BACKGROUND The diagnosis of polycystic ovary syndrome (PCOS) relies on clinical, biological and morphological criteria. With the advent of ultrasonography, follicle excess has become the main aspect of polycystic ovarian morphology (PCOM). Since 2003, most investigators have used a threshold of 12 follicles (measuring 2-9 mm in diameter) per whole ovary, but that now seems obsolete. An increase in ovarian volume (OV) and/or area may also be considered accurate markers of PCOM, yet their utility compared with follicle excess remains unclear. METHODS Published peer-reviewed medical literature about PCOM was searched using PubMed.gov online facilities and was submitted to critical assessment by a panel of experts. Studies reporting antral follicle counts (AFC) or follicle number per ovary (FNPO) using transvaginal ultrasonography in healthy women of reproductive age were also included. Only studies that reported the mean or median AFC or FNPO of follicles measuring 2-9 mm, 2-10 mm or <10 mm in diameter, or visualized all follicles, were included. RESULTS Studies addressing women recruited from the general population and studies comparing control and PCOS populations with appropriate statistics were convergent towards setting the threshold for increased FNPO at ≥25 follicles, in women aged 18-35 years. These studies suggested maintaining the threshold for increased OV at ≥10 ml. Critical analysis of the literature showed that OV had less diagnostic potential for PCOM compared with FNPO. The review did not identify any additional diagnostic advantage for other ultrasound metrics such as specific measurements of ovarian stroma or blood flow. Even though serum concentrations of anti-Müllerian hormone (AMH) showed a diagnostic performance for PCOM that was equal to or better than that of FNPO in some series, the accuracy and reproducibility issues of currently available AMH assays preclude the establishment of a threshold value for its use as a surrogate marker of PCOM. PCOM does not associate with significant consequences for health in the absence of other symptoms of PCOS but, because of the use of inconsistent definitions of PCOM among studies, this question cannot be answered with absolute certainty. CONCLUSIONS The Task Force recommends using FNPO for the definition of PCOM setting the threshold at ≥25, but only when using newer technology that affords maximal resolution of ovarian follicles (i.e. transducer frequency ≥8 MHz). If such technology is not available, we recommend using OV rather than FNPO for the diagnosis of PCOM for routine daily practice but not for research studies that require the precise full characterization of patients. The Task Force recognizes the still unmet need for standardization of the follicle counting technique and the need for regularly updating the thresholds used to define follicle excess, particularly in diverse populations. Serum AMH concentration generated great expectations as a surrogate marker for the follicle excess of PCOM, but full standardization of AMH assays is needed before they can be routinely used for clinical practice and research. Finally, the finding of PCOM in ovulatory women not showing clinical or biochemical androgen excess may be inconsequential, even though some studies suggest that isolated PCOM may represent the milder end of the PCOS spectrum.
Keywords: Anti-Müllerian hormone; antral follicle count; follicle number per ovary; ovarian volume; polycystic ovaries.