Anthropometric indices to predict insulin resistance in women with polycystic ovary syndrome in China

Research question: Which anthropometric index (waist-to-height ratio, waist circumference, waist-to-hip ratio and body mass index) is the best in predicting insulin resistance among Chinese women with polycystic ovary syndrome? Design: A total of 1124 patients with polycystic ovary syndrome at the Reproductive Endocrinology Division of West China Second University Hospital of Sichuan University were enrolled in this study. Identification of insulin resistance was based on homeostasis model assessment of insulin resistance scores 2.77 or over. Receiver operator characteristic analysis was carried out using the four anthropometric indices as the continuous variables and insulin resistance as the categorical variable to obtain the areas under the curve. Result: The area under the curve for the waist-to-height ratio (0.748 +/- 0.019) was greater than those for waist circumference (0.739 +/- 0.019), body mass index (0.738 +/- 0.017), and waist-to-hip ratio (0.659 +/- 0.020) in the prediction of insulin resistance. The waist-to-height ratio also had the highest Youden indices compared with those of waist circumference, body mass index, and waist-to-hip ratio; the waist-to-height ratio cut-off was 0.49. Conclusion: The waist-to-height ratio with a cut-off of 0.49 was the most accurate anthropometric indicator for predicting insulin resistance among Chinese women with polycystic ovary syndrome.