Clinical and biochemical presentation of polycystic ovary syndrome in women between the ages of 20 and 40

Human Reproduction, Vol.26, No.12 pp. 3443 3449, 2011 Advanced Access publication on September 16, 2011 doi:10.1093/humrep/der302 ORIGINAL ARTICLE Reproductive endocrinology Clinical and biochemical presentation of polycystic ovary syndrome in women between the ages of 20 and 40 So-Jung Liang 1,, Chun-Sen Hsu 1,, Chii-Ruey Tzeng 2, Chi-Huang Chen 2, and Ming-I Hsu 1, * 1 Department of Obstetrics and Gynecology, Wan Fang Hospital, Taipei Medical University, Taiwan No.111, Sec. 3, Xinglong Road, Taipei 11696, Taiwan 2 Department of Obstetrics and Gynecology, Taipei Medical University Hospital, Taipei Medical University, Taipei, Taiwan *Correspondence address. Tel: +886-2-29307930; Fax: +886-2-29300036; E-mail: hsumingi@yahoo.com.tw Submitted on April 13, 2011; resubmitted on August 8, 2011; accepted on August 12, 2011 background: The clinical features and metabolic complications of polycystic ovary syndrome (PCOS) may change with age. This study was designed to investigate the clinical and biochemical characteristics of PCOS patients between the ages of 20 and 40. methods: The study included 781 Taiwanese women, of whom 453 were diagnosed with PCOS and 328 were non-pcos controls. Anthroponmetric components, androgens, endocrine, insulin resistance, and metabolic components were measured and correlated with age. Above parameters were compared between younger and elder women with PCOS. results: Age had significant negative correlations with androgens (total testosterone and dehydroepiandrosterone sulfate), the modified Ferriman Gallwey score and the prevalence of acne and hirsutism. Age had significant positive correlations with fasting glucose, cholesterol, triglycerides and low-density lipoprotein. The 453 women who fulfilled diagnostic criteria for PCOS were classified by age into two groups: Group A (20 29 years old, n ¼ 294) and Group B (30 40 years old, n ¼ 159). Group A had significantly higher total testosterone levels than Group B. Group B had higher fasting insulin and glucose levels, triglycerides, body mass index and waist measurements and a higher incidence of obesity than Group A. The average ovarian volume was not significantly different among the two groups. conclusions: Increased age is accompanied by a decrease in the prevalence of both clinical and biochemical hyperandrogenism in women. Hyperandrogenism is the important factor for young women with PCOS; however, abdominal obesity and certain metabolic disturbances became major concerns for older women with PCOS. Key words: polycystic ovary syndrome / age / hyperandrogenism / metabolic syndrome Introduction Polycystic ovary syndrome (PCOS) is the most common endocrine disorder of reproductive-age women, affecting an estimated 9 18% of all women in this age group, depending on the diagnostic criteria applied (March et al., 2010). PCOS is considered to be a polygenic trait, and clinical features of this disorder may change with age, beginning in adolescence and ending with menopause (Pasquali and Grambineri, 2006a). Investigating PCOS in women of various ages is crucial because it is now believed to increase metabolic and cardiovascular risk and these risks probably depend on age (Cheung et al., 2008). According to the 2003 Rotterdam criteria, a PCOS diagnosis can be reached when at least two of the following three criteria are met: hyperandrogenism and/or hyperandrogenemia (HA); oligomenorrhea or amenorrhea or anovulation (Oligo-An); and polycystic ovary morphology (PCOM; Rotterdam ESHRE/ASRM-Sponsored PCOS Consensus Workshop Group, 2004). However, the clinical and biochemical presentation of PCOS may change over time in women of reproductive age (Pasquali and Grambineri, 2006b). Oligomenorrhea may be an early symptom of PCOS, especially in overweight girls with hirsutism or acne (Mastorakos et al., 2006). Hyperandrogenism partially resolves before menopause in women with PCOS (Winters et al., 2000). Ovarian volume and the number of follicles decrease with age in women with PCOS (Alsamarai et al., 2009). Furthermore, aging may also be associated with a defect in insulin action (Rowe et al., 1983). Therefore, the clinical features and metabolic consequences of PCOS may change with age (Rodríguez-Morán and Guerrero-Romero, 2003), and these age-related changes may affect These authors contributed equally to this work. & The Author 2011. Published by Oxford University Press on behalf of the European Society of Human Reproduction and Embryology. All rights reserved. For Permissions, please email: journals.permissions@oup.com

3444 Liang et al. the observed incidence of PCOS (Bili et al., 2001). Unfortunately, there have been few reports examining the differences in the phenotype and clinical presentation of PCOS based on age (Pasquali and Grambineri, 2006a). The best way to study the aging process in women with PCOS is to perform a longitudinal study with the same subjects, but it is difficult to follow such patients for more than 20 years without introducing study bias. Therefore, we attempted to address the issue by performing a cross-sectional study to evaluate the clinical and biochemical characteristics of women, of different ages, who were diagnosed with PCOS. Materials and Methods This study was approved by the Institutional Review Board of the Wan Fang Medical Center at Taipei Medical University, Taipei, Taiwan. We retrospectively reviewed the medical records of female patients who visited the Reproductive Endocrinology Clinic at the Wan Fang Medical Center at Taipei Medical University from 1 April 2004 to 31 December 2010. The following subjects were excluded from the study: (i) women who had been diagnosed with other etiologies that should be excluded when establishing a PCOS diagnosis, such as hyperprolactinemia, hypogonadotropic hypogonadism, premature ovarian failure, congenital adrenal hyperplasia, androgen-secreting tumors, Cushing s syndrome, disorders of the uterus (e.g. Asherman s syndrome and Mullerian agenesis) and chromosomal anomalies (e.g. Turner s syndrome); (ii) women who were,20 or.40 years of age; (iii) women who experienced menarche,3 years prior to the start of the study or who had Day 3 FSH.15 miu/ml; (iv) women with inadequate clinical/biochemical records and (v) women who had had ovarian cysts or ovarian tumors in an ultrasonographic examination. After these exclusions, a total of 781 subjects were enrolled in the study. The subject s medical histories included a detailed menstrual history and data on body weight, height, hirsutism and the presence or the absence of clinical acne. Additionally, the levels of serum FSH, LH, prolactin, total testosterone, fasting insulin and fasting glucose were available. Waist girth was defined as the middle circumference between the iliac crest and the lateral costal margin. Hip circumference was defined as the largest measurement over the buttocks. The waist-to-hip ratio was defined as waist girth/hip girth. Body mass index (BMI) was defined as body weight in kilograms divided by body height in meters squared (kg/m 2 ). Obesity was defined as BMI 25 kg/m 2, according to the Asia-Pacific definition. The number of menstrual cycles in the previous 12 months was recorded. An ultrasound examination of the pelvis, preferably transvaginal, was performed to detect polycystic ovaries. The protocol for evaluating PCOS patients at our clinics was expanded in 1 June 2009. After this time, the following data were also available: (i) serum androgens, including androstenedione, dehydroepiandrosterone sulfate (DHEA-S), 17-a-OH progesterone and the free androgen index; (ii) the 2-h oral glucose tolerance test and glucose sex hormone-binding globulin (SHBG) levels; (iii) lipid profiles, including total cholesterol, triglycerides, highdensity lipoprotein (HDL) and low-density lipoprotein (LDL) and (iv) liver function and inflammatory markers, including glutamic oxaloacetic transaminase, glutamic pyruvic transaminase and high-sensitivity C-reactive protein. Therefore, 283 of 781 cases had this additional information. The serum FSH, LH and prolactin were measured using enzyme immunoassay (AxSym System, Abbott Laboratories, IL, USA); serum total testosterone, DHEA-S, 17-a-OH progesterone and androstenedione concentrations were measured by radioimmunoassay (Diagnostic Systems Laboratories, Webster, TX, USA); SHBG was measured using an electrochemiluminescence immunoassay (Roche Diagnostics, Indianapolis, IN, USA); the fasting insulin concentrations were measured by radioimmunoassay (Siemens Medical Solutions Diagnostics, Los Angeles, CA, USA) and the fasting glucose was measured using the Beckman Coulter Synchron LX (Fullerton, CA, USA). Hirsutism was defined as a modified Ferriman Gallwey (mf-g) score of 6. The patients suspected of clinical androgen excess showed manifestations of hirsutism and/or acne. Biochemical HA was defined as total serum testosterone of 2.78 mmol/l (0.80 ng/ml, 1 ng/ml ¼ 3.47 mmol/l). The study population Women with and without PCOS were included in this study. PCOS was diagnosed according to the 2003 Rotterdam criteria. The definitions of Oligo-An, HA and PCOM have been described in detail previously (Hsu et al., 2009). The eventual study population consisted of a total of 781 women (average age 28.6 + 5.0 and range 20 40 years old), which included 453 women with PCOS (27.9 + 4.7, 20 40 years old) and 328 women without PCOS (average age 29.5 + 5.2, 20 40 years old). Overall, 283 of the 781 cases were investigated after 1 June 2009, and additional information was obtained for this subgroup. We further classified the 453 women with PCOS into two subgroups, Group A consisted of women in their 20s (n ¼ 294, 25.0 + 2.8, 20 29 years old), and Group B consisted of women of 30 40 years old (n ¼ 159, 33.2 + 2.6). Of the 453 subjects with PCOS, 145 were recruited after 1 June 2009. Of these cases, 85 were in Group A, and 60 were in Group B. The insulin sensitivity index was evaluated by the homeostasis model assessment insulin resistance (HOMA-IR) index using the following formula HOMA IR = Fasting insulin (mu/ml) Fasting glucose (mg/dl). 405 Metabolic syndrome has been defined as the presence of at least three of the following criteria, according to the third report of the National Cholesterol Education Program-Adult Treatment Panel III (ATP-III): abdominal obesity (a waist circumference.80 cm in women); serum TG 150 mg/dl or taking any medication for hypertriglyceridemia; serum HDL, 50 or taking any medication to increase the HDL level; BP 130 or 85 mmhg or taking any medication for hypertension; and fasting blood glucose 100 mg/dl or taking any medication for hyperglycemia. Statistical analysis SPSS 13.0 for Windows (SPSS, Inc., Chicago, IL, USA) was used for statistical analysis. We evaluated the correlation between age and PCOSrelated parameters using Pearson s correlation coefficients with a twotailed significance test; the results appear in Tables I and II. In Tables III and IV, the data are presented as the mean + SD. We used the x 2 and Fisher s exact tests to compare the categorical variables and an ANOVA model to compare the continuous variables in Tables III and IV. The means of more than two groups were compared using with the F-test and Dunnett s test with unequal variances for post hoc comparison. The difference between the groups were considered significant if the P-value was,0.05. Results Tables I and II show the correlation between ages and the PCOSrelated parameters for all of the study subjects. With all 781 of the study subjects included, Table I shows that age was negatively correlated with the total testosterone level (r ¼ 20.153, P, 0.001), acme, hirsutism and mf-g score (r ¼ 20.232, P, 0.001) and positively correlated with BMI, waist measurement, the fasting glucose

Clinical and biochemical presentation 3445 Table I The correlations between age and the clinical and biochemical parameters of PCOS (n 5 781). Total (n 5 781)... PCOS (n 5 453)... Non-PCOS (n 5 328)... Correlation P-value Correlation P-value Correlation P-value BMI 0.088 0.014* 0.149 0.001* 0.072 0.191 Waist 0.103 0.005* 0.166 <0.001*** 0.064 0.253 Number of menstrual cycle per years 0.088 0.014* 20.037 0.483 0.087 0.115 Average ovarian volume 20.054 0.144 20.047 0.328 20.023 0.690 Total testosterone 20.153 <0.001*** 20.143 0.002** 20.072 0.195 Acne 20.166 <0.001*** 20.100 0.034* 20.200 <0.001*** Hirsutism 20.178 <0.001*** 20.105 0.025* 20.255 <0.001*** mf-g score 20.232 <0.001*** 20.147 0.002** 20.326 <0.001*** Fasting insulin 20.051 0.285 0.091 0.054 20.045 0.420 Fasting glucose 0.205 <0.001*** 0.165 <0.001*** 0.289 <0.001*** HOMA-IR 0.032 0.370 20.005 0.918 0.123 0.026* HOMA-IR, homeostasis model assessment insulin resistance index. Bold values indicate the parameter s significant correlation with age. **P, 0.01. ***P, 0.001. Table II The correlations between age and additional clinical and biochemical parameters of PCOS (n 5 283, data obtained after 1 June 2009). Total (n 5 283)... PCOS (n 5 145)... Non-PCOS (n 5 138)... Correlation P-value Correlation P-value Correlation P-value Androstenedione 20.088 0.140 20.011 0.894 20.151 0.078 DHEA-S 20.235 <0.001*** 20.282 0.001** 20.194 0.021* 17-a-OH progesterone 20.018 0.762 20.014 0.871 20.027 0.752 Cholesterol 0.221 <0.001** 0.154 0.025* 0.312 <0.001*** Triglyceride 0.139 0.007* 0.139 0.044* 0.292 <0.001*** HDL 20.031 0.567 20.049 0.513 0.001 0.994 LDL 0.158 0.007* 0.147 0.076 0.191 0.024* 2-h Glucose 0.185 0.002** 0.187 0.025* 0.193 0.025* SHBG 20.046 0.440 20.130 0.118 0.001 0.990 Anti-Müllerian hormone 20.140 0.019** 20.106 0.205 20.197 0.021* Follicles number per ovary 20.018 0.769 0.076 0.366 20.002 0.981 DHEA-S, dehydroepiandrosterone sulfate; HDL, high-density lipoprotein; LDL, low-density lipoprotein; SHBG, sex hormone-binding globulin. Bold values indicate the parameter s significant correlation with age. **P, 0.01. ***P, 0.001. level (r ¼ 0.205, P, 0.001) and the number of menstrual cycles per year. Table II shows that age was negatively correlated with DHEA-S (r ¼ 20.235, P, 0.001) and anti-müllerian hormone (r ¼ 20.140, P ¼ 0.019) in the 283-case subgroup (of the 781 study subjects) and positively correlated with cholesterol (r ¼ 0.221, P, 0.001), triglycerides (r ¼ 0.139, P ¼ 0.007) and LDL (r ¼ 0.158, P ¼ 0.007). The prevalence of acne, hirsutism, the mf-g score and serum DHEA-S level were negatively correlated with age in both the PCOS and the non-pcos women. The cholesterol, triglyceride and fasting glucose levels were significantly positively correlated with age in both the PCOS and the non-pcos women. In Tables III and IV, all of the women with PCOS are classified into two groups according their ages: Group A (n ¼ 294, 20 29 years old) and Group B (n ¼ 159, 30 40 years old). Of the three PCOS criteria, PCOM was the most predominant characteristic of the women with PCOS. For the women who fulfilled the PCOS diagnostic criteria,

3446 Liang et al. Table III A comparison of clinical characteristics of the women with PCOS in the 20 29 and 30 40 years old groups. Total Group A (20 29 years old) Group B (30 40 years old) P-value (n 5 453) (n 5 294) (n 5 159) (A versus B) Age (years) 27.9 + 4.7 25.0 + 2.8 33.2 + 2.6 Menarche (years) 12.7 + 1.7 12.6 + 1.8 12.7 + 1.7 0.756 PCOS criteria Oligo- and/or anovulation (%) 74 71 79 0.083 Clinical and/or biochemical hyperandrogenism (%) 84 87 80 0.055 PCOM (%) 90 89 92 0.359 Anthroponmetric components BMI (kg/m 2 ) 25.7 + 6.6 25.0 + 6.5 27.1 + 6.5 0.001** Weight (kg) 66.5 + 17.9 65.0 + 18.1 69.4 + 17.3 0.011* Height (cm) 160.7 + 5.3 161.1 + 5.4 160.0 + 5.1 0.028* Waist (cm) 79.0 + 17.4 76.7 + 17.3 83.3 + 16.7 <0.001*** Hip (cm) 96.3 + 12.9 95.4 + 13.0 98.0 + 12.5 0.041* Waist-to-hip ratio 0.81 + 0.09 0.80 + 0.08 0.85 + 0.09 <0.001*** Obesity (%) 47 42 55 0.006* Androgens Biochemical HA (%) 53 54 49 0.342 Acne (%) 56 58 51 0.141 Hirsutism (%) 25 27 21 0.150 mf-g score 3.8 + 3.8 4.0 + 4.0 3.3 + 3.4 0.055 Testosterone (mmol/l) 2.9 + 1.2 3.0 + 1.3 2.7 + 1.1 0.005* Endocrine Prolactin (miu/ml) 13.9 + 5.0 14.2 + 4.9 13.3 + 5.1 0.091 LH (miu/ml) 10.2 + 7.6 10.5 + 8.1 9.7 + 6.8 0.355 FSH (miu/ml) 5.9 + 1.9 6.1 + 1.9 5.7 + 1.8 0.052 LH-to-FSH ratio 1.7 + 1.1 1.8 + 1.2 1.7 + 1.0 0.862 Insulin resistance Fasting Insulin (miu/ml) 10.4 + 10.8 9.6 + 10.6 11.8 + 10.9 0.037* Fasting glucose (mg/dl) 93.8 + 15.9 92.0 + 11.9 97.0 + 21.0 0.001** HOMA-IR 1.7 + 2.1 1.5 + 2.2 2.0 + 3.1 0.848 Average ovarian volume (ml) 11.3 + 5.4 11.3 + 5.6 11.3 + 4.9 0.972 Number of menstrual cycles per year 6.6 + 3.7 6.6 + 3.8 6.4 + 3.5 0.567 Data are either mean + SD or percentage. HOMA-IR, homeostasis model assessment insulin resistance index. Bold values indicate the parameter s significant difference between two groups. **P, 0.01. ***P, 0.001. Group A had a significantly higher total testosterone levels than Group B. Group B had higher levels of some metabolic disturbances (elevated fasting glucose and insulin levels, BMI, waist measurement and triglycerides and a higher incidence of obesity) than Group A. The serum LH, FSH levels and average ovarian volume were not significantly different among two groups. Discussion Clinical and biochemical androgen excesses are major characteristics of women with PCOS. An age-related decrease in androgen secretion occurs in women with PCOS, as it does in normal women. Ovarian steroid secretion capacity starts to decline as early as age 30 (Piltonen et al., 2004), and hyperandrogenism partially resolves before menopause in women with PCOS (Winters et al., 2000). The presentation of androgen excess was highly age-dependent both in the current study and in previous reports. Bili et al. (2001) studied 472 oligomenorrheic and amenorrheic infertile patients and found that age was inversely correlated with testosterone, androstenedione and dehydroepiandrosterone. Morán et al. (1999) studied 145 hyperandrogenic women and reported a negative association between DHEA-S levels and age.

Clinical and biochemical presentation 3447 Table IV A comparison of addition clinical characteristics of the women with PCOS in the 20 29 and 30 40 years old groups (n 5 145, data obtained after 1 June 2009). Total Group A (20 29 years old) Group B (30 40 years old) P-value (n 5 145) (n 5 85) (n 5 60) (A versus B) Age (years) 27.8 + 4.7 25.0 + 2.6 33.2 + 2.7 Androgens Androstenedione (ng/dl) 3.2 + 1.2 3.2 + 1.1 3.2 + 1.4 0.985 DHEA-S (ng/dl) 204.5 + 99.5 217.6 + 94.2 177.8 + 91.1 0.006* 17-OH PRG (ng/dl) 1.2 + 0.9 1.2 + 1.0 1.2 + 0.8 0.844 SHBG (nmol/l) 36.9 + 23.2 39.8 + 26.3 32.9 + 17.4 0.077 Oral glucose tolerance test 2-h glucose (mg/dl) 119.5 + 49.1 113.8 + 48.2 127.4 + 49.7 0.102 Lipids Cholesterol (mg/dl) 188.7 + 32.0 188.7 + 37.1 191.3 + 27.3 0.604 Triglyceride (mg/dl) 111.2 + 196.9 78.8 + 55.5 122.5 + 108.2 0.019* HDL (mg/dl) 51.9 + 28.5 55.4 + 18.0 47.5 + 13.4 0.901 LDL (mg/dl) 115.1 + 31.9 112.5 + 35.7 117.9 + 25.1 0.369 Inflammatiom hscrp (mg/dl) 0.36 + 0.43 0.30 + 0.42 0.32 + 0.31 0.818 Liver enzyme GOT (IU/l) 25.2 + 15.7 25.1 + 18.4 24.3 + 12.4 0.610 GPT (IU/l) 27.9 + 24.2 25.4 + 25.1 23.5 + 17.6 0.953 Blood pressure Systolic pressure (mmhg) 115.3 + 17.8 112.3 + 16.9 114.7 + 21.2 0.549 Diastolic pressure (mmhg) 78.14 + 13.3 78.4 + 12.6 81.2 + 12.9 0.069 Metabolism syndrome Metabolic syndrome (%) 30 24 38 0.055 Hypertension (%) 29 26 33 0.333 HDL, 50 mg/dl (%) 53 45 65 0.016* Triglyceride. 150 mg/dl (%) 14 9 20 0.069 Waist. 80 cm (%) 58 48 72 0.005* Data are either mean + SD or percentage. DHEA-S, dehydroepiandrosterone sulfate; 17-OH PRG, 17-a-OH progesterone; HDL, high-density lipoprotein; LDL, low-density lipoprotein; SHBG, sex hormone-binding globulin; GOT, glutamic oxaloacetic transaminase; GPT, glutamic pyruvic transaminase; hscrp, high-sensitivity C-reactive protein. Bold values indicate the parameter s significant difference between two groups. Winters et al. (2000) studied 84 women with PCOS and found that levels of total testosterone and non-shbg-bound testosterone were lower in older women with PCOS. Our study found that serum androgen markers (total testosterone and DHEA-S) were significantly negatively correlated with age. Furthermore, we found the prevalence of acne and hirsutism, and the mf-g score had significant negative associations with age in the women with and without PCOS. Accordingly, the prevalence of both clinical hyperandrogenism and biochemical HA should decrease as women age. Some studies have found that morphological ultrasonographic features consistent with polycystic ovaries are common in postmenopausal women and that these features are associated with higher than normal testosterone levels and metabolic alterations (Birdsall and Farquhar, 1996). Our results did not indicate any significant correlation between age and ovarian volume. However, the Pavlik et al. (2000) study of age and ovarian volume demonstrated a stable ovarian volume to age 35, the greatest decline between 35 and 55 years old and a minor decline after age 55. In normally ovulating women, ovarian volume decreases with age, with measurable decreases across each decade of reproductive life starting at age 35 40 (Pavlik et al., 2000; Oppermann et al., 2003). Furthermore, the decrease in ovarian volume is less pronounced in the women with PCOS than in the controls (Alsamarai et al., 2009). Our subjects consisted of 781 cases with average age of 28.6 (all of whom were,40 years old), and the ovarian volumes were relatively stable at this stage. Previous studies have suggested that menstrual cycles may become regular with age in women with PCOS (Birdsall and Farquhar, 1996; Elting et al., 2003). Our results also showed that the number of menstrual cycles per year was positively correlated with age. These results are consistent with previous reports.

3448 Liang et al. As described above, age can influence the status of women diagnosed with PCOS. Women diagnosed with PCOS, by 2003 Rotterdam criteria may have different phenotypes depending on age. In patients with PCOS, a PCOM was the most prevalent criterion for women age 20. While the prevalence of clinical and/or biochemical hyperandrogenism significantly decreases with age and the severity of menstrual disturbances may improve after age 20, ovarian volume and ovarian morphology seem to be relatively stable for women younger than 35 (Pavlik et al., 2000). Therefore, a PCOM became the prevalent diagnostic criterion for PCOS in women over 20 years old. Age is also an important risk factor for developing metabolic disorders and cardiovascular events. In the correlation analysis, we found that advanced age was associated with increased cholesterol, triglyceride and LDL levels. Furthermore, aging may also be associated with a defect in insulin action that is manifested by decreased whole-body tissue sensitivity to insulin without a change in tissue responsiveness (Rowe et al., 1983). The significant positive correlation between fasting glucose levels and age in the women with and without PCOS may reflect a glucose intolerance that develops as part of the aging process. In elderly subjects, the severity of carbohydrate intolerance correlates directly with the degree of peripheral insulin resistance (Fink et al., 1983). Our data showed that age positively correlated with BMI in women with PCOS, which was not true in the non-pcos groups. Furthermore, the PCOS women in Group A (20 29 years old) and Group B (30 40 years old) had a similar prevalence of hyperandrogenism, but Group B had significantly higher BMI scores and incidence of obesity than Group A. This finding is an interesting phenomenon. It is known that as age increases, serum total testosterone levels decrease. Furthermore, serum testosterone was strongly positively correlated with BMI in our study and previous reports (Pasquali et al., 2006). If an older group fulfills the set diagnostic criteria for PCOS, they may present with more obesity and higher BMI scores than a younger age group. In comparison to younger women with PCOS, therefore, older PCOS patients should face both age- and BMI-related increases in risk of insulin resistance and metabolic disturbance. For women of age 30, the negative impact of PCOS on health is clear. Compared with younger women, we found that PCOS patients of age 30 had higher fasting glucose levels, cholesterol, triglycerides, BMI and waist measurements consistent with Gülekli s suggestion that PCOS patients are prone to gain weight as they get older (Gülekli et al., 1993). This result implies that the association between abdominal obesity and metabolic disturbance in women with PCOS becomes clear later in life. Adolescent oligomenorrhea may spontaneously resolve and may modify the diagnosis. Diagnosing PCOS during adolescence using the Rotterdam criteria may not be appropriate (Carmina et al., 2010). Therefore, our study excluded women,20 years old. We evaluated the clinical and biochemical characteristics of the women with PCOS according to the age group; we also analyzed the correlation between age and PCOS-related markers. The results of our study suggested that some of the diagnostic criteria for PCOS might be modified when applied to different age groups. In conclusion, a PCOM was the most common PCOS diagnostic criteria in women over age 20. The severity of clinical and biochemical hyperandrogenism decreased with age. But after age 29, abdominal obesity and certain metabolic disturbances became more prevalent in women with PCOS. Authors roles S.-J.L. wrote this revision. C.-S.H. wrote the manuscript. C.-R.T. provided the concept. C.-H.C. analyzed the data. M.-I.H. designed and approved the manuscript. Funding This work was supported by a grant from the National Science Council (NSC 98-2629-B-038-001-MY3) (2-3) and by Taipei Medical University and Wan Fang Hospital (99-TMU-WFH-03-01). References Alsamarai S, Adams JM, Murphy MK, Post MD, Hayden DL, Hall JE, Welt CK. Criteria for polycystic ovarian morphology in polycystic ovary syndrome as a function of age. J Clin Endocrinol Metab 2009; 94:4961 4970. Bili H, Laven J, Imani B, Eijkemans MJ, Fauser BC. Age-related differences in features associated with polycystic ovary syndrome in normogonadotrophic oligo-amenorrhoeic infertile women of reproductive years. Eur J Endocrinol 2001;145:749 755. Birdsall MA, Farquhar CM. Polycystic ovaries in pre and post-menopausal women. Clin Endocrinol (Oxf) 1996;44:269 276. Carmina E, Oberfield SE, Lobo RA. The diagnosis of polycystic ovary syndrome in adolescents. Am J Obstet Gynecol 2010; 203:201.e1 201.e5. Cheung LP, Ma RC, Lam PM, Lok IH, Haines CJ, So WY, Tong PC, Cockram CS, Chow CC, Goggins WB. Cardiovascular risks and metabolic syndrome in Hong Kong Chinese women with polycystic ovary syndrome. Hum Reprod 2008;23:1431 1438. Elting MW, Kwee J, Korsen TJ, Rekers-Mombarg LT, Schoemaker J. Aging women with polycystic ovary syndrome who achieve regular menstrual cycles have a smaller follicle cohort than those who continue to have irregular cycles. Fertil Steril 2003;79:1154 1160. Fink RI, Kolterman OG, Griffin J, Olefsky JM. Mechanisms of insulin resistance in aging. J Clin Invest 1983;71:1523 1535. Gülekli B, Turhan NO, Senöz S, Kükner S, Oral H, Gökmen O. Endocrinological, ultrasonographic and clinical findings in adolescent and adult polycystic ovary patients: a comparative study. Gynecol Endocrinol 1993;7:273 277. Hsu MI, Liou TH, Liang SJ, Su HW, Wu CH, Hsu CS. Inappropriate gonadotropin secretion in polycystic ovary syndrome. Fertil Steril 2009; 91:1168 1174. March WA, Moore VM, Willson KJ, Phillips DI, Norman RJ, Davies MJ. The prevalence of polycystic ovary syndrome in a community sample assessed under contrasting diagnostic criteria. Hum Reprod 2010; 25:544 551. Mastorakos G, Lambrinoudaki I, Creatsas G. Polycystic ovary syndrome in adolescents: current and future treatment options. Paediatr Drugs 2006; 8:311 318. Morán C, Knochenhauer E, Boots LR, Azziz R. Adrenal androgen excess in hyperandrogenism: relation to age and body mass. Fertil Steril 1999; 71:671 674. Oppermann K, Fuchs SC, Spritzer PM. Ovarian volume in pre- and perimenopausal women: a population-based study. Menopause 2003; 10:209 213. Pasquali R, Gambineri A. Polycystic ovary syndrome: a multifaceted disease from adolescence to adult age. Ann N Y Acad Sci 2006; 1092:158 174.

Clinical and biochemical presentation 3449 Pasquali R, Gambineri A, Pagotto U. The impact of obesity on reproduction in women with polycystic ovary syndrome. BJOG 2006; 113:1148 1159. Pavlik EJ, DePriest PD, Gallion HH, Ueland FR, Reedy MB, Kryscio RJ, van Nagell JR Jr. Ovarian volume related to age. Gynecol Oncol 2000; 77:410 412. Piltonen T, Koivunen R, Perheentupa A, Morin-Papunen L, Ruokonen A, Tapanainen JS. Ovarian age-related responsiveness to human chorionic gonadotropin in women with polycystic ovary syndrome. J Clin Endocrinol Metab 2004;89:3769 3775. Rodríguez-Morán M,Guerrero-Romero F.Insulin resistance is independently related to age in Mexican women. J Endocrinol Invest 2003;26:42 48. 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 2004;81:19 25. Rowe JW, Minaker KL, Pallotta JA, Flier JS. Characterization of the insulin resistance of aging. J Clin Invest 1983;71:1581 1587. Winters SJ, Talbott E, Guzick DS, Zborowski J, McHugh KP. Serum testosterone levels decrease in middle age in women with the polycystic ovary syndrome. Fertil Steril 2000;73:724 729.