Background: Abdominal obesity has an important biological and epidemiological relationship to disease. The gold standard for measurement of visceral adipose tissue (VAT) is assessment by computerized tomography (CT) or magnetic resonance imaging (MRI), but because of simplicity and ease in collection, anthropometric variables are a desirable alternative to estimate VAT.
Objective: To compare the abilities of a single slice CT scan through the L4-L5 interspace (L4-L5 VAT), sagittal diameter, and body mass index (BMI) to estimate total volume VAT. Total volume VAT (the gold standard) was measured by total abdominal CT scanning, with a mean of 42 CT slices per patient. Estimation of VAT in subjects of similar body size was emphasized.
Design: Retrospective study of subjects undergoing complete abdominal and pelvic CT scanning for clinical reasons.
Subjects: 40 subjects (20 men and 20 women) mean age 56.5y, with a balanced selection for BMI < 27 and > 27.
Results: In univariate regression models, L4-L5 VAT explained the largest proportion of the variance in total VAT (R2=0.87 [P<0.001]), though age (R2=0.11 [P=0.04]), BMI (R2=0.37 [P<0.001]), and sagittal diameter (R2=0.50 [P < 0.001]) were also statistically significantly related to total VAT. When limited to individuals with a BMI > or= 27 however, L4-L5 VAT explained a large proportion of the variance in total VAT (R2=0.87 [P < 0.001]) whereas sagittal diameter was only of borderline significance (R2=0.20 [P=0.06]), and BMI was not associated with total VAT (R2=0.04 [P=NS]). In multiple regression analyses, L4-L5 VAT area explained a large proportion of the variance (0.84-0.90), and once in the model, BMI, sagittal diameter, and age did not additionally contribute significantly to the explained variance in total VAT.
Conclusions: Abdominal sagittal diameter is poorly correlated to total VAT for men and women with a BMI > 27. Within a 2 cm range of sagittal diameter, there is nearly a three-fold variability in total VAT.