Prediction of visceral fat area at the umbilicus level using fat mass of the trunk: The validity of bioelectrical impedance analysis

The aims of this study were to determine the validity of fat mass of the trunk as a predictor for visceral fat area at the umbilicus level and to develop equations to predict visceral fat mass at the umbilicus level using fat mass of the trunk measured by dual-energy X- ray absorptiometry ( DXA) and bioelectrical impedance analysis ( BIA). The participants were 121 normal Japanese adults ( 69 males, 52 females). Another 60 volunteer adults ( 34 males, 26 females) were recruited for examination of cross-validity. Altogether, 41 adults ( 15 males, 26 females) in the original group and 19 adults ( 7 males, 12 females) in the cross-validity group received BIA measurement. We measured fat mass by DXA and the BIA system, which was a single- frequency BIA with 8- point contact electrodes, and visceral fat area by computed tomography. We observed significant correlations for visceral fat area in waist circumference ( 0.56) and fat mass of the trunk measured by DXA ( 0.64). There was no significant difference in fat mass of the trunk between the DXA and BIA systems, but the BIA system tended to provide an underestimate compared with DXA. With combined fat mass of the trunk measured by DXA and waist circumference as predictors, visceral fat area was estimated by equation ( 1) ( R = 0.87, R-2 = 0.76, standard error of the estimate 20.9 cm(2)). When substituting fat mass of the trunk measured by BIA into equation ( 1), there was no significant difference in visceral fat area between the reference and predicted values. An equation using fat mass of the trunk measured by BIA ( equation 2) was obtained ( R = 0.89, R-2 = 0.78, standard error of the estimate 20.7 cm(2)), but a systematic error was found for the males. There was cross- validity in both equations. In conclusion, fat mass of the trunk is an effective predictor for the visceral fat area at the umbilicus level. Fat mass of the trunk measured by BIA might be a valid method to predict visceral fat, although further studies with larger samples taking into account the extent and type of obesity are required.