Background & aims

Body composition measurement is a valuable tool for assessing nutritional status and physical fitness in a variety of clinical settings. Although bioimpedance analysis (BIA) can easily assess body composition, its accuracy remains unclear. We examined the accuracy of direct segmental multi-frequency BIA technique (DSM-BIA) in assessing different body composition parameters, using dual energy X-ray absorptiometry (DEXA) as a reference standard.

Methods

A total of 484 middle-aged participants from the Leiden Longevity Study were recruited. Agreements between DSM-BIA and DEXA for total and segmental body composition quantification were assessed using intraclass correlation coefficients and Bland-Altman plots.

Results

Excellent agreements were observed between both techniques in whole body lean mass (ICC female = 0.95, ICC men = 0.96), fat mass (ICC female = 0.97, ICC male = 0.93) and percentage body fat (ICC female = 0.93, ICC male = 0.88) measurements. Similarly, Bland-Altman plots revealed narrow limits of agreements with small biases noted for the whole body lean mass quantification but relatively wider limits for fat mass and percentage body fat quantifications. In segmental lean muscle mass quantification, excellent agreements between methods were demonstrated for the upper limbs (ICC female≥0.91, ICC men≥0.87) and lower limbs (ICC female≥0.83, ICC male≥0.85), with good agreements shown for the trunk measurements (ICC female = 0.73, ICC male = 0.70).

Conclusions

DSM-BIA is a valid tool for the assessments of total body and segmental body composition in the general middle-aged population, particularly for the quantification of body lean mass.

Copyright © 2011 Elsevier Ltd and European Society for Clinical Nutrition and Metabolism. All rights reserved.