Some problems in fiber determination of a tannin-rich forage (Acacia saligna leaves) and their implications in in vivo studies

The detergent method for determination of cell-wall constituents has several applications in nutritional studies and was used for Acacia saligna leaves, a tannin-rich material, and for fecal samples of sheep and goats fed these leaves as a sole feed. The samples were analysed by the standard method which uses a separate sample for neutral detergent fiber (NDF) and acid detergent fiber (ADF) determinations and by the sequential approach in which NDF is subjected to acid detergent solution to obtain ADF. In both cases, the acid detergent lignin (ADL) is determined from ADF. For A. saligna leaves, sequentially determined ADF and ADL were higher compared with the directly determined ADF or ADL, whereas a reverse trend was observed for the fecal samples of both sheep and goats, suggesting a different nature of fibers in feed and in fecal samples. The hemicellulose (HC) content (NDF-ADF) for the leaves was 14.8% and 1.5% respectively, depending on whether the direct or sequential method was applied. Similarly, there was a difference in the HC contents of fecal samples when the fiber was determined directly or sequentially (sheep 5.2% and 8.9% and goats 5.5% and 9.4% respectively). There was a comparatively large difference in the values for cellulose (ADF-ADL) determined by the direct and sequential method in the leaves (14.3% vs. 19.4%) as compared with that in sheep feces (14.9% vs. 14%) and goat feces (15.3% vs. 14.1%). In leaves, both protein and condensed tannins (CT) were much higher in the sequentially determined fiber as compared with those values obtained for the leaves by determining fibers directly, explaining higher ADF and ADL values obtained sequentially. Also, the fiber fractions of fecal samples had a substantial amount of both protein and CT. The in vivo digestibilities of fiber and fiber bound proteins varied substantially and were negative in most cases when using the fiber values obtained. The results suggest that detergent extraction techniques do not predict or determine the in vivo digestibility of cell-wall constituents accurately and that these techniques are not suitable for tannin-rich forages. Further studies are required to account for the artifacts arising due to tannins, proteins, and their complexes in the fiber fractions.