A novel procedure for predicting energy digestibility in feedstuffs fed to pigs

Abstract Text: A general 3-step in vitro system using conical flasks has been widely employed to simulate the digestion and absorption procedure of pigs. Much time and effort is needed in this conventional procedure. Using a Daisy^II incubator, in vitro digestibility of nutrients in a great number of samples may be measured relatively easily and efficiently. The objective of this study was to develop a novel in vitro system using the Daisy^II incubator for estimating energy digestibility of feedstuffs for pigs. The in vivo energy digestibility values of 11 feed ingredients including almond meal, barley, 2 different sources of corn gluten feed, corn gluten meal, lupin hulls, lupin kernel, rice bran, soybean meal, wheat, and wheat bran and 12 diets containing the aforementioned ingredients were determined in a previous experiment employing a Latin square design with 12 animals and 12 periods. The ingredient and diet samples were incubated under the condition of enzymes (pepsin, pancreatin, and a multi-enzyme complex: Viscozyme) with the conventional 3-step (2, 4, and 18 h of incubation respectively) in vitro system. Additionally, the Daisy^II incubator was tested as an in vitro digestion system with 3 replicates of each sample. Equations for predicting in vivo energy digestibility were generated using in vitro dry matter disappearance (IVDMD) as an independent variable. To determine optimum incubation time for each digestion step, multiple digestion periods were tested and non-linear regression models with incubation time as an independent variable were employed. The IVDMD based on the conventional in vitro procedure (IVDMD_Flask) provided fairly reasonable estimates of the energy digestibility: DE:GE = 0.01 + (0.0093 × IVDMD_Flask, %) with P < 0.001 and r² = 0.96. However, the IVDMD using the Daisy^II incubator system (IVDMD_Daisy) with 2, 4, and 18 h of stepwise incubation resulted in less accurate estimates for in vivo energy digestibility (r² = 0.54). A modified in vitro procedure using Daisy^II incubator with 2, 14, and 18 h for the step 1, 2, and 3, respectively, resulted in fairly reasonable estimates of in vivo energy digestibility: DE:GE = 0.02 + (0.0093 × IVDMD_Daisy, %) with r² = 0.88, P < 0.001. In conclusion, IVDMD_Daisy can be used for prediction of the energy digestibility of a large number of feedstuffs and may replace IVDMD_Flask.

Keywords: Daisy^II incubator, in vitro dry matter disappearance, swine