Monozygotic Co-Twin Analyses Of Body Composition Measurements And Serum Lipids

Binyan Wang,Jonathan Necheles,Fengxiu Ouyang,Wei Ma,Zhiping Li,Xue Liu,Jianhua Yang,Houxun Xing,Xiping Xu,Xiaobin Wang
DOI: https://doi.org/10.1016/j.ypmed.2007.07.014
IF: 5.1
2007-01-01
Preventive Medicine
Abstract:Objective. To evaluate BMI and direct measures of body fat (BF) and lean body mass (LBM) in relation to fasting serum lipid profiles in a large Chinese population based twin sample using a monozygotic (MZ) co-twin analysis.Methods. Adiposity measures collected 1998-2000 on 987 MZ female 20-60 year old twin pairs (n=1974) included BMI, waist circumference (WC), waist/hip ratio (WHR), LBM, trunk fat (TF), %TF, total BF, and % total BF (measured by DEXA). Serum lipids included total cholesterol (TC), triglycerides (TG), LDL, and HDL. Co-twin analyses and conventional regression analyses were used to assess the association between individual adiposity and LBM measures, and serum lipids.Results. In this lean population with a mean BMI 21.8 (2.8), we observed considerable variability in adiposity measures and serum lipids. A positive linear association between all adiposity measures with LDL, TC, and TG, and a negative linear association with HDL was observed. A 1-unit z-score increase of adiposity measures, reflecting fat distribution, was associated with increases in (mmol/L) TC (0.063 to 0.164), LDL (0.064 to 0.131), TG (0.049 to 0.164), and a decrease in HDL (0.021 to 0.038) while controlling for matched factors within twin pairs (i.e., age and unmeasured confounders). However, similar associations were not observed for LBM.Conclusion. It is the BF (not LBM) that appears to be associated with serum lipid profiles. This study underscores that in populations where BMI is highly correlated with BF, BMI can be used as a surrogate for BF in evaluating risk of dyslipidemia. Otherwise, direct measures of BF are needed. (c) 2007 Elsevier Inc. All rights reserved.
What problem does this paper attempt to address?