Comparison of multi- and single-frequency bioelectrical impedance analysis with dual-energy X-ray absorptiometry for assessment of body composition in post-menopausal women: effects of body mass index and accelerometer-determined physical activity

Abstract

Background: Bioelectric impedance analysis (BIA) is commonly used in research to assess body composition. However, studies that validate the accuracy of BIA exclusively in post-menopausal women are lacking. The main purpose of the present study was to evaluate the agreement of multi-frequency (MF)-BIA and single-frequency (SF)-BIA with dual-energy X-ray absorptiometry (DXA) in the estimation of fat mass (FM) and fat-free mass (FFM) among post-menopausal women with variation in body mass index (BMI) and physical activity (PA).

Methods: FM and FFM were estimated by BIA and DXA in 146 post-menopausal women with a mean (SD) age of 62.8 (5.2) years. PA was determined by an accelerometer.

Results: The mean (SD) difference between MF-BIA and DXA was -1.8 (1.8) kg (P = 0.08) and 1.3 (1.8) kg (P = 0.01) for FM and FFM, respectively. SF-BIA provided a significantly lower estimate of FM [-2.0 (2.2) kg; P = 0.04] and a higher estimate of FFM [1.8 (2.4) kg; P < 0.01] compared to DXA. MF-BIA provided significantly better estimates of FM and FFM with narrower limits of agreement than SF-BIA in obese and insufficiently active subjects. In other BMI and PA groups, both BIA devices showed a similar deviation from DXA.

Conclusions: BIA tends to underestimate FM and overestimate FFM relative to DXA. MF-BIA appears to be a more appropriate method for the assessment of body composition than SF-BIA in post-menopausal woman with BMI >30 kg/m(2) and in those who are insufficiently active.

Keywords: Bland-Altman analysis; fat mass; fat-free mass; moderate physical activity; obesity; validity.