Comparison of Bioelectrical Impedance Analysis, Slaughter Skinfold-Thickness Equations, and Dual-Energy X-ray Absorptiometry for Estimating Body Fat Percentage in Colombian Children and Adolescents with Excess of Adiposity

Dual-energy X-ray absorptiometry (DXA) has been considered a reference method for measuring body fat percentage (BF%) in children and adolescents with an excess of adiposity. However, given that the DXA technique is impractical for routine field use, there is a need to investigate other methods that can accurately determine BF%. We studied the accuracy of bioelectrical impedance analysis (BIA) technology, including foot-to-foot and hand-to-foot impedance, and Slaughter skinfold-thickness equations in the measurement of BF%, compared with DXA, in a population of Latin American children and adolescents with an excess of adiposity. A total of 127 children and adolescents (11-17 years of age; 70% girls) from the HEPAFIT (Exercise Training and Hepatic Metabolism in Overweight/Obese Adolescent) study were included in the present work. BF% was measured on the same day using two BIA analysers (Seca((R)) 206, Allers Hamburg, Germany and Model Tanita((R)) BC-418((R)), TANITA Corporation, Sportlife Tokyo, Japan), skinfold measurements (Slaughter equation), and DXA (Hologic Horizon DXA System((R)), Quirugil, Bogota, Columbia). Agreement between measurements was analysed using t-tests, Bland-Altman plots, and Lin's concordance correlation coefficient (c). There was a significant correlation between DXA and the other BF% measurement methods (r > 0.430). According to paired t-tests, in both sexes, BF% assessed by BIA analysers or Slaughter equations differ from BF% assessed by DXA (p < 0.001). The lower and upper limits of the differences compared with DXA were 6.3-22.9, 2.2-2.8, and -3.2-21.3 (95% CI) in boys and 2.3-14.8, 2.4-20.1, and 3.9-18.3 (95% CI) in girls for Seca((R)) mBCA, Tanita((R)) BC 420MA, and Slaughter equations, respectively. Concordance was poor between DXA and the other methods of measuring BF% (c < 0.5). BIA analysers and Slaughter equations underestimated BF% measurements compared to DXA, so they are not interchangeable methods for assessing BF% in Latin American children and adolescents with excess of adiposity.