腰围/身高比值:预测糖尿病和高血压的有效的腹型肥胖指标

Waist/height ratio: an effective index for abdominal obesity predicting diabetes and hypertension

目的 探讨和评价腰围 /身高比值 (WHtR)作为腹型肥胖指标预测糖尿病、高血压风险的可行性和价值。方法 利用 1998年广东省糖尿病流行病学调查的 11742例人群数据 ,分析WHtR、体重指数 (BMI)、腰围 (WC)和腰臀围比 (WHR )与餐后血糖 ( 2hPG )、血压的相关关系 ;受试者工作特性曲线(ROC)判断最佳切点 ;多元logisitc回归分析比较四个指标对糖尿病、高血压风险的预测价值。结果 ( 1)WHtR与BMI的人群分布特征均无明显性别差异。 ( 2 )WHtR与WC呈高度正相关性 (r =0 .94,P <0 .0 0 0 1) ,WHtR与身高相关性比WC明显减弱 (r =-0 .0 8vsr =0 .2 5 ) ;WHtR与 2hPG、血压的相关性和WC相近。 ( 3 )四个指标中 ,ROC曲线下面积WHtR最大 ,最佳切点为 0 .5 0 ;与WC和WHR相比 ,WHtR有更好的诊断敏感性和特异性 ,尤其在BMI <2 5的人群。 ( 4 )多因素调整后 ,WHtR≥ 0 .5的糖尿病、高血压风险比值比分别为 :1.90 ( 1.3 8～ 2 .63 )和 2 .13 ( 1.77～ 2 .5 6) ,WHtR与BMI是四个指标中最重要的预测因素。结论WHtR计算简单 ,无性别差异 ,与腰围高度相关 ,受身高影响小 ;WHtR≥ 0 .5可有效预测糖尿病、高血压的患病风险。在更进一步的探讨和分析基础上 。

Objective To evaluate the feasibility and value of waist-height ratio (WHtR) as an abdominal obesity index in predicting the risk of diabetes and hypertension. Methods Data came from the 1998 Guangdong provincial diabetes survey, a population-based cross-sectional study, and 11742 residents aged 20-74 years (male 5450, female 6292, age 43.8±13.6 years) were enrolled. Partial correlation analysis of WHtR, waist cincumference (WC), body mass index (BMI), waist/hip ratio (WHR) and postprandial blood glucose (2hPG), systolic and diastolic blood pressure, ROC curve analysis in diabetes and hypertension, and multivariate logistic regression analysis were conducted. Results In presented population, the distribution characteristics of WHtR and BMI were identical in both gender population, not WC and WHR. After adjusted age and sex, WHtR showed positive correlation with WC (r=0.94, P<0.0001) significantly, the correlation between WC and body height was significantly higher than WHtR (r=0.25 vs r=-0.08). The relationship between SBP, DBP, 2hPG and WHtR were similar to WC. Of the four anthropometric indices, WHtR was found to have the largest areas under the ROC curve relative to diabetes(0.751,95%CI 0.727-0.775)and hypertension (0.716,95%CI 0.703-0.729). The best cut-off point of WHtR was 0.5 in diabetes and in hypertension(0.49 in male, 0.50 in female), WC was 78 cm in hypertension( 80 cm in male, 76 cm in female),80 cm in diabetes(82 cm in male, 78 cm in female), WHR was 0.86 in hypertension( 0.87 in male, 0.84 in female), 0.87 in diabetes (0.88 in male,0.85 in female), BMI was 23 in both diseases and gender. The sensitivity and specificity for identification of diabetes and hypertension of WHtR at 0.5 were better than WC and WHR, especially in BMI<25 population. Multivariate logistic regression analysis showed that, after controlling age, history, physical activity, sex, BMI, WHR and WC, the diabetes andhypertensionriskoddsratio(OR)ofWHtR≥0.5 were 1.90(1.38-2.63) and 2.13(1.77-2.56). WHtR and BMI were the most important predictive factors of diabetes and hypertension risk among the four anthropometric indexes. Conclusion WHtR is simple and no significantly difference between male and female, it correlates with WC significantly and keeps its characteristics, and is less affected by body height. WHtR≥0.5 may effectively predict the risk of diabetes and hypertension, its predicting power is closer to or better than WC and WHR. This result suggests that WHtR may be a simple and effective index of abdominal obesity.