Abstract P105: Early Adolescent Waist Size and Body Mass Index are Stronger Predictors of Lipid Levels After Ten Years Than Percent Body Fat in Black and White Girls

Abstract

Cardiometabolic risk begins in childhood and poses additional hazards for later chronic disease. In this study, we assessed the hypothesis that pre-adolescent measures of body fat (ages 9-10 years) would predict later adolescent lipid levels (ages 17-20 years) and that racial differences may exist in the efficacy of particular measures of body fat for predicting future lipid levels. Data from the NHLBI’s Growth and Health Study were used to evaluate the independent effects of early adolescent body mass index (BMI), waist circumference (WC), and percent body fat (%BF) - all distinct measures of body composition - in this cohort of Caucasian (white) and African-American (black) girls, initially 9-10 years of age. Girls with complete data (n=634, white; n=738, black) for body fat, relevant confounders, and follow-up lipid levels (i.e., HDL, LDL, and triglycerides) at 17-20 years of age were included. In late adolescence, there were statistically significant trends of increasing LDL and triglycerides and decreasing HDL levels (p<0.05 for all) across quintiles of all body fat measures at 9-10 years of age. By 17-20 years, white girls (vs. black girls) in the highest BMI quintile had slightly higher mean LDL levels (112.7 vs. 105.1), lower HDL levels (48.0 vs. 51.4), and much higher triglyceride levels (109.8 vs. 76.8 mg/dL). After adjusting for age, SES, height, age at menarche, physical activity, television viewing, fruit and vegetable consumption, and dietary fiber, WC was the strongest predictor of late-adolescent triglycerides and HDL levels in both whites and blacks while BMI was an equally-good predictor of LDL. Percent body fat using bioelectrical impedance analysis was generally a weaker predictor of lipid outcomes. Girls in the highest quintile (Q5) of WC had mean triglyceride levels that were 31.1 and 14.9 mg/dL higher than those in the lowest quintile (Q1), for whites and blacks, respectively; mean differences (Q5-Q1) in TG associated with BMI were 27.7 and 12.0 mg/dL, and for %BF were 13.6 and 7.2 mg/dl, respectively. White and black girls, respectively, in Q5 of WC at 9-10 years of age had HDL levels that were 7.8 mg/dL lower in late adolescence than those of girls in Q1. These results suggest that WC and BMI as simple anthropometric measures of body fat in pre-adolescent girls may be useful in identifying girls who are at risk for dyslipidemia by the time of later adolescence.