Hypoadiponectinaemia enhances waist circumference as a predictor of glucose intolerance and clustering of risk factors in Chinese men.

This study aimed to confirm the hypothesis that adiponectin levels can enhance the discriminative value of waist circumference (WC) in predicting impaired glycaemic status and clustering of risk factors.We used receiver operating characteristic (ROC) curve analysis to define the optimal cut-off value of adiponectin to predict diabetes in Chinese men with no relevant past medical history. This value was combined with WC to increase its discriminative power in ascertaining impaired glycaemic status and various cardiovascular risk factors.In 360 men (mean+/-S.D.; age: 41.3+/-9.2 years), the mean+/-S.D. adiponectin level was 5.2+/-2.7 microg/mL. Based on oral glucose tolerance tests (OGTTs), 84 men (23.3%) had undiagnosed diabetes, 52 (14.4%) had impaired glucose tolerance (IGT) and 224 (62.3%) had normal glucose tolerance. On ROC analysis, 5.7 microg/mL was the optimal cut-off value of adiponectin in this population to predict diabetes. Compared with subjects who had normal WC (defined as less than 90 cm) and high adiponectin levels (> or =5.7 microg/mL), the likelihood ratio of diabetes was 2.54 in those with central obesity and hypoadiponectinaemia.The combined use of low adiponectin levels and large WC measures has greater discriminative power than using either index alone to identify subjects at particular risk of glucose intolerance and clustering of risk factors.