Standardization of dual‐energy x‐ray visceral adipose tissue measures for comparison across clinical imaging systems
Jonathan P. Bennett, Bo Fan, En Liu, Leila Kazemi, Xianping Wu, Hou‐De Zhou, Ying Lü, John Shepherd
Abstract
OBJECTIVE: Excess visceral adipose tissue (VAT) is a major risk factor for metabolic syndrome (MetS) and clinical guidelines have been proposed to define VAT levels associated with increased risk. The aim was to standardize VAT measures between two dual-energy x-ray absorptiometry (DXA) manufacturers who provide different VAT estimates to support standardization of measures across imaging modalities. METHODS: Scans from 114 individuals (ages 18-81 years) on GE HealthCare (GEHC) and Hologic DXA systems were compared via Deming regression to standardize VAT between the two systems, validated in a separate sample (n = 15), with κ statistics to assess agreement of VAT measurements for classifying patients into risk categories. RESULTS: = 0.97). VAT area measures substantially agreed for metabolic risk classification (weighted κ = 0.76) with no significant differences in the population mean values. CONCLUSIONS: VAT measures can be estimated from GEHC and Hologic scans that classify individuals in a substantially similar way into metabolic risk categories, and systematic bias between the measures can be removed using simple regression equations. These findings allow for DXA VAT measures to be used in complement to other imaging modalities, regardless of whether scans used GEHC or Hologic systems.