Waist-to-Height Ratio and BMI Predict Different Cardiovascular Risk Factors in Chinese Children

Recently, waist-to-height ratio (WHR) has been documented as a better indicator of obesity and cardiovascular risk than BMI in several different pediatric populations. In the present study, we, for the first time, compared the association of nine cardiovascular risk factors with BMI, waist circumference (WC), and WHR in Chinese children. A random sample of 389 Han and 272 Uygur ethnic children aged 7–18 years (mean ± SE 10.7 ± 3.2) were selected stratified by age, sex, and ethnicity from a school-based cross-sectional study performed in Xinjiang Uygur Autonomous Region (the prevalence of obesity was 2.5 and 5.4% for Uygur and Han, respectively). Uygur and Han are the two major ethnic groups in this region (45% for each, 2003 national census).

Body weight, height, WC, systolic blood pressure (SBP), dystolic blood pressure (DBP), fasting serum triglyceride, total cholesterol, HDL, LDL, glucose, and insulin were measured or determined by standard methods. Homeostasis model assessment (HOMA) was computed by using the formula fasting glucose (mmol/l) × fasting insulin (μU/ml)/22.5.

This study was approved by the local ethics committee. Written informed consent was obtained from each participant or his/her parents. Partial correlations (r) between cardiovascular risk factors and anthropometric indices were calculated after adjusting for age, sex, and ethnicity. We observed that WHR showed the highest correlations with total cholesterol (r = 0.244), HDL (r = 0.163), and LDL (r = 0.231), while BMI showed the highest correlations with SBP (r = 0.272) and DBP (r = 0.104). WC displayed the highest correlations with fasting glucose (r = 0.105). All three indices showed similar correlations with triglycerides (r = 0.308–0.312), fasting insulin (r = 0.224–0.232), and HOMA (r = 0.210–0.229). Linear regression was also performed to calculate the variation of cardiovascular risk factors explained by different anthropometric indices (R²) after adjusting for age, sex, and ethnicity. The results confirmed our findings in the partial correlation analyses. The possible sex difference was also tested by including interaction between sex and anthropometric indices into the regression model. It showed that the effect of the three obesity indices on SBP differed between sexes (P < 0.001 for WC and BMI; P = 0.04 for WHR); however, further analyses in boys and girls separately showed that BMI explained the greater variation of SBP both in boys (R² = 7.9%) and girls (R² = 6.1%), than WC and WHR. Consistent with previous studies (1–3), we conclude that WHR was a better indicator of lipids and BMI was a better indicator of blood pressure in Chinese children. WC was shown as the best indicator of fasting glucose.