1602
VALIDATION OF A NEW APPROACH TO ESTIMATE TOTAL TRACT FIBER DIGESTIBILITY FROM IN VITRO NDFD VALUES

Monday, July 21, 2014
Exhibit Hall AB (Kansas City Convention Center)
Fernanda Lopes , University of Wisconsin, Madison, WI
Kathryn Ruh , University of Wisconsin, Madison, WI
David K. Combs , Department of Dairy Science University of Wisconsin, Madison, WI
Luiz F Ferraretto , University of Wisconsin, Madison, WI
Shane M Fredin , University of Wisconsin, Madison, WI
Claudia Arndt , University of Wisconsin, Madison, WI
Randy D Shaver , University of Wisconsin, Madison, WI
Louis E. Armentano , University of Wisconsin, Madison, WI
Abstract Text: The objective was to validate an in vitro model to predict the total tract fiber digestibility (TTNDFD) in dairy cattle. Nineteen diets from six different trials conducted at University of Wisconsin-Madison were analyzed for fiber digestibility using the in vitro standardized model (Goeser and Combs, 2009). Forages varied amongst diets (corn, alfalfa, tall-fescue and meadow fescue and wheat straw silages) and nutrient composition (NDF ranges from 22.5 to 32.1 %, CP 15.8 to 18.9 % and NFC 38.0 to 51.0 %). Total NDF digestibility observed from the in vivo trials was calculated using indigestible NDF or lignin as marker analyzed in fecal, diet and orts samples. The in vitro TTNDFD model predicts total tract fiber digestibility from the rate of pdNDF degradation (kd, ranges from 1.5 to 4.8 %/h), the rate of passage of pdNDF (kp, ranges from 2.5 to 2.8 %/h) and the proportion of total NDF that is potentially digestible. The kd is calculated from in vitro NDFD measurements taken at 24, 30 and 48 h of incubation using first order kinetics model with an indigestible fraction (Mertens, 1993). Passage of potentially digestible fiber is predicted from a regression model (Krizsan et al., 2010) for iNDF which is adjusted to account for the selective retention of pdNDF (Lund et al., 2006). The pool of indigestible fiber was estimated from 240 h in vitro NDF residues. Data were analyzed using SAS procedure of logistic regression. The coefficient of determination (R2) was used to measure the proportion of variation explained by the model. The range of in vivo TTNDFD was 26.3 to 55.6 % compared to 33.8 to 52.8% for predicted in vitro TTNDFD. The relationship between predicted in vitro TTNDFD and in vivo TTNDFD was TTNDFD in vivo = -5.7531 + 1.1561 TTNDFD in vitro predicted with R² of 61.6%, Root-MSE of 4.3% and p-value of <0.001. The in vitro test of diets from six different trials demonstrated that TTNDFD model can provide important insights into fiber utilization by dairy cattle that could be used in the field. The TTNDFD value can also be used as a stand-alone value to index forages, as already shown in other publications from our lab. The ability to predict total tract fiber digestibility from a model based on in vitroNDF degradation and incorporate this information into rations could improve our ability to optimize forage utilization and milk production.

Keywords: iNDF-Fiber-Digestibility