OBJECTIVE: To investigate which of the currently applied parameters to assess childhood overweight best predict cardiovascular risk factors. DESIGN: Cross-sectional study comparing five different methods to define overweight with respect to their power to predict cardiovascular risk factors. SUBJECTS: A total of 838 healthy children from the Prevention-Education-Program (Nuremberg, Germany; age 4-9 y, 405 boys, 433 girls). MEASUREMENTS: Obesity parameters¾body mass index (BMI), ponderal index (PI), the sum of triceps and subscapular skinfold thickness (SFT), percentage body fat (%BF) using SFT and two different regression formulas (Slaughter, %BF-SL; Dezenberg, %BF-DZ). Overweight defined by the 90th age- and sex-specific percentile of each obesity parameter. Comparison of LDL- and HDL-cholesterol, apolipoprotein-B (apo-B), triglycerides (TG), fibrinogen and blood pressure values (SBP/DBP) between normal-weight and overweight children. RESULTS: When overweight is defined by BMI or PI, all cardiovascular risk factors are significantly (P<0.01) different between overweight and normal-weight children (BMI: TG+20.5%, HDL-chol.-8.6%, LDL-chol.+9.6%, apo-B+6.8%, SBP+7.4%, DBP+8.6%, fibrinogen+13.2%; PI: TG+24.3%, HDL-chol.-6.1%, LDL-chol.+9.0%, apo-B+7.4%, SBP+5.9%, DBP+6.7%, fibrinogen+13.9%), while SFT, %BF-SL and %BF-DZ did not predict all cardiovascular risk factors. A sex-specific analysis showed that in girls BMI and PI both predict cardiovascular risk factors, while in boys this is only valid for BMI. CONCLUSION: In prepubescent children, height-to-weight indices such as BMI or PI better predict cardiovascular risk factors than obesity parameters using skinfold measurement. The BMI may be superior to the PI as the association between BMI and cardiovascular risk factors is less affected by gender. International Journal of Obesity (2001) 25, 830-837 |
, K G Parhofer and P Schwandt