Glucose Metabolism and Insulin Resistance in Women with Polycystic ...

0021-972X/03/$15.00/0 Printed in U.S.A.

The Journal of Clinical Endocrinology & Metabolism 88(8):3621–3625 Copyright © 2003 by The Endocrine Society doi: 10.1210/jc.2003-030328

Glucose Metabolism and Insulin Resistance in Women with Polycystic Ovary Syndrome during Therapy with Oral Contraceptives Containing Cyproterone Acetate or Desogestrel ` , M. PILLONI, G. B. MELIS, A. CAGNACCI, A. M. PAOLETTI, A. RENZI, M. ORRU

AND

A. VOLPE

Institute of Obstetrics and Gynecology, Policlinico di Modena (A.C., A.R., A.V.), 41100 Modena, Italy; and Institute of Obstetrics and Gynecology, University of Cagliari (A.M.P., M.O., M.P., G.B.M.), 09124 Cagliari, Italy Oral contraceptives slightly deteriorate insulin sensitivity. The present study investigated whether they may further unbalance the glucose metabolism of lean women with polycystic ovary syndrome (PCOS). Women with PCOS were assigned to receive for 6 months the biphasic association of 40/30 ␮g ethinyl estradiol (EE) and 25/125 ␮g desogestrel (DSG; n ⴝ 10) or the monophasic association of 35 ␮g EE and 2 mg cyproterone acetate (CPA; n ⴝ 10). Glucose tolerance was investigated by an oral glucose tolerance test (OGTT). Glucose utilization dependent [insulin sensitivity (SI)] or independent (Sg) of insulin was investigated by the minimal model method applied to a frequently sampled iv glucose tolerance test. EE/ DSG increased the response of C peptide to OGTT (1413 ⴞ 113

P

OLYCYSTIC OVARY SYNDROME (PCOS) is one of the most common reproductive endocrinological disorders, affecting 6 –10% of women (1). It is characterized by a heterogeneous clinical presentation including menstrual dysfunction, high androgen concentration, and ultrasound evidence of ovarian cysts (2, 3). A frequent finding in nonobese and obese women with PCOS is increased peripheral insulin resistance with secondary abnormal insulin secretion and hyperinsulinemia (4, 5). The major consequence of these abnormalities is a high and earlier incidence of impaired glucose tolerance and type 2 diabetes (6 – 8). Indeed, about 30 – 40% women of reproductive age with PCOS have impaired glucose tolerance, and about 10% suffer from type 2 diabetes. Insulin resistance and hyperinsulinemia not only lead to an earlier onset of diabetes, but also cluster with hypertension, dyslipidemia, atherosclerosis (9), and endothelial dysfunction (10, 11). Accordingly, women with PCOS are at a substantially increased risk for cardiovascular disease (12). Combined oral contraceptives (OCs) are frequently used for the long-term management of women with PCOS. Indeed, they allow normalization of menstrual irregularities, reduction of androgen production by the ovary, reduction of free androgen fraction, and reduction of peripheral androgen effects. Apart from these positive effects, OCs are believed to Abbreviations: AUC, Area under the curve; CPA, cyproterone acetate; DSG, desogestrel; EE, ethinyl estradiol; FSIGT, frequently sampled iv glucose tolerance test; OC, oral contraceptive; PCOS, polycystic ovary syndrome; Sg, glucose utilization-independent on insulin; SI, insulin sensitivity.

vs. 2053 ⴞ 213 area under the curve; P < 0.009) and the C peptide/insulin ratio (0.085 ⴞ 0.01 vs. 0.134 ⴞ 0.01 area under the curve; P < 0.003). It also increased the Sg (0.026 ⴞ 0.002 vs. 0.034 ⴞ 0.003; P < 0.04) and decreased the SI (2.40 ⴞ 0.26 vs. 1.68 ⴞ 0.27; P < 0.01). EE/CPA did not modify responses to OGTT of glucose, insulin, C peptide, or C peptide/insulin ratio. It did not modify Sg and significantly increased SI (1.47 ⴞ 0.38 vs. 3.27 ⴞ 0.48; P < 0.04). The present study indicates that EE/CPA improves SI, whereas EE/DSG impairs SI, but improves insulin clearance. The long-term metabolic effects of these two compounds on women with PCOS require further investigations. (J Clin Endocrinol Metab 88: 3621–3625, 2003)

slightly deteriorate insulin sensitivity (13–15), and for this reason their use may aggravate the already impaired glucose metabolism of women with PCOS. However, studies evaluating the role of OCs on glucose tolerance (16) or insulin sensitivity (17–23) in women with PCOS have obtained conflicting results, and either no effect (17–20) or a negative modification (16, 21–23) was reported. In the present study we investigated the effects exerted by two different OCs on the glucose-insulin metabolism of lean women with PCOS. Subjects and Methods Twenty lean young women with PCOS, who had not received OCs or endocrine and metabolic active treatments for at least 8 months, were enrolled in the study. Each woman gave her informed consent to the study, which had been previously approved by our local ethics committee and institutional review board. PCOS was defined as persistent amenorrhea or oligomenorrhea of perimenarchal onset. Ovarian hyperandrogenism was defined by elevated levels of total testosterone, free testosterone, or androstenedione; ultrasound evidence of PCOS; an LH/ FSH ratio greater than 1.0; and a Ferriman and Gallwey hirsutism score greater than 10 (24). Glucose tolerance was investigated using an oral glucose tolerance test (OGTT). Insulin sensitivity (SI) and glucose utilization independent of insulin (Sg) were investigated by a frequently sampled iv glucose tolerance test (FSIGT) associated with the minimal model method (25). Investigation was performed at baseline during the early follicular phase (4 –7 d after spontaneous or progestin-induced menstruation). FSIGT and OGTT were performed in a randomized order and 1 d apart. After the baseline evaluation each woman was randomized to receive for 6 months either 1) the biphasic OC containing, for the first 7 d, 40 ␮g ethinyl estradiol (EE) plus 25 ␮g desogestrel (DSG) and, for the subsequent 14 d, 30 ␮g EE plus 125 ␮g DSG (Gracial, Organon, Oss, The Netherlands; n ⫽ 10); or 2) the monophasic OC containing 35 ␮g EE and 2 mg cyproterone acetate (CPA; Diane 35, Schering AG, Berlin, Germany;

3621

3622

J Clin Endocrinol Metab, August 2003, 88(8):3621–3625

n ⫽ 10). Each woman was assigned to treatment following a single computer-generated list of randomization. During the 6 months of study, the women were requested not to modify their lifestyles or dietary habits. Investigations were repeated during the first 7 d of the last month of treatment.

Cagnacci et al. • OCs and Insulin in PCOS

testosterone, free testosterone, androstenedione, SHBG, and dehydroepiandrosterone sulfate were also analyzed in the first sample of OGTT by RIA (24). To avoid interassay variability, samples from each subject were analyzed together in the same assay.

Statistical analyses

OGTT evaluation Each woman was hospitalized at 0700 h after a 12-h overnight fast and 3 d of a diet containing at least 200 g/d carbohydrates. Women were kept at bed rest. A polyethylene catheter was inserted in an antecubital vein and was kept patent by a slow infusion of saline solution. A glucose load of 75 g was given orally at 0900 h. Samples of arterialized blood, obtained by forearm warming, were collected at ⫺15, 0, 30, 60, 90, 120, and 180 min after glucose administration.

FSIGT evaluation Each woman was hospitalized at 0700 h after a 12-h overnight fast and 3 d of a diet containing at least 200 g/d carbohydrates. Women were maintained at bed rest. Two polyethylene catheters were placed in two antecubital veins and were kept patent by a slow infusion of saline solution. One catheter was used for iv glucose or insulin administration, and the other for blood collection. At 0900 h, glucose (0.3 g/kg) was injected over 1 min iv, followed 20 min later by an iv insulin bolus (0.03 U/kg). Samples of arterialized blood were collected at ⫺5, ⫺1, 2, 4, 8, 20, 22, 30, 40, 60, 70, 100, and 180 min after the glucose load (26).

Assays For both the OGTT and FSIGT investigations, blood samples were collected in tubes and placed on ice. Blood was immediately centrifuged. An aliquot of serum was immediately tested for glucose levels, whereas another aliquot was immediately frozen at –25 C until assayed. Glucose was determined by the glucose oxidase method. Insulin levels were assayed in all samples in duplicate by RIA methods using commercial kits (Biodata, Guidonia Montecelio, Rome, Italy). The intra- and interassay coefficients of variation of the assay were 6.2% and 7%, respectively, and sensitivity was 14.35 pmol/liter. C Peptide levels were analyzed in all OGTT samples and in the samples collected during the first 20 min of the FSIGT. Assays were performed in duplicate using commercial RIA kits (Biodata, Guidonia Montecelio). Intra- and interassay coefficients of variation were 3.2% and 8.5%, respectively, and sensitivity was 33.1 pmol/liter. Circulating levels of LH, FSH, PRL, estradiol, total

Assuming that the difference induced by treatment was equal to 1 sd of the difference and by setting type I error at 0.05 and type II error at 0.20, eight subjects were sufficient in each group to detect a statistically significant modification in either the OGTT or FSIGT investigation. The responses of glucose, insulin, and C peptide observed during the OGTT and the first 20 min of the FSIGT were reported as absolute values and area under the curve (AUC), calculated by the trapezoid method and expressed in arbitrary units (units per minute). To obtain an index of hepatic insulin clearance, the C peptide/insulin ratio of absolute and integrated responses to glucose challenges was calculated (27, 28). Glucose and insulin values obtained during the FSIGT were used to calculate the SI (25, 26, 29) and Sg by a computerized algorithm (MIN-MOD). SI is expressed as units ⫻ 10⫺4 per minute ⫻ microunits per milliliter, and Sg as units ⫻ 10⫺4 per minute. Data were blindly analyzed by one of us (A.R.) using the statistical package StatView 5.0.1 for Apple MacIntosh (SAS Institute, Inc., Cary, NC). A t test was used to compare baseline data from the two groups. The t test for paired data was used to analyze variations induced by treatment within each group. Two-way ANOVA for repeated measures (treatment ⫻ time, with subjects as replicates) was also used to evaluate differences between the two treatment groups. All results are expressed as the mean ⫾ se.

Results

Basal conditions and modifications during treatment in body mass index, waist to hip ratio, and circulating levels of glucose, insulin, C peptide, C peptide/insulin ratio, PRL, SHBG, total testosterone, free testosterone, androstenedione, cortisol, 17-hydroxyprogesterone, and dehydroepiandrosterone sulfate of women receiving the two OCs are reported in Table 1. No significant difference was observed in the baseline values of the two groups, except for SHBG, which was significantly lower in the EE/DSG group (P ⬍ 0.04). At

TABLE 1. Mean (⫾SE) age, body mass index (BMI), waist to hip ratio (WHR), glucose, insulin, C peptide, C peptide/insulin, LH, FSH, prolactin (PRL), SHBG, total testosterone (tT), free testosterone (fT), androstenedione (A), cortisol, 17-hydroxyprogesterone (17OH-P), and dehydroepiandrosterone sulfate (DHEAS) in hyperandrogenic women with polycystic ovary syndrome before and after 6 months of therapy with the biphasic combination of EE/DSG or the monophasic combination of EE/CPA EE/DSG

Age (yr) BMI (kg/m2) WHR Glucose (mg/dl) Insulin (␮U/ml) C peptide (ng/ml) C peptide/insulin LH (mIU/ml) FSH (mIU/ml) PRL (ng/ml) SHBG (ng/ml) tT (ng/ml) fT (pg/ml) A (ng/ml) Cortisol (ng/ml) 17OH-P (ng/liter) DHEAS (ng/ml)

EE/CPA

Before

During

Pa

Before

During

Pa

22.7 ⫾ 0.7 23.5 ⫾ 1.9 0.7 ⫾ 0.02 84.5 ⫾ 2.9 15.7 ⫾ 1.1 1.57 ⫾ 0.22 0.082 ⫾ 0.016 8.5 ⫾ 1.7 6.0 ⫾ 0.5 19.8 ⫾ 2.8 43.2 ⫾ 6.9 0.6 ⫾ 0.01 1.4 ⫾ 0.09 3.8 ⫾ 0.2 166.6 ⫾ 23.4 1.3 ⫾ 0.04 1.9 ⫾ 0.4

22.5 ⫾ 1.4 0.7 ⫾ 0.02 85.2 ⫾ 2.4 19.6 ⫾ 1.1 2.8 ⫾ 0.17 0.145 ⫾ 0.009 3.1 ⫾ 1.1 2.3 ⫾ 0.6 22.2 ⫾ 6.3 177.2 ⫾ 1.9 0.36 ⫾ 0.07 1.0 ⫾ 0.1 1.6 ⫾ 0.3 282.0 ⫾ 27.3 0.9 ⫾ 0.2 1.8 ⫾ 0.3

NS NS NS ⬍0.01 ⬍0.01 ⬍0.02 ⬍0.01 ⬍0.01 NS ⬍0.01 ⬍0.01 ⬍0.05 ⬍0.05 ⬍0.01 ⬍0.05 NS

21.8 ⫾ 0.8 22.6 ⫾ 0.9 0.7 ⫾ 0.1 81.4 ⫾ 3.6 17.9 ⫾ 2.4 2.26 ⫾ 0.19 0.129 ⫾ 0.008 7.9 ⫾ 1.4 3.9 ⫾ 0.3 14.9 ⫾ 5.3 77.6 ⫾ 14.1 0.7 ⫾ 0.6 1.4 ⫾ 0.2 3.4 ⫾ 0.2 164.4 ⫾ 25.7 1.3 ⫾ 0.1 2.2 ⫾ 0.3

21.1 ⫾ 0.6 0.7 ⫾ 0.1 78.1 ⫾ 2.0 18.7 ⫾ 2.9 2.54 ⫾ 0.3 0.143 ⫾ 0.005 4.5 ⫾ 1.1 2.1 ⫾ 0.6 16.8 ⫾ 1.8 152.2 ⫾ 18.5 0.3 ⫾ 0.04 0.86 ⫾ 0.1 1.8 ⫾ 0.08 200.00 ⫾ 15.6 0.9 ⫾ 0.4 2.2 ⫾ 0.4

NS NS NS NS NS NS ⬍0.05 ⬍0.02 NS ⬍0.04 ⬍0.01 ⬍0.05 ⬍0.05 NS ⬍0.05 NS

Conversion factors to Systeme Internationale units are: glucose, ⫻ 0.05551; insulin, ⫻ 7.175; C peptide, ⫻ 331; LH ⫻ 1; FSH, ⫻ 1; PRL, ⫻ 1; T, ⫻ 3.467; A, ⫻ 0.0349; cortisol, ⫻ 27.59; 17OH-P ⫻ 0.03026; DHEAS, ⫻ 2.714. a P By paired t test before vs. during, within each group.


baseline, one subject in the EE/DSG group and two subjects in the EE/CPA group showed signs of glucose intolerance. All enrolled women completed the study and did not experience side-effects requiring dose adjustment. EE/DSG

In women receiving the biphasic OC containing EE plus DSG, responses to OGTT of glucose (17,456 ⫾ 897 vs. 17,398 ⫾ 1,048 AUC) and insulin (18,914 ⫾ 2,424 vs. 15,442 ⫾ 963 AUC) were not modified. On the other hand, the responses of C peptide (1,413 ⫾ 113 vs. 2,053 ⫾ 213 AUC; P ⬍ 0.009) and C peptide/insulin ratio (0.085 ⫾ 0.01 vs. 0.134 ⫾ 0.01 AUC; P ⬍ 0.003) were significantly increased (Fig. 1). The minimal model evaluation of the FSIGT results showed a significant increase in Sg (0.026 ⫾ 0.002 vs. 0.034 ⫾ 0.003; P ⬍ 0.04) and a significant decrease in SI (2.40 ⫾ 0.26 vs. 1.68 ⫾ 0.271; P ⬍ 0.01) after 6 months of OC administration (Fig. 2).

J Clin Endocrinol Metab, August 2003, 88(8):3621–3625 3623

EE/CPA

Responses to OGTT were not significantly modified by the administration of EE plus CPA. Integrated responses to OGTT of glucose (19,348 ⫾ 1,924 vs. 18,177 ⫾ 1,048 AUC), insulin (22,207 ⫾ 5,217 vs. 17,320 ⫾ 2,125 AUC), C peptide (2,541 ⫾ 389 vs. 2,006 ⫾ 183 AUC), and C peptide/insulin ratio (0.135 ⫾ 0.01 vs. 0.122 ⫾ 0.01 AUC) were similar before and after treatment (Fig. 1). The minimal model evaluation of the FSIGT results showed no significant modification of Sg (0.03 ⫾ 0.003 vs. 0.026 ⫾ 0.004) and a significant increase in SI (1.47 ⫾ 0.38 vs. 3.27 ⫾ 0.48; P ⬍ 0.04; Fig. 2). EE/DSG vs. EE/CPA

No difference was observed between the modifications induced by the two OCs in the responses of glucose, insulin, and C peptide to OGTT or FSIGT. Similarly, modification of Sg was similar during the two OCs. On the other hand, the increase in the C peptide/insulin ratio during the OGTT induced by EE/DSG was significantly higher than the modification induced by EE/CPA (P ⬍ 0.02). An interaction was also observed for SI (P ⬍ 0.03), which increased during EE/CPA and decreased during EE/DSG. Discussion

FIG. 1. Mean (⫾SE) glucose, insulin, C peptide, and C peptide/insulin responses to an OGTT in hyperandrogenic women with PCOS before and after 6 months of therapy with the biphasic combination of EE/ DSG (n ⫽ 10; left) or the monophasic combination of EE/CPA (n ⫽ 10; right). Arrows indicate the time of glucose administration. Conversion factor to Systeme Internationale units are: glucose, ⫻0.05551; insulin, ⫻7.175; C peptide, ⫻331.

In the present study we report the effect of two OCs on glucose-insulin metabolism in lean women with PCOS. The results show that the two OCs act differently. The biphasic combination of EE/DSG significantly decreased SI. This effect was associated with increased Sg and increased insulin production (C peptide) and clearance (C peptide/insulin ratio) during the OGTT. A reduction of Sg, for unknown reasons, has been described in women with PCOS (30 –32), and its improvement during EE/DSG might be beneficial. The mechanism through which EE/DSG increases the pancreatic ␤-cell response to glucose and insulin clearance is not known, but it resembles the effect induced by the administration of oral estrogens to postmenopausal women (27). Accordingly, this effect may be the consequence of the highly estrogenic equilibrium of this OC. An amplification of the pancreatic ␤-cell response to glucose may represent a positive effect, contributing to more efficient glucose control. Similarly, the induced increase in insulin clearance is probably beneficial because it indirectly indicates an improvement of insulin action on the liver, a key organ in carbohydrate metabolism (33). All of these mechanisms seem to counterbalance the reduction of SI induced by EE/DSG. Indeed, in the OGTT, glucose and peripheral insulin levels tend to be lower during than before EE/DSG administration. In the present study EE/CPA did not induce any modification of OGTT parameters, whereas significantly increased SI. EE/CPA was reported to decrease glucose tolerance, but not SI (20), in obese women with PCOS, in whom ovarian hyperandrogenism was not a necessary prerequisite for enrolment. Similarly, in other studies SI, analyzed by the fasting homeostasis model assessment evaluation (34) or by the clamp (18 –20, 22), was not influenced by EE/CPA. The difference between the present and previous results may derive from differences in the doses of EE/CPA used, in the

3624

J Clin Endocrinol Metab, August 2003, 88(8):3621–3625


FIG. 2. Mean (⫾SE) peripheral SI (units ⫻ 10⫺4 per minute ⫻ microunits per milliliter; left) and glucose-dependent glucose utilization (Sg; units ⫻ 10⫺4 per minute; right) in hyperandrogenic women with PCOS before (䡺) and after (o) 6 months of therapy with the biphasic combination of EE/DSG (n ⫽ 10) and the monophasic combination of EE/CPA (n ⫽ 10). *, P ⬍ 0.05; ***, P ⬍ 0.01 (vs. before).

selection of the subjects, and in the method used to evaluate insulin resistance (29). Greater doses of EE and/or CPA than those used here were employed in two previous studies (17, 18). Women with a body mass index comparable to those of our subjects were included only in one previous clamp study, in which the diagnosis of PCOS was not based on hyperandrogenism (22). Finally, the present report is the only study in which SI was investigated by the minimal model method instead of the clamp. Both are believed to reliably define insulin resistance (29), but they act differently. The clamp does not distinguish between SI and Sg and investigates the effect of insulin in a steady state, reached very slowly (26, 35). By contrast, the minimal model method evaluates the effect of insulin in a more dynamic situation. Accordingly, it is possible that slightly different results are achieved by application of the two methods. The reasons for decreased SI induced by EE/DSG and its improvement during EE/CPA are not clear. The different progestin molecules may play a role, but it cannot be excluded that the slightly higher dose of EE of the EE/DSG combination has an influence, considering that estrogens at high levels reduce SI (36). The single report of a reduction of abdominal fat in women with PCOS treated with EE/CPA or EE/DSG for a mean period of 10 yr indicates long-term beneficial metabolic effects of these OCs in women with PCOS (37). Similarly, our present results do not support a harmful effect of these two OCs in women with PCOS. Further studies are needed to evaluate whether they may contribute to reduce or delay the onset of diabetes in these women. Acknowledgments Received February 26, 2003. Accepted May 1, 2003. Address all correspondence and requests for reprints to: Dr. Angelo Cagnacci, Dipartimento Misto Materno Infantile, Ginecologia e Ostetricia, Policlinico di Modena, Via del Pozzo 71, 41100 Modena, Italy. E-mail: [email protected].

References 1. Carmina E, Lobo RA 1999 Polycystic ovary syndrome (PCOS): arguably the most common endocrinopathies associated with significant morbidity in women. J Clin Endocrinol Metab 84:1897–1899 2. Dunaif A 1994 Polycystic ovary syndrome. Curr Ther Endocrinol Metab 5:222– 229 3. Franks S 1995 Polycystic ovary syndrome. N Engl J Med 333:853– 861 4. Dunaif A, Finegood D 1996 ␤-cell dysfunction independent of obesity and glucose intolerance in the polycystic ovary syndrome. J Clin Endocrinol Metab 81:942–947 5. Dunaif A 1997 Insulin resistance and the polycystic ovary syndrome: mechanism and implications for pathogenesis. Endocr Rev 18:774 – 800

6. Legro RS 2001 Polycystic ovary syndrome: The new millenium. Mol Cell Endocrinol 184:87–93 7. Norman RJ, Masters L, Milner CR, Wang JX, Davies MJ 2001 Relative risk of conversion from normoglycemia to impaired glucose tolerance or non-insulin dependent diabetes mellitus in polycystic ovarian syndrome. Hum Reprod 16:1995–1998 8. Cibula D., Cifkova R., Fanta M, Poledne R, Zivny J, Skibova J 2000 Increased risk of non-insulin dependent diabetes mellitus, arterial hypertension and coronary artery disease in perimenopausal women with a history of the polycistic ovary syndrome. Hum Reprod 15:785–789 9. Nestler JE 1997 Insulin Regulation of human ovarian androgens [Review]. Hum Reprod 12(Suppl 1):53– 62 10. Paradisi G, Steimberg HO, Hempfling A, Cronin J, Hook G, Shepard MK, Baron AD 2001 Polycystic ovary syndrome is associated with endothelial dysfunction. Circulation 13:1410 –1415 11. Kelly CJG, Speirs A, Gould GW, Petrie JR, Lyall H, Connell JMC 2002 Altered vascular function in young women with polycystic ovary syndrome. J Clin Endocrinol Metab 87:742–746 12. Mather KJ, Kwan F, Corenblum B 2000 Hyperinsulinemia in polycystic ovary syndrome correlates with increased cardiovascular risk independent of obesity. Fertil Steril 73:150 –156 13. Kasdorf G, Kalkhoff RK 1988 Prospective studies of insulin sensitivity in normal women receiving oral contraceptive agents. J Clin Endocrinol Metab 66:846 – 852 14. Skouby SO, Andersen O, Saurberey N, Kuhl C 1987 Oral contraception and insulin sensitivity: in vivo assessment in normal women and women with previous gestational diabetes. J Clin Endocrinol Metab 64:519 –523 15. Godsland IF, Walton C, Felton C, Proudler A, Patel A, Wynn V 1992 Insulin resistance, secretion, and metabolism in users of oral contraceptives. J Clin Endocrinol Metab 74:64 –70 16. Nader S, Riad-Gabriel MG, Saad MF 1997 The effect of a desogestrel-containing oral contraceptive on glucose tolerance and leptin concentrations in hyperandrogenic women. J Clin Endocrinol Metab 82:3074 –3077 17. Pasquali R, Fabbri R, Venturoli S, Paradisi S, Antenucci D, Melchionda N 1986 Effect of weight loss and antiandrogenic therapy on sex hormone blood levels and insulin resistance in obese patients with polycystic ovaries. Am J Obstet Gynecol 154:139 –114 18. Dahlgren E, Landin K, Krotkiewski M, Holm G, Janson PO 1998 Effects of two antiandrogen treatments on hirsutism and insulin sensitivity in women with polycystic ovary syndrome. Hum Reprod 13:2706 –2711 19. Armstrong VL, Wiggan MI, Ennis CN, Sheridan B, Traub AI, Atkinson AB, Bell PM 2001 Insulin action and insulin secretion in polycystic ovary syndrome treated with ethinyl oestradiol/cyproterone acetate. Q J Med 94:31–37 20. Morin-Papunen LC, Vauhkonen I, Koivunen RM, Ruokonen A, Martikainen K, Tapanainen JS 2000 Endocrine and metabolic effects of metformin versus ethinyl estradiol-cyproterone acetate in obese women with polycystic ovary syndrome: a randomized study. J Clin Endocrinol Metab 85:3161–3168 21. Cibula D, Fanta M, Hill M, Sindelka G, Skrha J, Zivny J 2002 Insulin sensitivity in non-obese women with polycystic ovary syndrome during treatment with oral contraceptives containing low-androgenic progestin. Hum Reprod 17:76 – 82 22. Morin-Papunen LC, Vauhkonen I, Koivunen RM, Ruokonen A, Martikainen K, Tapanainen JS 2003 Metformin versus ethinyl estradiol-cyproterone acetate in the treatment of nonobese women with polycystic ovary syndrome: a randomized study. J Clin Endocrinol Metab 88:148 –156 23. Korytkowski M, Mokan M, Horwitz M, Berga S 1995 Metabolic effects of oral contraceptives in women with polycystic ovary syndrome. J Clin Endocrinol Metab 80:3327–3334 24. Paoletti AM, Cagnacci A, Soldani R, Orru` M, Ajossa S, Pittorra G, Mulas P, Melis GB 1995 Evidence that an altered prolactin release is consequent to abnormal ovarian activity in polycystic ovary syndrome. Fertil Steril 64:1094 – 1098 25. Welch S, Gebhart SSP, Bergman RN, Phillips LS 1990 Minimal model anal-


26.

27. 28.

29. 30. 31.

ysis of intravenous glucose tolerance test-derived insulin sensitivity in diabetics subjects. J Clin Endocrinol Metab 71:1508 –1518 Cagnacci A, Arangino S, Renzi A, Cagnacci P, Volpe A 2001 Insulin sensitivity in women: a comparison among values derived from intravenous glucose tolerance tests with different sampling frequency, oral glucose tolerance test or fasting. Eur J Endocrinol 145:281–287 Cagnacci A, Soldani R, Carriero PL, Paoletti AM, Fioretti P, Melis GB 1992 Effect of low doses of transdermal 17␤-estradiol on carbohydrate metabolism in postmenopausal women. J Clin Endocrinol Metab 74:1396 –1400 Cagnacci A, Tuveri F, Cirillo R, Setteneri AM, Melis GB, Volpe A 1998 The effect of 17␤-estradiol on glucose metabolism of postmenopausal women is evident during the oral but not the intravenous glucose administration. Maturitas 28:163–167 American Diabetes Association 1998 Consensus development conference on insulin resistance. Diabetes Care 21:310 –314 Ciaraldi TP, el-Roeiy A, Madar Z, Reichart D, Olefsky JM, Yen SSC 1992 Cellular mechanisms of insulin resistance in polycystic ovary sindrome. J Clin Endocrinol Metab 75:577–583 Falcone T, Little AB, Morris D 1992 Impaired glucose effectiveness in patients with polycystic ovary syndrome. Hum Reprod 7:922–925

J Clin Endocrinol Metab, August 2003, 88(8):3621–3625 3625

32. Cagnacci A, Paoletti AM, Arangino S, Melis GB, Volpe A 1999 Effect of ovarian suppression on glucose metabolism of young lean women with and without ovarian hyperandrogenism. Hum Reprod 14:893– 897 33. Bonora E, Zavaroni I, Coscelli C, Butturini U 1983 Decreased hepatic insulin extraction in subject with mild glucose intolerance Metabolism 32:438 – 446 34. Elter K, Imir G, Durmusoglu F 2002 Clinical, endocrine and metabolic effects of metformin added to ethinyl estradiol-cyproterone acetate in non-obese women with polycystic ovarian syndrome: a randomized controlled study. Hum Reprod 17:1729 –1737 35. Godsland IP, Stevenson JC 1995 Insulin resistance: syndrome or tendency? Lancet 346:100 –103 36. Lindheim SR, Presser SC, Ditkoff EC, Vijod MA, Stanczyk FZ, Lobo RA 1993 A possible bimodal effect of estrogen on insulin sensitivity in postmenopausal women and the attenuating effect of added progestin. Fertil Steril 60:664 – 667 37. Pasquali R, Gambineri A, Anconetani B, Vicennati V, Colitta D, Caramelli E, Casimirri F, Morselli-Labate AM 1999 The natural history of the metabolic syndrome in young women with the polycystic ovary sindrome and the effect of long-term oestrogen-progestagen treatment. Clin Endocrinol (Oxf) 50:517– 527