Share this article on:

Validation Of A Multi-electrode Bioelectrical Impedance Analyzer With A Dual-energy X-ray Absorptiometer: 3668 Board #115 June 3 8: 00 AM - 9: 30 AM

Meier, Nathan1; Bai, Yang2; Lee, Duck-chul FACSM1

Medicine & Science in Sports & Exercise: May 2017 - Volume 49 - Issue 5S - p 1050
doi: 10.1249/01.mss.0000519883.22275.c8
G-33 Free Communication/Poster - Research Methods Saturday, June 3, 2017, 7:30 AM - 11:00 AM Room: Hall F

1Iowa State University, Ames, IA. 2University of Vermont, Burlington, VT. (Sponsor: Duck-chul Lee, FACSM)


Reported Relationships: N. Meier: Contracted Research - Including Principle Investigator; Biospace Co, Ltd.

Sarcopenia, the loss of muscle mass, strength, and function due to ageing, is a major health concern for the growing older adult population. One challenge for prevention, diagnosis, and treatment of sarcopenia is the need for a dual-energy X-ray absorptiometry (DXA) analyzer to measure appendicular lean mass (ALM). DXA is the recommended measurement of muscle mass for sarcopenia, but is expensive and exposes subjects to radiation. Therefore, inexpensive, safe, and widely available alternative measurements, such as bioelectrical impedance analysis (BIA), need to be identified and validated to be practically utilized in clinical settings.

PURPOSE: The purpose of this study is to validate the multi-frequency BIA with 8 tactile electrodes (InBody 720) with the gold-standard DXA scan (Hologic Horizon W).

METHODS: Participants were 277 older adults from 65 to 96 years old without history of cancer and severe medical or mental conditions. Individuals completed a 12-hour fast, refrained from activity that morning and wore scrubs. BIA and DXA analyses were taken immediately after each other.

RESULTS: Correlation between the two methods for fat free mass (FFM), ALM, and percent body fat (PBF) were 0.93, 0.86 and 0.92, respectively, after adjusting for age and sex. Mean Percent Error (MPE) (DXA - InBody) and Mean Absolute Percent Error (MAPE), measures of prediction accuracy, were -13% and 13% for FFM, -12% and 13% for ALM, and 16% and 17% for PBF. Prediction equations were developed for improved estimation, in which age was coded in years and sex was coded as 1 for male and 0 for female:

DXA FFM= 0.83 (BIA FFM) + 0.025 (Age) + 2.0 (Sex) + 0.36 (R2=0.96)

DXA ALM= 0.74 (BIA ALM) - 0.025 (Age) + 1.84 (Sex) + 4.15 (R2=0.92)

DXA PBF= 0.71 (BIA PBF) - 0.089 (Age) - 3.3 (Sex) + 23.5 (R2=0.91)

DISCUSSION: The BIA body composition variables are highly correlated with DXA variables. However, we found consistent overestimation of FFM and ALM and underestimation of PBF in BIA compared to DXA based on MPE and MAPE analyses, which were incorporated in the development of FFM, ALM, and PBF estimation equations.

Supported by unrestricted research grant by Biospace Co, Ltd.

© 2017 American College of Sports Medicine