Validation of ultrasound estimates of visceral fat in black South African adolescents

Accurate quantification of visceral adipose tissue (VAT) is needed to understand ethnic variations and their implications for metabolic disease risk. The use of reference methods such as computed tomography (CT) and magnetic resonance imaging (MRI) is limited in large epidemiological studies. Surrogate measures such as anthropometry and dual-energy X-ray absorptiometry (DXA) do not differentiate between VAT and subcutaneous abdominal adipose tissue (SCAT). Ultrasound provides a validated estimate of VAT and SCAT in white populations. This study aimed to validate the use of ultrasound-based assessment of VAT in black South African adolescents. One hundred healthy adolescents (boys = 48, girls = 52) aged 18-19 years participating in the birth to twenty cohort study had VAT and SCAT measured by single slice MRI at L4. These MRI "criterion measures" were related to ultrasound VAT and SCAT thickness, anthropometry (BMI, waist and hip circumferences), and DXA android region fat. Ultrasound VAT thickness showed the strongest correlations with MRI VAT (Spearman's correlation coefficients: r = 0.72 and r = 0.64; in boys and girls, respectively), and substantially improved the estimation of MRI VAT compared to anthropometry and DXA alone; in regression models the addition of ultrasound VAT thickness to models containing BMI, waist, and DXA android fat improved the explained variance in VAT from 39% to 60% in boys, and from 31% to 52% in girls. In conclusion, ultrasound substantially increased the precision of estimating VAT beyond anthropometry and DXA alone. Black South African adolescents have relatively little VAT compared to elderly whites, and we therefore provide new ultrasound-based prediction equations for VAT specific to this group.