Estimation of NDF pool in the rumen of cattle using fecal excretion and diet characteristics

Abstract Text: The objective of this work was to estimate ruminal pool of NDF in cattle from the fecal excretion of fiber and diet characteristics using a meta-analytical approach. The dataset was compiled from 60 experiments with lactating cows and growing cattle carried out in Europe and Brazil, totaling 227 treatment means. The dataset was analyzed using linear mixed models according to meta-analysis techniques. The variation among experiments was considered as a random effect in the models. An unstructured (co)variance matrix was used. The evaluation of the effect of animal category was accomplished on intercept and slopes through a “dummy” variable D (i.e. binary variable),where D = 0 for lactating dairy cows and D = 1 for growing cattle. All analyses were performed using the MIXED procedure of SAS 9.2.Different models were adjusted considering two different units for NDF pool (pNDF): kg and g/kg BW. The independent variables were selected based on biological relevance and also on estimates of partial Pearson’s correlation with the dependent variable. In this way, the independent variables were: fecal excretion of NDF (fNDF), representing the undigested fiber pool; dietary NDF content (dNDF), representing the dietary input of fiber; and the ratio between dietary iNDF (indigestible NDF) and dNDF (iNDF/dNDF), representing the quality of dietary fiber.The adjusted models were: pNDF (kg) = 2.95 -2.75 × D + 1.217 ×fNDF(kg) – 0.08034 × fNDF² (kg) + 0.003×dNDF(g/kg DM) – 1.606 × (iNDF/dNDF), R² = 0.95, s_XY = 0.54, where D = 0 for lactating cows and D = 1 for growing cattle; and pNDF (g/kg BW) = 5.20 – 3.62 × D + 1.067 × fNDF (g/kg BW) – 0.03324 × fNDF² (g/kg BW) + 0.005 × dNDF (g/kg DM) – 2.809 × (iNDF/dNDF), R² = 0.93, s_XY = 0.94. All terms were found significant (P<0.05) and there was no lack of fit for both models (P>0.63). The meta-analysis results brought into evidence that NDF rumen pool can be estimated from fecal and diet characteristics avoiding the utilization of invasive techniques.

Keywords: meta-analysis, mixed models, ruminal evacuation