Assessment of body composition by dual-energy X-ray absorptiometry, bioimpedance analysis and anthropometrics in children: the Physical Activity and Nutrition in Children study

TY - JOUR

T1 - Assessment of body composition by dual-energy X-ray absorptiometry, bioimpedance analysis and anthropometrics in children

T2 - the Physical Activity and Nutrition in Children study

AU - Tompuri, Tuomo T

AU - Lakka, Timo A

AU - Hakulinen, Mikko

AU - Lindi, Virpi

AU - Laaksonen, David E

AU - Oskari Kilpeläinen, Tuomas

AU - Jääskeläinen, Jarmo

AU - Lakka, Hanna-Maaria

AU - Laitinen, Tomi

N1 - © 2013 Scandinavian Society of Clinical Physiology and Nuclear Medicine. Published by John Wiley & Sons Ltd.

PY - 2015

Y1 - 2015

N2 - Objective and methods: We compared InBody720 segmental multifrequency bioimpedance analysis (SMF-BIA) with Lunar Prodigy Advance dual-energy X-ray absorptiometry (DXA) in assessment of body composition among 178 predominantly prepubertal children. Segmental agreement analysis of body compartments was carried out, and inter-relationships of anthropometric and other measures of body composition were defined. Moreover, the relations of different reference criteria for excess body fat were evaluated. Results: The prevalence of excess body fat varies greatly according to the used criteria. Intraclass and Pearson's correlations between SMF-BIA and DXA were >0·92 in total body and >0·74 in regional measures. SMF-BIA underestimated percentage body fat (%BF) and fat mass (FM), and overestimated lean mass (LM) and percentage LM with significant offset trend bias. Higher adiposity increased offsets, and overall agreement was poorer in girls. On average, %BF offsets (girls/boys) and limits of agreement (LA) were 3·9/1·6% [(−)1·4–9·2%/(−)3·4–6·7%]. Interestingly percentage offsets of fat content (%BF: 18·9/10·1%, FM: 18·8/11·1%) showed no significant bias trends indicating that the corresponding absolute methodological offset depends on the amount of fat content. The smallest percentage offset was found with LM: 4·3/0·1%, referring offset (LA) of 0·88/0·03 kg (±2·05/±1·71 kg). Correspondingly, segmental LM had poorer agreement than total body LM. All anthropometrics except for the waist-to-hip ratio showed strong correlations (r = 0·76–0·95) with abdominal and total body fat. Conclusion: Segmental multifrequency bioimpedance analysis is precise enough for total-LM analysis and had also sufficient trueness for total body composition analysis to be used in epidemiological purposes. There is need to generate scientifically and clinically relevant criteria and reference values for excess body fat.

AB - Objective and methods: We compared InBody720 segmental multifrequency bioimpedance analysis (SMF-BIA) with Lunar Prodigy Advance dual-energy X-ray absorptiometry (DXA) in assessment of body composition among 178 predominantly prepubertal children. Segmental agreement analysis of body compartments was carried out, and inter-relationships of anthropometric and other measures of body composition were defined. Moreover, the relations of different reference criteria for excess body fat were evaluated. Results: The prevalence of excess body fat varies greatly according to the used criteria. Intraclass and Pearson's correlations between SMF-BIA and DXA were >0·92 in total body and >0·74 in regional measures. SMF-BIA underestimated percentage body fat (%BF) and fat mass (FM), and overestimated lean mass (LM) and percentage LM with significant offset trend bias. Higher adiposity increased offsets, and overall agreement was poorer in girls. On average, %BF offsets (girls/boys) and limits of agreement (LA) were 3·9/1·6% [(−)1·4–9·2%/(−)3·4–6·7%]. Interestingly percentage offsets of fat content (%BF: 18·9/10·1%, FM: 18·8/11·1%) showed no significant bias trends indicating that the corresponding absolute methodological offset depends on the amount of fat content. The smallest percentage offset was found with LM: 4·3/0·1%, referring offset (LA) of 0·88/0·03 kg (±2·05/±1·71 kg). Correspondingly, segmental LM had poorer agreement than total body LM. All anthropometrics except for the waist-to-hip ratio showed strong correlations (r = 0·76–0·95) with abdominal and total body fat. Conclusion: Segmental multifrequency bioimpedance analysis is precise enough for total-LM analysis and had also sufficient trueness for total body composition analysis to be used in epidemiological purposes. There is need to generate scientifically and clinically relevant criteria and reference values for excess body fat.

U2 - 10.1111/cpf.12118

DO - 10.1111/cpf.12118

M3 - Journal article

C2 - 24325400

SN - 1475-0961

VL - 35

SP - 21

EP - 33

JO - Clinical Physiology and Functional Imaging

JF - Clinical Physiology and Functional Imaging

IS - 1

ER -