Passage kinetics of fibre in dairy cows obtained from duodenal and faecal ytterbium excretion: Effect of forage type

Passage kinetics of ytterbium labelled fibre was evaluated in four experiments with fistulated dairy cows. Eight forages (grass hay, early cut grass silage, late cut grass silage, whole crop barley silage, maize silage, lucerne hay, whole crop pea silage and clover-grass silage) were fed ad libitum as the sole feed or supplemented with 5.8 kg DM of concentrate. Rumen passage kinetics was determined algebraically and using age-dependent one compartment (Gn) and two compartment (GnG1) models based on duodenal and faecal excretion of Yb. model provided the best fit of the one compartment models tested, whereas the G3G1 model was the overall superior model. The G4G1 model may be preferred for data sets based on faecal output, indicating an additional minor post-duodenal compartment. Using an algebraic approach total mean retention time averaged 39 h (duodenum) and 45 h (faeces), however the algebraic approach underestimated total mean retention time with on average 4 h compared to the G3G1 model. Mean retention time in the rumen compartments based on faecal sampling was on average only 1.2 h longer than based on duodenal sampling, and therefore a model combining duodenal and faecal samples is used for estimation of passage kinetics. Mean retention time in the rumen compartments varied from 32 h for supplemented pea silage to 48 h for unsupplemented late cut grass silage, but absolute values should be used with care, due to the risk for marker migration and uneven distribution of marker between particles. ion time in the first non-escapable age-dependent compartment was always shorter than in the subsequent age-independent compartment. When forage was fed as the sole feed, the proportion of the mean retention time spent in the first compartment averaged 0.39, but varied from 0.32 (lucerne hay) to 0.47 (grass hay). When forage was supplemented, the proportion of the mean retention time spent in the first compartment averaged 0.33 but varied from 0.22 (whole crop barley silage) to 0.43 (early cut grass silage). Based on the present data, a fixed average distribution of mean retention time between the non-escapable and the escapable rumen compartments of 0.36:0.64 seems to be the best estimate for modelling purposes in future dynamic feed evaluation systems. On average 0.82 of the total tract retention time could be assigned to the pre-duodenal part of the GI-tract and 0.18 to the lower gut.