Abstract
OBJECTIVE: We compared various adiposity indices (AIs)-body mass index (BMI), waist circumference (WC), visceral adiposity index (VAI), lipid accumulation product, and the triglyceride-glucose (TyG), TyG-BMI, and TyG-WC indices-regarding their associations with insulin resistance (IR) among women with polycystic ovary syndrome (PCOS). METHODS: This cross-sectional investigation included 632 reproductive-aged women with PCOS, aligning with the 2018 International Evidence-based Guideline. Clinical characteristics, including height, weight, and WC, were recorded. Fasting blood samples were analyzed to measure insulin, lipid, and glucose levels. AIs were then calculated using these measurements along with various formulas. The homeostatic model assessment for IR was utilized, with a value of 2.6 or higher indicating IR. An area under the curve (AUC) above 0.8 was considered clinically useful. RESULTS: Approximately 69.1% of participants exhibited IR. All AIs were significant predictors of IR, regardless of androgen status. VAI and the TyG index were identified as poor markers of IR among women with PCOS (AUC, 0.6 to 0.69), while the remaining indices were considered fair markers (AUC, 0.7 to 0.79). VAI underperformed all other AIs except the TyG index, which was notably inferior to the TyG-BMI and TyG-WC indices. Nevertheless, the AUCs of TyG-BMI and TyG-WC were similar to those of BMI and WC. CONCLUSION: Although AIs exhibit androgen-independent relationships with IR, their usefulness as indicators of IR in PCOS is not adequately supported by our results (AUC <0.8). Furthermore, combining the TyG index with BMI and WC does not significantly improve the diagnostic utility of IR.