POLYCYSTIC OVARY SYNDROME

Transcription

1 POLYCYSTIC OVARY SYNDROME The impact of testosterone administration to female-to-male transsexuals on insulin resistance and lipid parameters compared with women with polycystic ovary syndrome Susanne Cupisti, M.D., Ph.D., a Erik J. Giltay, M.D., Ph.D., b Louis J. Gooren, M.D., Ph.D., c Desiree Kronawitter, M.D., a Patricia G. Oppelt, M.D., a Matthias W. Beckmann, M.D., Ph.D., a Ralf Dittrich, Ph.D., a and Andreas Mueller, M.D., Ph.D. a a Department of Obstetrics and Gynecology, Erlangen University Hospital, Erlangen, Germany; b Department of Psychiatry, Leiden University Medical Center, Leiden, the Netherlands; and c Department of Endocrinology, Free University Medical Center (VUMC), Amsterdam, the Netherlands Objective: To evaluate the impact of testosterone (T) administration to female-to-male transsexuals (FtMs) on insulin resistance and lipid parameters and to compare the effects with women with polycystic ovary syndrome (PCOS). Design: Cohort analysis. Setting: University hospital. Patient(s): Twenty-nine FtMs and 240 women with PCOS. Intervention(s): Screening panel, ultrasound of the ovaries, hormone, lipid, and glucose and insulin measurements. Main Outcome Measure(s): Endocrine, metabolic parameters, and insulin resistance. Result(s): The PCOS women had significantly higher fasting, 1-h, and 2-h insulin levels and a significantly lower insulin sensitivity index compared with FtMs before and after their T treatment. There were higher triglyceride levels and lower high-density lipoprotein cholesterol levels upon T treatment in FtMs compared with the PCOS women. Women with PCOS had higher body mass index (BMI) values. Positive correlations between insulin resistance indices and BMI were found only in women with PCOS. Conclusion(s): Testosterone administration by itself showed little detrimental influence on insulin resistance indices, but it had significant effects on lipid profiles. Compared with T, BMI had a greater impact on insulin resistance in women with PCOS. (Fertil Steril Ò 2010;94: Ó2010 by American Society for Reproductive Medicine.) Key Words: Insulin resistance, PCOS, female-to-male transsexuals, testosterone, lipid profiles Polycystic ovary syndrome (PCOS) is the most frequent androgen excess disorder in women and has been defined by the Androgen Excess Society (AES) as a predominantly hyperandrogenic syndrome (1). Women with hyperandrogenic syndrome are at risk for chronic anovulation, infertility, and early pregnancy loss. In addition, many are obese, and the number of obese individuals among patients with hyperandrogenic syndrome is increasing (2). They suffer from insulin resistance (IR) and other features of metabolic syndrome, which increase the risk for type 2 diabetes mellitus and cardiovascular disease (3, 4). In nonobese women with PCOS, altered insulin sensitivity is also found along with hyperandrogenemia (5). Even Received December 22, 2009; revised March 7, 2010; accepted March 15, 2010; published online May 7, S.C. has nothing to disclose. E.G. has nothing to disclose. L.G. has nothing to disclose. D.K. has nothing to disclose. P.O. has nothing to disclose. M.B. has nothing to disclose. R.D. has nothing to disclose. A.M. has nothing to disclose. Reprint requests: Andreas Mueller, M.D., PhD., Department of Obstetrics and Gynecology, Erlangen University Hospital, Universitaetsstrasse 21 23, D Erlangen, Germany (FAX: þ ; andreas.mueller@uk-erlangen.de). when no biochemical evidence of IR is found, abnormal vascular function has been reported (6). Hyperandrogenic syndrome may show more IR than would be expected on the basis of age and body mass index (BMI) (7). The mechanisms underlying this phenomenon are not fully understood. Although hyperandrogenemia may play a role (8, 9), other factors may also be involved (10). Many studies of treatment with metformin in PCOS patients have reported a combined effect on insulin parameters as well as androgenic parameters, indicating a causative role for IR in the development of hyperandrogenemia (11 13). Conversely, studies of treatment with oral contraceptives, known to lower androgenic parameters in PCOS patients, have failed to show a consistent effect on insulin parameters (14). Testosterone (T) is a risk factor for dyslipidemia (15, 16). Some studies have not shown any correlation with hyperandrogenemia in PCOS patients (17), whereas others have reported a correlation with hyperinsulinemia (18). There is evidence that obesity, rather than other aspects of PCOS, is responsible for the deterioration of insulin sensitivity (19). However, there is also evidence that triglycerides (TG) correlate positively and that high-density lipoprotein (HDL) cholesterol correlates /$36.00 Fertility and Sterility â Vol. 94, No. 7, December doi: /j.fertnstert Copyright ª2010 American Society for Reproductive Medicine, Published by Elsevier Inc.

2 TABLE 1 Comparison of age, body mass index (BMI), and endocrine variables in female-to-male transsexuals (FtMs) before and after 1 year of testosterone (T) treatment, compared with the data for women with polycystic ovary syndrome (PCOS). FtMs before (n [ 29) FtMs after (n [ 29) PCOS women (n [ 240) P value* Age (y) 29.9 ( ) 30.7 ( ) 28.0 ( ).06 Range BMI (kg/m 2 ) 23.5 ( ) a 24.2 ( ) a 27.9 ( ) b.001 Range LH (IU/L) 5.9 ( ) 4.7 ( ) a 6.7 ( ) b.04 FSH (IU/L) 5.9 ( ) a 3.1 ( ) b 5.8 ( ) a <.001 PRL (ng/ml) 9.3 ( ) 8.6 ( ) 8.4 ( ).53 E 2 (pmol/l) 136 ( ) 116 (95 141) 143 ( ).14 Total T (nmol/l) 1.8 ( ) a 20.4 ( ) b 2.5 ( ) c <.001 DHEAS (pmol/l) 5.8 (4.8 7) 5.4 ( ) 6.1 ( ).44 SHBG (nmol/l) 45.1 ( ) a 22.7 ( ) b 34.0 ( ) c <.001 Note: Data are presented as mean (95% confidence interval) for age or back-transformed geometric mean (95% confidence interval) for other values. * Analysis of variance for difference between the three groups. a,b,c Values in the same row with different letters are significantly different (P <.05; post hoc Sidak test). negatively with free Tand with the free androgen index in adolescent girls with PCOS (16). To assess the causative effect of hyperandrogenemia on insulin and lipid parameters, the present study investigated female-tomale transsexuals (FtMs) who were receiving long-term T treatment. The first aim of the study was to analyze the impact of T treatment on IR indices and lipid parameters in FtMs receiving treatment with T. The second aim was to compare the endocrine and metabolic profiles in women with PCOS and FtMs before and after 1 year of T treatment. MATERIALS AND METHODS Patients Twenty-nine FtMs who had completed their first year of T treatment between 2006 and 2008 were studied. A total of 240 women with PCOS served as control group. The data of the PCOS women were collected from 2006 to 2008; all met the definition of PCOS as described by the AES (1). No FtMs suffering from PCOS were included in the study. They were treated with 1,000 mg testosterone undecanoate, following the recommended dosage schedule with injections every 12 weeks. Before receiving androgen therapy, all FtMs were examined prospectively regarding body weight, blood pressure, and metabolic and hormonal status. All FtMs and women with PCOS underwent a complete screening panel, including physical examination, weight and height measurement, and ultrasound examination of the ovaries, and their body mass index (BMI) was calculated. After 1 year of T treatment, the FtMs were evaluated again, and the analyzed parameters were compared with baseline values and with the corresponding parameters in women with PCOS who had not yet received any treatment. Evidently, serum T levels after T administration were much higher (>8-fold) than in women with PCOS. In addition, whereas FtMs maintained a normal BMI while receiving T, women with PCOS had a higher range of BMIs. Therefore, the FtMs were also matched to women with PCOS with BMIs and ages similar to the BMIs and ages encountered in FTMs. Hirsutism Hirsutism was scored in accordance with the modified Ferriman-Gallwey score, by which a score of R6 was classified as hirsutism (1, 20, 21). Assessment of ovulation status The interval between bleeding episodes was assessed. The major clinical signs oligomenorrhea or amenorrhea vary in duration but are generally unambiguous (22). Women with regular bleeding episodes between 26 and 35 days were categorized as eumenorrheic. Women with cycles >35 days were categorized as oligomenorrheic. In these women, serum was obtained between days 3 and 5 of the cycle. In women with amenorrhea within the preceding 6 months, blood was sampled randomly (23 25). Polycystic ovaries Polycystic ovaries were diagnosed on ultrasound when at least one ovary had a volume of >10 ml or there were R12 follicles measuring 2 9 mm in diameter (26). The study was approved by the local Institutional Review Board. Each of the subjects provided written informed consent and completed a standard medical history questionnaire with an emphasis on menstrual dating and regularity, hirsutism, acne, gynecologic history, history of infertility, medications, and family history. Exclusion of Related Disorders The exclusion criteria were: 21-hydroxylase deficient nonclassic adrenal hyperplasia, hyperandrogenic insulin resistance acanthosis nigricans syndrome, an androgen-secreting neoplasm, Cushing syndrome, or any history of manifest thyroid dysfunction, surgery, or thyroid hormone medication. Women receiving hormonal therapy, including oral contraceptives or steroid medications, in the preceding 6 months were also excluded (27). Briefly, for evaluation of an androgen-secreting neoplasm, the total T cut-off value used was >7.0 nmol/l. To exclude 21-hydroxylase deficiency in patients with a 17-hydroxyprogesterone (17-OHP) level above 6.0 nmol/l, 17-OHP levels stimulated by ACTH were measured (25, 28). Procedures and Definitions Calculation of IR Patients followed no diet. After an overnight fast, an oral glucose (75 g) tolerance test (OGTT) was carried out, with glucose (mg/dl) and insulin (miu/ml) measured at 0, 60, and 120 minutes. Assessment of IR followed (29, 30) the homeostatic model assessment of insulin resistance (HOMA-IR), the homeostatic model for assessment of B-cell function (HOMA-B) (31), and the insulin sensitivity index (ISI) (32, 33). Biochemical measurements All hormone assays and total cholesterol, low-density lipoprotein (LDL) cholesterol, HDL cholesterol, and TG (measured after an overnight fasting period of 12 h) were measured in a routine diagnostic endocrine laboratory. All samples were obtained between 8:00 and 10:00 a.m Cupisti et al. Insulin resistance and testosterone in transsexuals and women with PCOS Vol. 94, No. 7, December 2010

3 TABLE 2 Comparison of metabolic variables in female-to-male transsexuals before and after 1 year of testosterone treatment in comparison with the data for women with polycystic ovary syndrome (PCOS). FtMs before (n [ 29) FtMs after (n [ 29) PCOS women a P value b P value c Total cholesterol (mg/dl) Unadjusted 184 ( ) 190 ( ) 195 ( ) Adjusted 183 ( ) 188 ( ) 196 ( ) LDL cholesterol (mg/dl) Unadjusted 116 ( ) 130 ( ) 121 ( ) Adjusted 117 ( ) 129 ( ) 121 ( ) HDL cholesterol (mg/dl) Unadjusted 54 (49 60) 43 (39 48) 54 (51 57).91 <.001 Adjusted 51 (46 55) 41 (37 45) 56 (53 59).06 <.001 Triglycerides (mg/dl) Unadjusted 129 ( ) 150 ( ) 122 ( ) Adjusted 144 ( ) 164 ( ) 117 ( ) Fasting glucose (mg/dl) Unadjusted 83 (80 86) 83 (80 87) 85 (84 86) Adjusted 84 (81 87) 84 (81 87) 85 (84 86) Glucose after 1 h (mg/dl) Unadjusted 113 ( ) 121 ( ) 117 ( ) Adjusted 121 ( ) 127 ( ) 116 ( ) Glucose after 2 h (mg/dl) Unadjusted 92 (83 103) 98 (88 109) 100 (97 104) Adjusted 98 (89 109) 103 (94 114) 99 (96 102) Fasting insulin (mu/ml) Unadjusted 5.4 ( ) 6.5 ( ) 9.2 (8.4 10) < Adjusted 7.0 ( ) 8.3 ( ) 8.7 ( ) Insulin after 1 h (mu/ml) Unadjusted 40.9 ( ) 43.1 ( ) 75.3 ( ) <.001 <.001 Adjusted 51.4 ( ) 53.2 ( ) 71.5 ( ) Insulin after 2 h (mu/ml) Unadjusted 29.8 ( ) 31 ( ) 57.1 ( ) <.001 <.001 Adjusted 37.6 ( ) 38.6 ( ) 54.1 ( ) HOMA-IR Unadjusted 1.17 ( ) 1.40 ( ) 2.08 ( ) < Adjusted 1.58 ( ) 1.81 ( ) 1.96 ( ) HOMA-B Unadjusted 99 (77 126) 119 (93 152) 154 ( ) Adjusted 121 (97 151) 146 ( ) 147 ( ) ISI Unadjusted 9.9 ( ) 9.8 ( ) 5.4 (4.9 6) <.001 <.001 Adjusted 7.7 ( ) 7.9 ( ) 5.7 ( ) Note: Data are presented as estimated marginal mean of back-transformed values (95% confidence interval). HDL ¼ high-density lipoprotein; HOMA-B ¼ homeostatic model for assessment of b-cell function; HOMA-IR ¼ homeostatic model for assessment of insulin resistance; ISI ¼ insulin sensitivity index; LDL ¼ low-density lipoprotein; other abbreviations as in Table 1. a Lipid measurements: n ¼ 116; Glucose and insulin measurements: n ¼ 240, b Comparison between FtMs before treatment and PCOS women by analysis of (co)variance, unadjusted and adjusted for age and BMI. c Comparison between by FtMs after 1 year of treatment and PCOS women by analysis of (co)variance, unadjusted and adjusted for age and BMI. Total testosterone (TT), DHEAS, SHBG, E 2, PRL, LH, FSH, and insulin were measured with chemiluminescent enzyme immunoassays (Immulite 2000; Siemens Medical Solutions Diagnostics, Bad Nauheim, Germany), as described in detail previously (23 25, 28, 29, 30, 34, 35). Plasma glucose was measured with the glucose oxidase method, (Immulite 2000; Siemens Medical Solutions Diagnostics). Statistical Analysis Most variables, except age, had positively skewed distributions and were logarithmically transformed to approach normalized data. To compare variables, we used analysis of variance or analysis of covariance and presented backtransformed geometric mean values that were unadjusted or adjusted for age and BMI. Intercorrelations were analyzed with the Pearson correlation coefficient. In a sensitivity analysis, we repeated the analyses in 29 FtMs compared with 29 age- and BMI-matched PCOS subjects. Two-tailed P values of <.05 were considered to be statistically significant. The software used was SPSS 17.0 (SPSS, Chicago, IL). RESULTS Changes in the endocrine and metabolic parameters are presented in Tables 1 and 2. The values were adjusted for age and BMI. Fasting insulin, HOMA-IR, and HOMA-B were significantly higher and ISI lower in the PCOS women compared with FtMs Fertility and Sterility â 2649

4 2650 Cupisti et al. Insulin resistance and testosterone in transsexuals and women with PCOS Vol. 94, No. 7, December 2010 TABLE 3 Intercorrelations between age, BMI, and metabolic variables in FtMs before and after 1 year of T treatment and the data for women with PCOS. TT Age BMI TC LDL HDL TG HOMA-IR HOMA-B FtMs before Age 0.38 (.04) BMI 0.12 (.54) 0.31 (.10) TC 0.61 (<.001) 0.27 (.15) 0.09 (.65) LDL 0.56 (.002) 0.55 (.002) 0.03 (.87) 0.66 (<.001) HDL 0.13 (.51) 0.07 (.71) 0.21 (.27) 0.03 (.87) 0.02 (.92) TG 0.08 (.67) 0.33 (.08) 0.11 (.56) 0.17 (.39) 0.57 (.001) 0.18 (.37) HOMA-IR 0.23 (.24) 0.03 (.90) 0.15 (.43) 0.08 (.69) 0.26 (.18) 0.08 (.67) 0.03 (.89) HOMA-B 0.01 (.95) 0.04 (.85) 0.10 (.62) 0.15 (.45) 0.15 (.43) 0.14 (.48) 0.08 (.70) 0.71 (<.001) ISI 0.21 (.28) 0.03 (.89) 0.22 (.25) 0.10 (.62) 0.32 (.09) 0.14 (.48) 0.13 (.52) 0.89 (<.001) 0.66 (<.001) FtMs after Age 0.10 (.61) BMI 0.06 (.75) 0.33 (.08) TC 0.18 (.36) 0.37 (.049) 0.23 (.24) LDL 0.04 (.849) 0.46 (.01) 0.29 (.13) 0.75 (<.001) HDL 0.33 (.08) 0.14 (.48) 0.33 (.08) 0.13 (.52) 0.40 (.03) TG 0.03 (.87) 0.07 (.74) 0.29 (.12) 0.51 (.005) 0.50 (.006) 0.40 (.03) HOMA-IR 0.21 (.27) 0.01 (.96) 0.15 (.45) 0.09 (.66) 0.17 (.38) 0.00 (.99) 0.18 (.35) HOMA-B 0.27 (.17) 0.21 (.29) 0.01 (.97) 0.09 (.63) 0.31 (.11) 0.14 (.46) 0.02 (.91) 0.58 (.001) ISI 0.27 (.16) 0.04 (.84) 0.09 (.65) 0.12 (.55) 0.01 (.98) 0.17 (.39) 0.41 (.03) 0.86 (<.001) 0.56 (.001) PCOS women Age 0.20 (.002) BMI 0.13 (.04) 0.06 (.33) TC 0.02 (.81) 0.22 (.02) 0.03 (.71) LDL 0.00 (1.00) 0.15 (.11) 0.16 (.09) 0.84 (<.001) HDL 0.08 (.37) 0.20 (.03) 0.47 (<.001) 0.12 (.21) 0.19 (.04) TG 0.08 (.39) 0.04 (.65) 0.43 (<.001) 0.23 (.01) 0.29 (.002) 0.39 (<.001) HOMA-IR 0.24 (<.001) 0.07 (.26) 0.68 (<.001) 0.01 (.88) 0.16 (.09) 0.37 (<.001) 0.42 (<.001) HOMA-B 0.14 (.04) 0.23 (<.001) 0.51 (<.001) 0.05 (.57) 0.07 (.47) 0.31 (.001) 0.31 (.001) 0.72 (<.001) ISI 0.19 (.003) 0.07 (.30) 0.67 (<.001) 0.02 (.86) 0.14 (.14) 0.35 (<.001) 0.46 (<.001) 0.91 (<.001) 0.67 (<.001) Note: Pearson correlation coefficients are shown on transformed data (with P values in parentheses). TC ¼ total cholesterol; TG ¼ triglycerides; TT ¼ total testosterone; other abbreviations as in Tables 1 and 2.

5 TABLE 4 Comparison of metabolic variables in 29 FtMs before and after 1 year of treatment and in 29 age- and BMI-matched women with PCOS. FtMs before FtMs after PCOS women P value a P value b Age 29.9 ( ) 30.7 ( ) 30.1 ( ) BMI 23.7 ( ) 24.2 ( ) 23.6 ( ) TC (mg/dl) 184 ( ) 190 ( ) 196 ( ) LDL cholesterol (mg/dl) 116 ( ) 130 ( ) 121 ( ) HDL cholesterol (mg/dl) 54 (50-59) 43 (40-47) 65 (58-73).02 <.001 TG (mg/dl) 129 ( ) 150 ( ) 101 (76-133) Fasting glucose (mg/dl) 83 (80-85) 83 (81-86) 86 (84-89) Glucose after 1 h (mg/dl) 113 ( ) 121 ( ) 113 ( ) Glucose after 2 h (mg/dl) 92 (84-102) 98 (89-108) 97 (88-107) Fasting insulin (mu/ml) 5.4 ( ) 6.5 ( ) 6.6 ( ) Insulin after 1 h (mu/ml) 4.9 ( ) 43.1 ( ) 58.4 ( ) Insulin after 2 h (mu/ml) 29.8 ( ) 31.0 ( ) 42.2 ( ) HOMA-IR 1.17 ( ) 1.40 ( ) 1.50 ( ) HOMA-B 99 (79-123) 119 (96-148) 104 (84-129) ISI 9.9 ( ) 9.8 ( ) 7.3 ( ) Note: Data are estimated marginal means of back-transformed values (with 95% confidence intervals). Abbreviations as in Tables 1 3. a Comparison between FtMs before treatment and PCOS women by analysis of variance. b Comparison between by FtMs after 1 year of treatment and PCOS women by analysis of variance. before and after T treatment, but these differences were attenuated after adjustment for BMI and age. Before T administration, TG and HDL cholesterol levels were not different between PCOS women and FtMs, but after 1 year of T treatment, TG were higher and HDL cholesterol levels lower in FtMs than in PCOS women. Further adjustment for age and BMI did not affect these differences in lipid levels. The PCOS women showed significantly higher BMI values compared with the FtMs before and after their T treatment (Tables 1 and 2). FSH and LH levels were significantly higher in women with PCOS than in FtMs after T treatment. SHBG levels were significantly lower in PCOS women than in FtMs before T treatment, but were lower (not significantly) after T treatment in FtMs. Testosterone levels were higher in PCOS women than in FtMs before T treatment. No differences were found regarding age, PRL, E 2, or DHEAS levels. Intercorrelation Analysis in the Two Groups The intercorrelations between T, age, BMI, and different metabolic variables in FtMs before and after T treatment compared with PCOS women are presented in Table 3. In the PCOS women, TT showed only weak correlations with all other parameters. In FtMs at baseline and after T treatment, these relationships were not significant, whereas the relationships reached a level of significance in PCOS women. There were no significant correlations with age. In the PCOS women, BMI showed the most strongly significant correlations with IR parameters (i.e., HOMA- IR, HOMA-B, and ISI). In a sensitivity analysis to substantiate this finding, the FtMs were also compared with women with PCOS matched to them for BMI and age (Table 4). This extra analysis confirmed that T is a stronger determinant of dyslipidemia but not of IR, because in PCOS women, markers of IR were no longer statistically significantly different from those found in FtMs when the influence of BMI and age was accounted for. In contrast, neither TT, age, nor BMI showed significant correlations with IR parameters in FtMs either before or after 1 year of T treatment. In fact, there were no marked differences in the strength and direction of the correlations between TT and IR parameters between the groups. There was a close correlation between total cholesterol and LDL cholesterol in all groups, and between TG and total cholesterol or LDL cholesterol in the FtMs, as was expected. In addition, the close correlation between values for HOMA-IR, HOMA-B, and ISI in the groups can be explained on the basis of the calculation method used for these parameters. DISCUSSION This study investigated the role of Ton IR and lipids in FtMs before and after their T treatment and compared these parameters with women with PCOS. The prevalence of PCOS is not significantly increased in FtMs (36). Neither in women with PCOS nor in FtMs before or after T treatment was there any close correlation between T levels and IR parameters. If T indeed is causal in the development of IR in women, then IR in FtMs receiving high-dose T treatment should worsen, which was not found in this study. Instead, IR parameters were more strongly associated with obesity than attributable to T levels, as was also apparent from the sensitivity analysis of women with PCOS comparing FtMs with PCOS women of similar BMI and age. It has been reported that T administration worsens cardiovascular risk in FtMs, though it was not confirmed in another study (40, 41). Testosterone administration reduced HDL cholesterol and increased TG, although LDL cholesterol and insulin sensitivity remained unaffected (25, 37, 38). Low HDL cholesterol values are also common in women with PCOS (16, 29, 30). Additionally, T induced a modest gain in weight in FtMs (25, 38), partially explained by an increase in lean body and muscle mass. Adiponectin, however, was significantly reduced (39), which is considered to increase cardiovascular risk (40). In studies with clinical end points, no increases in morbidity or mortality were observed in transsexuals receiving cross-sex hormones (42, 43). In the present study, the lipid profiles in FtMs after T administration resembled the significant decrease in HDL cholesterol and (nonsignificant) increase in TG in women with PCOS, possibly Fertility and Sterility â 2651

6 attributable to T, although T levels were much higher in T-treated FtMs than in PCOS women, indicating a possible limitation of this study. Contradicting this assumption of the role of T is the lack of correlation between IR indices and free T, on the one hand, and the lipid profile in women with PCOS on the other (44). Furthermore, administration of oral contraceptives to PCOS women decreases T levels, with no change of fasting insulin and IR indices (45, 46). The present findings strongly support that hyperandrogenemia may not cause IR in women. In this study, even after adjustment for BMI, ISI values were significantly lower in PCOS women than in FtMs, reflecting a greater disturbance in IR in women with PCOS. Interestingly, ISI is the only IR parameter that has a dynamic impact, influenced both by fasting insulin and by 1-h and 2-h insulin values after an oral glucose load. Moreover, after intercorrelation analysis, no close correlation was found between TT and IR parameters in FtMs or in PCOS women.. Only BMI correlated significantly with IR parameters in the PCOS group, with no correlation in the FtMs. In our earlier analysis of 184 women with PCOS, only BMI of R25 kg/m 2 and amenorrhea were associated with severe endocrine and metabolic abnormalities (30). In contrast, hirsutism, oligomenorrhea, and polycystic ovaries on ultrasound were not associated with IR in PCOS women (30). Other factors may have an additional effect on IR in women (47, 48). In the present study, higher T levels were not associated with IR, unless the confounding effects of higher BMIs were not accounted for. We conclude that BMI is a more important determinant of IR in PCOS women than high T levels. Testosterone predominantly had a deleterious effect on lipid profiles, which confirms findings from other studies in PCOS women (16, 49, 50 53). In conclusion, the present study comparing T-treated FtMs and women with PCOS found that hyperandrogenemia is not an important causative factor in the development of IR in women. However, significant deleterious effects of increasing T on HDL cholesterol and TG levels were evident. REFERENCES 1. Azziz R, Carmina E, Dewailly D, Diamanti- Kandarakis E, Escobar-Morreale HF, Futterweit W, et al. Position statement: criteria for defining polycystic ovary syndrome as a predominantly hyperandrogenic syndrome: an Androgen Excess Society guideline. J Clin Endocrinol Metab 2006;91: Yildiz BO, Knochenhauer ES, Azziz R. Impact of obesity on the risk for polycystic ovary syndrome. J Clin Endocrinol Metab 2008;93: Dunaif A. Insulin resistance and the polycystic ovary syndrome: mechanism and implications for pathogenesis. Endocr Rev 1997;18: Azziz R, Sanchez LA, Knochenhauer ES, Moran C, Lazenby J, Stephens KC, et al. Androgen excess in women: experience with over 1000 consecutive patients. J Clin Endocrinol Metab 2004;89: Nardo LG, Yates AP, Roberts SA, Pemberton P, Laing I. The relationships between AMH, androgens, insulin resistance and basal ovarian follicular status in nonobese subfertile women with and without polycystic ovary syndrome. Hum Reprod 2009;24: Cussons AJ, Watts GF, Stuckey BG. Dissociation of endothelial function and arterial stiffness in nonobese women with PCOS. Clin Endocrinol (Oxf) 2009;71: Barber TM, McCarthy MI, Wass JA, Franks S. Obesity and polycystic ovary syndrome. Clin Endocrinol (Oxf) 2006;65: Moghetti P, Tosi F, Castello R, Magnani CM, Negri C, Brun E, et al. The insulin resistance in women with hyperandrogenism is partially reversed by antiandrogen treatment: evidence that androgens impair insulin action in women. J Clin Endocrinol Metab 1996;81: M ohlig M, J urgens A, Spranger J, Hoffmann K, Weickert MO, Schl osser HW, et al. The androgen receptor CAG repeat modifies the impact of testosterone on insulin resistance in women with polycystic ovary syndrome. Eur J Endocrinol 2006;155: Janssen OE, Mehlmauer N, Hahn S, Offner AH, G artner R. High prevalence of autoimmune thyroiditis in patients with polycystic ovary syndrome. Eur J Endocrinol 2004;150: Krstevska B, Dimitrovski CH, Pemovska G, Misevska S, Dimova Z, Simeonova S, et al. Metformin improves menstrual patterns, endocrine and metabolic profile in obese hyperinsulinemic women with a polycystic ovary syndrome. Prilozi 2006;27: Palomba S, Falbo A, Zullo F, Orio F Jr. Evidencebased and potential benefits of metformin in the polycystic ovary syndrome: a comprehensive review. Endocr Rev 2009;30: Oppelt PG, Mueller A, Jentsch K, Kronawitter D, Reissmann C, Dittrich R, et al. The effect of metformin treatment for 2 years without caloric restriction on endocrine and metabolic parameters in women with polycystic ovary syndrome. Exp Clin Endocrinol Diabetes DOI: / s Orbetsova M, Kamenov Z, Kolarov G, Zakharieva S, Khristov V, Atanasova I, et al. [Effect of 6-month treatment with oral antiandrogen alone and in combination with insulin sensitizers on body composition, hormonal and metabolic parameters in women with polycystic ovary syndrome (PCOS) in order to determine therapeutic strategy]. Akush Ginekol (Sofiia) 2006;45: Saad F, Gooren L. The role of testosterone in the metabolic syndrome: a review. J Steroid Biochem Molec Biol 2009;114: Fruzzetti F, Perini D, Lazzarini V, Parrini D, Genazzani AR. Adolescent girls with polycystic ovary syndrome showing different phenotypes have a different metabolic profile associated with increasing androgen levels. Fertil Steril 2009;92: Rizzo M, Berneis K, Hersberger M, Pepe I, Di Fede G, Rini GB, et al. Milder forms of atherogenic dyslipidemia in ovulatory versus anovulatory polycystic ovary syndrome phenotype. Hum Reprod 2009;24: Kalra A, Nair S, Rai L. Association of obesity and insulin resistance with dyslipidemia in Indian women with polycystic ovarian syndrome. Indian J Med Sci 2006;60: Rossi B, Sukalich S, Droz J, Griffin A, Cook S, Blumkin A, et al. Prevalence of metabolic syndrome and related characteristics in obese adolescents with and without polycystic ovary syndrome. J Clin Endocrinol Metab 2008;93: Rotterdam ESHRE/ASRM-Sponsored PCOS Consensus Workshop Group. Revised 2003 consensus on diagnostic criteria and long-term health risks related to polycystic ovary syndrome. Fertil Steril 2003;81: Archer JS, Chang RJ. Hirsutism and acne in polycystic ovary syndrome. Best Pract Res Clin Obstet Gynaecol 2004;18: Souter I, Sanchez LA, Perez M, Bartolucci AA, Azziz R. The prevalence of androgen excess among patients with minimal unwanted hair growth. Am J Obstet Gynecol 2004;191: Norman RJ, Dewailly D, Legro RS, Hickey TE. Polycystic ovary syndrome. Lancet 2007;370: Mueller A, Dittrich R, Binder H, Hoffmann I, Beckmann MW, Cupisti S. Evaluation of recommended markers for assessing hyperandrogenemia in women classified as having polycystic ovary syndrome (PCOS) according to the revised 2003 diagnostic criteria. Eur J Med Res 2006;11: Mueller A, Cupisti S, Binder H, Hoffmann I, Beckmann MW, Dittrich R. The role of albumin in the calculation of free and bioavailable testosterone in women with hyperandrogenemia. In Vivo 2006;20: Mueller A, Cupisti S, Binder H, Hoffmann I, Kiesewetter F, Beckmann MW, et al. Endocrinological markers for assessment of hyperandrogenemia in hirsute women. Horm Res 2007;67: Yilmaz M, Bukan N, Ersoy R, Karakocx A, Yetkin I, Ayvaz G, et al. Glucose intolerance, insulin resistance and cardiovascular risk factors in first degree relatives of women with polycystic ovary syndrome. Hum Reprod 2005;20: Mueller A, Dittrich A, Cupisti S, Beckmann MW, Binder H. Is it necessary to measure free testosterone to assess hyperandrogenemia in women? The role of calculated free and bioavailable testosterone. Exp Clin Endocrinol Diabetes 2006;114: Kajaia N, Binder H, Dittrich R, Oppelt PG, Flor B, Cupisti S, et al. Low sex hormone binding globulin as a predictive marker for insulin resistance in women with hyperandrogenic syndrome. Eur J Endocrinol 2007;157: Cupisti S, Kajaia N, Dittrich R, Duezenli H, Beckmann MW, Mueller A. Body mass index and ovarian function are associated with endocrine and metabolic abnormalities in women with hyperandrogenic syndrome. Eur J Endocrinol 2008;158: Matthews DR, Hosker JP, Rudenski AS, Naylor BA, Treacher DF, Turner RC. Homeostasis model assessment: insulin resistance and beta-cell function from fasting plasma glucose and insulin concentrations in man. Diabetologia 1985;28: Matsuda M, DeFronzo RA. Insulin sensitivity indices obtained from oral glucose tolerance testing: comparison with the euglycemic insulin clamp. Diabetes Care 1999;22: Cupisti et al. Insulin resistance and testosterone in transsexuals and women with PCOS Vol. 94, No. 7, December 2010

7 33. Radikova Z, Koska J, Huckova M, Ksinantova L, Imrich R, Vigas M, et al. Insulin sensitivity indices: a proposal of cut-off points for simple identification of insulin-resistant subjects. Exp Clin Endocrinol Diabetes 2006;114: Cupisti S, Dittrich R, Binder H, Beckmann MW, Mueller A. Evaluation of biochemical hyperandrogenemia and body mass index in women presenting with amenorrhea. Exp Clin Endocrinol Diabetes 2007;115: Cupisti S, Dittrich R, Binder H, Kajaia N, Hoffmann I, Maltaris T, et al. Influence of body mass index on calculated androgen parameters in adult women with hirsutism and PCOS. Exp Clin Endocrinol Diabetes 2007;115: Mueller A, Gooren LJ, Naton-Sch otz S, Cupisti S, Beckmann MW, Dittrich R. Prevalence of polycystic ovary syndrome (PCOS) and hyperandrogenemia in female-to-male transsexuals. J Clin Endocrinol Metab 2008;93: Elbers JM, Giltay EJ, Teerlink T, Scheffer PG, Asscheman H, Seidell JC, et al. Effects of sex steroids on components of the insulin resistance syndrome in transsexual subjects. Clin Endocrinol (Oxf) 2003;58: Gooren LJ, Giltay EJ. Review of studies of androgen treatment of female-to-male transsexuals: effects and risks of administration of androgens to females. J Sex Med 2008;5: Berra M, Armillotta F, D Emidio L, Costantino A, Martorana G, Pelusi G, et al. Testosterone decreases adiponectin levels in female to male transsexuals. Asian J Androl 2006;8: Resmini E, Andraghetti G, Rebora A, Cordera R, Vera L, Giusti M, et al. Leptin, ghrelin, and adiponectin evaluation in transsexual subjects during hormonal treatments. J Androl 2008;29: Meriggiola MC, Armillotta F, Costantino A, Altieri P, Saad F, Kalhorn T, et al. Effects of testosterone undecanoate administered alone or in combination with letrozole or dutasteride in female to male transsexuals. J Sex Med 2008;5: Asscheman H, Gooren LJ, Eklund PL. Mortality and morbidity in transsexual patients with cross-gender hormone treatment. Metabolism 1989;38: Van Kesteren PJ, Asscheman H, Megens JA, Gooren LJ. Mortality and morbidity in transsexual subjects treated with cross-sex hormones. Clin Endocrinol (Oxf) 1997;47: Kuba VM,Cavalieri PM,Christoforo AC, Faria Junior R, Caetano R, Coeli CM, et al. [Insulin resistance and metabolic profile in lean and overweight/obese polycystic ovary syndrome patients]. Arq Bras Endocrinol Metabol 2006;50: Costello MF, Shrestha B, Eden J, Johnson NP, Sjoblom P. Metformin versus oral contraceptive pill in polycystic ovary syndrome: a Cochrane review. Hum Reprod 2007;22: Costello M, Shrestha B, Eden J, Sjoblom P, Johnson N. Insulin-sensitising drugs versus the combined oral contraceptive pill for hirsutism, acne and risk of diabetes, cardiovascular disease, and endometrial cancer in polycystic ovary syndrome. Cochrane Database Syst Rev 2007;24:CD Dittrich R, Kajaia N, Cupisti S, Hoffmann I, Beckmann MW, Mueller A. Association of thyroidstimulating hormone with insulin resistance and androgen parameters in women with PCOS. Reprod Biomed Online 2009;19: Mueller A, Sch ofl C, Dittrich R, Cupisti S, Oppelt PG, Schild RL, et al. Thyroid-stimulating hormone is associated with insulin resistance independently of body mass index and age in women with polycystic ovary syndrome. Hum Reprod 2009;24: Nisenblat V, Norman RJ. Androgens and polycystic ovary syndrome. Curr Opin Endocrinol Diabetes Obes 2009;16: Mueller A, Kiesewetter F, Binder H, Beckmann MW, Dittrich R. Long-term administration of testosterone undecanoate every 3 months for testosterone supplementation in female-to-male transsexuals. J Clin Endocrinol Metab 2007;92: Shroff R, Syrop CH, Davis W, Van Voorhis BJ, Dokras A. Risk of metabolic complications in the new PCOS phenotype based on the Rotterdam criteria. Fertil Steril 2007;88: Amato MC, Galluzzo A, Finocchiaro S, Criscimanna A, Giodano C. The evaluation of metabolic parameters and insulin sensitivity for a more robust diagnosis of the polycystic ovary syndrome. Clin Endocrinol 2008;69: Mueller A, Dittrich R, Binder H, K uhnel W, Maltaris T, Hoffmann I, et al. High dose estrogen treatment increases bone mineral density in maleto-female transsexuals receiving gonadotropinreleasing hormone agonist in the absence of testosterone. Eur J Endocrinol 2005;153: Fertility and Sterility â 2653