What happens in the bag?

The most widely used method to estimate the rumen degradation of dietary components in feedstuffs is the in situ or in sacco method. This method is based on rumen incubation of substrate (feed) in nylon or dacron bags followed by rinsing and analysis of the residue. Small pores in the bag allow microbes to enter the bag whilst a variable portion of the feed is retained in the bag. The results are used to estimate the ruminal effective degradation (ED) that is used in several protein evaluation systems. The weaknesses of the in situ method are its low precision, the lack of standardization, and its inaccuracy. The accuracy can be divided in a bias with respect to the in situ method itself, and the difference between the in situ and in vivo degradation. The bias of the in situ method itself is related to several assumptions regarding the size and degradation rate of the washable fraction, secondary particle loss, and microbial contamination. The aim of this thesis was to examine possibilities to reduce this bias by modification of the in situ methodology.

The bias related to the assumptions regarding the washout fraction was successfully reduced by developing and using a modified rinsing method that involves less vigorously shaking conditions and a solvent which mimics the rumen pH and osmolality. This modified rinsing method markedly reduced the soluble (S) fraction of N, especially for legume seeds, and the non-soluble washout (W-S) fraction, especially for starch, compared to the conventional method. Consequently, the estimation of the ED became less dependent on the assumptions regarding the degradation of the S and W-S fraction. In vitro results did not support the assumption of a much faster degradation of the W-S fraction of starch than that of the non-washout fraction of starch. The modified rinsing method also allowed measuring the in situ degradation of products that contain mainly small particles such as wheat yeast concentrates.

The bias related to the breakdown of particles was also successfully reduced by development and application of a combination of the modified rinsing method and an in vitro method that simulates particulate matter loss during incubation. This in vitro method was based on in situ results obtained with an inert marker (i.e., silica gel) which showed that particulate matter loss during incubation was moderate and limited to particles smaller than approximately 40 μm. Correction for these losses decreased the estimated ED of feed ingredients used. This study also showed that the fractional degradation rate of starch in grains was strongly affected by the process of secondary particulate loss when using the conventional method, and applying the modified rinsing method markedly reduced the error due to secondary particulate loss.    .

In comparison to the conventional method, the modified method resulted in an increase of  the bias related to microbial contamination of the residues. The ratio between diaminopimelic acid (DAPA), a marker for bacterial protein, and N in the residues was higher when using the modified method than that in the conventional method. The results obtained for the modified method also indicated lysis of bacterial cells during rinsing. This bias led to a lower ED when using the modified method with the impact greatly depending on the degree of lysis of bacterial cells.

In summary, the modified method increased the non-washout fraction of N and starch of various feed ingredients, which offers the possibility to use it for a larger range of feed ingredients, and reduced the bias related to assumptions on the washout fraction and the breakdown of particles compared to the conventional method. On the other hand, the modified method increased the bias related to microbial contamination and enlarged the difference between the in situ and in vivo degradation. .