Precision of Dual-Energy X-Ray Absorptiometry Measurements Obtained on Pork Half-Carcasses

Abstract Text: This study was conducted to investigate the repeatability and reproducibility of dual-energy X-ray absorptiometry (DXA) measurements taken in 9 pork half-carcasses with moderate weight (38.5 to 53.1 kg) and backfat thickness (14 to 21.5 mm) variability. A Lunar Prodigy Advance (software version 13.40.038) device was configured in the total body reading and standard mode. Repeatability conditions were created by scanning each carcass 10 times in the same position. Repeatability measures the error inherent to DXA readings. Reproducibility conditions were created by scanning each carcass once in 10 different positions. Reproducibility conditions include the repeatability and reproducibility sources of variation. Reproducibility measures the variation inherent to carcass positioning on DXA table. These 10 positions were obtained by turning the carcass (skin up vs. down), changing the scanning direction (head-to-hind foot vs. hind foot-to-head), the carcass alignment (straight vs. diagonal) and belly position (extended vs. folded). After the scans, the images were analysed using either a rectangular custom region of interest (ROI) or using as reference the standard grid for human body by placing the carcass within the arm ROI. These two region analysis were performed for all the images obtained in both repeatability and reproducibility conditions. The coefficients of variation (CV) were computed for each carcass individually and then combined assuming a normal variance distribution. DXA bone mineral density (BMD, g/cm²), bone mineral content (BMC, g), soft tissue (g), and fat (%) were considered in this study. Repeatability and reproducibility for custom ROI were of 0.61 vs. 0.88% for BMD, of 0.68 vs. 3.65% for BMC, of 0.07 vs. 0.29% for soft tissue, and of 0.81 vs. 2.44% for fat. Repeatability and reproducibility for arm ROI were of 0.53 vs. 0.85% for BMD, of 0.56 vs. 0.76% for BMC, of 0.05 vs. 0.23% for soft tissue, and of 0.78 vs. 2.75% for fat. In conclusion repeatability and reproducibility were considered high for custom and arm ROI, except for BMC from custom ROI in reproducibility condition.

Keywords: DXA, repeatability, reproducibility