Using cluster analysis to identify a homogeneous subpopulation of women with polycystic ovarian morphology in a population of non-hyperandrogenic women with regular menstrual cycles.

Can cluster analysis can be used to identify a homogeneous subpopulation of women with polycystic ovarian morphology (PCOM) within a very large population of control women in a non-subjective way? Identification and exclusion of the cluster corresponding to women with PCOM from controls improved the diagnostic power of serum anti-Müllerian hormone (AMH) level and follicle number per ovary (FNPO) in discriminating between women with or without polycystic ovary syndrome (PCOS). There is disagreement as to whether women with PCOM should be excluded from the control population when establishing FNPO and AMH diagnostic thresholds for the definition of PCOS and how to identify such women. It has been demonstrated that cluster analysis can detect women with PCOM within the control population through a set of classifying variables among which the most relevant was AMH. The adequacy of this approach has not been confirmed in other clinical settings. This was a retrospective study using clinical and laboratory data derived from the computerized database. The data were collected from March 2011 to May 2013. The study included 893 patients referred for routine infertility evaluation and treatment. The patients were divided into three groups: (i) the control group (n = 621) included women with regular menstrual cycles and no signs of hyperandrogenism (HA), (ii) the full-blown PCOS group (n = 95) consisted of women who were diagnosed as having PCOS based on the presence of both HA and oligo/amenorrhoea (OA), (iii) the mild PCOS group included women with only two items of the Rotterdam classification, i.e. PCOM at ultrasonography according to the FNPO threshold of 12 or more and either OA (n = 110) or HA (n = 67). After exclusion of women with PCOM from the controls, the AMH threshold of 28 pmol/l with specificity 97.5% and sensitivity 84.2% [area under the curve (AUC) 0.948 (95% confidence interval (CI) 0.915-0.982)] and FNPO threshold of 12 with specificity 92.5% and sensitivity 83.2% [AUC 0.940 (95% CI 0.909-0.971)] for identifying PCOS were derived from the receiver operating characteristic curve analysis. The AMH threshold of 28 pmol/l had the same specificity for discriminating the mild and the full-blown PCOS phenotypes from controls. There was no selection bias other than being evaluated for infertility treatment, however, this could also be considered as a limitation of the study as these women may not necessarily be a representative sample of the general population. The study demonstrated that serum AMH has intrinsically a very high potency to detect women with PCOM within a control group. However, the AMH threshold is specific for this method and clinical setting. AMH threshold value for the definition of PCOM is method specific and cannot be universally applied. This study confirmed the results of previous studies that cluster analysis can identify women with PCOM within the very large control population without using a predefined diagnostic threshold for FNPO, and that AMH can detect women with PCOM with a high specificity. The exclusion of PCOM women from the controls by using a cluster analysis should be considered when establishing reference intervals and decision threshold values for various parameters used to characterize PCOS. The authors have no funding/competing interest(s) to declare.

Dewailly D ，Alebić MŠ ，Duhamel A ，Stojanović N ... -  《-》

Polycystic ovarian morphology and the diagnosis of polycystic ovary syndrome: redefining threshold levels for follicle count and serum anti-Müllerian hormone using cluster analysis.

Can cluster analysis be used to differentiate between normo-ovulatory women with normal ovaries and normo-ovulatory women with polycystic ovarian morphology (PCOM) in a non-subjective manner? Cluster analysis can be used to accurately and non-subjectively differentiate between normo-ovulatory women with normal ovaries and normo-ovulatory women with PCOM. Currently, PCOM is diagnosed using a fixed threshold level, i.e. 12 or more follicles per ovary, and is one of the diagnostic criteria of polycystic ovary syndrome (PCOS). However, PCOM is also encountered in normo-ovulatory women, suggesting that it could just represent a normal variant. On the other hand, recent studies have shown subtle endocrine abnormalities in women with isolated PCOM that resemble those found in women with PCOS. Because of the strong correlation between anti-Müllerian hormone (AMH) and follicle number, a high serum AMH level has been proposed as a surrogate marker for PCOM and could, therefore, be integrated in the diagnostic classifications for PCOS. This was a retrospective observational cohort study. Original cohorts had been recruited for previous studies between 1998 and 2010. Two hundred ninety-seven regularly cycling women and 700 women with PCOS were eligible for inclusion. Cluster analysis was performed in 297 regularly cycling women. After exclusion of 'PCOM' clusters, each 'non-PCOM' cluster (young, n = 118 and old, n = 100) was included in the construction of a receiver operating characteristics curve to test the diagnostic performance of follicle number per ovary (FNPO) and serum AMH in discriminating similarly aged full-blown PCOS patients (n = 411 and 237, respectively) from normal regularly cycling non-PCOM women. The optimal number of clusters was four; age was the most important classifying variable, followed by the FNPO and serum AMH. Two distinct clusters of normo-ovulatory women with PCOM were isolated and differed solely by age, i.e. 'young' and 'old'. Both 'PCOM' clusters had their similarly aged counterpart of 'non-PCOM' clusters. Likewise, two clusters comprised women younger than 30 years, with (n = 28, 'PCOM regularly cycling women') or without (n = 118, 'normal regularly cycling women') features of PCOM (increased FNPO and/or serum AMH). The two other clusters in older women could be labelled 'normal regularly cycling women' or 'PCOM regularly cycling women' (n = 100 and 51, respectively). The prevalence of PCOM was significantly greater in old than in young regularly cycling women controls. In the young population, after exclusion of the 'PCOM regularly cycling women', the diagnostic performance of AMH, expressed by area under the curve (AUC) (AUC = 0.903; CI (0.876-0.930)) to differentiate PCOS women from normal regularly cycling women was similar to that using the FNPO (AUC = 0.915, CI (0.891-0.940)) (P = 0.25), confirming results from earlier studies. In the old population, the diagnostic performance of AMH was greater than that of FNPO (AUCs = 0.948 (0.927-0.970) vs 0.874 (0.836-0.912), respectively, P = 0.00035). Cut-off levels of AMH and antral follicle count distinguishing regularly cycling non-PCOM women from PCOS women were higher in young women than in older women. Data of normal women were obtained from earlier studies, aiming to measure normal endocrine values. Apparently, the strong effect of age in cluster analysis revealed a dichotomy in the age distribution among the cohort of regularly cycling women included. This was involuntary since in none of the original studies, eligibility was limited by age and there was considerable overlap in age ranges of the cohorts. Transvaginal ultrasound was performed using a 6.5-8 mHz probe and our data confirm that this threshold level for FNPO is still valid if using such probe frequencies, although the use of devices with a maximum frequency lower than 8 mHz has become obsolete. Obviously, newer ultrasound scanner using higher transducer frequency will facilitate the detection of more follicles. Our data support the use of AMH as a surrogate for ultrasound to define PCOM, which is one of the three items of the Rotterdam classification. They also show that age should be taken into account to define the optimal threshold. The fact that the prevalence of PCOM was increased in the older regularly cycling women, may be due to 'attenuated' PCOS, a phenomenon that has been described in ageing women with PCOS. These women might have had anovulatory cycles in the past and have become ovulatory with increasing age, and were, therefore, eligible for this study. However, since most women included at older age have had spontaneous pregnancies in the past, PCOM at older age may be associated with a subclinical form of PCOS, which may also be present in young regularly cycling women. No funding was received for this study. J.S.E.L. has received grants and support from Ferring, MSD, Organon, Merck-Serono, Schering Plough and Serono during recruitment and analysis of data for this study. S.L.F., A.D. and D.D. do not have any conflict of interest.

Lie Fong S ，Laven JSE ，Duhamel A ，Dewailly D ... -  《-》

Use of the serum anti-Müllerian hormone assay as a surrogate for polycystic ovarian morphology: impact on diagnosis and phenotypic classification of polycystic ovary syndrome.

Does the use of the serum anti-Müllerian hormone (AMH) assay to replace or complement ultrasound (U/S) affect the diagnosis or phenotypic distribution of polycystic ovary syndrome (PCOS)? Combining U/S and the serum AMH assay to define polycystic ovarian morphology (PCOM) diagnoses PCOS (according to the Rotterdam classification) in more patients than definitions using one or the other of these indicators exclusively. Since 2003, PCOM, as defined by U/S, is one of the three diagnostic criteria for PCOS. As it is closely correlated with follicle excess seen at U/S, an excessive serum AMH level could be used as a surrogate for PCOM. Single-center retrospective study from a database of prospectively collected clinical, laboratory and ultrasound data from patients referred for oligo-anovulation (OA) and/or hyperandrogenism (HA) between January 2009 and January 2016. The standard Rotterdam classification for PCOS was tested against two modified versions that defined PCOM by either excessive serum AMH level alone (AMH-only) or a combination (i.e. 'and/or') of the latter and U/S. The PCOS phenotypes were defined as A (full phenotype, OA+HA+PCOM), B (OA+HA), C (HA+PCOM) and D (OA+PCOM). PCOS was more frequently diagnosed when PCOM was defined as the combination 'positive U/S' and/or 'positive AMH' (n = 639) than by either only U/S-only (standard definition, n = 612) or by AMH-only (n = 601). With this combination, PCOM was recognized in 637 of the 639 cases that met the Rotterdam classification, and phenotype B practically disappeared. In this population, U/S and AMH markers were discordant for PCOM in 103 (16.1%) cases (9% U/S-only, 7.1% AMH-only, P = 0.159). The markers used had no other significant impact on the phenotypic distribution (except for phenotype B). However, the percentage of cases positive by U/S-only was significantly higher in phenotype D than in phenotype A (14.1% vs. 5.8%, P < 0.05). Furthermore, in the discordant cases, plasma LH levels were significantly higher in the AMH-only group than in the concordant cases, and fasting insulin serum levels tended to be higher in the U/S-only group. This is a retrospective study. A referral bias explains the relatively high proportion of patients with phenotype D (28%). PCOM was defined by in-house thresholds. The AMH assay used is no longer commercially available. Our results suggest that ideally both U/S data and serum AMH level should be integrated to define PCOM in the Rotterdam classification. In a cost-effectiveness approach, the choice of one or the other has little impact on the diagnosis and the phenotyping of PCOS. No external funding. The authors have no conflict of interest to declare.

Fraissinet A ，Robin G ，Pigny P ，Lefebvre T ，Catteau-Jonard S ，Dewailly D ... -  《-》

The phenotypic diversity in per-follicle anti-Müllerian hormone production in polycystic ovary syndrome.

Is intrinsic dysregulation of granulosa cells (GC) and consequent increases in the per-follicle production of anti-Müllerian hormone (AMH), correlated with the phenotypic presentation of women with polycystic ovaries? Involvement of intrinsic GC dysregulation in oligo-anovulation associated with polycystic ovary syndrome (PCOS) is likely because among women with PCOS, those with oligo-amenorrhea have higher per-follicle AMH production than those who ovulate normally, irrespective of their androgen and/or metabolic status. Women with PCOS have higher serum AMH level than non-PCOS women due to an increased follicle number and excessive AMH production per follicle, the latter reflecting a putative GC dysfunction that may vary between PCOS phenotypes. This is a retrospective analysis of data collected from 1021 women undergoing infertility evaluation from March 2011 to October 2013. The study included women with polycystic ovarian morphology (PCOM) who met the Rotterdam criteria for PCOS (n = 272), women with PCOM only (n = 168) and controls (n = 581). We used serum AMH to antral follicle count (AFC) ratio (AMH/AFC) as a marker of per-follicle AMH production and checked whether this ratio was associated with the PCOS phenotype and to the menstrual, androgen and metabolic status in women with PCOS, women with PCOM only and in controls. AMH/AFC was significantly higher in oligo-amenorrheic women with PCOS than in eumenorrheic women with PCOS or PCOM (P < 0.001) but also in the latter group compared with controls (P < 0.001) regardless of androgen status. Stepwise discriminant analysis yielded a significant score for the menstrual status with a discriminant power of 26.5% (P < 0.001). This score included AFC, AMH/AFC, waist circumference and LH with partial R(2) of 0.172, 0.042, 0.024 and 0.023, respectively. The AMH to AFC ratio as a surrogate marker for average AMH may be subject to error because follicles below the sensitivity limit of the ultrasonography used may also contribute to serum AMH concentration and secondly, AFC can be subjective. The higher AMH/AFC in women with PCOM only than in controls suggests that isolated PCOM may represent a PCOS-like phenotype in which an inherent dysfunction of GC exists but is too mild to affect the ovulatory process. No funding was obtained for this study. There are no conflicts of interest to be declared. Non-applicable.

Alebić MŠ ，Stojanović N ，Duhamel A ，Dewailly D ... -  《-》

Anti-Mullerian hormone in the diagnosis of polycystic ovary syndrome: can morphologic description be replaced?

Eilertsen TB ，Vanky E ，Carlsen SM 《-》