Visceral adipose tissue in obesity: A comparison between DXA and MRI measures

While magnetic resonance imaging (MRI) accurately measures visceral adipose tissue (VAT), it is costly and time-consuming. Dual X-ray absorptiometry (DXA) offers a faster, more affordable alternative. However, prior comparisons between DXA and MRI often involve different anatomical locations, limiting accuracy assessments. This study aims to compare VAT volume measurements from DXA and MRI at the same anatomical location (L3 vertebra). We hypothesized that both methods would yield similar VAT results. Methodology This is a retrospective, cross-sectional study that analyzed 50 pairs of matched MRI and DXA scans from 28 non-diabetic participants with obesity (18-65 years, BMI range of ≥30 to ≤55 kg/m2). VAT was measured at the L3 vertebral body region by manual manipulation of the android region of interest (ROI) within the DXA (Hologic Horizon A; Hologic Inc., Marlborough, MA, USA) scan, and manual segmentation of the MRI-VAT (3T MR system , Siemens Prisma, Erlangen, Germany). MRI-VAT was expressed as total segmented volume (MRI-VATtsv) and fat fraction-corrected volume (MRI-VATffv). A linear mixed-effects model (LMM) was used to compare VAT estimates, accounting for within-subject variability. Bland-Altman analysis assessed agreement and bias. Results DXA-VAT was significantly higher than both MRI-VATtsv (+68 mL; p = 0.0038) and MRI-VATffv (+154.7 mL; p < 0.001), despite showing a strong correlation (R = 0.80; p < 0.001). Bland-Altman analysis showed a positive bias for DXA-VAT compared to MRI-VATtsv (LoA: 442, -396 mL) and MRI-VATffv (LoA: 519, -209.5 mL). Conclusion In obese adults, DXA and MRI showed strong agreement for VAT measurement at the L3 vertebra, but DXA consistently yielded higher VAT values compared to MRI. These findings indicate that DXA offers a practical means of estimating VAT, but systematic bias and limited agreement relative to MRI highlight important differences in absolute VAT quantification.