Between the simple anthropometric values and measurement of VFA, it was proven that W had the closest relationship with VFA in both men (r=0.68) and women (r=0.65). The number of disorders was greater than 1.0 at 100 cm 2 of VFA and the best combination of the sensitivity and specificity for determining subjects with multiple risk factors was 100 cm 2 of VFA. The obese subjects with a BMI ≥25 had no correlation between BMI and VFA because of the wide individual variation of VFA. BMI showed a close positive correlation with SFA (r=0.82), even for BMI ≥25 (r=0.77), but had a weaker correlation with VFA (r=0.54). The best combination of the sensitivity and specificity for detecting subjects with multiple risk factors was a BMI of 25. The number of complications increased in accordance with BMI and the average value was greater than 1.0 at a BMI of 25. The relationship between each parameter and the prevalence of the complications was investigated. Hyperglycemia, dyslipidemia, and hypertension were evaluated as obesity-related complications. Anthropometric parameters, including BMI, waist circumference (W), waist/hip circumference (W/H), ratio and waist circumference/body height (W/BH) ratio, were measured. Visceral fat area (VFA) and subcutaneous fat area (SFA) were determined by computed tomography (CT) at the umbilical level.
The subjects were 1,193 Japanese subjects (775 men, 418 women age: 20-84 years old, body mass index (BMI): 14.9-56.4 kg/m 2) including subjects undergoing a health examination and obese subjects visiting an obesity clinic. The present study was designed to establish adequate criteria for categorizing `obesity disease' in Japan in relation to obesity-related complications.