The objective was to explore the clinical utility of triglyceride (TG) to high-density lipoprotein cholesterol (HDL-C) ratio in predicting insulin resistance (IR) in 4 ethnic groups and the relationship between IR and TG/HDL-C in comparison to that with other lipid measures. Apparently healthy Aboriginals, Chinese, Europeans, and South Asians (N = 784) were assessed for sociodemographics, lifestyle, anthropometry, lipids, glucose, and insulin. The homeostasis model assessment of IR was used as a measure of IR. Compared with other lipid parameters, TG/HDL-C was the highest correlate of the homeostasis model assessment of IR (age and sex adjusted) in Aboriginals (r = 0.499, P < .001), Chinese (r = 0.432, P < .001), Europeans (r = 0.597, P < .001), and South Asians (0.372, P < .001). For a 1-unit increase in TG/HDL-C, the odds of being insulin resistant increased about 4 times (odds ratio [OR], 3.95; 95% confidence interval [CI], 1.86-8.42; P < .001) in Aboriginals, 3.4 times in Chinese (OR, 3.44; 95% CI, 1.79-6.62; P < .001), 1.9 times in Europeans (OR, 1.94; 95% CI, 1.00-3.75; P = .049), and 1.8 times in South Asians (OR, 1.77; 95% CI, 0.91-3.45; P = .094) (age, sex, smoking, physical activity, body mass index, and waist circumference adjusted). Receiver operating characteristic curve analyses revealed areas under the curve (95% CI) of 0.777 (0.707-0.847) in Aboriginals, 0.723 (0.647-0.798) in Chinese, 0.752 (0.675-0.828) in Europeans, and 0.676 (0.590-0.762) in South Asians. Optimal cutoffs (sensitivity, specificity) of TG/HDL-C for identifying individuals with IR were 0.9 (93.0%, 51.9%), 1.1 (71.7%, 61.5%), 1.1 (73.5%, 70.9%), and 1.8 (52.0%, 77.9%) in Aboriginal, Chinese, European, and South Asian individuals, respectively. The TG/HDL-C ratio may be a good marker to identify insulin-resistant individuals of Aboriginal, Chinese, and European, but not South Asian, origin.
Copyright © 2012 Elsevier Inc. All rights reserved.