Using metabolic markers to identify insulin resistance in premenopausal women with and without polycystic ovary syndrome.

Background: Insulin resistance (IR) is associated with increased risk for type 2 diabetes mellitus and cardiovascular disease. Quantifying IR is invasive and time-consuming, and thus not routinely used in clinical practice. Simple metabolic markers to predict IR exist, but have not been validated in premenopausal women or women with polycystic ovary syndrome (PCOS).

Objective: To evaluate the ability of metabolic markers to identify premenopausal women with/without PCOS who are insulin resistant.

Design/setting: Cross-sectional analysis.

Participants: One hundred and seventy-one non-diabetic premenopausal overweight/obese women without PCOS and 71 women with PCOS.

Methods: IR was quantified by the steady-state plasma glucose during the modified insulin-suppression test. Metabolic markers (BMI, lipid/lipoprotein concentrations, and fasting glucose) were evaluated for their discriminative ability to identify IR, using area under the receiver-operating-characteristic curve (AUROC) analysis. Optimal cut-points were evaluated for predictive power.

Results: In the non-PCOS group, the triglyceride/HDL cholesterol ratio (TG/HDL-C) was the best marker (AUROC 0.73). Optimal diagnostic cut-point was 1.9. In the PCOS group, the TG/HDL-C ratio, cholesterol/HDL-C ratio (TC/HDL-C), and HDL-C performed well (AUROC > 0.80), with optimal cut-points for TG/HDL-C 1.3, TC/HDL-C 3.4, and HDL-C 52 mg/dL: TG/HDL-C was more sensitive, but HDL-C had a higher PPV for IR.

Conclusion: TG/HDL-C can identify IR in premenopausal women with and/without PCOS; diagnostic cut-points differ from those of men and postmenopausal women. HDL-C is an alternative predictor in women with PCOS. These simple metabolic markers, which are standardized between labs, inexpensive, and routinely measured, can be used to tailor lifestyle and medical interventions to improve health outcomes in insulin-resistant premenopausal women.