Prediction of in vivo starch digestion in cattle from in situ data

Rate and site of starch digestion are important in terms of nutrient availability in ruminants. The objective was to predict starch digestion in the rumen and intestines of cattle based on the in situ technique. A database with in vivo observations of starch digestibility was collected from 87 references involving 316 treatments on cattle (47% with lactating cows). In situ values of feeds from Offner et al. (2003) were used to calculate the effective degradability (ED, kg/kg starch) for each diet. For treatments with reliable ED (n = 179), the equation predicting starch digestibility in the rumen (Rd, kg/kg starch) from ED was: Rd = 0.302 + 0.59ED (r(2) = 0.28, RSD = 0. 13 8 kg/kg starch). Experiment explained about 50% of the residual variation: Rd = 0.263 +0.63ED (n = 179, n(exp) = 84, r(2) = 0.92, RSD = 0.061 kg/kg starch). Dry matter intake, expressed as a percentage of live weight (DMI, % LW), also improved the accuracy of the prediction: Rd = 0.439 + 0.68ED - 0.083DMI (n = 179, R-2 = 0.47, RSD = 0.119 kg/kg starch). The effect of other factors was investigated. Starch digestibility in the small intestine (SId, kg/kg RES) could be predicted from the amount of rumen escape starch (RES, kg/kg DM): SId = 0.740 - 1.22RES (n = 5 1, n(exp) = 18, r(2) = 0.91, RSD = 0.108 kg/kg RES). Digestion in this organ was influenced by both the quantity and the nature of rumen escape starch. The digestibility in the large intestine was around 0.5 kg/kg ileal starch and increased with DM intake. A compensatory role of the intestines was outlined in the study. Equations predicting starch digestibility in the rumen, small intestine and large intestine were then validated on the whole dataset. This study provides original information to allow empirical prediction of starch digestion in ruminants. (C) 2003 Published by Elsevier B.V.