Comparison of some aspects of the in situ and in vitro methods in evaluation of neutral detergent fiber digestion

The objective of the present study was to compare digestion rates (k(d)) of NDF for different feeds estimated with the in situ method or derived from an automated gas in vitro system. A meta-analysis was conducted to evaluate how in situ derived k(d) of NDF related to in vivo digestibility of NDF. Furthermore, in vitro true digestibility of the feed samples incubated within filter bags or dispersed in the medium was compared, and k(d) for insoluble and soluble components of those feeds were estimated. Four different concentrates and 4 forages were used in this study. Two lactating Swedish Red cows fed a diet of 60% grass silage and 40% concentrate on DM basis were used for in situ incubations and for collection of rumen fluid. The feed samples were ground through a 2.0-mm screen before the in situ incubations and a 1.0-mm screen before the in vitro gas incubations. In situ nylon bags were introduced into the rumen for determination of k(d) of NDF. Additional kinetic data were produced from isolated NDF and intact samples subjected to in vitro incubations in which gas production was recorded for 72 h. Samples were weighed in the bottles or within filter bags (for fiber and in vitro studies) that were placed in the bottles. The interaction between feed and method was significant (P < 0.01); k(d) of NDF for grass hay tended (P = 0.06) to be less whereas k(d) of NDF for alfalfa, barley grain, canola meal, and dried sugar beet pulp were greater (P < 0.01) when estimated with the in situ method than from gas production recordings. The meta-analysis suggested that in situ derived k(d) of NDF were biased and underestimated in vivo digestibility of NDF. Digestion rates of the intact samples were lower for all feeds, except for the hay, when incubated within the bags compared with dispersed in the medium (P < 0.01). Less OM and NDF were digested for all feeds when incubated within bags than dispersed in the medium (P < 0.01). It is concluded from the in vitro study that microbial activity within the bags is less than in the medium. Significant interactions between method (in situ vs. in vitro) and feed suggest that one or both methods result in biased estimates of digestion kinetics.