Background Polycystic ovary syndrome (PCOS) is considered predominantly as a hyperandrogenetic syndrome and the evaluation of metabolic parameters and insulin sensitivity is not mandatory.
Context PCOS diagnostic criteria [National Institute of Health (NIH), Rotterdam Consensus (ROT), Androgen Excess Society (AES)] are unanimous recognized. We aimed to assess in women with suspected PCOS whether the application of the three diagnostic criteria differently characterizes the metabolic profile and insulin sensitivity.
Design Retrospective study in a cohort of women admitted to our Outpatient Clinic for suspected PCOS.
Patients Two hundred and four women with suspected PCOS in comparison to a group of normal, age-matched Sicilian women (N = 34) without signs of metabolic syndrome.
Measurements We evaluated hyperandrogenaemia and clinical hyperandrogenism, ovarian morphology, hypothalamo–hypophyseal axis and metabolic syndrome parameters. An oral glucose tolerance test (OGTT; 75 g glucose) measured areas under the curve (AUC) for insulin, C peptide and homeostasis model assessment of insulin-resistance (HOMA-IR) were performed.
Results The prevalence of PCOS was 51% according to NIH, 83% to ROT and 70·6% to AES, and only 100 patients were qualified simultaneously under these three criteria. The prevalence of the metabolic syndrome in PCOS women was 26·92% (NIH), 21·77% (ROT) and 23·61% (AES), respectively. In comparison to healthy women, PCOS women showed increased fasting insulinaemia (PCOS/ROT: P = 0·028; PCOS/NIH: P = 0·007; PCOS/EAS: P = 0·023), 120 min insulin after OGTT insulinaemia (for the three criteria: P < 0·001), AUC2h insulin (for the three criteria: P < 0·001) and AUC2h C peptide (for the three criteria: P < 0·001).
Conclusions Our study highlights the fact that regardless of the diagnostic criteria used, evaluation of the metabolic parameters and insulin sensitivity is important for a correct diagnosis of PCOS and a therapeutic approach.