2012 Annual Report
1a.Objectives (from AD-416):
Quantify the impact of reducing oil in DDGS on the ME in growing pigs and poultry.
1b.Approach (from AD-416):
1. Analyze four samples of OE-DDGS (10.5, 7.5, 5.5 and 4.5% oil, as-is basis) for CP, EE, starch, NDF, acid detergent fiber (ADF), total dietary fiber (TDF), ash, and minerals. 2. Conduct energy (DE and ME) digestibility study with pigs fed OE-DDGS. 3. Conduct energy (ME only) digestibility study with broilers fed OE-DDGS.
4. Correlate DE and ME for swine and ME for broilers to OE-DDGS composition.
5. Compare DE and ME to previously determined regression equations and data.
Three groups of 24 finishing gilts (n = 72, body weight = 105.6 kg) were housed individually in metabolism crates that allowed for separate, but total, collection of feces and urine. Crates were equipped with a stainless steel feeder and a nipple waterer, to which the pigs had ad libitum access. Gilts were randomly assigned to 1 of 4 test diets or the basal diet with the basal diet replicated 4 times and each test diet replicated 5 times within each group, resulting in 12 and 15 replications for the basal and each test diet, respectively, over the entire experiment. The basal diet contained 96.7% corn and vitamins and minerals with corn being the sole energy-containing ingredient. Test diets contained 70% of the basal diet and 30% of each test ingredient. All diets were fed in a meal form. Feed was provided to the gilts twice daily (1.5 kg/meal) during the 9-day adaptation and the 4-day collection period. Total feed offered and unconsumed feed were weighed and recorded at the end of the 4-day collection period. Because feed intake may affect subsequent nutrient digestibility and digestible energy (DE) and metabolizable energy (ME) determinations, 3 pigs which refused > 20% of their diets relative to pigs within the same group were removed from the study (1 pig each for low, medium, and high crude fat products). During the time-based 4-day total fecal and urine collection period, stainless steel wire screens were placed under each metabolism crate for total fecal collection, while stainless steel buckets containing 30 mL of 6N HCl were placed under each crate for the total urine collection. Feces and urine were collected twice daily and stored at 0°C until the end of the collection period. At the end of the collection period, feces were pooled over the 4-day period, dried in a 70°C forced air oven, weighed, ground through a 1-mm screen, and a subsample was taken for analysis. Likewise, urine samples were pooled over the 4-day period, thawed at the end of the collection period, weighed, and a subsample collected for analysis. All reduced oil-distillers dried grains with soluble (RO-DDGS) products were ground through a 1-mm screen prior to chemical analysis. Samples were analyzed for various nutritional and physical components, and for mycotoxin contamination. Nutrient and energy intakes were calculated based upon actual feed intake over the 4-day collection period. The nutrient and energy digestibility of each test ingredient was calculated by subtracting the nutrient or energy contributed by the basal diet from the nutrient or energy of the diet containing that particular test ingredient and then dividing the result by the inclusion rate of the test ingredient in the diet. Using the individual pig as the experimental unit, data were subjected to analysis of variance with treatment in the model and treatment means are reported as least-square means. The experiment was conducted as a completely randomized design with nutrient or energy digestibility of the basal diet used as a covariate to nutrient and energy digestibility values among all groups of pigs. Final body weight and feed intake were also tested as covariates, but they were not found to be significant; and, thus, they were not included in the final model. Stepwise regression was used to determine the effect of the feedstuff composition on apparent nutrient and energy digestibility with variables having P-values = 0.15 being retained in the model.
Many analytes were relatively well related to the change in the ether extract (EE) of the RO-DDGS. As expected, as the percent EE in RO-DDGS was reduced, gross energy decreased while total dietary fiber, crude protein, and ash increased. In contrast, neutral detergent fiber appeared to decrease as oil was extracted. We cannot explain this apparent decrease as it goes against the logic that as oil is removed, other components would be concentrated. We also measured mycotoxin contamination in the basal diet and each RO-DDGS sample, and although some mycotoxins were detected, they were well below limits of concern.
Digestibility of various components in the RO-DDGS differed between treatments, but were not helpful in predicting DE or ME of the RO-DDGS samples. This was not expected based upon previous literature, and warrants further evaluation of the data.
The main focus of this experiment was to relate DE and ME to the percent EE in the RO-DDGS sample. As expected, gross energy was well related to EE content, but the simple linear relationship between DE or ME and percent EE was not obvious. The relationship between DE or ME (kcal/kg dry matter) and percent EE in the RO-DDGS (dry matter basis) could be graphed, suggesting a change in DE and ME by 32 and 46 kcal, respectively, per kg dry matter for each one percentage unit change in EE, but the coefficient of determination was poor, 0.22 and 0.32 for DE and ME, respectively. This lack of relationship can also be denoted by regression analysis demonstrating that the P value for DE and ME was nonsignificant, being 0.54 and 0.43, respectively. Calculating an expected change in DE and ME and using corn oil (DE = 8755, ME = 8405) and corn grits (DE = 3355, ME = 3210) as component "equivalents" for a RO-DDGS, suggests that a decrease in one percentage unit of EE relates to a loss of approximately 54 and 52 kcal/kg of DE an ME, respectively, slightly higher than what we found in the current experiment. Regardless of analysis for the RO-DDGS samples, no combination of nutrient analyses could be used to predict DE. In summary, relating DE and ME to only the percent EE in RO-DDGS was not a statistically sound method of analysis, with the data in the current experiment suggesting that additional parameters such as gross energy and "fiber" are necessary for improved prediction equations.