PT30. - BODY FAT ESTIMATES BY BIOELECTRICAL IMPEDANCE ANALYSIS IN COMPARISON WITH DUAL‑ENERGY X‑RAY ABSORPTIOMETRY: A CROSS-SECTIONAL STUDY IN COMMUNITY ADULTS FROM THE NUTRIFUNCTION PROJECT

PT30.

BODY FAT ESTIMATES BY BIOELECTRICAL IMPEDANCE ANALYSIS IN COMPARISON WITH DUAL‑ENERGY X‑RAY ABSORPTIOMETRY: A CROSS-SECTIONAL STUDY IN COMMUNITY ADULTS FROM THE NUTRIFUNCTION PROJECT

R. S. Guerra^1,2,3,*, M. Correira⁴, M. Rodrigues^2,5, A. Cardoso⁵, A. S. Sousa^1,3,6, J. Mendes^1,3, R. Valdiviesso^5,7, A. R. Sousa-Santos^8,9,10, C. Silva^1,3, N. Borges^5,7, A. Santos⁵, T. F. Amaral^2,5

¹FP-I3ID, FP-BHS, Faculty of Health Sciences, Fernando Pessoa University, ²INEGI – Institute of Science and Innovation in Mechanical and Industrial Engineering, LAETA - Associate Laboratory for Energy, Transports and Aerospace, ³RISE-Health, Fernando Pessoa University, Fernando Pessoa Teaching and Culture Foundation, ⁴Faculty of Health Sciences, Fernando Pessoa University, ⁵Faculty of Nutrition and Food Sciences, University of Porto, Porto, ⁶Center for Innovative Care and Health Technology (ciTechcare), Polytechnic of Leiria, Leiria, ⁷RISE-Health, Faculty of Nutrition and Food Sciences, University of Porto, Porto, ⁸University Institute of Health Sciences (IUCS - CESPU), ⁹Associate Laboratory i4HB - Institute for Health and Bioeconomy, University Institute of Health Sciences - CESPU, ¹⁰UCIBIO - Applied Molecular Biosciences Unit, Forensics and Biomedical Sciences Research Laboratory, University Institute of Health Sciences (1H-TOXRUN, IUCS-CESPU), Gandra, Portugal

Rationale: The multiplicity of Bioelectrical Impedance Analysis (BIA) devices, of equations based on impedance parameters for body composition estimates and heterogeneous findings hinder its use. This study aims to compare 4 BIA equations with DXA, in adults. 

Methods: A cross-sectional study was conducted in community adults. Information on sex, age, weight, stature, resistance and reactance was collected. Fat‑free mass was estimated with the equations of Kyle, Sun, Deurenberg and Lukaski; fat mass and % body fat (%BF) were calculated. %BF was determined by Hologic Horizon DXA. Differences between %BF by DXA and each equation were calculated and assessed with Wilcoxon test, for the whole sample, stratified by sex, by sex and age, and by sex and WHO BMI categories. Results are presented as median and interquartile range.

Results: Sample was composed of 667 participants (18-85 y, 67% women). ImpediMed SFB7 was used on 81% and Akern Quantum/S on 19%. %BF by DXA was 29.3(12.8), 33.1(10.3) for women and 21.8(10.0) for men. Kyle_Eq provided the smallest difference compared to DXA for total sample [0.00(4.56), p=0.351], when stratified by sex [women: ‑0.13(4.48), p=0.813; men: 0.09(4.40), p= 0.058)] and when stratified by sex and age classes, except for 18-24y women for whom Lukaski_Eq provided the smallest difference: -0.34(4.19), p=0.542. Kyle_Eq performed better for normal [0.28(4.31), p=0.093], pre-obese [0.39(5.34), p=0.269] and obese [-0.32(4.59), p=0.709] men, for normal [‑0.44(4.27), p=0.027] and pre-obese [0.84(3.68), p=0.004] women; Lukaski_Eq was better for underweight [1.03(3.75), p=0.865] and obese [‑0.01(4.14), p=0.873] women.

Conclusion: Kyle equation provided the best estimates compared to DXA. However, sex, age and BMI should be considered when choosing the equation to estimate %BF. UIDB/50022/2020

References:

Disclosure of Interest: None declared