Evaluating the accuracy of Dual Energy X-ray Absorptiometry (DEXA) in predicting carcass composition from varying pig populations

Traditional cut-out methods are time consuming and expensive, as well as susceptible to error due to staff dexterity and fatigue, yet to date, no suitable alternative method for accurately estimating carcass composition has been developed. To explore its applicability in the face of wide animal variations, dual energy X-ray absorptiometry (DEXA) was examined in the present study. A total of 558 pigs from three different breeds (Lacombe, Iberian and Duroc sire × commercial Large White × Landrace F1 dam), two sexes (barrow and gilt), two slaughter weights (115 and 135 kg) and three different diets (flaxseed, canola and control) were used in this study to comprehensively represent a wide range of pork. DEXA estimations were used to predict the dissected/chemical yield for lean, fat, and weight of carcass sides and primal cuts (ham, shoulder, loin and belly). The accuracy of the predictions was assessed based on coefficient of determination (R2) and residual standard deviation (RSD). The linear relationships between DEXA and weight (R2 = 0.99, P < 0.01), dissected fat (R2 = 0.98, P < 0.01) and lean (R2 = 0.95, P < 0.01) for all the primal cuts and side carcasses were high with low RSD (<1). Relationships between DEXA and chemical fat (R2 = 0.92, P < 0.01) and lean (R2 = 0.90, P < 0.01) of pork bellies were lower but still acceptable with low RSD (<2.9).With the exception of breed, the linear relationships with DEXA prediction remained high despite a wide variation in the pig population. The present results indicate that DEXA estimations could be reliably used to predict eclectic carcass composition and may be useful for quick assessment of suppliers' raw material against industrial compositional standards, which could enhance current payment systems.