Abstract: Objective: To evaluate the predictive value of the four anthropometric indexes, including body mass index (BMI), waist circumference (WC), waist-hip ratio (WHR) and waist-height ratio (WHtR), on the insulin resistance (IR) of polycystic ovary syndrome (PCOS). Methods: Retrospective analysis of 754 PCOS patients of childbearing age, receiver operating characteristic  curve (ROC), diagnostic test and binary logistic regression analysis were included in this study. Results: ①The BMI,WC,WHR，WHtR and the homeostasis model of assessment for insulin resistence index (HOMA-IR) in the PCOS with IR group were higher than those in the PCOS with non-IR group (all P＜0.05). ②The AUCs of ROC curves of BMI, WC, WHR and WHtR were 0.768, 0.769, 0.739 and 0.768. The optimal cutoff values of BMI, WC, WHR and WHtR for the prediction of IR in PCOS patients were 26.33 kg/m2, 81.15 cm, 0.85 and 0.53, respectively. ③The binary logistic regression analysis showed that BMI and WHR were risk factors of IR in PCOS patients. The risk of IR in PCOS patients with BMI≥26.33 kg/m2 was 5.189 times higher than that in PCOS patients with BMI＜26.33 kg/m2; and the risk of IR in PCOS patients with WHR≥0.85 was 1.903 times higher than that in PCOS patients with WHR＜0.85. The predictive value of BMI plus WHR was better than any one indicator (AUC of ROC curve was 0.778). Conclusions: Anthropometric indexes have predictive value for the IR of PCOS. WC or WHR can be used to predict IR in PCOS patients with abdominal obesity, and BMI can be used in PCOS patients with systemic obesity. The predictive value of BMI plus WHR is better. When BMI≥26.33 kg/m2 and/or WHR≥0.85, the risk of IR is significantly increased. The appropriate weight management in those obese PCOS patients will reduce their risk of IR.

Key words: Polycystic ovary syndrome, Body mass index;, Waist circumference, Waist-hip ratio, Waist-height ratio