Serum müllerian-inhibiting substance levels in adolescent girls with normal menstrual cycles or with polycystic ovary syndrome

Transcription

1 Serum müllerian-inhibiting substance levels in adolescent girls with normal menstrual cycles or with polycystic ovary syndrome Yong Siow, Ph.D., a Sari Kives, M.D., c Paige Hertweck, M.D., b Sally Perlman, M.D., b and Mary E. Fallat, M.D. a a Department of Surgery and b Department of Obstetrics, Gynecology and Women s Health, University of Louisville, Louisville, Kentucky; and c Department of Obstetrics and Gynecology, University of Toronto, Toronto, Ontario, Canada Objective: To compare serum müllerian-inhibiting substance (MIS) concentrations in adolescent girls with polycystic ovary syndrome (PCOS) or normal menstrual cycles. Design: Prospective study. Setting: University department of obstetrics and gynecology. Patient(s): Thirty-one girls (12 18 years old) with PCOS and 17 girls (12 19 years old) with normal menstrual cycles. Intervention(s): Serum was collected from girls with PCOS or normal cycles during the early follicular phase of the menstrual cycle, stored frozen until assayed. Main Outcome Measure(s): Serum levels of MIS, E 2, free-t, androstenedione, LH, and FSH. Result(s): Serum MIS levels in girls with PCOS were significantly higher compared with normal girls ( [SD] and ng/ml, respectively). The subjects were stratified for body mass index (BMI) ( and 25 kg/m 2 ). Serum MIS levels in PCOS girls ( ng/ml [BMI 25 kg/m 2 ] and [BMI 25 kg/m 2 ]) were significantly higher compared with normal girls ( and ng/ml, respectively). Conclusion(s): Adolescent girls with PCOS have significantly higher serum MIS levels compared with normally cycling girls. Serum MIS levels in PCOS were not influenced by BMI. Increased MIS production may represent an early manifestation of the disease. (Fertil Steril 2005;84: by American Society for Reproductive Medicine.) Key Words: Adolescent PCOS girls, serum MIS Polycystic ovary syndrome (PCOS) is a syndrome of ovarian dysfunction with clinical manifestations including menstrual irregularities, signs of androgen excess, and obesity (1). In the United States, the prevalence of PCOS is between 5% and 8%, making this disorder among the most common affecting women of reproductive age (2, 3). No single diagnostic criterion is sufficient for diagnosis. Polycystic ovary syndrome can be diagnosed when two of the following three criteria are present: oligo- or anovulation, biochemical or clinical signs of hyperandrogenism, and polycystic ovaries (1). The clinical diagnosis of PCOS remains a diagnosis of exclusion, and known disorders that mimic the PCOS phenotype, such as congenital adrenal hyperplasia, androgen secreting tumors, and Cushing syndrome, should be ruled out first. It is widely accepted that the etiology of PCOS is multifactorial and this syndrome is a heterogenous condition with Received August 27, 2004; revised and accepted February 9, Supported by grants from Kosair Charities and NortonHealth Care Community Trust, Louisville, Kentucky. Presented at the 59th Annual Meeting of the American Society for Reproductive Medicine, San Antonio, Texas, October 11 15, Reprint requests: Yong Siow, Ph.D., Medical and Dental Research Building, Room 330, 511 South Floyd Street, Louisville, Kentucky (FAX: ; ysiow@louisville.edu). a spectrum of clinical and biochemical features. The spectrum encompasses anovulatory subjects without hirsutism or acne who present with elevated serum T levels, and, conversely, hirsute women with regular ovulatory menses (4). Anovulation and signs of androgen excess such as hirsutism and acne are common among adolescent females (5 8), and polycystic ovaries are often found in teenagers with menstrual disturbances and/or hirsutism (9). Similarities in the relationship between polycystic ovaries and oligomenorrhea, hyperandrogenism, and high LH levels have been documented in both adults and adolescents (7). Although it has been suggested that PCOS may start before or in early puberty, and evolves through adolescence into adulthood, detection of ovarian dysfunction is difficult until late adolescence (7, 10). Current studies support the view that oligomenorrhea in adolescents is not a stage in the physiologic maturation of the hypothalamic pituitary-ovarian axis but an early sign of PCOS (6). Müllerian-inhibiting substance (MIS) is a 140-kDa polypeptide hormone, produced by ovarian granulosa cells from about week 36 of gestation to menopause (11, 12). The role of MIS in ovarian function has not been defined. Our earlier studies have shown serum MIS fluctuations during the normal menstrual cycle, suggesting that MIS may have a 938 Fertility and Sterility Vol. 84, No. 4, October /05/$30.00 Copyright 2005 American Society for Reproductive Medicine, Published by Elsevier Inc. doi: /j.fertnstert

2 regulatory role in folliculogenesis (13). In previously reported animal studies, investigators have demonstrated specific and marked changes in expression of MIS during the estrus cycles in rats (14), MIS inhibition of meiotic maturation of rat oocytes (15), and MIS regulation of follicle development in mice (16). In women undergoing ovulation induction in preparation for an IVF procedure, an association between early follicular phase serum MIS and the number of retrieved oocytes was demonstrated, suggesting that serum MIS may be used as a clinical marker for assessing ovarian reserve (17). Poor response to IVF, which is indicative of diminished ovarian reserve, was associated with reduced baseline serum MIS concentrations (18). We have reported that follicular fluid and sera from women with PCOS undergoing IVF procedures contained higher concentrations of MIS with correspondingly lower E 2 levels (19). The higher follicular fluid MIS levels in PCOS correlated well with a greater number of immature oocytes, supporting the contention that MIS suppresses oocyte maturation (20). We speculate that MIS may be an endogenous inhibitor of E 2 production. Our subsequent studies in women with untreated PCOS similarly reported the presence of higher serum MIS levels in women with PCOS (21), and this has been confirmed by other investigators (22). The aim of this study was to investigate whether serum MIS levels are also elevated in anovulatory or oligomenorrheic adolescent girls who prove to have PCOS. MATERIALS AND METHODS Subjects We studied 31 adolescents with PCOS and 17 girls with normal menstrual cycles. The research protocol was approved by the Human Protection Program Office, and informed written consent and assent forms were obtained from all participants and parents. All subjects were recruited from patients attending a university gynecology practice for menstrual problems (PCOS subjects) or nonmenstrual-related reasons (regular cycles). The diagnosis of PCOS was based on the revised consensus on diagnostic criteria (1). All subjects were either anovulatory or had chronic oligomenorrhea ( 6 menses/yr), had evidence of clinical hyperandrogenism or had elevated levels of serum free-t, and/or were hirsute. All subjects were studied before initiation of any therapy for PCOS. In the adolescents with normal regular menstrual cycles, the average cycle length ranged between 25 to 30 days. None of the subjects in this group were hirsute. Transabdominal ultrasonography was carried out in 15 of the 17 girls with normal cycles and 27 of the girls with PCOS. The subjects were stratified for body mass index (BMI) into nonobese and obese groups. Obesity is defined as a BMI greater than the 85th percentile for age and sex based on data from the National Health and Nutrition Examination Survey (23). All obese subjects had BMI values 25 kg/m 2, whereas all nonobese subjects had BMI values 25 kg/m 2. The BMI values were calculated using the method available at the Centers for Disease Control website, htm. Hirsutism was graded using the Ferriman & Gallwey (FG) scoring system (24). A woman was considered hirsute if the score was 8 counted from hormone-sensitive areas (i.e., face, lower abdomen, anterior thighs, chest, breasts, and pubic area). Blood Sampling and Hormone Assays Blood samples from the PCOS subjects were obtained during a clinic visit. Because the subjects were either anovulatory or oligomenorrheic, the collection of blood samples were not timed with respect to ovulatory activity. Blood sampling in the normal cycling adolescent was timed with respect to the menstrual cycle and was collected in the early follicular phase (day 2 or 3 following menses) of the menstrual cycle as previously described (13). Samples were collected during the early follicular phase at day 2 or 3 following menses in normal women because this phase of follicular development is likely most similar to the stage of arrested folliculogenesis in PCOS women. In addition, mean circulating serum MIS levels during the early follicular phase, midcycle, and midluteal phases of the normal menstrual cycle were similar, although significant statistical differences in the trend in MIS production within individual subjects during the menstrual cycle were found (13). Serum MIS concentrations were measured by an ELISA (19). The measurement of MIS was carried out in 2 separate assays. Each assay included samples from normal cycling and PCOS subjects. To assess assay variability, MIS levels were measured repeatedly in three different serum samples. Serum samples from one normal cycling subject containing 1.8 ng/ml MIS and two PCOS subjects with 4.4 and 7.2 ng/ml MIS concentrations were used. The coefficients of variation (CV) for the intra-assay variability for the samples determined in replicates of six were 16.8%, 13.6%, and 6.3%, respectively. The CV for the interassay variability determined in four separate assays were 17.4%, 10.4%, and 9.6%, respectively. Serum levels of free-t, androstenedione (A 4 ), E 2, LH, and FSH were determined using an enzyme immunoassay (EIA; Diagnostic System Laboratories, Webster, TX). Serum samples were stored at 80 C until assayed for MIS and other hormones within six months of collection. Statistical Analysis Statistical analysis was by the unpaired Student s t test using the SPSS software version 8.0 (SPSS, Chicago, IL). A P value of.05 was considered significant. Fertility and Sterility 939

3 TABLE 1 Clinical characteristics of study population. Regular cycles PCOS P< All subjects No. of subjects Age (yr) (12 19) (12 18) NS Age of menarche (yr) (11 13) (10 15) NS BMI (kg/m 2 ) ( ) ( ).004 No. of girls hirsute 0 9 Stratified for BMI 25 kg/m 2 No. of subjects Age (yr) (13 19) (12 17) NS Age of menarche (yr) (11 13) (11 14) NS BMI (kg/m 2 ) ( ) ( ) NS No. of girls hirsute 0 1 Stratified for BMI 25 kg/m 2 No. of subjects 6 19 Age (yr) ( ) (12 18) NS Age of menarche (yr) (11 12) (10 15) NS BMI (kg/m 2 ) ( ) ( ).01 No. of girls hirsute 0 8 Note: Values are mean SD (data range). The girls who are hirsute had FG score of 8. Statistical comparison was by Student s t test. NS not significant. RESULTS The clinical characteristics of all subjects with or without stratification for BMI are shown in Table 1. The average ages of 17 subjects with regular cycles and 31 girls with PCOS were similar and ranged from 10 to 19 years. The age of menarche for PCOS girls was similar to that for girls with regular menstrual cycles, and was not different when the groups of girls were stratified for BMI. Of the girls with PCOS, 4 (13%) were anovulatory and most were oligomenorrheic. Nine of the 31 girls with PCOS (29%) were hirsute with a mean FG score of None of the girls with regular cycles were hirsute. As a group, the girls with PCOS were obese, with a mean ( SD) BMI of kg/m 2, ranging from 17 to 48 kg/m 2 ; 19 of the girls in this group (61.3%) had a BMI 25 kg/m 2. Eleven of the girls with regular cycles (64.7%) were not obese, with a mean BMI of kg/m 2, ranging from 17.8 to 35.4 kg/m 2 ;6ofthe girls in this group (35.3%) had a BMI 25 kg/m 2. Five of the girls with normal menstrual cycles had multiple follicles (3 16) in either the left or right ovary that were between 4 and 10 mm in size. Of the 31 PCOS girls, 21 girls had between 3 and 10 follicles that were 4 to 10 mm in size. The data were analyzed for correlation between serum MIS concentrations and number of follicles. The correlation coefficients (r 2 ) for MIS levels and follicle numbers obtained for girls with normal cycles and PCOS girls were and , respectively. Sera concentrations of T, A 4, and E 2 are shown in Figure 1. Levels of T and A 4 are significantly higher in adolescents with PCOS compared with normal cycling girls. Despite the increased production of androgens in adolescent PCOS girls, serum levels of E 2 were similar to those in normal cycling girls. No relationships between these hormones and MIS were found. When steroid data were stratified for BMI ( 25 kg/m 2 or 25 kg/m 2 ), serum levels of T and A 4 in PCOS girls were higher compared with normal cycling girls (Fig. 2). Within the PCOS group, the levels of T and A 4 were not different in girls with a lower or high BMI. Serum E 2 levels in PCOS girls and normal cycling girls were not different irrespective of BMI (Fig. 2). Serum levels of FSH and LH and LH/FSH ratios are shown in Figure 3. Serum LH levels were elevated in PCOS girls, as were LH/FSH ratios compared to girls with normal menstrual cycles. When serum LH levels were stratified for BMI, serum LH levels in PCOS girls with BMIs 25 kg/m 2 and 25 kg/m 2 ( and miu/ml, respectively) were significantly higher compared with the respective normal groups ( and miu/ml, P.004 and P.02, respectively). The LH/FSH ratios in PCOS girls with BMI 25 kg/m 2 and 25 kg/m 2 ( and , respectively) were also significantly higher than the respective normal groups ( and , P.001 and P.02, respectively). Serum FSH levels in 940 Siow et al. Serum MIS levels in adolescence PCOS Vol. 84, No. 4, October 2005

4 FIGURE 1 Serum concentrations of androgens and E 2 in all normal and PCOS girls. Values are mean SD for normal cycling girls (red bars) and PCOS girls (black bars). a P.01; b P.001. PCOS subjects, with and without stratification for BMI, were not different compared with normal cycling girls (data not shown). Serum levels of MIS in PCOS girls were significantly higher compared with levels in girls who had normal menstrual cycles (Fig. 4). When stratified for BMI values, serum MIS levels remained significantly elevated in PCOS girls, irrespective of BMI status. DISCUSSION The diagnostic criteria described recently for PCOS (1) were adopted in this study. All of the PCOS girls in this study had at least two of the three following abnormalities associated with the syndrome: oligomenorrhea or anovulation, androgen excess, and polycystic ovaries. The primary clinical indicator of androgen excess was the presence of hirsutism. Using a Ferriman-Gallwey score of 8 as an indicator of the presence of hirsutism, only 9 of the 31 girls in the PCOS group (29%) were hirsute. The incidence of hirsutism among the PCOS subjects in our study is relatively low. None of these girls have received treatment for the hirsute condition. Hirsutism is generally less prevalent in adolescence (25). Testosterone circulates almost entirely bound, primarily to sex hormone binding protein (SHBP) (26). Although total T measurements have been used to help screen for hirsutism, the measurement of T is highly variable and imprecise (27). The concentration of unbound T or free-t depends directly on overall T production and level of SHBP, and free-t measurements are more likely to correlate with hirsutism (28, 29). Free or unbound T is biologically active, and free-t levels measured directly should better correlate with its endocrine effects and have a role in the detection of androgen excess and confirmation of PCOS (30). We have used free-t measurements to assess androgen production in this study. In our study, serum levels of free-t in girls with PCOS were significantly higher compared with girls with normal cycles. This finding is consistent with other adolescent PCOS studies (6, 31, 32), as well as with our study involving newly diagnosed older PCOS women (21). The concentrations of free-t in our studies are relatively lower than the levels reported in studies involving adult women (33, 34), which may be due to differences in methodology. Serum T levels appear to change with age. In a study of hirsute women between 25 and 52 years old, serum androgens were shown to negatively correlate with age; serum levels were significantly lower in women older than 41 years compared with women aged 25 or less (25). Serum levels of T and A 4 in nonobese or obese PCOS girls are higher compared with respective normal girls. This suggests that factors causing overproduction of androgens and associated with anovulatory cycles are present in both nonobese and obese adolescent girls. Within the anovulatory group, the levels of T and A 4 in Fertility and Sterility 941

5 FIGURE 2 Serum T, A 4, and E 2 levels in girls stratified for BMI. Values are mean SD in normal cycling girls (red bars) and girls with PCOS (black bars) stratified for BMI 25 kg/m 2 or 25 kg/m 2. a P.04; b P.04; c P.005. Androgens are produced in the ovarian theca cells, and are then converted to E 2 in neighboring granulosa cells by the enzyme cytochrome P450 aromatase, also known as CYP19 (35). Despite the increased production of androgens in adolescent PCOS girls, serum levels of E 2 were similar to those in normal girls. This contrasts with our previous findings in older PCOS women, where serum E 2 levels were significantly lower compared with normal women (21). This difference in E 2 levels between older and younger females with an anovulatory condition is an unexpected finding and may be due to differences in the maturity of the gonadal steroid synthetic mechanism and/or the responses to other hormones that modulate ovarian activities. Serum levels of MIS in PCOS adolescent girls are significantly higher compared with levels in girls who have normal menstrual cycles. This elevation of serum MIS in PCOS girls is similar to what we have described in women with newly diagnosed PCOS before treatment (21). These findings suggest that the cause or causes of the abnormal increase in MIS production in women with PCOS are already present during adolescence. Other studies have reported that the characteristic endocrine abnormalities of PCOS, such as overproduction of androgens (T and A 4 ), become clinically manifest during adolescence (9, 32). The current study is the first to show an increased MIS production in adolescent girls with PCOS-like symptoms. No relationship between MIS and E 2 levels was found in the current study. This contrasts with our previous findings in older PCOS women, where higher serum MIS levels in older PCOS women were negatively correlated with E 2 levels (21), suggesting that E 2 production was suppressed by FIGURE 3 Serum concentrations of FSH and LH in all normal and PCOS girls. Values are mean SD and were not stratified for BMI for normal cycling girls (red bars) and PCOS girls (black bars). a P.002; b P nonobese girls were not different from obese girls, suggesting that androgen production is independent of obesity status. These findings are consistent with studies by other investigators (32). 942 Siow et al. Serum MIS levels in adolescence PCOS Vol. 84, No. 4, October 2005

6 FIGURE 4 Serum MIS levels in PCOS girls (black) and girls with normal (red) menstrual cycles for all girls (top) and for girls stratified for BMI (bottom). Horizontal lines represent mean MIS concentrations. a P.002; b P 05; c P.003. MIS. In animal studies, MIS has been shown to suppress transcription of the CYP19 gene (36). This inverse relationship between serum MIS and E 2 was not found in a recent study (22). These discrepancies might simply reflect differences in the timing of blood sampling. In our previous study (21), blood from PCOS women was collected randomly and not timed with respect to the ovulatory stage. In contrast, when blood was sampled only in the early follicular phase after either spontaneous or progestin-induced menstruation, as described in the study by Pigny and colleagues (22), the MIS-E 2 relationship was not found. We had proposed that the reduction in E 2 production in older PCOS women is caused by inhibition of CYP19 by MIS (21). The effects of MIS on ovarian function have not been elucidated. Serum MIS levels have been shown to decrease over time and with advancing age in normo-ovulatory women, and a strong correlation between MIS concentrations with the number of antral follicles has been described, suggesting that MIS may be a useful marker of ovarian follicle pool and ovarian aging (37, 38). No relationship between MIS levels and follicle numbers were found in adolescent girls in our present study, and we found no correlation between MIS and FSH levels in either normal cycling or PCOS girls. This is in contrast to other studies involving older women ages years that showed a negative correlation between MIS and FSH (17, 18, 22). One reason for this age difference may be related to the substantial endocrinologic changes that are occurring in adolescents with respect to hormonal equilibrium and maturation of the hypothalamic-pituitary-ovarian axis which typically occurs 2 to 5 years after menarche (39). The higher serum levels of LH in PCOS girls compared with normal girls in our studies confirmed other reported studies involving adolescent girls (6, 40) and older women (41). Similar to these studies, serum FSH levels in our studies were not different in PCOS subjects, whereas LH/ FSH ratios were higher in both PCOS women and PCOS adolescents compared with those who had normal menstrual cycles. The relationship between FSH, LH, and MIS is not well known, but FSH may behave as a negative regulator of MIS synthesis in human adult ovary. In adult rat ovaries, FSH has been shown to down-regulate MIS and MIS type II receptor (14). But in the FSH-deficient (FSH / ) transgenic mouse, prepubertal testicular MIS production is increased by FSH stimulation via enhancement of MIS gene transcription (42). These conflicting FSH regulatory effects on MIS may be related to differences in sex, age, and animal species. In summary, our study shows that serum MIS levels in anovulatory adolescent girls who prove to have PCOS are elevated compared with normally cycling girls. Serum MIS levels were found to be elevated in our earlier study of older anovulatory women who proved to have PCOS. An increase in MIS production appears to be an early manifestation of the disease. Acknowledgments: We thank Kendal Stephens, M.D., for her assistance in the recruitment of subjects for the study, and Shen Guo, M.S., for technical assistance. REFERENCES 1. The Rotterdam ESHRE/ASRM-Sponsored PCOS Workshop Group. Revised 2003 consensus on diagnostic and long-term health risks related to polycystic ovary syndrome. Fertil Steril 2004;81: Knochenhauer ES, Key TJ, Kahsar-Miller M, Waggoner W, Boots LR, Azziz R. Prevalence of the polycystic ovary syndrome in unselected black and white women of the southeastern United States: a prospective study. J Clin Endocrinol Metab 1998;83: Michaelmore KF, Balen AH, Dunger DB, Vessey MP. Polycystic ovaries and associated clinical and biochemical features in young women. Clin Endocrinol 1999;51: Carmina E, Lobo RA. Polycystic ovaries in hirsute women with normal menses. Am J Med 2001;111: Apter D. Endocrine and metabolic abnormalities in adolescents with Fertility and Sterility 943

7 PCOS-like condition: consequences for adult reproduction. Trends Endocrinol Metabol 1998;9: van Hooff MHA, Voorhorst FJ, Kaptein MBH, Hirasing RA, Koppenaal C, Schoemaker J. Endocrine features of polycystic ovary syndrome in a random population sample of year-old adolescents. Hum Reprod 1999;14: Rosenfield RL, Ghai K, Ehrmann DA, Barnes RB. Diagnosis of the polycystic ovary syndrome in adolescence: comparison of adolescent and adult hyperandrogenism. J Pediatr Endocrinol Metab 2000;13: Driscoll D. Polycystic ovary syndrome in adolescence. Ann N Y Acad Sci 2003;997: Franks S. Adult polycystic ovary syndrome starts in childhood. Best Prac Res Clin Endocrinol Metab 2002;16: Ibanez L, de Zegher F, Potau N. Anovulation after precocious pubarche: early markers and time course in adolescence. J Clin Endocrinol Metab 1999;84: Lee MM, Donahoe PK, Hasegawa T, Silverman B, Crist GB, Best S, et al. Müllerian-inhibiting substance in humans: normal levels from infancy to adulthood. J Clin Endocrinol Metab 1996;81: Raipert-De Meyts E, Jorgensen N, Graem N, Muller J, Cate RL, Skakkebaek NE. Expression of antimüllerian hormone during normal and pathological gonadal development: association with differentiation of Sertoli and granulosa cells. J Clin Endocrinol Metab 1999;84: Cook CL, Siow Y, Taylor S, Fallat ME. Serum müllerian-inhibiting substance levels during normal menstrual cycle. Fertil Steril 2000;73: Baarends WM, Uilenbroek JT, Kramer P, Hoogerbrugge JW, van Leeuwen EC, Themmen AP, et al. Antimüllerian hormone and antimüllerian hormone type II receptor messenger ribonucleic acid expression in rat ovaries during postnatal development, the estrous cycle, and gonadotropin-induced follicle growth. Endocrinology 1995;136: Takahashi M, Koide S, Donahoe PK. Müllerian-inhibiting substance as oocyte meiosis inhibitor. Mol Cell Endocrinol 1986;47: Durlinger ALL, Kramer P, Karels B, de Jong FH, Uilenbroek JT, Grootegoed JA, et al. Control of primordial follicle recruitment by antimüllerian hormone in the mouse ovary. Endocrinology 1999;140: Seifer DB, MacLaughlin DT, Christian BP, Feng B, Shelden RM. Early follicular serum müllerian-inhibiting substance levels are associated with ovarian response during assisted reproductive technology cycles. Fertil Steril 2002;77: van Rooij IAJ, Broekmans FJM, te Velde ER, Fauser BCJM, Bancsi LFJMM, de Jong FH, et al. Serum antimüllerian hormone levels: a novel measure of ovarian reserve. Hum Reprod 2002;17: Fallat ME, Siow Y, Marra M, Cook C, Carrillo A. Müllerian-inhibiting substance in follicular fluid and serum a comparison of patients with tubal factor infertility, polycystic ovarian syndrome and endometriosis. Fertil Steril 1997;67: Ueno S, Kuroda T, MacLaughlin DT, Ragin RC, Manganaro TF, Donahoe PK. Müllerian-inhibiting substance in the adult rat ovary during various stages of the estrous cycle. Endocrinology 1989;125: Cook CL, Siow Y, Brenner AG, Fallat ME. Relationship between serum müllerian-inhibiting substance and other reproductive hormones in untreated women with polycystic ovary syndrome and normal women. Fertil Steril 2002;77: Pigny P, Merlen E, Robert Y, Cortet-Rudellic C, Decanter C, Jonard S, et al. Elevated serum level of antimüllerian hormone in patients with polycystic ovary syndrome: relationship to the ovarian follicle excess and to the follicular arrest. J Clin Endocrinol Metab 2003;88: Must A, Dallal GE, Dietz WH. Reference data for obesity: 85th and 95th percentiles of body mass index (wt/ht2): a correction. Am J Clin Nutr 1991;54: Ferriman D, Gallwey JD. Clinical assessment of body hair growth in women. J Clin Endocrinol Metab 1961;21: Ruutiainen K, Erkkola R, Gronroos MA, Irjala K. Influence of body mass index and age on the grade of hair growth in hirsute women of reproductive ages. Fertil Steril 1988;50: Vermeulen A, Verdonck L, Kaufman JM. A critical evaluation of simple methods for the estimation of free testosterone in serum. J Clin Endocrinol Metab 1999;84: Boots LR, Potter S, Potter HD, Azziz R. Measurement of total serum testosterone levels using commercially available kits: high degree of between-kit variability. Fertil Steril 1998;69: Vermeulen A, Verdonck L, Kaufman JM. A critical evaluation of simple methods for the estimation of free testosterone in serum. J Clin Endocrinol Metab 1999;84: Diamanti-Kandarakis E, Koulie CR, Bergiele AT, Filandra FA, Tsianateli TC, Spina GG, et al. A survey of the polycystic ovary syndrome in the Greek island of Lesbos: a hormonal and metabolic profile. J Clin Endocrinol Metab 1999;84: Azziz R. The time has come to simplify the evaluation of the hirsute patient. Fertil Steril 2000;74: Moll GW Jr, Rosenfield RL. Plasma free testosterone in the diagnosis of adolescent polycystic ovary syndrome. J Pediatr 1983;102: Silfen ME, Denburg MR, Manibo AM, Lobo RA, Jaffe R, Ferin M, et al. Early endocrine, metabolic, and sonographic characteristics of polycystic ovary syndrome (PCOS): comparison between nonobese and obese adolescents. J Clin Endocrinol Metab 2003;88: Maciel GAR, Soares JM Jr, da Motta ELA, Haidar MA, de Lima GR, Baracat EC. Nonobese women with polycystic ovary syndrome respond better than obese women to treatment with metformin. Fertil Steril 2004;81: Huppert J, Chiodi M, Hillard PJA. Clinical and metabolic findings in adolescent females with hyperandrogenism. J Pediatr Adolesc Gynecol 2004;17: Strauss JF, Hsueh AJW. Ovarian hormone synthesis. In: Degroot LJ, Jameson JL, editors. Endocrinology. Vol. 3. Philadelphia: Saunders, 2001: Josso N, Racine C, di Clemente N, Rey R, Xavier F. The role of antimüllerian hormone in gonadal development. Mol Cell Endocrinol 1998;145: de Vet A, Laven JSE, de Jong FH, Themmen APN, Fauser BCJM. Antimüllerian hormone serum levels: a putative marker for ovarian aging. Fertil Steril 2002;77: Durlinger AL, Gruijters MJ, Kramer P, Karels B, Ingraham HA, Nachtigal MW, et al. Anti-müllerian hormone inhibits initiation of primordial follicle growth in the mouse ovary. Endocrinology 2002; 143: Apter D. Development of the hypothalamic-pituitary-ovarian axis. Ann N Y Acad Sci 1997;816: van Hooff MHA, Voorhorst FJ, Kaptein MBH, Hirasing RA, Koppenaal C, Schoemaker J. Polycystic ovaries in adolescents and the relationship with menstrual cycle patterns, luteinizing hormone, androgens, and insulin. Fertil Steril 2000;74: Laven JSE, Mullders AGMGJ, Visser JA, Themmen AP, de Jong FH, Fauser BCJM. Anti-müllerian hormone serum concentrations in normoovulatory and anovulatory women of reproductive age. J Clin Endocrinol Metab 2004;89: Lukas-Croisier C, Lasala C, Nicaud J, Bedecarras P, Kumar TR, Dutertre M, et al. Follicle-stimulating hormone increases testicular antimüllerian hormone (AMH) production through Sertoli cell proliferation and a nonclassical cyclic adenosine 5=-monophosphate mediated activation of the AMH gene. Mol Endocrinol 2003;17: Siow et al. Serum MIS levels in adolescence PCOS Vol. 84, No. 4, October 2005