Is the PCOS diagnosis solved by ESHRE/ASRM 2003 consensus or could it include ultrasound examination of the ovarian stroma?

Transcription

1 Human Reproduction Vol.21, No.12 pp , 2006 Advance Access publication October 19, doi: /humrep/del306 Is the PCOS diagnosis solved by ESHRE/ASRM 2003 consensus or could it include ultrasound examination of the ovarian stroma? C.Belosi 1, L.Selvaggi 2, R.Apa 2, M.Guido 2, D.Romualdi 2, A.M.Fulghesu 3 and A.Lanzone 1,2,4,5 1 ISI, Istituto Scientifico Internazionale Paolo VI, 2 Unit of Physiopathology of Human Reproduction, Department of Obstetrics and Gynecology, Università Cattolica del Sacro Cuore, Rome, 3 Department of Obstetrics and Gynecology, Università degli Studi di Cagliari, Cagliari and 4 OASI Institute for Research, Troina (EN), Italy 5 To whom correspondence should be addressed at: Department of Obstetrics and Gynaecology, Università Cattolica del Sacro Cuore, L.go A. Gemelli 8, Rome, Italy. alanzone@rm.unicatt.it BACKGROUND: The clinical heterogeneity of polycystic ovary syndrome (PCOS) is mirrored by the unceasing debate on the most appropriate diagnostic criteria. METHODS AND RESULTS: To highlight differences and inconsistencies between NIH and ESHRE/ASRM criteria, we applied them to 375 patients with oligo/amenorrhoea and signs of hyperandrogenism. Among them, we identified 273 women with PCOS according to NIH, whereas up to 345 patients fulfilled ESHRE/ASRM criteria. The 72 patients, constituting the gap between the two classifications, exhibited a lower expression of clinical signs compared with the 273 patients matching both criteria. To the whole group, we then applied the ESHRE/ASRM criteria modified to include an easily reproducible ultrasound examination of the ovarian stroma (UCSC criteria). In this way, we identified 30 women who were healthy according to all criteria, 37 affected by PCOS according only to the ESHRE/ASRM Consensus, 35 affected according only to the UCSC and ESHRE/ASRM criteria and 273 who were considered to have PCOS by all criteria. These groups showed a progressively increasing expression of PCOS features. CONCLUSION: In the grey area between NIH and ESHRE/ASRM classifications, UCSC criteria could identify a subgroup of women, missed by NIH criteria, with more pronounced stigmas than those identified by ESHRE/ASRM criteria alone, and who may profit more from a targeted therapy. Key words: diagnosis/polycystic ovary syndrome/ultrasound criteria Introduction Polycystic ovary syndrome (PCOS) is an endocrine metabolic disorder affecting 5 10% of women of reproductive age. It is clinically characterized by anovulation, hyperandrogenism and oligoamenorrhoea (Franks, 1995). Owing to the typical heterogeneity in the combination of signs and symptoms, the definition of the syndrome is still not fully established. In 1935, Stein and Leventhal described the association between amenorrhoea and polycystic ovaries (PCO), which is now known as PCOS. To make a diagnosis of the syndrome, they required visualization of the ovaries by laparotomy and histological confirmation carried out by biopsy (Stein and Leventhal, 1935). Subsequently, as a result of the association between the endocrine alterations and the histological picture of PCOS, clinical attention aimed to define the biochemical criteria (Nestler et al., 1998). With the advent of ultrasounds and related endovaginal probes, a non-invasive technique became available to appraise the dimension and the morphology of the ovaries, and the information obtained by this method appeared to be consistent with laparoscopical (Eden, 1988) and histological (Saxton et al., 1990) findings. Yet, nowadays, it is broadly shown that the presence of PCO in the ultrasound scan (ultrasound-pco) is not sufficient for making diagnosis of PCOS, owing to the prevalence of ultrasound-pco, which is about 20% in a nonselected population (Loucks et al., 2000). Nevertheless, in patients who show one or more clinical signs of PCOS, the prevalence of ultrasound-pco is higher (Polson et al., 1988). According to the predominant American view, the 1990 National Institute of Health (NIH) Conference about PCOS recommended that diagnostic criteria should include the evidence of hyperandrogenism and ovulatory dysfunction (Zawadski and Dunaif, 1992) as an essential part of the diagnosis, without any regard to the morphological diagnosis of PCO by ultrasonography. Other investigators proposed a definition of PCOS requiring a typical ultrasound morphology of the ovary, associated with menstrual disorders and/or signs of hyperandrogenism; however, hormonal parameters were not identified as mandatory markers for making the diagnosis (Homburg, 2002) The Author Published by Oxford University Press on behalf of the European Society of Human Reproduction and Embryology. All rights reserved. For Permissions, please journals.permissions@oxfordjournals.org

2 Ovarian stroma examination in PCOS diagnosis In 2003, at the Rotterdam ESHRE/ASRM Consensus workshop, an attempt was made to standardize the working definition of the PCOS. Since then, the presence of two of three of the following criteria have been required for the diagnosis of PCOS: (i) oligo and/or anovulation, (ii) clinical and/or biochemical signs of hyperandrogenism and (iii) echographic PCO, after the exclusion of other pathologies with a similar clinical presentation such as congenital adrenal hyperplasia, Cushing s syndrome and androgen-secreting tumours (The Rotterdam ESHRE/ASRM Sponsored PCOS Consensus Workshop Group, 2004). According to the available literature (Balen et al., 2003), the echographic criteria fulfilling sufficient specificity and sensitivity to define ultrasound-pco are the following: presence of 12 or more follicles in each ovary measuring 2 9 mm in diameter and/or increased ovarian volume (>10 cm 3 ). The aim of this study was to apply to the same population several diagnostic criteria to investigate the different population clusters, along with their clinical and endocrine characteristics, to approach an the ideal PCOS definition. Moreover, we evaluate the impact of stroma analysis as described by Fulghesu et al. (2001) as a possible adjunctive marker of PCOS characterization. Materials and methods The population examined by this study consisted of 375 women aged between 17 and 36 years, referred to our department, showing at least one of the following symptoms: chronic anovulation, menstrual disorders (oligo/amenorrhoea), hirsutism or acne. All were healthy, euthyroid and normoprolactinaemic, with a spontaneous onset of puberty and a normal sexual development. None of them received medication known to affect carbohydrate metabolism and plasma sex steroids for at least 3 months before the study. Obesity was defined as a BMI >27 kg/m 2, as previously published (Ciampelli et al., 2002). The clinical studies were conducted during the early follicular phase (day 3 6 after menses) after a spontaneous or progestin-induced menstrual flow in the oligoamenorrhoeic patients. At 8 a.m., after fasting overnight for h, blood samples were obtained. Plasma levels of estradiol (E 2 ), testosterone, dehydroepiandrosterone sulphate (DHEAS), androstenedione, 17-hydroxyprogesterone (17-OHP), cortisol, LH, FSH, progesterone, prolactin, sex hormone-binding globulin (SHBG), tryglicerides, highdensity lipoprotein cholesterol (HDL-C), low-density lipoprotein cholesterol (LDL-C), very low-density lipoprotein cholesterol (VLDL-C) and non-esterified fatty acid (NEFA) were determined in basal condition. Patients underwent also an oral glucose tolerance test (OGTT): blood samples were collected before and 30, 60, 90 and 120 min after ingestion of 75 g glucose in 150 ml water. Insulin and glucose were assayed in all samples. The same day, a transvaginal pelvic ultrasound was performed on each patient using a 6.5 MHz endovaginal probe (Esaote, AUC5). The ultrasound examinations were performed by one of seven well-trained observers (A.M.F., Mario Ciampelli, M.G., R.A., Paola Villa, C.B. and L.S.), who were not aware of the patient s endocrine profiles. The following parameters were evaluated echographically: (i) Presence, number and disposition of follicles in each ovary. (ii) Ovarian volume, estimated according to the formula: π/6 (A B C), where A, B and C are the three ovarian diameters. (iii) The stromal/area ratio (S/A), calculated as ovarian stromal area, evaluated by outlining with the caliper the peripheral profile of the stroma identified by a central area slightly hyperechoic, with respect to the total area of the ovary evaluated by outlining with the caliper the external limits of the ovary in the maximum plane section. The mean ovarian volume, area, stroma and S/A ratio for each individual patient were calculated by adding the sizes of each ovary and then dividing by 2. For the echographic diagnosis of PCOS, we adapted the criteria of Adams to the quantitative evaluation of stroma: a threshold value for S/A ratio of 0.34 (Fulghesu et al., 2001) associated with 10 or more follicles with a diameter of 2 8 mm, arranged around an echodense central stroma (Adams et al., 1985), has been considered positive for PCO. A group of 50 normal ovulatory women served as controls; the mean (±SD) length of the menstrual cycle in these patients was 28.3 ± 1.2 days. Ovulatory cycles were previously confirmed by midluteal plasma progesterone values of >8 ng/ml for three consecutive cycles. These patients were used to define the threshold values for hormonal and ultrasound parameters as mean plus 2 SD of such control values. All hormone concentrations were determined by commercial radioimmunoassay kits. Gonadotrophins and insulin were assayed by a double-antibody method, and all steroids were assayed by the dextran charcoal separation technique. Plasma glucose was determined by the glucose oxidase method with a glucose analyzer (Beckman Instruments, Palo Alto, CA, USA). The intra-assay and inter-assay coefficients of variation were less than 8 and 15%, respectively, in all determinations. Total cholesterol and triglycerides concentrations were determined by an enzymatic assay (Bristol, Paris, France). HDL- C concentrations were determined after precipitation of chilomicrons, VLDL-C and LDL-C (Boehringer, Mannheim, Germany). VLDL-C was separated (as the supernatant) from LDL-C and HDL-C by lipoprotein ultracentrifugation. A magnesium chloride/phosphotungstick acid technique was used to precipitate LDL-C from the bottom fraction after ultracentrifugation. NEFAs were determined by an acylcoenzyme A oxidase-based colorimetric method. A normal glycemic response, impaired fasting glucose (IFG) and impaired glucose tolerance (IGT) to OGTT was defined according to the criteria of the American Diabetes Association (Wahl et al., 1998). OGTT data were analysed as the area under curve (AUC) after the glucose ingestion, calculated by the trapezoidal rule. In a group of 121 patients, the metabolic study was extended with the evaluation of the M value, an index of insulin sensitivity, calculated with a hyperinsulinaemic-euglycaemic clamp, as previously described (Ciampelli et al., 1997). The free androgen index (FAI) was estimated by the testosterone and SHBG plasma concentration according to this formula: [T] _ ( _log10[shbg]) (Rajkhowa et al., 1994). The population so examined has been considered according to NIH, ESHRE/ASRM and to our criteria (UCSC criteria) that differ from those of ESHRE/ASRM in the echographic definition of PCO indicated above. All results are expressed as the mean ± SD. Data were stored and analysed using the SPSS (Statistical Package for Social Science, release 5.0; SPSS, Chicago, IL, USA), on an IBM-compatible computer. Statistical analysis was performed by one-way analysis of variance (ANOVA) and Student s t-test for unpaired data, when necessary. Pearson s correlation coefficient was used to study the relationship between variables. The analysis of contingency was made with Fisher s exact test. A P-value <0.05 was considered statistically significant. Informed consent was obtained from each patient; the study protocol had been previously approved by our Institutional Review Board. Results Figure 1 indicates the classification of patients based on the different criteria used. Starting from the overall 375 consecutively studied patients, in 254 of them, we identified the presence of three diagnostic parameters (oligo- or anovulation, 3109

3 C.Belosi et al. 375 patients 345 PCOS diagnosis ESHRE/ASRM CONSENSUS 30 No PCOS diagnosis ESHRE/ASRM Consensus 273 PCOS diagnosis NIH Consensus 72 No PCOS diagnosis NIH Consensus Figure 1. Patient s population according to different criteria for polycystic ovary syndrome (PCOS) diagnosis used. clinical and/or biochemical signs of hyperandrogenism and polycystic ultrasound morphology of the ovaries, according to ESHRE/ASRM criteria), whereas in 19 women, oligo- or anovulation was associated to signs of hyperandrogenism; 46 patients showed oligo- or anovulation and ultrasound- PCO, and 26 women showed biochemical signs of hyperandrogenism and ultrasound PCO. Finally, 19 patients had only oligo- or anovulation, and 11 patients showed only signs of hyperandrogenism. If we apply the NIH criteria to this population, we can identify 273 PCOS patients (72.8%) showing both menstrual disorders and sign of hyperandrogenism. By using the ESHRE/ASRM criteria, we were able to identify 345 PCOS patients (92%) presenting at least two of the three proposed parameters (Figure 1). According to the ESHRE/ASRM criteria, PCOS has been diagnosed in 72 patients, who would not be included on the basis of the NIH definition (19.2% of the studied population); these women presented only one clinical symptom and ultrasound PCO. We have then evaluated the clinical characteristics of such 72 patients (PCOS-Rotterdam) and have compared them with those of the 273 PCOS patients who fulfilled both the NIH and the ESHRE/ASRM criteria (PCOS-Rotterdam/NIH). They exhibited significantly lower prevalence of hirsutism (30.6 versus 75.4%; P < 0.05), oligomenorrhoea (55.6 versus 93.7%; P < 0.05), amenorrhoea (23.0 versus 52.0%; P < 0.05) and acne (18.1 versus 48.7%; P < 0.05). Concerning the anthropometric and biochemical characteristics (Table I), the 72 patients presented a significantly lower BMI and waist to hip ratio (WHR), a significantly lower FAI and testosterone circulating concentrations. Probably owing to a less, although not significant, prevalence of obesity in such group, both fasting and after-load insulin pts Oligo/anovulation Hyperandrogenism Polycystic ovaries 19 pts Oligo/anovulation Hyperandrogenism 46 pts Oligo/anovulation Polycystic ovaries 19 out of 46 pts 26 pts Hyperandrogenism Polycystic ovaries 16 out 26 pts 311 PCOS diagnosis UCSC Criteria 19 pts Oligo/anovulation 11 pts Hyperandrogenism 27 No PCOS diagnosis All Criteria 2 out of 19 pts 1 out of 11 pts plasma levels were lower in the 273 PCOS-Rotterdam/NIH patients, whereas the insulin sensitivity index (M) was higher. Concerning the ultrasound characteristics of the ovaries, we noted that the PCOS-Rotterdam/NIH patients showed a higher ovarian volume and a higher stroma/area ratio compared with that of the PCOS-Rotterdam patients (Table I). Moreover, as indicated in Table II, it was observed in PCOS-Rotterdam/NIH group a greater incidence of abnormal OGTT as well as of FAI values or plasma testosterone and androstenedione concentrations above the normal threshold values. No differences between the two groups were observed in lipid profile (data not shown). We then moved to study the same population using the echographic criteria of Adams integrated with the quantitative evaluation of stroma, as previously described (UCSC criteria) (Figure 1). All the 273 PCOS-Rotterdam/NIH patients fulfilled UCSC definition; when we examined the 72 PCOS-Rotterdam patients, 35 of them showed a diagnostic accordance with USCS criteria (PCOS-Rotterdam/UCSC group), whereas the other 37 women were classified as PCOS only according to ESHRE/ASRM criteria (PCOS-Rotterdam/No-UCSC group): in fact, they showed only one clinical sign with augmented ovarian volume, without showing an increased S/A ratio. There were also 27 patients for whom all the classification considered as not affected by PCOS (No-PCOS group). Three patients were defined PCOS based on S/A ratio [without increased ovarian volume (>10 ml) or more than 12 follicles] and only one clinical sign: owing to the limited number, these patients were not included in the statistical analysis.

4 Ovarian stroma examination in PCOS diagnosis Table I. Clinical, biochemical and hormonal characteristics of the 273 NIH polycystic ovary syndrome (PCOS) patients compared with the 72 No-NIH Rotterdam PCOS patients PCOS-Rotterdam/NIH (n = 273) PCOS-Rotterdam (n = 72) P-value Age (years) ± ± 6.65 NS Weight (kg) ± ± <0.05 BMI (kg/m 2 ) ± ± 4.75 <0.05 WHR 0.81 ± ± 0.06 <0.05 FSH (mui/ml) 5.36 ± ± 2.33 <0.05 LH (mui/ml) 8.71 ± ± 5.74 NS LH/FSH 1.68 ± ± 0.78 <0.05 Testosterone (ng/ml) 0.70 ± ± 0.21 <0.05 Androstenedione (ng/ml) 2.85 ± ± 1.03 NS 17-OHP (ng/ml) 1.40 ± ± 0.67 NS DHEAS (ng/ml) ± ± <0.05 SHBG (nmol/l) ± ± <0.05 FAI 9.60 ± ± 3.02 <0.05 Fast insulin (μui/ml) ± ± 4.72 <0.05 AUC insulin (μui/ml 240!) ± ± <0.05 M (mg/kg/min) a 3.80 ± ± 3.03 <0.05 Ovarian total volume (ml) ± ± 3.69 <0.05 Ovarian stroma/area ratio 0.40 ± ± 0.08 < OHP, 17-hydroxyprogesterone; AUC, area under curve; DHEAS, dehydroepiandrosterone sulphate; FAI, free androgen index; M, insulin sensibility index; NS, not significant; SHBG, sex hormone-binding globulin; WHR, waist to hip ratio. PCOS-Rotterdam/NIH: patients defined as affected by PCOS according to both NIH and ESHRE/ASRM criteria. PCOS-Rotterdam: patients defined affected by PCOS according only ESHRE/ASRM criteria. a Calculated on 95 of 273 PCOS-Rotterdam/NIH and on 26 of 72 PCOS-Rotterdam/No-PCOS-NIH patients. Table II. Anthropometrical and hormonal characteristics of the polycystic ovary syndrome (PCOS)-Rotterdam/ NIH and PCOS-Rotterdam groups Tables III and IV summarize the clinical and hormonal parameters of these three groups, as well as the distribution of their abnormal clinical and hormonal characteristics. As expected, the 27 patients not affected by PCOS showed significantly lower androgen plasma levels and incidence of abnormal clinical and hormonal features when compared with the patients defined as PCOS according to both Rotterdam and UCSC criteria. Furthermore, no significant differences were found in the distribution of clinical and hormonal abnormal values and in circulating androgens in the unaffected group compared with the PCOS-Rotterdam/No-UCSC group. This last group showed a significantly lower prevalence of FAI and LH over the respective threshold values, as well as a lower presence of clinical signs, when compared with PCOS-Rotterdam/ UCSC women (Table IV). Moreover, in addition to the difference PCOS-Rotterdam/NIH (n = 273) PCOS-Rotterdam/ (n = 72) BMI (40.3) 23 (31.9) Altered OGTT a 17 (8.5) 0* FAI 7 (%) 174 (63.7) 11 (15.3)* Testosterone 0.6 ng/ml (%) 178 (65.2) 29 (40.3)* Androstenedione 3.0 ng/ml (%) 119 (43.5) 15 (20.8)* FAI + testosterone (%) 134 (49.1) 9 (12.5)* FAI + androstenedione (%) 84 (30.1) 4 (5.5)* FAI + testosterone + androstenedione (%) 75 (27.4) 3 (4.2)* Testosterone + androstenedione (%) 98 (35.9) 13 (18.1)* FAI, free androgen index; OGTT, oral glucose tolerance test. PCOS-Rotterdam/NIH: patients defined as affected by PCOS according to both NIH and ESHRE/ASRM criteria. PCOS-Rotterdam: patients defined affected by PCOS according to only ESHRE/ASRM criteria. The threshold values were defined as mean plus 2 SD of the control population. a Calculated on 199 of 273 PCOS-Rotterdam/NIH patients and on 49 of 72 PCOS-Rotterdam patients. *P < Group I versus Group II. in S/A ratio and ovarian volume were significantly greater in PCOS-Rotterdam/UCSC than in PCOS-Rotterdam/No-PCOS UCSC group; both groups also showed significantly greater ovarian volume as compared with No-PCOS patients (Table III). The linear relationship between the ultrasound parameters and the endocrine and metabolic data are shown in Table V. The total area of evaluation did not show a significant correlation with any hormonal or metabolic variable. The S/A ratio showed the most significant correlation with the androgens levels and insulin secretion. Discussion The PCOS has been described, since 1935, as an endocrine disorder characterized by amenorrhoea and PCO. In these

5 C.Belosi et al. Table III. Clinical, echographic and hormonal parameters of the polycystic ovary syndrome (PCOS)-Rotterdam group according to stroma evaluation PCOS-Rotterdam/UCSC (n = 35) PCOS-Rotterdam No-PCOS UCSC (n = 37) No-PCOS (n = 27) Age (years) ± ± ± 5.63 WHR 0.76 ± ± ± 0.07 Weight ± ± ± 9.64*, ** BMI (kg/m 2 ) ± ± ± 3.30*, ** FAI 5.56 ± ± 1.86* 3.48 ± 1.74* Testosterone (ng/ml) 0.55 ± ± ± 0.15*, ** Androstenedione (ng/ml) 2.39 ± ± ± OHP (ng/ml) 1.37 ± ± ± 0.30*, ** LH (mui/ml) 9.61 ± ± 4.24* 5.05 ± 2.63*, ** LH/FSH 1.51 ± ± ± 0.47*, ** M (mg/kg/min) a 5.87 ± ± 1.80 AUC I (μui/ml 240!) ± ± ± AUC Pep C ± ± * ± Ovarian total volume (ml) ± ± 2.99* 8.28 ± 1.28*, ** Ovarian stroma/area ratio 0.41 ± ± 0.06* 0.27 ± 0.05* 17-OHP, 17-hydroxyprogesterone; AUC, area under curve; FAI, free androgen index; WHR, waist to hip ratio. PCOS-Rotterdam/UCSC: patients defined as affected by PCOS according ESHRE/ASRM criteria and according UCSC criteria but not according to NIH criteria. PCOS-Rotterdam/No-PCOS UCSC: patients defined as affected by PCOS according ESHRE/ASRM criteria but not according to UCSC criteria and NIH criteria. No-PCOS: patients defined as not affected by PCOS according to all three classifications. a Calculated on 18 of 35 PCOS-Rotterdam/UCSC and on 8 of 37 PCOS-Rotterdam/No-PCOS UCSC. *P < 0.05 versus PCOS-Rotterdam/UCSC. **P < 0.05 versus PCOS-Rotterdam/No-PCOS UCSC. Table IV. Distribution of clinical and hormonal characteristics of the polycystic ovary syndrome (PCOS) Rotterdam group according the stroma evaluation PCOS-Rotterdam/ UCSC, n = 35 (%) PCOS-Rotterdam/No-PCOS UCSC, n = 37 (%) Oligomenorrhoea/amenorrhoea (%) 29 (82.8) 28 (75.6) 18 (66.6) Hirsutism (%) 15 (42.8) 7 (18.9)* 3 (11.1)* Acne (%) 10 (28.5) 3 (8.1)* 6 (22.2) One clinical sign 16 (45.7) 34 (91.9)* 27 (100)* Two clinical signs 19 (54.3) 3 (8.1)* 0* Three clinical signs Altered OGTT FAI 7 7 (20) 4 (10.8) 0 Testosterone 0.6 ng/ml 17 (48.5) 12 (32.4) 0 Androstenedione 3.0 ng/ml 11 (31.4) 4 (10.8) 0 FAI 7 and testosterone 0.6 ng/ml 5 (14.3) 4 (10.8) 0 FAI 7 and androstenedione 3.0 ng/ml 4 (11.4) 0 0 FAI 7 and testosterone 0.6 ng/ml and androstenedione 3.0 ng/ml 3 (8.5) 0 0 Testosterone 0.6 ng/ml and androstenedione 3.0 ng/ml 10 (28.5) 3 (8.1) 1 (3.7)* No-PCOS, n = 27 (%) FAI: free androgen index; OGTT, oral glucose tolerance test. PCOS-Rotterdam/UCSC: patients defined as affected by PCOS according to ESHRE/ASRM criteria and according to UCSC criteria but not according to NIH criteria. PCOS-Rotterdam/No-PCOS UCSC: patients defined as affected by PCOS according to ESHRE/ASRM criteria but not according to UCSC criteria and NIH criteria. No-PCOS: patients defined as not affected by PCOS according to all three classifications. *P < 0.05 versus PCOS-Rotterdam/UCSC. Table V. Linear relationships between ultrasound parameters and clinical, endocrine and metabolic data in the studied population 17-OHP, 17-hydroxyprogesterone; AUC, area under curve; DHEAS, dehydroepiandrosterone sulphate; FAI, free androgen index; NS, not significant Ovarian volume Stromal area Stromal area/total area ratio Total area R P R P R P NS BMI NS NS NS NS 0.28 <0.01 NS Testosterone 0.27 < < < NS Androstenedione NS NS 0.30 < < NS 17-OHP NS NS NS NS 0.29 < NS DHEAS NS NS NS NS NS NS NS Cortisol NS NS NS NS NS NS NS FAI NS NS 0.30 < < NS Fasting insulin NS NS NS NS 0.25 < NS AUC insulin NS NS NS NS 0.25 < NS Fasting glucose/fasting insulin NS NS NS NS NS NS NS

6 Ovarian stroma examination in PCOS diagnosis years, many different definitions have been proposed and introduced in the clinical practice for the diagnosis of PCOS, but no general consensus has been obtained. These controversies could be in part ascribed either to the intrinsic characteristics of the syndrome, such as the heterogeneity of symptoms and their variability during the course of the reproductive life, or to the overlapping of the instrumental and laboratory diagnostic criteria resulting in the lack of generally accepted threshold values to be used (Balen et al., 1995; Franks, 1995). This has confused and seriously hindered the clarification of the genetics, aetiology, clinical associations, treatment and late sequelae of the syndrome (Geisthovel, 2003). In 2003, to standardize the working definition of the PCOS, the diagnostic criteria of PCOS were revised by ESHRE/ASRM workshop. As can be easily noted, the NIH criteria are included in the ESHRE/ASRM criteria when the diagnosis of PCOS is based only on the clinical and biochemical criteria. However, because the ESHRE/ASRM evaluation is based on two of the three abovementioned criteria (clinical, biochemical and ultrasonographic criteria), the corresponding diagnosis of PCOS encompasses a higher proportion of women. Since the advent of ultrasound, numerous parameters have been proposed to morphologically define PCO, but there is still no consensus about their diagnostic value. Despite a large overlap with control population, the ultrasound-pco features are significantly associated with the typical symptoms, such as oligomenorrhoea, hirsutism or acne (Adams et al., 1986; Bunker et al., 1989; O Driscoll et al., 1994; Balen et al., 1995). Adams et al. (1985) proposed the following definition: presence of more then 10 cysts measuring 2 8 mm in diameter arranged peripherally around a dense core of stroma or scattered through an increased amount of stroma. In 1994, Dewailly et al. (Dewailly et al., 1994) proposed the use of ovarian hypertrophy (i.e. an ovarian area >5.5 cm 2 unilaterally or bilaterally) as a morphological indicator of PCOS, because it can be easier to quantify and to correlate with the stromal hypertrophy. Other authors came to the same conclusions by using ovarian volume (Pache et al., 1993; Takahashi et al., 1995; van Santbrink et al., 1997). Recently, Jonard et al. (2003) modified the ultrasound definition of PCO indicated by Adams to: increased ovarian area (>5.5 cm 2 ) or volume (>11 ml) and/or presence of >12 follicles measuring 2 9 mm in diameter (mean of both ovaries). The ESHRE workshop realized that ultrasound PCO should be considered as one of the possible criteria for PCOS and that the criteria fulfilling sufficient specificity and sensitivity to define PCO should be the following: presence of 12 or more follicles in each ovary measuring 2 9 mm in diameter and/or increased ovarian volume (>10 ml). The evaluation of the follicular distribution was omitted as well as the increase in echogenicity and volume of the ovarian stroma, even if the latter is specific for PCOS (Buckett et al., 1999). This resolution was based on the contention that the evaluation of the ovarian stroma could be too subjective or not adaptable to the routine daily practice. With the improvement in ultrasound software, the brightness or echogenicity of the ovarian stroma can be determined much more objectively, and therefore, the quantification of ovarian stroma by computerized reading of ultrasound images has revealed that stromal hypertrophy is a frequent and specific feature in ovarian androgenic dysfunction. However, the weak point in this evaluation could be its application in routine general practice (Balen et al., 2003). On the basis of these considerations, we have proposed an easily reproducible diagnostic criterion on ultrasound analysis of PCOS patients. We found that the evaluation of S/A ratio can allow the selection of such patients, with a substantial absence of false-positives and false-negatives. This index is strictly correlated to androgen plasma concentrations and presents the advantage of a low intra- and inter-observer coefficient of variation. In our assessment of the relationship between ultrasound and hormonal data, we found that the S/A ratio was significantly related to testosterone, androstenedione and 17-OHP plasma levels as well as to FAI values. The bulk of these data allowed us to define a threshold value of this ratio, able to discriminate ultrasound-pco from normal or multifollicular ovaries, with a very high sensitivity and specificity. Starting from these considerations, if only the NIH criteria would be taken into account, we could exclude from the diagnosis a group of women who are affected by hyperandrogenism or oligoamenorrhoea, and consequently, who might benefit from medical therapy. Moreover, these patients ought to undergo a correct diagnosis also to prevent the long-term health risks correlated with the syndrome: diabetes, cardiovascular disease and endometrial cancer (Holte et al., 1994; Pierpoint et al., 1998; Ehrmann et al., 1999). On the contrary, the ESHRE/ASRM Consensus might include in the PCOS condition also women who show only one clinical symptom (anovulation or hyperandrogenism), associated to an ultrasound ovarian polycystic morphology. The possibility to enrol in the PCOS group women that show at least one of the two ultrasound characteristics, theoretically, means that the increased ovarian volume alone should be assumed as an exclusive criterion independent of the other ultrasound features. Is it possible to identify further clusters among these patients defined PCOS beyond NIH criteria but in accordance to Rotterdam Consensus? When we used the S/A ratio associated with the criteria of Adams, we were able to define two subset of PCOS patients with different clinical and biochemical characteristics: in fact, significant differences concerning the prevalence of symptoms and a trend towards a higher prevalence of abnormally elevated androgens were observed in those patients defined PCOS according to both Rotterdam and UCSC criteria. On the contrary, the clinical and hormonal profile of the PCOS-Rotterdam/No-UCSC group resembled that observed in No-PCOS women. Accordingly, in a recent article, Azziz (2005) concluded that the widespread adoption of the diagnostic criteria proposed in the 2003 Rotterdam meeting proceedings should be considered premature, particularly concerning the available data and the phenotypic heterogeneity of the syndrome. Figure 2 depicts a theoretical scheme indicating the presence in the overall PCOS population of subsets with different clinical 3113

7 C.Belosi et al. Figure 2. Summary of clinical and biochemical characteristics for different diagnostic subgroups of polycystic ovary syndrome (PCOS). and biochemical features. Classical Stein Leventhal PCOS patients fulfilling all the possible diagnostic criteria showed the largest prevalence of clinical and biochemical signs. However, among those patients defined as PCOS also by ultrasound PCO visualization, the stroma evaluation was able to target a further group of women whose features were pronounced as shown by the prevalence of abnormal androgen markers and clinical symptoms. In conclusion, the previous and recent definitions of PCOS should have considered more adequately the typical criteria for ultrasound diagnosis of PCO. In the NIH classification, the problem regarding ultrasound aspects has not been included; in the ESHRE/ASRM definition the ultrasound criteria may be too large with relatively low specificity; therefore, multifollicular ovaries risk being evaluated as PCO. It appears that the NIH criteria identify individuals who are more metabolically affected than those identified by the Rotterdam criteria, which may also include subjects who 3114 are minimally affected. Further analysis of ultrasound parameters such as stroma evaluation, which seems more strictly related to the metabolic and endocrine imbalance of the syndrome, may offer the opportunity to better define the various phenotypes of PCOS and to consider the different clinical impacts to optimize the patients management and medical treatment. References Adams J, Franks S, Polson DW, Mason HD, Abdulwahid N, Tucker M, Morris DV, Price J and Jacobs HS (1985) Multifollicular ovaries: clinical and endocrine features and response to pulsatile gonadotropin releasing hormone. Lancet 2, Adams J, Polson DW and Frank S (1986) Prevalence of polycystic ovaries in women with anovulation and idiopathic hirsutism. Br Med J 293, Azziz R (2005) Diagnostic criteria for polycystic ovary syndrome: a reappraisal. Fertil Steril 83, Balen AH, Conway GS, Kaltsas G, Techatrasak K, Manning PC, West C and Jacobs HS (1995) Polycystic ovary syndrome: the spectrum of the disorder in 1741 patients. Hum Reprod 10,

8 Ovarian stroma examination in PCOS diagnosis Balen AH, Laven JSE, Tan SL and Dewailly D (2003) Ultrasound assessment of the polycystic ovary: international consensus definitions. Hum Reprod Update 9, Buckett WM, Bouzayen R, Watkin KL, Tulandi T and Tan SL (1999) Ovarian stromal echogenicity in women with normal and polycystic ovaries. Hum Reprod 14(3), Bunker CB, Newton JA, Kilborn J, Patel A, Conway GS, Jacobs HS, Greaves MW and Dowd PM (1989) Most women with acne have polycystic ovaries. Br J Dermatol 121, Ciampelli M, Fulghesu AM, Cucinelli F, Pavone V, Caruso A, Mancuso S and Lanzone A (1997) Heterogeneity in beta cell activity, hepatic insulin clearance and peripheral insulin sensitivity in women with polycystic ovary syndrome. Hum Reprod 12(9), Ciampelli M, Leoni F, Lattanzi F, Guido M, Apa R and Lanzone A (2002) A pilot study of the long term effects of acipimox in polycystic ovarian syndrome. Hum Reprod 17, Dewailly D, Robert Y, Helin I, Ardaens Y, Thomas-Desrousseaux P, Lemaitre L and Fossati P (1994) Ovarian stromal hypertrophy in hyperandrogenic women. Clin Endocrinol (Oxf) 41(5), Eden JA (1988) Which is the best test to detect the polycystic ovary? Aust NZ J Obstet Gynaecol 28, Ehrmann DA, Barnes RB, Rosenfield RL, Cavaghan MK and Imperial J (1999) Prevalence of impaired glucose tolerance and diabetes in women with polycystic ovary syndrome. Diabetes Care 22, Franks S (1995) Polycystic ovary syndrome. N Engl J Med 333, Fulghesu AM, Ciampelli M, Belosi C, Apa R, Pavone V and Lanzone A (2001) A new ultrasound criterion for the diagnosis of polycystic ovary syndrome: the ovarian stroma/total area ratio. Fertil Steril 76, Geisthovel F (2003) A comment on the European Society of Human Reproduction and Embryology/American Society for Reproductive Medicine consensus of the polycystic ovarian syndrome. Reprod Biomed Online 7, Holte J, Bergh T, Berne C and Lithell H (1994) Serum lipoprotein lipid profile in women with the polycystic ovary syndrome: relation to anthropometric, endocrine and metabolic variables. Clin Endocrinol (Oxf ) 41, Homburg R (2002) What is polycystic ovarian syndrome? A proposal for a consensus on the definition and diagnosis of polycystic ovarian syndrome. Hum Reprod 17, Jonard S, Robert Y, Corter-Rudelli C, Pigny P, Decanter C and Dewailly D (2003) Ultrasound examination of polycystic ovaries: is it worth counting the follicles? Hum Reprod 18, Loucks TL, Talbott EO, McHugh KP, Keelan M, Berga SL and Guzick DS (2000) Do polycystic-appearing ovaries affect the risk of cardiovascular disease among women with polycystic ovary syndrome? Fertil Steril 74, Nestler JE, Jakubovicz DJ, de Vargas AF, Brik C, Quintero N and Medina F (1998) Insulin stimulates testosterone biosynthesis by human thecal cells from women with polycystic ovary syndrome by activating its own receptor and using inositol glycan mediators as the signal transduction system. J Clin Endocrinol Metab 83, O Driscoll JB, Mamtora H, Higginson J, Pollock A, Kane J and Anderson DC (1994) A prospective study of the prevalence of clear-cut endocrine disorders and polycystic ovaries in 350 patients presenting with hirsutism or androgenic alopecia. Clin Endocrinol (Oxf) 41, Pache TD, de Jong FH, Hop WC and Fauser BC (1993) Association between ovarian changes assessed by transvaginal sonography and clinical and endocrine signs of the polycystic ovary syndrome. Fertil Steril 59(3), Pierpoint T, McKeigue PM, Isaacs AJ, Wild SH and Jacobs HS (1998) Mortality of women with polycystic ovary syndrome at long-term follow-up. J Clin Epidemiol 51, Polson DW, Adams J, Wadsworth J and Franks S (1988) Polycystic ovaries a common finding in normal women. Lancet 1, Rajkhowa M, Bicknell J, Jones M and Clayton RN (1994) Insulin sensitivity in women with polycystic ovary syndrome: relationship to hyperandrogenemia. Fertil Steril 61, van Santbrink EJ, Hop WC and Fauser BC (1997) Classification of normogonadotropic infertility: polycystic ovaries diagnosed by ultrasound versus endocrine characteristics of polycystic ovary syndrome. Fertil Steril 67(3), Saxton DW, Farquhar CM, Rae T, Beard RW, Anderson MC and Wadsworth J (1990) Accuracy of ultrasound measurements of female pelvic organs. Br J Obstet Gynaecol 97, Stein IF and Leventhal ML (1935) Amenorrhoea associated with bilateral polycystic ovaries. Am J Obst Gynecol 29, Takahashi K, Okada M, Ozaki T, Uchida A, Yamasaki H and Kitao M (1995) Transvaginal ultrasonographic morphology in polycystic ovarian syndrome. Gynecol Obstet Invest 39(3), The Rotterdam ESHRE/ASRM Sponsored PCOS Consensus Workshop Group (2004) Revised 2003 consensus on diagnostic criteria and long-term health risks related to polycystic ovary syndrome. Fertil Steril 81, Wahl PW, Savage PJ, Psaty BM, Orchard TJ, Robbins JA and Tracy RP (1998) Diabetes in older adults: comparison of 1997 American Diabetes Association classification of diabetes mellitus with 1985 WHO classification. Lancet 352(9133), Zawadski JK and Dunaif A (1992) Diagnostic criteria for polycystic ovary syndrome: towards a rational approach. In Dunaif A, Givens JR and Haseltine F (eds) Polycystic Ovary Syndrome. Blackwell Scientific, Boston, pp Submitted on January 24, 2006; resubmitted on May 9, 2006 and June 14, 2006; accepted on July 5,