Raising threshold for diagnosis of polycystic ovary syndrome excludes population of patients with metabolic risk.

Objective: To characterize the population of patients excluded from a diagnosis of polycystic ovary syndrome (PCOS) when follicle number criteria are increased to 25 per ovary as suggested by the Androgen Excess and Polycystic Ovary Syndrome Society's recent task force.

Design: Cross-sectional study.

Setting: Tertiary academic center.

Patient(s): A total of 259 women with PCOS according to Rotterdam criteria who were systematically examined from 2007 to 2015, with 1,100 ovulatory women participating in the Ovarian Aging (OVA) Study as controls.

Intervention(s): Anthropometric measurements, serum testing, ultrasonic imaging, and comprehensive dermatologic exams.

Main Outcome Measure(s): Body mass index (BMI), waist to hip ratio (WHR), serum cholesterol, fasting glucose and insulin, follicle count per ovary, biochemical hyperandrogenemia, and hirsutism.

Result(s): Forty-seven of 259 women meeting the Rotterdam criteria (18.1%) were excluded from a diagnosis of PCOS when the follicle number criteria was increased to 25. These women had clinical evidence of hyperandrogenism (68.1%) and biochemical hyperandrogenemia (44.7%), although fewer reported oligoanovulation (26.8%). The excluded women had elevated total cholesterol, fasting insulin, and homeostatic model of insulin resistance (HOMA-IR) when compared with controls despite controlling for age and BMI.

Conclusion(s): The women excluded from the PCOS diagnosis by raising the threshold of follicle number per ovary to ≥25 continue to show evidence of metabolic risk.