Assessment of insulin resistance in lean women with polycystic ovary syndrome

Abstract

Objective: To develop and validate a specific simple measure of insulin sensitivity using oral glucose tolerance test (OGTT) values for lean polycystic ovary syndrome (PCOS) women.

Design: Retrospective study.

Setting: Gynecologic Outpatient Clinic of University Hospital, affiliated with Unit of Gynecologic Endocrinology.

Patient(s): Totals of 201 lean and 198 overweight/obese (ov-ob) nondiabetic PCOS patients were retrospectively selected.

Intervention(s): None.

Main outcome measure(s): All patients underwent OGTT, euglycemic-hyperinsulinemic clamp, and androgenic and biochemical assays. The predictive performance of each insulin resistance (IR) index was analyzed with the use of receiver operating characteristic (ROC) curves.

Result(s): Higher correlation coefficients with clamp studies were obtained with the Belfiore Area (RS=0.579) and the homeostasis-model assessment (HOMA)-M120 (RS=-0.576) in lean PCOS patients and with the Sib (RS=0.697) in ov-ob PCOS patients. The best predictive index of IR in lean PCOS was a HOMA-M120 value of ≥12.8 or more (area under the ROC curve [AUC] 92.4%). In the ov-ob PCOS population, the best predictive performance was obtained by a Sib of ≤10.2 or less (AUC 85.7%).

Conclusion(s): IR should be assessed in all PCOS women, both lean and ov-ob subjects. The HOMA-M120 resulted as a very simple tool, validated specifically for the lean PCOS woman whose cardiometabolic impairment is more frequently misunderstood.

Keywords: HOMA-M(120); Polycystic ovary syndrome; clamp; insulin resistance.