Genauigkeit der Dualenergie-Röntgenabsorptiometrie (DXA) zur Ermittlung der Körperzusammensetzung von Schweinen in vivo

Abstract. Title of the paper: Accuracy of dual energy X-ray absorptiometry (DXA) for the determination of the body composition of pigs in vivo
New methods for the measurement of the body composition in vivo (or post mortem) like DXA need to be evaluated before routine use in livestock research or performance testing. Two body weight groups of pigs I) 30–50 kg (n=61) and II) 70–90 kg (n=62) were studied in this analysis. The whole bodies of the pigs were analyzed by a GE Lunar DPX-IQ scanner in vivo for the amount and percentage of fat, lean tissue, bone mineral, and the bone mineral density as well. Three days later, pigs were slaughtered and after a 24 h cooling period, the reference parts (shoulder, loin, side, and ham) of the left carcass half were dissected into lean meat, fat, bone, tendons, connective tissue and skin according to the EU reference method. Generally, DXA variables explained a high proportion of variation (R²>0.8; √MSE=70–784 g) in the tissue masses of the reference parts (fat, muscle, bone) in both weight groups. The fat percentage of the reference parts and the DXA whole body results (in vivo) showed a very high relationship with a maximum R²=0.82 (√MSE=1.47%) for group II. Slightly lower was the relationship between the lean meat content (%) of the reference parts and DXA results in vivo (R2=0.72; √MSE=1.75) for group II. The bone percentage was predicted with R²=0.3 (√MSE=0.64), caused by the very different analysis methods. Generally, DXA is a suitable method for growth studies in the above body weight range. In order to predict the percentage of muscle (lean meat) or fat tissue accurately by DXA, pigs should weigh >50 kg, since the soft tissue percentages of group I resulted in R²≤0.38 (√MSE≤2.55).