Importance of Image Review for Accurate Reporting of Hologic DXA Visceral Adipose Tissue

Aims: Increased visceral adipose tissue (VAT) volume is strongly associated with several serious metabolic diseases including heart disease and diabetes, and also to fracture risk. Accurate measure of this soft tissue compartment is imperative if it is to be included in clinical risk models. Hologic software, version Apex 4.0, now includes a VAT estimate from whole body scans. Although the analysis is automated, the region is visible and allows for manual adjustment if necessary. The aim of our study was to evaluate how often the VAT region needed adjustment and what would be the error if no manual adjustment is made. Methods: We reanalyzed a subset of the NHANES study whole body scans (20112012) to include VAT volume to the study results. These NHANES participants are considered a representative sample of the US population from ages 18 years to 59 years old. All scans were analyzed using Hologic version Apex 4.0 auto-analysis mode and manually reviewed for proper VAT region placement. VAT regions needing correction were manually adjusted. Paired T-tests were used to determine the difference in error between the auto and manual analyses for VAT and subcutaneous adipose tissue volume (SAT). Results: A total 3283 adult whole body scans were reviewed. VAT boundary marker adjustment was required on 247, or approximately 8% of the scans. The average difference between manual and auto analysis for both VAT and SAT were significantly different (p!0.0001) with VAT being -17.9 34.8 cm and SAT higher +18.0 35.5 cm. Absolute differences ranged from -213.7 to 145.8 cm and -145.8 to 213.7 cm for VAT and SAT, respectively. Conclusion: Our results confirm that for VAT and SAT regions, it is imperative to visualize the and check the region demarcations to assure high accuracy. Failure to do this necessary quality control review could lead to clinically inaccurate results in 1 out of 12 patients. Disclosure of Interest: None Declared