that hyperandrogenism resolves partially in women with PCOS in perimenopause (14). Adrenal androgens in PCOS remain high until menopause, whereas adrenal hyperandrogenism is not well documented in postmenopausal women with PCOS (15). Further investigation is warranted in this area. The aim of this study was to investigate hyperandrogenism in PCOS after menopause. To do so, we measured ovarian and adrenal androgen levels at baseline, after ACTH stimulation, and after 3-d oral dexamethasone administration. Hypothalamic-pituitary-adrenal axis response was also evaluated with CRH stimulation.

H


TABLE 1. Clinical characteristics (mean ⫾ SD) of postmenopausal PCOS and control women Characteristics Age (yr) Years since menopause Weight (kg) BMI Waist circumference (cm) Waist to hip ratio Total body fat (%)

PCOS (n ⴝ 20) 54.87 ⫾ 4.42 4.57 ⫾ 2.81 75.14 ⫾ 8.60 28.80 ⫾ 2.63 99.5 ⫾ 10.96

Controls (n ⴝ 20) P 58.45 ⫾ 6.59 NS 5.45 ⫾ 3.72 NS 65.18 ⫾ 9.86 NS 26.03 ⫾ 2.87 NS 88.70 ⫾ 8.99 ⬍0.05

0.90 ⫾ 0.10 43.61 ⫾ 4.32

0.84 ⫾ 0.06 ⬍0.05 40.38 ⫾ 4.89 NS

NS, Nonsignificance.

record of their menses and their clinical and biochemical status of hyperandrogenism. Diagnosis of the menopausal status was based on a history of more than 12 months of amenorrhea after the time of the final menstrual period and on FSH serum levels greater than 30 IU/liter. Diagnosis of PCOS was confirmed retrospectively and was based on the National Institutes of Health criteria established in 1990 (16). All subjects with a history of oligomenorrhea (less than eight menses per year) and signs of clinical and/or biochemical hyperandrogenism in their record were diagnosed as patients with PCOS. All control subjects were characterized by regular menses and absence of clinical and biochemical signs of hyperandrogenism during their reproductive years. Women with specific entities that mimic PCOS such as hyperprolactinemia, nonclassic adrenal hyperplasia, Cushing’s syndrome, and ovarian or adrenal androgen-secreting tumors were excluded. Women with elevated blood pressure, diabetes (based on 75 g, 2 h oral glucose tolerance test), history of oophorectomy or hysterectomy, or use of any hormonal medication for at least 1 yr before the study were also excluded. All subjects were in good health. The study was approved by our institution’s ethics committee and all subjects gave written informed consent. There were no dropouts during the study period.

Protocol

Patients and Methods

At the first visit, a complete medical history was obtained and standard anthropometric examination including weight, height, and waist and hip circumference measurements was performed. Skinfold thickness at four body sites (bicep, tricep, suprailiac, and subcscapular skinfold) was also measured. Blood samples were obtained for measurement of FSH, LH, prolactin, total T, ⌬4A, DHEAS, SHBG, progesterone, 17-hydroxyprogesterone (17-OHP) and estradiol levels at baseline. At 0800 h subjects underwent a rapid ACTH stimulation test. One week later, subjects collected a 24-h urine sample for measurement of cortisol and 17-hydroxycorticosteroid (17-OHCS) levels before their second visit. At their second visit, subjects had a CRH stimulation test at 1800 h. One week later, subjects underwent a 3-d adrenal suppression test with 0.5 mg dexamethasone every 6 h, and blood samples were taken at 0800 h of the fourth day (third visit).

Patients and control subjects

Methods

Twenty postmenopausal women with PCOS and 20 age- and body mass index (BMI)-matched healthy controls were included in this study (Table 1). All women were followed up in our obstetrics/gynecology department before and after their pregnancies until after their menopause. Thus, each subject had a detailed

Anthropometry Weight in kilograms was measured with an electronic scale. Height in meters was measured to the nearest millimeter with a stadiometer, and BMI in kilograms per square meter was calcu-


lated (Table 1). Maximum hip and waist circumference in centimeters was measured in duplicate with a flexible tape and waist to hip ratio was calculated. Skinfold thickness was measured on the right side of the body with a Harpenden skinfold caliper (Assist Creative Resources Ltd., Wrexham, UK) in triplicate to the nearest 0.1 mm. Bicep and tricep skinfold thicknesses were measured at the midpoint of the upper arm, between the acromion process and the tip of the bent elbow. Suprailiac skinfold was pinched at 2–3 cm above the iliac crest on the lateral side and midaxillary line. Subscapular thickness was measured at the natural fold approximately 2–3 cm below the shoulder blade at an oblique angle. Percentage of total body fat was estimated from skinfold thicknesses (17).

ACTH stimulation test At 0800 h, subjects were placed at bed rest and were given an iv bolus of 250 ␮g ACTH (Synactene 0.25 mg/ml; Novartis Pharma S.A.S., Rueil Malmaison, France) over 1 min (18). Blood samples for the measurement of cortisol, 17-OHP, ⌬4A, DHEAS, and total T serum levels were drawn 1 min before and 30 and 60 min after ACTH administration.


625

cence polarization immunoassay (Abbott Laboratories); the detection limit was 10 ng/ml; the total CV was less than 12%. 17-OHCS levels were measured by a chromatographic-spectrophotometric method (BioSystems SA, Barcelona, Spain); the detection limit was 1.3 mg/liter; the intraassay CV was 9.3%.

Statistical analysis Hormonal and clinical parameters were compared with a Student’s t test for normally distributed independent samples and with the Mann-Whitney U test for variables with a skewed distribution. When comparing androgen levels, to adjust for waist circumference, a multiple regression analysis was performed, taking as dependent variable each androgen alone and as independent variables the PCOS/control status and the waist circumference. By using the trapezoid method, ACTH and cortisol responses to CRH stimulation and cortisol, 17-OHP, ⌬4A, DHEAS, and total T responses to ACTH stimulation were expressed as area under the curve (AUC). Cortisol, 17-OHP, ⌬4A, DHEAS, and total T responses to ACTH stimulation were also

CRH stimulation test At 1700 h, subjects were placed at bed rest, and an iv catheter was inserted 45– 60 min before CRH administration. At 1800 h, an iv bolus of 1 ␮g/kg human CRH was administered over 1 min (19). Blood samples for the measurement of ACTH and cortisol levels were drawn 15 and 1 min before and 5, 15, 30, 60, 90, and 120 min after CRH administration.

Dexamethasone suppression test All subjects were given 0.5 mg dexamethasone orally every 6 h for 3 consecutive days starting at 0800 h of the first day (20). At 0800 h of the fourth day, blood samples were collected for measurement of cortisol, 17-OHP, ⌬4A, DHEAS, total T, and LH serum levels.

Hormone assays Plasma and serum from all blood samples were separated in a centrifuge within 30 min from blood withdrawal, stored in polystyrene tubes, and frozen at ⫺70 C until assayed. DHEAS, SHBG, and ACTH were measured by an electrochemiluminescence immunoassay (Roche Diagnostics, Mannheim, Germany); the detection limit was 0.1 ␮g/dl, 0.35 nmol/liter, and 1 pg/ml, respectively; the total coefficient of variation (CV) was less than 4.7%, less than 5.6% and less than 5.4%, respectively. Total T was measured by a microparticle enzyme immunoassay (Abbott Laboratories, Abbott Park, IL); the detection limit was 0.1 ng/ml; the total CV was less than 12%. Free androgen index (FAI) was calculated as the ratio of total T (nanomoles per liter) to SHBG (nanomoles per liter) levels ⫻ 100. ⌬4A was measured by an ELISA (IBL-America, Spring Lake Park, MN); the detection limit was 2 ng/ml; the total CV was less than 12%. 17-OHP was measured by ELISA (BioSource, Nivelles, Belgium); the detection limit was 0.03 ng/ml; the total CV was less than 8%. FSH, LH, estradiol, progesterone, and prolactin were measured by microparticle enzyme immunoassay (Abbott Laboratories); the analytical sensitivity was 0.4 mIU/ml, 0.5 mIU/ml, 20 pg/ml, 0.2 pg/ml, and 0.6 ng/ml, respectively; the total imprecision was 5.1–10.1, 5.2–10, 4.4 –13.1, 3.4 –11.7, and 3.4 – 6.3%, respectively. Serum and urinary cortisol were measured by a fluores-

FIG. 1. ACTH (A) and cortisol (B) levels (mean ⫾ SE) during CRH stimulation in postmenopausal PCOS (straight line) and control (dotted line) women. There were no statistically significant differences regarding ACTH and cortisol levels at each time point between postmenopausal PCOS and control women (one factor ANOVA repeated measures).

626

Markopoulos et al.



ANOVA repeated measures. Before (baseline) and after the dexamethasone suppression test, variables were compared between postmenopausal PCOS and control women by two-way ANOVA followed by Fisher’s least significant difference post hoc test. The change ⌬ of cortisol, ⌬4A, DHEAS, 17OHP, total T, FAI, and LH levels measured during the dexamethasone suppression test was defined as baseline levels minus postdexamethasone levels for each parameter (⌬cortisol, ⌬⌬4A, ⌬DHEAS, ⌬17-OHP, ⌬total T, ⌬FAI, and ⌬LH, respectively). The percentage reduction between baseline and postdexamethasone LH levels was calculated. The linear backward regression analysis was used to define the predictive variables. P ⬍0.05 was considered statistically significant. All the statistical analyses were performed using Statistica 6.0 software (StatSoft. Inc., Tulsa, OK).

Results Anthropometrics and baseline hormonal profile Baseline values of anthropometric characteristics of postmenopausal PCOS and control women are shown in Table 1. Waist circumference and waist to hip ratio values were higher in postmenopausal PCOS women compared with controls (P ⬍ 0.05). Age, number of years since menopause, weight, BMI, and percentage of total body fat did not differ significantly between postmenopausal PCOS and control women. Baseline values of hormonal characteristics of postmenopausal PCOS and control women are presented in Table 2. Baseline progesterone (P ⬍ 0.05), 17OHP (P ⬍ 0.05), ⌬4〈 (P ⬍ 0.05), DHEAS (P ⬍ 0.05), total T (P ⬍ 0.05), and FAI (P ⬍ 0.001) levels were higher, whereas SHBG levels (P ⬍ 0.01) were FIG. 2. ⌬4〈 (A), DHEAS (B), and total T (C) levels (mean ⫾ SE) in postmenopausal PCOS (straight line) and control (dotted line) women during ACTH stimulation. The asterisk (*) indicates a lower in postmenopausal PCOS than statistically significant difference (P ⬍ 0.05) from controls at the same time point. The cross (⫹) control women. When comparisons of indicates a statistically significant difference (P ⬍ 0.05) from the respective baseline levels. The ⌬4〈, DHEAS, total T, FAI, progesterAUCs and ⌬AUCs for ⌬4〈 (A1 and A2, respectively), DHEAS (B1 and B2, respectively), and total T one, 17-OHP, and SHBG baseline lev(C1 and C2, respectively) during ACTH stimulation in postmenopausal PCOS (black columns) and control (lineated columns) women are shown. The asterisk (*) indicates a statistically significant els, between PCOS and control difference (P ⬍ 0.05) from controls. women, were adjusted for waist circumference by multiple regression assessed by the AUC calculation after substraction of their reanalysis, their differences did not change (P ⬍ 0.05 for spective baseline values 关delta (⌬) AUC兴. ACTH and cortisol each parameter, respectively). Baseline LH, FSH, estralevels during CRH stimulation test were compared between postmenopausal women with PCOS and controls by one-way diol, prolactin, urinary cortisol, and 17-OHCS did not


FIG. 3. Cortisol (A), LH (B), DHEAS (C), and total T (D) levels (mean ⫾ SE) at baseline (d 1) and after 3-d (d 4) dexamethasone suppression in postmenopausal PCOS (straight line) and control (dotted line) women.The cross (⫹) indicates a statistically significant difference (P ⬍ 0.05) from the respective baseline levels. The asterisk (*) indicates a statistically significant difference (P ⬍ 0.05) from controls at the same time point. In their respective inserts, box and whisker plots (median, 25th to 75th quartile, nonoutlier range) of ⌬cortisol (A), ⌬LH (B), ⌬DHEAS (C), and ⌬total T (D) levels in postmenopausal PCOS and control women after 3-d dexamethasone suppression test are shown. The asterisk (*) indicates a statistically significant difference (P ⬍ 0.05) from controls.


627

FIG. 4. 17-OHP (A), ⌬4〈 (B), and FAI (C) levels (mean ⫾ SE) at baseline (d 1) and after 3-d (d 4) dexamethasone suppression in postmenopausal PCOS (straight line) and control (dotted line) women. The cross (⫹) indicates a statistically significant difference (P ⬍ 0.05) from the respective baseline levels. The asterisk (*) indicates a statistically significant difference (P ⬍ 0.05) from controls at the same time point. In their respective inserts, box and whisker plots (median, 25th to 75th quartile, nonoutlier range) of ⌬17-OHP (A), ⌬⌬4〈 (B), and ⌬FAI (C) levels in postmenopausal PCOS and control women after 3-d dexamethasone suppression test are shown. The asterisk (*) indicates a statistically significant difference (P ⬍ 0.05) from controls.

differ significantly between postmenopausal PCOS and control women. Backward multiple regression analysis performed in postmenopausal PCOS women revealed that among weight, waist to hip ratio, and baseline total T, DHEAS and ⌬4〈 levels, baseline total T levels were the best positive predictor (P ⬍ 0.05, ␤ ⫽ 0.542) of the PCOS status after menopause.

628

Markopoulos et al.



TABLE 2. Hormonal characteristics 关median (25th to 75th quartile)兴 of postmenopausal PCOS and control women Hormone FSH (mIU/ml) LH (mIU/ml) Estradiol (pg/ml) Prolactin (ng/ml) Progesterone (ng/ml) 17-OHP (ng/ml) ⌬4A (ng/dl) DHEAS (ng/ml) Total T (ng/ml) SHBG (nmol/liter) FAI Urinary cortisol (␮g per 24 h) Urinary cortisol (␮g per 24 h)/creatinine (g per 24 h) 17-OHCS (mg per 24 h) 17- OHCS (mg per 24 h)/creatinine (g per 24 h)

PCOS (n ⴝ 20) 72.8 (47.2–90.3) 37.8 (26.1– 42.2) 24 (20 –26) 8.2 (6.6 –12.3) 0.3 (0.3– 0.4) 0.5 (0.48 – 0.55) 205 (145–349) 1430 (591–1560) 0.47 (0.39 – 0.59) 43.2 (25.1–51.1) 3.94 (3.3–5.4) 46 (38 –95.2) 49.3 (34.4 –126.2) 4.7 (3.3–5.6) 3.3 (2.2–5.6)

Controls (n ⴝ 20) 77.1 (41.3– 88.6) 19.3 (18.9 –39.7) 29 (26 –39.6) 7.1 (4.9 –9) 0.2 (0.1– 0.2) 0.33 (0.23– 0.4) 107 (80 –144) 602 (390 – 889) 0.37 (0.18 – 0.42) 67.7 (54.5–98.3) 1.48 (1.05–1.6) 62.7 (34.5– 86) 57.1 (51.3–101.6) 3.7 (2.8 –5.8) 4.4 (2.2–5.9)

P NS NS NS NS ⬍0.05 ⬍0.05 ⬍0.05 ⬍0.05 ⬍0.05 ⬍0.01 ⬍0.001 NS NS NS NS

NS, Nonsignificance.

Hormonal tests measurements CRH test (Fig. 1) ACTH (Fig. 1A) and cortisol (Fig. 1B) levels at each time point during CRH test as well as levels of the AUC of the same hormones (data not shown) did not differ between postmenopausal PCOS and control women. ACTH test (Fig. 2) ⌬4〈 levels at 30 and 60 min after ACTH stimulation were higher (P ⬍ 0.05) and tended to be higher (P ⱖ 0.08), respectively, in postmenopausal PCOS than control women (Fig. 2A). DHEAS (Fig. 2B) and total T (Fig. 2C) levels were higher at 30 (P ⬍ 0.05) and 60 min (P ⬍ 0.05) after ACTH administration in PCOS than control women, whereas cortisol and 17-OHP levels did not differ significantly between these two groups (data not shown). ⌬4〈, DHEAS, and total T AUCs (Fig. 2, A1, B1, and C1, respectively) were higher (P ⬍ 0.05) in postmenopausal PCOS than control women, whereas cortisol and 17-OHP AUCs did not differ significantly between these two groups (data not shown). However, after adjusting for baseline levels, the AUCs for ⌬4〈, DHEAS, and total T did not differ significantly between postmenopausal PCOS and control women (Fig. 2, A2, B2, and C2, respectively). Dexamethasone suppression test (Figs. 3 and 4) At baseline, postmenopausal PCOS women had higher 17-OHP (P ⬍ 0.05), ⌬4A (P ⬍ 0.05), DHEAS (P ⬍ 0.01), total T (P ⬍ 0.01), and FAI (P ⬍ 0.01) levels than controls (two way ANOVA followed by Fisher’s least significant difference post hoc test). After 3-d dexamethasone suppression, cortisol, 17-OHP, ⌬4〈, DHEAS, total T, and FAI levels were significantly suppressed (P ⬍ 0.05) in both

postmenopausal PCOS and control women (Figs. 3 and 4). After dexamethasone suppression, 17-OHP, ⌬4〈, and FAI levels were significantly higher (P ⬍ 0.05, P ⬍ 0.05, and P ⬍ 0.01, respectively) in postmenopausal PCOS than control women, whereas there was no statistically significant difference in cortisol, DHEAS, and total T levels between the two groups. The changes in cortisol and 17OHP levels from baseline did not differ significantly between postmenopausal PCOS and control women (Figs. 3A and 4A, inserts, respectively), whereas the changes in ⌬4〈, DHEAS, total T, and FAI levels from baseline were significantly greater (P ⬍ 0.05) in PCOS than controls (Figs. 4B, 3C, 3D, and 4C, inserts, respectively). In both postmenopausal PCOS and control women, LH levels were not significantly different between baseline and after 3 d of dexamethasone (P ⫽ 0.12 and P ⫽ 0.41, respectively) (Fig. 3B). Furthermore, LH levels at baseline and after 3 d of dexamethasone were not significantly different between PCOS and control women (P ⫽ 0.08 and P ⫽ 0.57, respectively) (Fig. 3B). Similarly, there was no statistically significant difference in the change in LH levels from baseline (Fig. 3B) between postmenopausal PCOS and control women expressed as absolute circulating concentration (mean ⫾ SE, 8.78 ⫾ 3.12 vs. 8.56 ⫾ 2.45 mIU/ ml, respectively, P ⫽ 0.95, Student’s t test) or as percentage suppression from baseline (25 ⫾ 13 vs. 27 ⫾ 14%, respectively, P ⫽ 0.86, Student’s t test).

Discussion We found that postmenopausal overweight women with PCOS had higher abdominal fat accumulation than ageand BMI-matched controls, without difference in percent-


age of total body fat and weight. PCOS is frequently characterized by abdominal adiposity, regardless of BMI (21), whereas aging and possibly menopause are associated with an increase in fat mass and a redistribution of fat to the abdominal area (22, 23). In addition, postmenopausal women with PCOS had higher ⌬4〈, DHEAS, total T, FAI, progesterone, and 17-OHP baseline levels, whereas SHBG baseline levels were significantly lower than in control women. Abdominal obesity is associated with a more androgenic hormone profile and low SHBG levels in pre- and postmenopausal non-PCOS (12, 24) and premenopausal PCOS women (25). In this study, when comparisons of androgen levels between postmenopausal PCOS and control women were adjusted for waist circumference, their differences did not change. Furthermore, backward regression analysis demonstrated that baseline total T levels and not waist to hip ratio or weight were the most important predictor of the PCOS status after menopause. Abdominal obesity has been associated with enhanced cortisol secretion in obese premenopausal women (26). In this study, however, there was no difference in 24-h urinary cortisol and 17-OHCS levels between the PCOS and control women. Regarding hyperandrogenism in postmenopausal PCOS women, Winters et al. (14) found that although total and non-SHBG bound T levels decreased in peri- and a few postmenopausal women with PCOS, the latter had higher total and non-SHBG bound T levels when compared with non-PCOS age-matched women. In another study, Birdsall and Farquhar (13) found higher total T levels in postmenopausal women with ultrasonographic polycystic ovarian morphology when compared with postmenopausal women with normal ovaries. In addition, among 390 postmenopausal women, a subgroup of 103 with PCOS characteristics during reproductive years had ⌬4〈 and free T levels at the top quartile of the whole group of women (27). Puurunen et al. (15) studied different groups of women with and without PCOS from young reproductive age to postmenopause. The investigators concluded that there is no significant change in androgens with aging in women with PCOS, whereas total T, ⌬4〈, and DHEAS levels were higher in the six postmenopausal PCOS women studied than in the age- but not BMI-matched controls. Enhanced adrenal endocrine function in women with PCOS could be the result of altered pituitary ACTH secretion. In this study, post-CRH stimulation ACTH and cortisol responses did not differ between postmenopausal PCOS and control women. Similar findings were previously reported in young women with PCOS (5, 6). Others have suggested enhanced ACTH and cortisol responses in young women with PCOS after CRH stimulation (28).


629

Kondoh et al. (29) classified women of reproductive age with PCOS as normal, exaggerated, and poor responders to CRH stimulation. Furthermore, we found that after ACTH stimulation ⌬4〈, DHEAS, and total T levels remained higher in postmenopausal PCOS than control women. These differences are apparently due to the elevated baseline androgen levels in postmenopausal PCOS women as indicated by the absence of significant difference between the ⌬AUCs of the adrenal androgens in the two groups. Previous studies in PCOS before menopause reported increased adrenal androgen response to ACTH. Enhanced secretory response of ⌬4〈 and dehydroepiandrosterone after incremental ACTH stimulation was reported in young women with PCOS (6). In that study women were treated with 1 mg dexamethasone the night before ACTH testing, whereas ⌬4〈 levels were higher at baseline and the hormonal differences among the groups after ACTH stimulation were subtle. Moran et al. (7) reported higher activities of 17hydroxylase, 17,20-lyase, and 3␤-hydroxysteroid dehydrogenase, estimated from product to precursor ratios in response to ACTH stimulation in young PCOS women with adrenal androgen excess. Wu et al. (30) reported significantly higher 17-OHP and ⌬4〈 responses to exogenous ACTH administration in young women with PCOS, a finding confirmed by Lanzone et al. (31) in young PCOS women with hyperinsulinemia. Furthermore, Puurunen et al. (15) reported enhanced T, ⌬4〈, and DHEAS responses to ACTH stimulation in six postmenopausal women with PCOS. Of note, Erel et al. (32) demonstrated that adrenal androgen response to ACTH stimulation was similar between women with PCOS and controls of reproductive age. In this study, to reveal the extent of the ovarian contribution to the circulating androgen pool in PCOS women after menopause, adrenal steroid secretion was suppressed with 0.5 mg dexamethasone administration every 6 h for 3 d. LH levels at baseline and after dexamethasone administration were not significantly different in postmenopausal PCOS compared with control women. Because LH secretion is pulsatile, the importance of this finding might be limited by the single measurement of LH levels performed before and after dexamethasone suppression. Similarly, dexamethasone administration at the same regimen for 2 d did not suppress LH levels significantly in hirsute premenopausal women (20), whereas daily treatment of PCOS women of reproductive age with 1 mg dexamethasone for 1 month did not result in gonadotropin level reduction (33). In this study, suppressed but significantly higher 17-OHP and ⌬4〈 levels were found after dexamethasone administration in women with PCOS, suggesting that the postmenopausal ovary in PCOS women

630

Markopoulos et al.


produces more 17-OHP and ⌬4〈 than in control women. Dexamethasone administration suppressed DHEAS and total T in PCOS women to levels not significantly different from control women. According to this result (assuming that LH secretion was not disrupted by dexamethasone administration), together with the elevated baseline levels of these androgens in PCOS women, it seems that the adrenal fraction of these steroids is greater in postmenopausal women with PCOS than in controls. The greater change from baseline in ⌬4〈, DHEAS, and total T levels after adrenal steroidogenesis suppression by dexamethasone in postmenopausal PCOS than in control women underlines the adrenal contribution to hyperandrogenism in PCOS after menopause. Interestingly, during reproductive age dexamethasone suppression results in higher total T and DHEAS levels in PCOS than non-PCOS women, indicating the ovarian and/or peripheral contribution to these androgen levels (34, 35). In this study, FAI levels were suppressed after dexamethasone administration but remained higher in postmenopausal PCOS than control women. In view of the absence of difference in total T levels after dexamethasone suppression between PCOS and control women, it is possible that the higher FAI levels in the former could be attributed to the lower SHBG levels in PCOS than control women. Because young women with PCOS have normal cortisol levels, increased zona reticularis mass has been proposed as the cause of adrenal androgen hypersecretion (36). On the other hand, aging in healthy non-PCOS women is associated with diminished DHEAS but not ⌬4〈 adrenal production, which could be attributed to an age-related mass reduction of the zona reticularis (37). According to Puurunen et al. (15), adrenal androgen production in PCOS is sustained over time during the reproductive years until menopause. Molecular abnormalities resulting in increased 17-hydroxylase activity have been implicated in exaggerated adrenal androgen synthesis and secretion in women with PCOS (38). 〈 defective serine kinase action may induce posttranscriptional hyperphosphorylation of P450c17␣ serine residues leading to increased 17,20 lyase activity (39). However, others failed to link the CYP17 gene, which encodes for P450c17␣, with the pathogenesis of hyperandrogenism in PCOS (40). In conclusion, postmenopausal overweight women with PCOS are characterized by abdominal adiposity, a feature that apparently persists after menopause. At baseline, total T, ⌬4〈, DHEAS, and FAI levels are higher in postmenopausal PCOS than in non-PCOS women, whereas after ACTH stimulation, total T, ⌬4〈, and DHEAS levels remain higher in postmenopausal PCOS than non-PCOS women. The dexamethasone suppression results in postmenopausal PCOS women suggest that DHEAS and total T are partially of adrenal origin. This


adrenal androgen hypersecretion does not seem to be related to a centrally originating hypothalamic-pituitaryadrenal axis dysfunction, as indicated by the normal ACTH and cortisol responses to CRH stimulation. Ovarian androgen production also persists after menopause in women with PCOS as reflected by the higher ⌬4〈 levels after dexamethasone suppression, although the ovarian contribution was not fully assessed (i.e. by hypothalamicpituitary-gonadal axis suppression with a GnRH analog). Thus, adrenal and ovarian hyperandrogenism persist in PCOS women for at least 5 yr after menopause continuing the exposure of these women to excess androgen, possibly resulting in multiple adverse clinical sequelae. Future studies assessing fully the androgenic profile in postmenopausal PCOS women could be more informative (eventually with selective venous sampling of the adrenals and the ovaries or another novel assessment).

Acknowledgments Address all correspondence and requests for reprints to: Dr. George Mastorakos, 3 Neofytou Vamva Street, Athens 10674, Greece. E-mail: [email protected]. Disclosure Summary: The authors have nothing to disclose.

References 1. Azziz R, Carmina E, Dewailly D, Diamanti-Kandarakis E, EscobarMorreale HF, Futterweit W, Janssen OE, Legro RS, Norman RJ, Taylor AE, Witchel SF 2006 Criteria for defining polycystic ovary syndrome as a predominantly hyperandrogenic syndrome: an androgen excess society guideline. J Clin Endocrinol Metab 91:4237– 4245 2. Gilling-Smith C, Story H, Rogers V, Franks S 1997 Evidence for a primary abnormality of thecal cell steroidogenesis in the polycystic ovary syndrome. Clin Endocrinol (Oxf) 47:93–99 3. Kumar A, Woods KS, Bartolucci AA, Azziz R 2005 Prevalence of adrenal androgen excess in patients with the polycystic ovary syndrome (PCOS). Clin Endocrinol (Oxf) 62:644 – 649 4. Endoh A, Kristiansen SB, Casson PR, Buster JE, Hornsby PJ 1996 The zona reticularis is the site of biosynthesis of dehydroepiandrosterone and dehydroepiandrosterone sulfate in the adult human adrenal cortex resulting from its low expression of 3␤-hydroxysteroid dehydrogenase. J Clin Endocrinol Metab 81:3558 –3565 5. Carmina E, Lobo RA 1990 Pituitary-adrenal responses to ovine corticotropin-releasing factor in polycystic ovary syndrome and in other hyperandrogenic patients. Gynecol Endocrinol 4:225–232 6. Azziz R, Black V, Hines GA, Fox LM, Boots LR 1998 Adrenal androgen excess in the polycystic ovary syndrome: sensitivity and responsivity of the hypothalamic-pituitary-adrenal axis. J Clin Endocrinol Metab 83:2317–2323 7. Moran C, Reyna R, Boots LS, Azziz R 2004 Adrenocortical hyperresponsiveness to corticotropin in polycystic ovary syndrome patients with adrenal androgen excess. Fertil Steril 81:126 –131 8. Judd HL, Judd GE, Lucas WE, Yen SS 1974 Endocrine function of the postmenopausal ovary: concentration of androgens and estrogens in ovarian and peripheral vein blood. J Clin Endocrinol Metab 39:1020 –1024


9. Fogle RH, Stanczyk FZ, Zhang X, Paulson RJ 2007 Ovarian androgen production in postmenopausal women. J Clin Endocrinol Metab 92:3040 –3043 10. Judd HL, Lucas WE, Yen SS 1974 Effect of oophorectomy on circulating testosterone and androstenedione levels in patients with endometrial cancer. Am J Obstet Gynecol 118:793–798 11. Couzinet B, Meduri G, Lecce MG, Young J, Brailly S, Loosfelt H, Milgrom E, Schaison G 2001 The postmenopausal ovary is not a major androgen-producing gland. J Clin Endocrinol Metab 86: 5060 –5066 12. Burger HG, Dudley EC, Cui J, Dennerstein L, Hopper JL 2000 A prospective longitudinal study of serum testosterone, dehydroepiandrosterone sulfate, and sex hormone-binding globulin levels through the menopause transition. J Clin Endocrinol Metab 85: 2832–2838 13. Birdsall MA, Farquhar CM 1996 Polycystic ovaries in pre- and post-menopausal women. Clin Endocrinol (Oxf) 44:269 –276 14. Winters SJ, Talbott E, Guzick DS, Zborowski J, McHugh KP 2000 Serum testosterone levels decrease in middle age in women with the polycystic ovary syndrome. Fertil Steril 73:724 –729 15. Puurunen J, Piltonen T, Jaakkola P, Ruokonen A, Morin-Papunen L, Tapanainen JS 2009 Adrenal androgen production capacity remains high up to menopause in women with polycystic ovary syndrome. J Clin Endocrinol Metab 94:1973–1978 16. Zawadski JK, Dunaif A 1992 Diagnostic criteria for polycystic ovary syndrome: towards a rational approach. In: Dunaif A, Givens JR, Haseltine F, eds. Polycystic ovary syndrome. Boston: Blackwell Scientific; 377–384 17. Durnin JV, Womersley J 1974 Body fat assessed from total body density and its estimation from skinfold thickness: measurements on 481 men and women aged from 16 to 72 years. Br J Nutr 32:77–97 18. Kehlet H, Binder C 1973 Value of an ACTH test in assessing hypothalamic-pituitary-adrenocortical function in glucocorticoidtreated patients. Br Med J 2:147–149 19. Chrousos GP, Schulte HM, Oldfield EH, Gold PW, Cutler Jr GB, Loriaux DL 1984 The corticotropin-releasing factor stimulation test. An aid in the evaluation of patients with Cushing’s syndrome. N Engl J Med 310:622– 626 20. Karpas AE, Rodriguez-Rigau LJ, Smith KD, Steinberger E 1984 Effect of acute and chronic androgen suppression by glucocorticoids on gonadotropin levels in hirsute women. J Clin Endocrinol Metab 59:780 –784 21. Carmina E, Guastella E, Longo RA, Rini GB, Lobo RA 2009 Correlates of increased lean muscle mass in women with polycystic ovary syndrome. Eur J Endocrinol 161:583–589 22. Cartwright MJ, Tchkonia T, Kirkland JL 2007 Aging in adipocytes: potential impact of inherent, depot-specific mechanisms. Exp Gerontol 42:463– 471 23. Tchernof A, Poehlman ET 1998 Effects of the menopause transition on body fatness and body fat distribution. Obes Res 6:246 –254 24. Haffner SM, Katz MS, Dunn JF 1991 Increased upper body and overall adiposity is associated with decreased sex hormone binding globulin in postmenopausal women. Int J Obes 15:471– 478 25. Cosar E, Uc¸ok K, Akgu¨n L, Ko¨ken G, Sahin FK, Arioz DT, Bas¸ O 2008 Body fat composition and distribution in women with polycystic ovary syndrome. Gynecol Endocrinol 24:428 – 432 26. Mårin P, Darin N, Amemiya T, Andersson B, Jern S, Bjo¨rntorp P 1992 Cortisol secretion in relation to body fat distribution in obese premenopausal women. Metabolism 41:882– 886


631

27. Shaw LJ, Bairey Merz CN, Azziz R, Stanczyk FZ, Sopko G, Braunstein GD, Kelsey SF, Kip KE, Cooper-Dehoff RM, Johnson BD, Vaccarino V, Reis SE, Bittner V, Hodgson TK, Rogers W, Pepine CJ 2008 Postmenopausal women with a history of irregular menses and elevated androgen measurements at high risk for worsening cardiovascular event-free survival: results from the National Institutes of Health-National Heart, Lung, and Blood Institute sponsored Women’s Ischemia Syndrome Evaluation. J Clin Endocrinol Metab 93: 1276 –1284 28. Lanzone A, Petraglia F, Fulghesu AM, Ciampelli M, Caruso A, Mancuso S 1995 Corticotropin-releasing hormone induces an exaggerated response of adrenocorticotropic hormone and cortisol in polycystic ovary syndrome. Fertil Steril 63:1195–1199 29. Kondoh Y, Uemura T, Ishikawa M, Yokoi N, Hirahara F 1999 Classification of polycystic ovary syndrome into three types according to response to human corticotropin-releasing hormone. Fertil Steril 72:15–20 30. Wu XK, Zhou SY, Sallinen K, Po¨lla¨nen P, Erkkola R 2000 Ovarianadrenal cross-talk in polycystic ovary syndrome: evidence from wedge resection. Eur J Endocrinol 143:383–388 31. Lanzone A, Fulghesu AM, Guido M, Fortini A, Caruso A, Mancuso S 1992 Differential androgen response to adrenocorticotropic hormone stimulation in polycystic ovarian syndrome: relationship with insulin secretion. Fertil Steril 58:296 –301 32. Erel CT, Senturk LM, Oral E, Colgar U, Ertungealp E 1998 Adrenal androgenic response to 2-hour ACTH stimulation test in women with PCOS. Gynecol Endocrinol 12:223–229 33. Lachelin GC, Judd HL, Swanson SC, Hauck ME, Parker DC, Yen SS 1982 Long term effects of nightly dexamethasone administration in patients with polycystic ovarian disease. J Clin Endocrinol Metab 55:768 –773 34. Fassnacht M, Schlenz N, Schneider SB, Wudy SA, Allolio B, Arlt W 2003 Beyond adrenal and ovarian androgen generation: increased peripheral 5␣-reductase activity in women with polycystic ovary syndrome. J Clin Endocrinol Metab 88:2760 –2766 35. Fulghesu AM, Lanzone A, Fortini A, Guido M, Caruso A, Mancuso S 1993 Evaluation of ovarian and adrenal sources of androgens in women with polycystic ovary syndrome. Dexamethasone and GnRH-agonist administration. J Reprod Med 38:387–392 36. Yildiz BO, Azziz R 2007 The adrenal and polycystic ovary syndrome. Rev Endocr Metab Disord 8:331–342 37. Parker Jr CR, Slayden SM, Azziz R, Crabbe SL, Hines GA, Boots LR, Bae S 2000 Effects of aging on adrenal function in the human: responsiveness and sensitivity of adrenal androgens and cortisol to adrenocorticotropin in premenopausal and postmenopausal women. J Clin Endocrinol Metab 85:48 –54 38. Wickenheisser JK, Quinn PG, Nelson VL, Legro RS, Strauss 3rd JF, McAllister JM 2000 Differential activity of the cytochrome P450 17␣-hydroxylase and steroidogenic acute regulatory protein gene promoters in normal and polycystic ovary syndrome theca cells. J Clin Endocrinol Metab 85:2304 –2311 39. Zhang LH, Rodriguez H, Ohno S, Miller WL 1995 Serine phosphorylation of human P450c17 increases 17,20-lyase activity: implications for adrenarche and the polycystic ovary syndrome. Proc Natl Acad Sci USA 92:10619 –10623 40. Urbanek M, Legro RS, Driscoll DA, Azziz R, Ehrmann DA, Norman RJ, Strauss 3rd JF, Spielman RS, Dunaif A 1999 Thirty-seven candidate genes for polycystic ovary syndrome: strongest evidence for linkage is with follistatin. Proc Natl Acad Sci USA 96:8573– 8578