| Keywords |
Body composition ; Bioelectrical impedance analysis ; Body fat ; Obese |
| Abstract |
The objective of this study was to determine the accuracy of three kind of bioelectrical impedance analysis (BIA) to assess body composition (BC) in overweight and obese adults. Fifty-three healthy obese people (ages 26 to 75 years; BMI, 25 to 32 kg/m2) underwent BIA assessment of BC using the hand-foot BIA (Zeus) (H-FIMP) and handhand BIA (Omron 305) (H-HIMP) and the foot-foot BIA (Tanita 611) (F-FIMP). Measures of validity were determined by Pearson correlations between BIA and DXA, paired t-tests to assess mean differences, as well as biases and limits of agreement. All BIA methods showed good relative agreement with DXA [H-FIMP: fat mass (FM), r2=0.81; fat-free mass (FFM), r2=0.81; percentage body fat (BF%), r2=0.69; H-HIMP: FFM, r2=0.76; BF%, r2=0.40; F-FIMP: FFM, r2=0.79, BF%, r2=0.44; all p<0.001]. Absolute agreement between DXA and H-HIMP was moderately good, as indicated by a small bias and wide limits of agreement (bias, +/?1.96 standard deviation; FFM, 1.0, +/?4.9 kg; BF%, ?1.3, +/?7.5%). H-FIMP: and F-FIMP also exhibited a smaller bias but wide limits of agreement (H-FIMP: FFM, 0.7, +/?5.3 kg; BF%, ?0.9, +/?6.9%; F-FIMP: FFM, 0.3, +/?5.7 kg; BF%, 00.5, +/?7.8%). Compared with DXA, H-HIMP produced large bias and wide limits of agreement, and its accuracy estimating body composition in obese men was poor. H-FIMP and FFIMP demonstrated little bias and may be useful for group comparisons, but their utility for assessment of body composition in individuals is limited. |