Determination and prediction of energy values in corn distillers dried grains with solubles sources with varying oil content for growing pigs

This study was conducted to determine the DE and ME content of 25 samples of corn distillers dried grains with solubles (DDGS) fed to growing pigs and to generate prediction equations for DE and ME based on chemical analysis. The 25 samples included 15 full-oil (no oil extracted; ether extract [EE] > 8%) DDGS and 10 reduced-oil (oil extracted; EE < 8%) DDGS collected from 17 ethanol plants in China. A corn–soybean meal diet constituted the basal diet and the other 25 diets replaced a portion of the corn, soybean meal, and lysine of the basal diet with 28.8% of 1 of the 25 corn DDGS sources. Seventy-eight barrows (initial BW = 42.6 ± 6.2 kg) were used in the experiment conducted over 2 consecutive periods (n = 6 per treatment) using a completely randomized design. For each period, pigs were placed in metabolism cages for a 5-d total collection of feces and urine following a 7-d adaptation to the diets. Among the 25 corn DDGS samples, EE, NDF, DE, and ME content (DM basis) ranged from 2.8 to 14.2%, 31.0 to 46.6%, 3,255 to 4,103 kcal/kg, and 2,955 to 3,899 kcal/kg, respectively. Using a stepwise regression analysis, a series of DE and ME prediction equations were developed not only among all 25 DDGS but also only within 15 full-oil DDGS and 10 reduced-oil DDGS samples. The best fit equations of DE (kcal/kg DM) for the complete set of 25 DDGS, 15 full-oil DDGS, and 10 reduced-oil DDGS were 2,064 – (38.51 × % NDF) + (0.64 × % GE) – (39.70 × % ash), –(87.53 × % ADF) + (1.02 × % GE) – (22.99 × % hemicellulose), and 3,491 – (40.25 × % NDF) + (46.95 × % CP), respectively. The best fit equations for ME (kcal/kg DM) for the complete set of 25 DDGS, 15 full-oil DDGS, and 10 reduced-oil DDGS were 1,554 – (44.11 × % NDF) + (0.77 × % GE) – (68.51 × % ash), 7,898 – (42.08 × % NDF) – (136.17 × % ash) + (101.19 × % EE) (103.83 × % CP), and 4,066 – (46.30 × % NDF) + (45.80 × % CP) – (106.19 × % ash), respectively. Using the sum of squared residuals to compare the accuracy of the 3 groups of prediction equations revealed that separate equations for full-oil DDGS and reduced-oil DDGS each provided a better fit than a single equation for the entire set of DDGS sources. These results indicated that the DE and ME values in corn DDGS are related to the chemical composition, primarily the EE and fiber concentrations. Specific prediction equations derived from full-oil and reduced-oil DDGS are better than equations derived from the entire set of DDGS.