Method: Body composition assessment of sows using dual-energy X-ray absorptiometry

For about 30 years, the introduction of dual X-ray absorptiometry (DXA) scanners in swine research has enabled the non-invasive study of body composition kinetics in animals. So far, the use of DXA technology in swine was focused on piglets, growing pigs up to about 140 kg of BW, as well as carcasses. Due to their size and weight, measuring a sow’s body composition is beyond the technical limits of the device. Furthermore, the chemical composition derived from DXA values is based on equations developed for pigs weighing between 20 and 100 kg. The present aim was to focus on the sow to (1) present a standard operation procedure to obtain the body composition of sows by DXA, and (2) assess the ability of available equations to predict a sow’s chemical body composition. For (1), a study investigated the effect of the animal’s position on DXA body composition. A total of 58 DXA acquisitions of sows were obtained on the standard ventral position (front and back legs extended) and on the lateral position (on left flank with right legs placed inward and left legs placed outward). The predicted BW, lean tissue mass, fat tissue mass, bone mineral content, bone area, and bone mineral density of the standard ventral position from the obtained lateral position resulted in root mean square prediction errors expressed as a percentage of the observed mean value of 0.5, 1.9, 5.0, 2.7, 3.1 and 3.5%, respectively. For (2), 3 sows were scanned alive and then slaughtered to measure chemical composition, then, these results were compared with equations based on growing pig data. The chemical composition of the carcass was predicted more accurately than that of the empty body. Regarding minerals, the Ca and P contents of the empty body were overestimated (12 and 3% respectively), as with the Ca content of the carcass (6%), while the P content of the carcass was underestimated (5%). In conclusion, the proposed material and operation procedure enables the scanning of sows which exceed the maximal specification of a DXA device. Furthermore, before concluding the accuracy of the chemical body composition prediction equations based on DXA data for pigs weighing between 20 and 100 kg, additional data are required to determine their applicability to sows.