Résumé
<p><b>OBJECTIVE</b>To explore the appropriate waist-to-hip ratio (WHR) cutoffs to identify people at high risk of cardiovascular disease of Uygur population aged 35 years and over in Xinjiang.</p><p><b>METHODS</b>The cardiovascular risk survey (CRS) in Xinjiang was conducted from October 2007 to March 2010, using 4-stagestratified random sampling method and 14 618 representative participated this survey, and the questionnaire survey, anthropometric data, blood pressure, serum total cholesterol, triglyceride, low density lipoprotein (LDL-C), high density lipoprotein (HDL-C) and fasting glucose were measured. A total of 4 657 participants aged 35 years and over with complete anthropometric data were analyzed. The sensitivity, specificity and distance on the receiver operating characteristic (ROC) curve of different WHR levels predicting risk factors of cardiovascular disease were calculated. The analysis method of ROC curve was used to determine the optimum cut-off point of WHR predicting risk factors of cardiovascular disease.</p><p><b>RESULTS</b>(1) There were significantly differences in prevalence of hypertension, diabetes mellitus, hypercholesterolemia, low HDL-C level, and hypertriglyceridemia between WHR < 0.75,0.75 ≤ WHR < 0.80,0.80 ≤ WHR < 0.85,0.85 ≤ WHR < 0.90,0.90 ≤ WHR < 0.95,0.95 ≤ WHR < 1.00, WHR ≥ 1.00 in male participants (P < 0.01 or 0.05), LDL-C level was similar among groups in males (P = 0.139). There were significantly differences in prevalence of hypertension, diabetes mellitus, hypercholesterolemia and hypertriglyceridemia between WHR < 0.75,0.75 ≤ WHR < 0.80,0.80 ≤ WHR < 0.85,0.85 ≤ WHR < 0.90,0.90 ≤ WHR < 0.95,0.95 ≤ WHR < 1.00, WHR ≥ 1.00 in female participants (all P < 0.01), and there were no significantly differences in prevalence of high LDL-C level and low HDL-C level among groups in females (both P > 0.05). (2) ROC analysis for hypertension, dyslipidemia, diabetes and ≥ 2 of these risk factors suggested a WHR cutoff of 0.92 for men and 0.90 for women as the optimal cutoff value for predicting high risk of cardiovascular disease of Uygur population aged 35 years and over in Xinjiang.</p><p><b>CONCLUSION</b>Higher WHR cutoffs are needed for screening people at high risk of cardiovascular disease among Uygur population aged 35 years and over in Xinjiang.</p>