Developments in estimating visceral fat area from medical examination data.

AIM The aim of this study is to develop a simple method for estimating visceral fat area (VFA) using medical examination data. METHODS The study subjects were 100 males who had undergone medical examinations and computed tomography (CT) at the umbilicus level. Multiple regression analysis was conducted to estimate VFA. The Bland & Altman Method was used to examine the tendency for mean difference between VFA observed by CT and VFA estimated by medical examination data. We calculated cross-validation and sensitivity and specificity at VFA > or = 100 cm2. RESULTS As a result of multiple regression analysis, the waist-height ratio (WHtR) and triglyceride (TG) were taken as independent variables (r=0.910). The Bland & Altman Method showed 0.00+/-63.88 cm2. The cross-validation regression equation was r=0.889. Sensitivity was 0.833 and specificity was 0.900. CONCLUSION WHtR is a simple index used to diagnose the accumulation of VFA. It has been reported that TG, which increases with accumulating VFA, is a factor in hyperlipidemia; therefore, we consider the obtained independent variables to be appropriate. In addition, the regression equation showed high correlation and good results by cross-validation, the Bland & Altman Method, sensitivity, and specificity. We assert that VFA can be estimated using this method.