POLYCYSTIC OVARY SYNDROME

Transcription

1 FERTILITY AND STERILITY VOL. 77, NO. 5, MAY 2002 Copyright 2002 American Society for Reproductive Medicine Published by Elsevier Science Inc. Printed on acid-free paper in U.S.A. POLYCYSTIC OVARY SYNDROME Androgen and lipid profiles in adolescents with polycystic ovary syndrome who were treated with two forms of combined oral contraceptives George Mastorakos, M.D., Carolina Koliopoulos, M.D., and George Creatsas, M.D. 2nd Department of Obstetrics and Gynecology, University of Athens, Aretaieion Hospital, Athens, Greece Objective: To compare the effects of cyproterone acetate and desogestrel, as part of combined oral contraceptives, on lipid metabolism and hirsutism of adolescents with polycystic ovary syndrome (PCOS). Design: Prospective randomized clinical trial. Setting: Outpatient gynecology clinic (referral center) of a university. Patient(s): Twenty-eight adolescent girls with clinical and biological hyperandrogenism and six or less menses during the past 12 months. Intervention(s): Group A (n 14) received 0.15 mg of desogestrel plus mg of ethinyl estradiol daily. Group B (n 14) received 2 mg of cyproterone acetate plus mg of ethinyl estradiol daily. Treatment was given for 21 days followed by a 7-day rest for a period of 12 months. Main Outcome Measure(s): Hirsutism and lipid profile were evaluated before initiation and at 3, 6, 9, and 12 months of treatment. Androgen profile was evaluated before and at 12 months of treatment. Result(s): A significant decline of the Ferriman-Gallway hirsutism score was observed from the sixth month of therapy in both groups. During therapy, the levels of testosterone, free testosterone, 4 -androstenedione, and 17OH-progesterone decreased significantly, whereas sex hormone-binding globulin (SHBG) increased significantly in both groups. The level of total cholesterol and low density lipoprotein (LDL) cholesterol increased significantly, whereas high density lipoprotein (HDL) cholesterol and apolipoprotein A-I increased significantly from the third month of therapy in both groups. Total cholesterol/hdl cholesterol and LDL cholesterol/hdl cholesterol ratios remained unchanged. The levels of triglycerides increased significantly in the cyproterone acetate treated group after the third month. Conclusion(s): Treatment of adolescent girls with PCOS with the two studied formulations is comparably effective in decreasing hirsutism and androgen levels. Both combined oral contraceptives are associated with an increase of total cholesterol, LDL cholesterol, and HDL cholesterol levels and no change of the total cholesterol/hdl cholesterol and LDL cholesterol/hdl cholesterol ratios. Treatment with the cyproterone acetate combined oral contraceptive is associated with a tendency toward increasing the levels of triglycerides. (Fertil Steril 2002;77: by American Society for Reproductive Medicine.) Key Words: Adolescence, hirsutism, polycystic ovary syndrome (PCOS), oral contraceptives, desogestrel, cyproterone acetate, lipid metabolism Received February 12, 2001; revised and accepted November 2, Reprint requests: George Mastorakos, 3, Neofytou Vamva, Athens, Greece (FAX: ; mastorak@mail. kapatel.gr) /02/$22.00 PII S (02)02993-X Polycystic ovary syndrome (PCOS) is among the most common disorders of the premenopausal women, affecting 5% to 10% of this population (1). It is characterized by hyperandrogenism, chronic anovulation, and insulin resistance, which results in hirsutism, irregular menses, and infertility, as well as a higher risk for diabetes mellitus and cardiovascular disease (2). There are studies suggesting a strong familial component in PCOS (3). It is now proposed that PCOS is an oligogenic disorder in which a small number of key genes interact with environmental factors (notably dietary), resulting in the abnormalities associated with this syndrome (4). Polycystic ovary syndrome has a pubertal onset (5). Adolescent girls affected by PCOS present with menstrual irregularity and often weight gain immediately or shortly after menarche. They develop signs of hyperandrogenism, associated with augmented LH pulsatility and increased LH/FSH ratio (5). These 919

2 patients demonstrate increased ovarian volume, due to increased ovarian stroma; these multiple small follicles are histologically atretic, and are located peripherally along with a thickened sclerotic ovarian cortex (6, 7). They present also with hyperinsulinemia, reflecting insulin resistance and reduced levels of insulin growth factor binding protein 1 (IGFBP-1) and sex hormone-binding globulin (SHBG) (8). In certain girls with premature adrenarche, hyperandrogenism may be the first sign of PCOS and/or insulin resistance (9). The link between these disorders could be serine phosphorylation of the 17,20-lyase activity of P450c17 (10) and/or of the insulin receptor (11). Different therapeutic approaches have been proposed for the treatment of PCOS-related hirsutism. The antiandrogenic progestogen cyproterone acetate in combined oral contraceptives is employed in the therapy of the PCOS-related hirsutism in some parts of the world such as Europe (12). On the other hand, new progestogens such as desogestrel, employed in the new combined oral contraceptives, have a very low androgenic activity and may prove to be effective in the treatment of hirsutism (13). Dyslipidemia characterized by increased total cholesterol, low density lipoprotein (LDL) cholesterol, and triglycerides, as well as decreased high density lipoprotein (HDL) cholesterol is found frequently in women with PCOS independently of the presence of excess weight (14 18). Thus, the study of the effect of treatment of PCOS on the lipid profile of these patients, especially at the age of adolescence, is important. The metabolic effects of the new combined oral contraceptives when employed for contraception by healthy adult women are minimal (19 21). To compare the effects on hirsutism score, androgen, and lipid profile of the two treatments with a combined oral contraceptive containing cyproterone acetate and another containing desogestrel, we employed a prospective randomized study of 28 adolescent PCOS girls during a 12-month period of treatment with either of these two formulations. MATERIAL AND METHODS TABLE 1 Clinical, hormonal, and metabolic characteristics of adolescent girls with PCOS before initiation of treatment with desogestrel/ethinyl estradiol (group A) or cyproterone acetate/ethinyl estradiol (group B) (mean SE). Characteristic Group A Desogestrel/ethinyl estradiol (n 14) Group B Cyproterone acetate/ ethinyl estradiol (n 14) Age (y) Hirsutism score Waist/hip ratio BMI Testosterone (ng/ml) Free testosterone (pg/ml) androstenedione (ng/ml) DHEAS (ng/ml) SHBG (nmol/l) Total cholesterol (mg/dl) LDL cholesterol (mg/dl) HDL cholesterol (mg/dl) Triglycerides (mg/dl) Apolipoprotein A-I (mg/dl) Apolipoprotein A-II (mg/dl) Apolipoprotein B (mg/dl) Lp( ) (mg/dl) Total/HDL cholesterol LDL/HDL cholesterol Fasting glucose (mg/dl) Fasting insulin ( IU/mL) Fasting glucose/insulin Note: There is no statistically significant difference in any value between the two groups. Patients Twenty-eight adolescent girls (age range: 14 to 19 years) with clinically evoked and biologically confirmed hyperandrogenism and six or fewer menses during the past 12 months indicating chronic anovulation were consecutively recruited for this study in the outpatient adolescent gynecology clinic of our university hospital (Table 1). The diagnosis of PCOS in these patients was based on the criteria established at the 1990 PCOS conference organized at the National Institutes of Health (Bethesda, Maryland) by NICHD (National Institute of Child Health and Development) (1). The above criteria, along with the exclusion of specific disorders such as nonclassic adrenal 21-hydroxylase deficiency, hyperprolactinemia, or androgen-secreting neoplasms, define PCOS (1). Hormone measurements and one dynamic test were performed on the patients before initiation of the protocol as part of the exclusion criteria to rule out other pathologic entities that might be themselves responsible for a PCOSlike syndrome. Thus, prolactin plasma levels were evaluated to rule out hyperprolactinemia. Thyroid-stimulating hormone and free triiodothyronine plasma levels were evaluated to rule out a functional thyroidopathy. To rule out 21- hydroxylase deficiency associated nonclassic congenital adrenal hyperplasia, each patient had a 250- g ACTH stimulation test IV followed by measurement of 17OHprogesterone. Patients were all in good health, euthyroid, and had normal plasma prolactin levels. The response of 17OHprogesterone to the ACTH stimulation test in all patients was within normal range. No participants had used any hormonal medication, including combined oral contraceptives, for at least 6 months before the study. Body mass index (BMI) and waist/hip ratio were calculated for each patient (see Table 1). Seventeen 920 Mastorakos et al. PCOS in adolescence Vol. 77, No. 5, May 2002

3 (60.71%), 8 (28.57%), and 3 (10.71%) patients presented with a waist/hip ratio 0.80, between 0.80 and 0.85, and 0.85, respectively. Hirsutism was evaluated according to Ferriman-Gallway criteria (see Table 1) (22). The investigators who determined the Ferriman-Gallway scores were blinded to the participant s treatment arm. The patients and their parents were informed about the study and gave their consent. The study was approved by the ethics committee of our institution and institutional review board approval was obtained. Protocol At admission, blood samples were obtained on days 3 to 7 from cycling patients or on a random day for amenorrheic patients for measurement of levels of testosterone, free testosterone, 4 -androstenedione, dehydroepiandrosterone sulfate (DHEAS), 17OH-progesterone and SHBG, total cholesterol, LDL cholesterol and HDL cholesterol subfractions, triglycerides, apolipoproteins (A-I, A-II, B) and Lp( ). On the same day, an abdominal ultrasound with a 5-MHz transducer was performed to assess the morphology and volume of the uterus and ovaries. Subsequently, patients were randomly assigned to one of the following treatment groups. Group A was 14 patients receiving 0.15 mg of desogestrel plus mg of ethinyl estradiol daily. Group B was 14 patients receiving 2 mg of cyproterone acetate plus mg of ethinyl estradiol daily. Treatment was administered daily for 21 days, followed by a 7-day rest, over a period of 1 year. Hirsutism and levels of androgens and lipids were reevaluated trimonthly after initiation of therapy till completion of the 12-month study period. To avoid a factor of confusion in the interpretation of data, the patients were not advised to lose weight during the study period. We decided to study two commonly used combined oral contraceptive formulations, which contain slightly different amounts of ethinyl estradiol, to draw some practical conclusions on their usefulness for the treatment of PCOS. Hormone Assays Testosterone was measured by a radioimmunoassay (RIA) method (SORIN, Saluggia, Italy), with a sensitivity of 0.05 ng/ml, and coefficient of intra-assay and interassay variability of 9.4% and 10.0%, respectively. Free testosterone was measured by an RIA method (DPC, Los Angeles, CA), with sensitivity of 0.1 pg/ml, and coefficient of intraassay and interassay variability of 4.3% and 6.5%, respectively. Sex hormone binding globulin was measured by an IRMA method (Diagnostic Systems Laboratories, Webster, TX), with a sensitivity of 3 nmol/l, and coefficient of intra- and interassay variability of 3.7% and 11.5%, respectively. Androstendione was measured by an RIA method (Diagnostic Systems Laboratories), with sensitivity of 0.02 ng/ml, and coefficient of intra-assay and interassay variability of 5.9% and 7.0%, respectively. Lipid Assays Serum total cholesterol and triglycerides were measured by enzymatic procedures. Precipitation was used to measure HDL cholesterol; LDL cholesterol was calculated according to this equation: (LDL cholesterol) Total cholesterol Triglycerides/5 (HDL cholesterol). Apolipoproteins A-I, A-II, and B were measured by tholosimetria (Biodesign, Kennebunkport, ME), with coefficient of intra-assay and interassay variability of 7.9% and 10.1%, respectively, for A1; 7% and 9%, respectively, for A2; and 6.5% and 8.2%, respectively, for B. An ELISA method with coefficient of intra-assay and interassay variability of 6.0% and 12.7%, respectively, was used to measure Lp( ) (Innotest Lp( ); Innogenetics, Belgium). Statistical Analysis Statistical analysis was performed by multiple regression analysis and ANOVA repeated measures. The LSD post-hoc test was applied, and P.05 was considered statistically significant. RESULTS For the initial values see Table 1, and for correlations of clinical and biological parameters before initiation of treatment see Table 2. The initial clinical characteristics, and androgens and lipids levels did not differ between the treatment groups. Fasting glucose and insulin levels as well as the fasting glucose/insulin ratio before treatment were not statistically different between the two groups, indicating that insulin sensitivity was comparable between them (see Table 1). BMI correlated positively with total cholesterol/hdl cholesterol and LDL cholesterol/hdl cholesterol ratios, as well as with levels of Lp( ) and free testosterone (P.05). These total cholesterol/hdl cholesterol and LDL cholesterol/hdl cholesterol ratios were positively correlated with waist/hip ratio (P.05). Sex hormone binding globulin correlated negatively with BMI, waist/hip ratio, and total cholesterol/hdl cholesterol and LDL cholesterol/hdl cholesterol ratios (P.05); it was positively correlated with HDL cholesterol (P.05). There was also a negative but not statistically significant correlation of HDL cholesterol with the Ferriman-Gallway score (P.08). Apolipoprotein A-I correlated negatively with free testosterone (P.05). No other correlation was observed between androgens (testosterone, free testosterone, 4 -androstenedione, and DHEAS) and lipids. The level of 17OH-progesterone was positively correlated with 4 -androstenedione and DHEAS (P.05). A strong correlation between testosterone and free testosterone was observed (r 0.895, P.001), serving as internal control for the correlation analyses (see Table 2). FERTILITY & STERILITY 921

4 TABLE 2 Statistically significant (P.05) coefficients of correlation (r) of clinical and biological parameters before initiation of treatment. Parameter 1 BMI BMI BMI BMI BMI BMI Parameter 2 Waist/Hip Free testosterone SHBG Lp( ) Total cholesterol/hdl-c LDL-C/HDL-C r Parameter 1 Waist/Hip Waist/Hip Waist/Hip Parameter 2 SHBG Total cholesterol/hdl cholesterol LDL cholesterol/hdl cholesterol r Parameter 1 Testosterone Testosterone 4 -androstenedione DHEAS Parameter 2 Free testosterone Apolipoprotein A-I 17OH-progesterone 17OH-progesterone r Parameter 1 SHBG SHBG SHBG Parameter 2 HDL cholesterol Total cholesterol/hdl cholesterol LDL cholesterol/hdl cholesterol r Clinical and Biological Parameters During Treatment From the clinical point of view, both formulations were well tolerated by PCOS adolescent patients, ensuring cyclicity. During treatment, the BMI and waist/hip ratio remained unchanged in both groups of adolescent girls with PCOS. Hirsutism Score and Androgens Levels During Treatment A significant decline in the Ferriman-Gallway hirsutism score was observed from the sixth month of therapy with both formulations (P.05) (Fig. 1). The hirsutism score continued to decline until the 12th month of treatment (P.05). No difference was observed between the 9th and 12th month of treatment. When the hirsutism score was compared between the two groups during treatment with one-factor ANOVA repeated measures method, no statistically significant difference was found at any time point (see Fig. 1). A statistically significant decline in testosterone, free testosterone, 4 -androstenedione, and 17OH-progesterone levels were observed after treatment with both formulations. During treatment, SHBG levels increased significantly in both groups (P.05), but DHEAS levels did not present any significant change (see Table 3). When testosterone, free testosterone, 4 -androstenedione, 17OH-progesterone, DHEAS, and SHBG levels were compared between the two groups during treatment with one-factor ANOVA repeated measures method no statistically significant difference was found at any time point. Lipids Levels During Treatment During treatment with both formulations, the total cholesterol levels increased at the third month of therapy. A significant rise of LDL cholesterol levels was observed at the twelfth month of treatment with both formulations as compared to their respective baseline. Total cholesterol and LDL cholesterol levels were not different at any time point of observation between the two groups. The serum levels of HDL cholesterol increased significantly from the third month of therapy with both formulations (P.05) (see Table 4). Total cholesterol/hdl cholesterol and LDL cholesterol/ HDL cholesterol ratios did not demonstrate any statistically significant change during the 12-month treatment in both groups when compared to their respective baseline or between them at any time point (data not shown). Regarding apolipoproteins, only A-I levels increased from the third up to the 12th month of treatment for the desogestrel-treated group and at the third month only for the cyproterone acetate treated group (P.05). No statistically significant change was observed on A-II and B apolipoprotein levels as well as Lp( ) levels in either group during the treatment period (data not shown). When total cholesterol, LDL cholesterol, HDL cholesterol, and apolipoprotein A-I levels were compared between the two groups with onefactor ANOVA repeated measures during treatment, no statistically significant difference was found at any time point. A statistically significant increase in triglycerides as compared to baseline was observed from the third month of therapy with the cyproterone acetate combined oral contraceptive (P.05) whereas triglycerides did not increase in the desogestrel-treated group during the whole time of the study (Fig. 2). When triglycerides were compared between the two groups during treatment, a tendency for significant difference was found in the cyproterone acetate as compared to the desogestreltreated group at the 12th month of therapy (P.08). DISCUSSION The role of combined oral contraceptives in the treatment of adult women with PCOS is well documented (2). In this 922 Mastorakos et al. PCOS in adolescence Vol. 77, No. 5, May 2002

5 FIGURE 1 Ferriman-Gallway hirsutism score in the two groups of PCOS adolescent girls during the 12-month period of treatment with a combined oral contraceptive containing desogestrel (D, solid line) or cyproterone acetate (CA, dashed line). The asterisk indicates a statistically significant difference (P.05) from the baseline value of each group. NS indicates that the difference between the two groups was not statistically significant during the whole 12-month period of treatment. Statistical analysis was performed by ANOVA repeated measures. Values are given as mean SE. study we found that the administration of two different formulations of combined oral contraceptives (desogestrel/ ethinyl estradiol and cyproterone acetate/ethinyl estradiol) to adolescent girls with PCOS are effective at ensuring normal menstrual cycles and reducing hirsutism after the first 6 months of therapy and at least up to 12 months as indicated TABLE 3 Serum total testosterone (Te, ng/ml), free testosterone (fte, pg/ml), 4 -androstenedione ( 4 A, ng/ml), 17OH-progesterone (17OHP, ng/ml), and sex hormone binding globulin (SHBG, nmol/l) values in the two groups of PCOS adolescent girls at baseline (0) and after 12 months (12) of treatment with a desogestrel combined oral contraceptive (group A) or a cyproterone acetate combined oral contraceptive (group B). Te (0) Te (12) fte (0) fte (12) 4 A(0) 4 A(12) Group A * (33.66%) * (48.46%) * (40.58%) Group B * (43,44%) * (45.62%) * (31.60%) DHEAS (0) DHEAS (12) 17OHP(0) 17OHP(12) SHBG(0) SHBG(12) Group A (11.08%) * (40.12%) * (473.61%) Group B (19.41%) * (53.94%) * (464.29%) Note: Values are expressed as mean SE. The asterisk (*) indicates statistically significant difference (P.05) from the baseline value for the corresponding hormone. Statistical analysis was performed by ANOVA repeated measures. Percentage in parentheses indicate change from baseline. No statistically significant difference was observed between the hormone values of the two treatment groups at any time point. FERTILITY & STERILITY 923

6 FIGURE 2 Serum triglyceride values in the two groups of PCOS adolescent girls during the 12-month period of treatment with a combined oral contraceptive containing desogestrel (D, solid line) or cyproterone acetate (CA, dashed line). The asterisk indicates a statistically significant difference (P.05) from the baseline value of each group. NS indicates that the difference between the two groups was not statistically significant during the whole 12-month period of treatment. Statistical analysis was performed by ANOVA repeated measures. Values are given as mean SE. by the Ferriman-Gallway hirsutism score. The latter reflects the significant reduction of testosterone, free testosterone, 4 -androstenedione, and 17OH-progesterone as well as the increase of SHBG found after treatment by both formulations. These results are in accordance with studies in adult patients with PCOS and/or hirsutism showing that these two combined oral contraceptives are equally effective in treating hirsutism, lowering androgens, and raising SHBG (23 27). More specifically, Porcile et al. (23) treated adult hirsute women with an oral contraceptive containing either desogestrel or cyproterone acetate. Initial Ferriman-Gallway hirsutism scores declined with both treatments to levels similar to those of the control group. No differences were observed between the two treatment groups. Regarding the antiandrogenic effect of combined oral contraceptives containing desogestrel, in a recent study the treatment of hirsute adult women with such a combined oral contraceptive (containing 30 mg of ethinyl estradiol and 150 g of desogestrel) significantly reduced the hirsutism score and hyperandrogenemia (27). Moreover, the same oral contraceptive employed by healthy adult women as contraception significantly reduced androgens such as DHEAS, testosterone, dihydrotestosterone, and 4 -androstenedione (28). Furthermore, treatment of 19 hyperandrogenic adolescent girls for 12 months with a similar combined oral contraceptive improved acne; reduced hirsutism in some patients (58%); reduced significantly plasma levels of LH, total and free testosterone; increased levels of SHBG; and decreased ovarian volume (29). These effects are attributed to both the estrogenic and progestogenic components of these combined oral contraceptives. The ethinyl estradiol contained in combined oral contraceptives increases the levels of SHBG, resulting in lower free androgen levels (30). The third generation of combined oral contraceptive formulations containing desogestrel blocks the estrogen-mediated increase of SHBG to a lesser degree than older combined oral contraceptives (31). Desogestrel and cyproterone acetate, as progestogens, inhibit LH-induced androgen production (30 32). Furthermore, the antiandrogenic progestogen cyproterone acetate exerts its 924 Mastorakos et al. PCOS in adolescence Vol. 77, No. 5, May 2002

7 TABLE 4 Serum total cholesterol (mg/dl), HDL cholesterol (mg/dl), LDL cholesterol (mg/dl), and apolipoprotein A-I (mg/dl) values in the two groups of PCOS adolescent girls at 0, 3, 6, 9, and 12 months of treatment with a desogestrel combined oral contraceptive (group A) or a cyproterone acetate combined oral contraceptive (group B). Time (months) (0) (3) (6) (9) (12) Total cholesterol Group A * * * Group B * * * * HDL cholesterol Group A * * * Group B * * * LDL cholesterol Group A * Group B * Apolipoprotein A-I Group A * * * * Group B * Note: Values are expressed as mean SE. The asterisk (*) indicates statistically significant difference (P.05) from baseline. Statistical analysis was performed by ANOVA repeated measures. No statistically significant difference was observed between the hormone values of the two treatment groups at any time point. effect by competing with androgens at their receptor sites (32). Regarding the lipid profile, we found that total cholesterol as well as LDL cholesterol increased significantly in both groups. Total cholesterol/hdl cholesterol and LDL cholesterol/hdl cholesterol ratios, however, did not change significantly during the 12-month follow-up period. These ratios are considered to be a rather reliable measure of the individual s risk for cardiovascular disease. Levels of HDL cholesterol increased significantly in adolescent PCOS patients treated with both formulations. In adult PCOS patients, some investigators have documented a rise of HDL cholesterol along with a rise of total cholesterol, suggesting an estrogenic dominance of the treatment (33, 34). However, Prelevic et al. (35) observed an increase in LDL cholesterol levels. Porcile et al. (23) observed a rise in HDL cholesterol in adult PCOS patients who were receiving the same desogestrel combined oral contraceptive only after the second year of treatment, whereas Cullberg et al. (36) did not observe any change at all in LDL cholesterol or HDL cholesterol levels. The A-I apolipoprotein levels, the major apolipoprotein of HDL cholesterol, increased from the third up to the 12th month of treatment for the desogestrel-treated group and for the third month only for the cyproterone acetate-treated group. Triglycerides increased during the 12- month follow-up period in the cyproterone acetate treated group of patients. This is in accordance with most studies on lipid metabolism among adult women with PCOS reporting a rise of triglycerides and/or hirsutism treated by a combined oral contraceptive containing cyproterone acetate (23, 33). On the other hand, desogestrel-treated adolescent PCOS patients demonstrated no significant rise of triglycerides. Escobar-Morreale et al. (27) reported that triglycerides remained unchanged in adult hirsute women who were treated by a combined oral contraceptive containing desogestrel, whereas other studies have documented a rise of triglycerides (36, 37). Adult PCOS women at baseline show reduced HDL cholesterol, and increased total cholesterol, LDL cholesterol, and triglycerides serum levels when compared with healthy controls (38 40). These differences are more pronounced in PCOS patients less than 40 years of age. This is probably related to the same age-associated LDL cholesterol rise in healthy controls (41). The abnormal lipid profile associated with hyperinsulinemia, hypertension, and obesity often observed in PCOS women suggests that adult PCOS patients may be at risk for cardiovascular disease (38 40). In this study, BMI and waist/hip ratio, before initiation of treatment, are positively correlated to each other, indicating that weight increase in adolescent PCOS girls favors the android type of obesity. Furthermore, BMI and waist/hip ratio, before initiation of treatment, correlate positively with total cholesterol/hdl cholesterol and LDL cholesterol/hdl cholesterol ratios, which is in accordance with observations indicating an association of central obesity with myocardial infarction, angina pectoris, and stroke (16, 42). A longitudinal study from Gotenborg, Sweden, demonstrated that waist/ hip ratio is positively associated with a 12-year incidence of myocardial infarction, angina pectoris, stroke, and death for adult women (43). Another unfavorable change in the lipid profile of obese adolescent girls with PCOS is demonstrated in our study by the positive correlation of BMI and Lp( ). Moreover, the FERTILITY & STERILITY 925

8 negative and positive correlation of BMI to SHBG and free testosterone, respectively, provides the biochemical proof of the progressing hyperandrogenism and hirsutism in parallel with the increase of obesity in adolescent girls with PCOS. At baseline, SHBG correlates negatively with total cholesterol/hdl cholesterol, LDL cholesterol/hdl cholesterol, and waist/hip ratios, and positively with HDL cholesterol. These findings are in accordance with previous reports where low SHBG is associated not only with increased risk for cardiovascular disease, excess body fat, and abdominal obesity but also with hyperinsulinemia and type 2 diabetes (44). Moreover, we observed that free testosterone correlates negatively with apolipoprotein A-I. It should be noted that both 4 -androstenedione and DHEAS levels are positively correlated with 17OH-progesterone levels, indicating that both ovarian and adrenal androgen steroid metabolisms are involved in the hyperandrogenism of PCOS adolescent girls, as previously reported for adult women with PCOS (45). In conclusion, both combined oral contraceptive formulations are effective at ensuring normal menstrual cycles and reducing hirsutism in adolescent PCOS patients. Both of the treatments studied with combined oral contraceptives had similar impacts on lipids, although the cyproterone acetate combined oral contraceptive increased levels of triglycerides more. It seems that both formulations are suitable for longterm treatment of young PCOS patients. The association of increased BMI with multiple parameters of lipid and androgen metabolism of adolescent girls with PCOS suggests that obese teenage patients should be encouraged to lose weight and exercise to improve the clinical and metabolic features of the syndrome. Furthermore, psychological support should be available. Acknowledgments: The authors thank Analytiki Laboratories for the measurement of androgens and lipids. References 1. Dunaif A, Givens JK, Haseltine F, Merriam GR. The polycystic ovary syndrome. Cambridge, MA: Blackwell Scientific, Creatsas G, Koliopoulos C, Mastorakos G. Combined oral contraceptive treatment of adolescent girls with polycystic ovary syndrome. Lipid profile. Ann N Y Acad Sci 2000;900: Legro RS. Polycystic ovary syndrome. Phenotype to genotype. Endocrinol Metab Clin North Am 1999;28: Franks S, Gharani N, McCarthy M. Genetic abnormalities in polycystic ovary syndrome. Ann Endocrinol (Paris) 1999;60: Apter D, Butzow T, Laughlin GA, Yen SS. Accelerated 24-hour LH pulsatile activity in adolescent girls with ovarian hyperandrogenism: relevance to the developmental phase of polycystic ovarian syndrome. J Clin Endocrinol Metab 1994;79: Herter LD, Magalhaes JA, Spritzer PM. Relevance of the determination of ovarian volume in adolescent girls with menstrual disorders. J Clin Ultrasound 1996;24: Venturoli S, Porcu E, Fabbri R, Pluchinotta V, Ruggeri S, Macrelli S, et al. Longitudinal change of sonographic ovarian aspects and endocrine parameters in irregular cycles of adolescence. Pediatr Res 1995; 38: Apter D, Butzow T, Laughlin GA, Yen SS. Metabolic features of polycystic ovary syndrome are found in adolescent girls with hyperandrogenism. J Clin Endocrinol Metab 1995;80: Miller WL. The molecular basis of premature adrenarche: an hypothesis. Acta Paediatr Suppl 1999;88: Zhang LH, Rodriguez H, Ohno S, Miller WL. Serine phosphorylation of human P450c17 increases 17,20-lyase activity: implications for adrenarche and the polycystic ovary syndrome. Proc Natl Acad Sci USA 1995;92: Dunaif A, Xia J, Book CB, Schenker E, Tang Z. Excessive insulin receptor serine phosphorylation in cultured fibroblasts and in skeletal muscle. A potential mechanism for insulin resistance in the polycystic ovary syndrome. J Clin Invest 1995;96: Creatsas G, Hassan E, Deligeoroglou E, Tolis G, Aravantinos D. Treatment of polycystic ovarian disease during adolescence with ethinyl estradiol: cyproterone acetate versus a D-Tr-6 LHRH analog. Int J Gynaecol Obstet 1993;42: Speroff L, Glass RH, Nathan GK. Clinical gynecologic endocrinology and infertility. Baltimore, MD: Williams & Wilkins, Robinson S, Henderson AD, Gelding SV, Kiddy D, Niththyananthan R, Bush A, et al. Dyslipidaemia is associated with insulin resistance in women with polycystic ovaries. Clin Endocrinol (Oxf) 1996;44: Norman RJ, Masters SC, Hague W, Beng C, Pannall P, Wang IX. Metabolic approaches to the subclassification of the polycystic ovary syndrome. Fertil Steril 1995;63: Wild RA. Obesity, lipids, cardiovascular disease and androgen excess. Am J Med 1995;98: Holte J, Bergh T, Berne C, Lithell H. Serum lipoprotein lipid profile in women with the polycystic ovary syndrome:relation to anthropometric, endocrine and metabolic variables. Clin Endocrinol (Oxf) 1994;41: Conway GS, Agrawal R, Betteridge DJ, Jacobs HS. Risk factors for coronary artery disease in lean and obese women with the polycystic ovary syndrome. Clin Endocrinol (Oxf) 1992; 37: Godsland IF, Crook D, Simpson R, Proudler T, Felton C, Lees B, et al. The effects of different formulations of oral contraceptive agents on lipid and carbohydrate metabolism. New Engl J Med 1990;323: Crook D, Godsland I. Safety evaluation of modern oral contraceptives. Effects on lipoprotein and carbohydrate metabolism. Contraception 1998;57: Lobo RA, Skinner JB, Lippman JS, Cirillo SJ. Plasma lipids and desogestrel and ethinyl estradiol: a metaanalysis. Fertil Steril 1996;65: Ferriman D, Gallwey JD. Clinical assessment of body hair growth in women. J Clin Endocrinol Metab 1961;21: Porcile A, Gallardo E. Longterm treatment of hirsutism: desogestrel compared with cyproterone acetate in oral contraceptives. Fertil Steril 1991;55: Gjonnaess H. Comparison of ovarian electrocautery and oral contraception in the treatment of hyperandrogenism in women with polycystic ovary syndrome. Acta Obstet Gynecol Scand 1999;78: Charoenvisal C, Thaipisuttikul Y, Pinjaroen S, Krisanapan O, Benjawang W, Koster A, Doesburg W. Effects on acne of two oral contraceptives containing desogestrel and cyproterone acetate. Int J Fertil Menopausal Stud 1996;41: Van der Vange N, Blankenstein MA, Kloosterboer HJ, Haspels AA, Thijssen JH. Effects of seven low-dose combined oral contraceptives on sex hormone binding globulin, corticosteroid binding globulin, total and free testosterone. Contraception 1990;41: Escobar-Morreale HF, Lasuncion MA, Sancho J. Treatment of hirsutism with ethinyl estradiol-desogestrel contraceptive pills has beneficial effects on the lipid profile and improves insulin sensitivity. Fertil Steril 2000;74: Coenen CM, Thomas CM, Borm GF, Hollanders JM, Rolland R. Changes in androgens during treatment with four low-dose contraceptives. Contraception 1996;53: Venturoli S, Porcu E, Gammi L, Fabbri R, Paradisi R, Flamigni C, et al. The effects of desogestrel and ethinylestradiol combination in normal and hyperandrogenic young girls: speculations on contraception in adolescence. Acta Eur Fertil 1988;19: Murphy AA, Cropp CS, Smith BS, Burkman RT, Zacar HA. Effect of low dose oral contraceptive on gonadotropins, androgens and sex hormone binding globulin in non-hirsute women. Fertil Steril 1990;53: Dewis P, Petsos P, Newman M, Anderson DC. The treatment of hirsutism with a combination of desogestrel and ethinyl oestradiol. Clin Endocrinol 1985;22: Mastorakos et al. PCOS in adolescence Vol. 77, No. 5, May 2002

9 32. Miller JA, Jacobs HS. Treatment of hirsutism and acne with cyproterone acetate. Cin Endocrinol Metab 1986;15: Falssetti L, Pasinetti E. Effects of long term administration of an oral contraceptive containing ethinylestradiol and cyproterone acetate on lipid metabolism in women with polycystic ovary syndrome. Acta Obstet Gynecol Scand 1995;74: Vexiau P, Vexiau-Robert D, Martineau I, Hardy N, Villette JM, Fiet J, Cathelineau G. Metabolic effect at six and twelve months of cyproterone acetate (2 mg) combined with ethinyl estradiol (35 micrograms) in 31 patients. Horm Metab Res 1990;22: Prelevic GM, Wurzburger MI, Trpkovic D. Effects of a low dose estrogen-antiandrogen combination (Diane-35) on lipid and carbohydrate metabolism in patients with polycystic ovary syndrome. Gynecol Endocrinol 1990;4: Cullberg G, Hamberger L, Mattsson LA, Mobacken H, Samsioe G. Lipid metabolic studies in women with a polycystic ovary syndrome during treatment with a low-dose desogestrel-ethinylestradiol combination. Acta Obstet Gynecol Scand 1985;64: Rojanasakul A, Chailurkit L, Sirimongkolkasem R, Chaturachinda K. Effects of combined desogestrel-ethinylestradiol treatment on lipid profiles in women with polycystic ovarian disease. Fertil Steril 1987; 48: Wild RA. Metabolic aspects of polycystic ovary syndrome. Semin Reprod Endocrinol 1997;15: Talbott E, Guzick D, Clerici A, Berga S, Detre K, Wiemer K, Kuller L. Coronary heart disease risk factors in women with polycystic ovary syndrome. Arterioscler Thromb Vask Biol 1995;15: Wild RA, Painter PC, Coulson PB, Carruth KB, Ranney GB. Lipoprotein lipid concentrations and cardiovascular risk in women with polycystic ovary syndrome. J Clin Endocrinol Metab 1985;61: Talbott E, Clerici A, Berga SL, Kuller L, Guzick D, Detre K, et al. Adverse lipid and coronary heart disease risk profiles in young women with polycystic ovary syndrome: results of a case control study. J Clin Epidemiol 1998;51: Pasquali R, Cassimirri F, Venturoli S, Antonio M, Morselli L, Reho S, et al. Body fat distribution has weight-independent effects on clinical, hormonal and metabolic features in women with the polycystic ovary syndrome. Metabolism 1994;43: Lapidus L, Bengtsson C, Larsson B, Kjell P, Rybo E, Sjostrom L. Distribution of adipose tissue and risk of cardiovascular disease and death: a 12 year follow up of participants in the population study of women in Gothenburg, Sweden. BMJ 1984;289: Tchernof A, Toth MJ, Poehlman ET. Sex hormone-binding globulin levels in middle-aged premenopausal women. Associations with visceral obesity and metabolic profile. Diabetes Care 1999;22: Ehrmann DA, Rosenfield RL, Barnes RB, Brigell DF, Sheikh Z. Detection of functional ovarian hyperandrogenism in women with androgen excess. N Engl J Med 1992;327: FERTILITY & STERILITY 927