Accurate screening for insulin resistance in PCOS women using fasting insulin concentrations

The aims of this cross-sectional study were to evaluate the relative agreement of both static and dynamic methods of diagnosing IR in women with polycystic ovary syndrome (PCOS) and to suggest a simple screening method for IR. All participants underwent serial blood draws for hormonal profiling and lipid assessment, a 3 h, 75 g load oral glucose tolerance test (OGTT) with every 15 min measurements of glucose and insulin, and an ACTH stimulation test. The prevalence of IR ranged from 12.2% to 60.5%, depending on the IR index used. Based on largest area under the curve on receiver operating curve (ROC) analyses, the dynamic indices outperformed the static indices with glucose to insulin ratio and fasting insulin (fInsulin) demonstrating the best diagnostic properties. Applying two cut-offs representing fInsulin extremes (<7 and >13 mIU/l, respectively) gave the diagnosis in 70% of the patients with high accuracy. Currently utilized indices for assessing IR give highly variable results in women with PCOS. The most accurate indices based on dynamic testing can be time-consuming and labor-intensive. We suggest the use of fInsulin as a simple screening test, which can reduce the number of OGTTs needed to routinely assess insulin resistance in women with PCOS.