Relationship between indices of obesity obtained by anthropometry and dual-energy X-ray absorptiometry: The Fourth and Fifth Korea National Health and Nutrition Examination Survey (KNHANES IV and V, 2008–2011). Issue 5 (September 2015)

• Record Type: Journal Article
• Authors: Kim, Seul Gi; Ko, Ki dong; Hwang, In Cheol; Suh, Heuy Sun; Kay, Shelley; Caterson, Ian; Kim, Kyoung Kon
• Is Part Of: Obesity research & clinical practice. Volume 9:Issue 5(2015:Sep./Oct.)
• Page Start: 487
• Page End: 498
• Publication Date: 2015-09
• Subjects: Dual-energy X-ray absorptiometry -- Body mass index -- Waist circumference -- Body composition -- Adiposity; Obesity -- Research -- Periodicals; Obesity -- Treatment -- Periodicals; Obesity -- Periodicals; Obésité -- Recherche -- Périodiques; Obésité -- Traitement -- Périodiques; Obesity -- Research; Obesity -- Treatment; Electronic journals; Periodicals; 616.398
• DOI: 10.1016/j.orcp.2014.11.002
• Languages: English
• ISSNs: 1871-403X

Abstract

Summary: Background: Body mass index (BMI), waist circumference (WC), and even dual-energy X-ray absorptiometry (DXA) are used for obesity diagnosis. However, it is not known which DXA-derived index of obesity correlates best with BMI and/or WC and it is not clear whether such an index is accurate or not. Objectives: The aim of this study is to show the relationship between anthropometric measurements (BMI, WC) and body fat indices from DXA and to determine which DXA indices are strongly related to BMI and WC. Subjects: This study was based on data obtained from the Fourth and Fifth Korea National Health and Nutrition Examination Survey (KNHANES IV-V). DXA measurements were performed on survey subjects over 10 years old from July 2008 through to May 2011. Of these, 18 198 individuals, aged 19 years and older for whom DXA data were available, were included. Methods: Weighted Pearson's correlated coefficients ( r ) were calculated among indices, according to sex, age group and menopause, and the coefficients were compared with each other. Results: BMI correlates most with trunk body fat mass in kg ( r = 0.831) and then with total body fat in kg ( r = 0.774, P < 0.00043 for difference of r ). In the older age group, BMI correlates with total body fat mass ( r = 0.822) better than with trunk fat mass ( r = 0.817, P < 0.00043). WC correlates with trunk body fat mass most in both genders and all age groups (0.804 ≤ r ≤ 0.906). Correlations of BMI ( r = 0.645 for men, 0.689 for women) and WC ( r = 0.678 for men, 0.634 for women) to body fat percentages (%) were less robust than those to body fat mass. Conclusions: BMI and WC reflect trunk and total body fat in kg more than body fat percentage derived by DXA.