Prediction of whole-body and segmental body composition by bioelectrical impedance in morbidly obese subjects

Obes Surg. 2012 Apr;22(4):587-93. doi: 10.1007/s11695-011-0570-3.

Abstract

Background: Validated equations for body composition analysis using bioelectrical impedance (BIA) in morbidly obese (MO) subjects are scarce. Thus, our aim was todevelop new equations from physical and BIA parameters to estimate whole-body and segmental body composition inMO subjects, with dual-energy X-ray absorptiometry(DXA) as the reference method.

Methods: A cross-sectional study on 159 Caucasian MO subjects (female 78%, age 43.5 ± 11.8 years, BMI 45.6 ± 4.9 kg/m2) divided in two groups was conducted: model building cohort (n = 110) and model validation cohort (n 0 49). Stepwise regression analysis was used to develop specific fat free mass (FFM) and fat mass (FM) equations.

Results: Gender, body weight, and height2/impedance accounted, respectively, for 89.4% (p < 0.001) and 89.3% (p < 0.001) of the variability of DXA-total FFM in the two cohorts. Using the new equation, the mean difference between the DXA-FFM and BIA-FFM estimates was +0.180 kg (95% CI: -0.34 to +0.7 kg, p 0 NS), and the resulting limits of agreement were +6.76 and −6.40 kg. Similarly, good estimates of DXA truncal-, android-, and gynoid-FM from anthropometric and BIA parameters could be obtained from weight, height2/impedance, and waist and hip circumferences (respectively, R2 adjusted: 0.657, 0.776, and 0.770; p < 0.001).

Conclusions: The new equations derived from physical and BIA parameters provide accurate estimates of body composition in MO subjects.

Publication types

  • Comparative Study
  • Research Support, Non-U.S. Gov't
  • Validation Study

MeSH terms

  • Absorptiometry, Photon*
  • Adult
  • Algorithms
  • Body Composition*
  • Cohort Studies
  • Cross-Sectional Studies
  • Electric Impedance*
  • Female
  • Humans
  • Male
  • Obesity, Morbid / epidemiology
  • Obesity, Morbid / metabolism*
  • Predictive Value of Tests
  • Reproducibility of Results
  • Spain / epidemiology
  • Statistics, Nonparametric