Unravelling the phenotypic map of polycystic ovary syndrome (PCOS): a prospective study of 634 women with PCOS

Background The phenotypic spectrum of PCOS has been broadened but the prevalence and clinical significance of PCOS phenotypes continue to challenge the scientific community. Objective Analysis of the phenotypic spectrum of PCOS and determination of the association between metabolic, hormonal and new ultrasonographic criteria. Design Clinical prospective study. Patients Six hundred and thirty-four women with PCOS (18-35 years) and comparable body mass index (BMI), diagnosed by ESHRE/ASRM criteria, were categorized into two major groups of phenotypes: classic (NIH) and nonclassic (ESHRE/ASRM), and then subdivided into phenotypes with all possible combinations of diagnostic characteristics. One hundred and eight healthy women of reproductive age, matched for BMI, were recruited as the control group. MeasurementsUltrasonographic (ovarian follicle number and volume), hormonal and metabolic parameters. ResultsClassic PCOS phenotype was more frequent (85.96%) than the nonclassic phenotype (14.04%). The patients were also classified according to the presence of biochemical (80.60%) or clinical hyperandrogenaemia (12.62%), and, in the newly introduced group, to absence of evidence of hyperandrogenaemia (6.78%). Subjects with classic PCOS with biochemical hyperandrogenaemia (74.76%) were more insulin resistant than controls. In women with PCOS, follicle number was positively related to both insulin resistance and biochemical hyperandrogenaemia. Additionally, women with PCOS had a 6.58-fold increased relative risk (RR) of > 12 follicles on intravaginal ultrasound and a 2.94-fold increased RR of ovarian volume > 10 cm(3) compared to controls. Conclusion The most frequent PCOS phenotype is the subgroup with biochemical hyperandrogenaemia only within the classic phenotype, which is more insulin resistant than the control group. The follicle number, a newly introduced ultrasonographic marker, is related to both metabolic and hormonal features of PCOS.