In vivo prediction of goat kids body composition from the deuterium oxide dilution space determined by isotope-ratio mass spectrometry

Deuterium oxide dilution space (D2OS) determination is one of the most accurate methods for in vivo estimation of ruminant body composition. However, the time-consuming vacuum sublimation of blood preceding infrared spectroscopy analysis, which is traditionally used to determine deuterium oxide (D2O) concentration, limits its current use. The use of isotope-ratio mass spectrometry (IRMS) to determine the deuterium enrichment and thus quantify D2O in plasma could counteract this limitation by reducing the sample preparation for plasma deproteinisation through centrifugal filters. The aim of this study was to validate the D2OS technique using IRMS in growing goat kids to establish in vivo prediction equations of body composition. Seventeen weaned male Alpine goat kids (8.6 wk old) received a hay-based diet supplemented with 2 types of concentrates providing medium (n = 9) or high (n = 8) energy levels. Kids were slaughtered at 14.0 (n = 1, medium-energy diet), 17.2 (n = 4, medium-energy diet, and n = 4, high-energy diet), or 21.2 wk of age (n = 4, medium-energy diet, and n = 4, high-energy diet). Two days before slaughter, D2OS was determined after an intravenous injection of 0.2 g D2O/kg body mass (BM) and the resulting study of D2O dilution kinetics from 4 plasma samples (+5, +7, +29, and +31 h after injection). The deuterium enrichment was analyzed by IRMS. After slaughter, the gut contents were discarded, the empty body (EB) was minced, and EB water, lipid, protein, ash, and energy contents were measured by chemical analyses. Prediction equations for body components measured postmortem were computed from in vivo BM and D2OS. The lack of postmortem variation of fat-free EB composition was confirmed (mean of 75.3% [SD 0.6] of water), and the proportion of lipids in the EB tended (P = 0.06) to be greater for the high-energy diet (13.1%) than for the medium-energy diet (11.1%). There was a close negative relationship (residual CV [rCV] = 3.9%, R-2 = 0.957) between EB water and lipid content, whereas D2OS was closely related to total body water (rCV = 2.9%, R-2 = 0.944) but D2OS overestimated it by 5.8%. Adding D2OS to BM improved the in vivo predictions of EB lipid and energy content (rCV = 13.1% and rCV = 7.9%, respectively) but not those of protein or ash. Accuracy of the obtained prediction equations was similar to those reported in studies determining D2OS by infrared spectroscopy. Therefore, the use of IRMS to quantify D2OS provides a highly accurate measure of the in vivo body composition in goat kids.