Skip to main content
ARS Home » Midwest Area » Madison, Wisconsin » U.S. Dairy Forage Research Center » Cell Wall Biology and Utilization Research » Research » Publications at this Location » Publication #286104

Title: Comparisons of in vitro fermentation and high moisture forage processing methods for determination of neutral detergent fiber digestibility.

item Hall, Mary Beth

Submitted to: Animal Feed Science and Technology
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 11/22/2014
Publication Date: 1/1/2015
Publication URL:
Citation: Hall, M. 2015. Comparisons of in vitro fermentation and high moisture forage processing methods for determination of neutral detergent fiber digestibility. Animal Feed Science and Technology. 199:127-136.

Interpretive Summary: Every year, thousands of feed samples for commercial farming and research are analyzed for fiber digestibility to determine the feeding and economic value of those feeds for livestock production. Making it easier to handle samples could increase the throughput of the laboratories. Understanding how the method of processing samples for analysis affects the results will also tell us the best way to handle samples. In one experiment, we evaluated two systems for analyzing fiber digestibility. The first was the conventional and more labor-intensive flask in vitro fermentation system with continuous gassing and manual mixing of samples. The second was a new and less labor-intensive system that allows handling of fermentation vessels in tube racks, seals vessels with crown caps, and does not require manual sample handling during the fermentation. Both systems gave similar fiber digestibility values for fibrous feeds, except for corn silage, for which the flask system gave greater values. The results from the tube system showed less variation than did those from the flask system. The tube fermentation system may be useful to achieve more consistent results and easier sample handling, but may have its best use with feeds other than corn silage. In another experiment, in vitro fiber digestibility was determined on forages that were prepared similarly except for their final drying step. In this step, samples were kept frozen and wet, oven-dried at 55 degrees C, or freeze-dried. We hypothesized that keeping a sample frozen and undried might show advantages for testing fresh pasture grasses. For dried samples, some showed no difference between the methods. Others varied as to which gave a greater fiber digestibility value, with the oven-dried sample generally giving the highest fiber digestibilities. The frozen wet samples showed no advantage over the dried samples. These results suggest that the oven-drying methods presently used by most laboratories are acceptable to give reliable fiber digestibility values.

Technical Abstract: In vitro fermentation systems are widely used to measure fiber digestibility of feedstuffs in order to assess feeding values for use in diet formulation and feed valuation. In this study, an in vitro fermentation method with the potential to improve sample throughput and ease of handling was investigated. Additionally, methods of forage sample preparation and their effects on fermentation results were explored. In the first experiment, a commonly used fermentation system to determine in vitro neutral detergent fiber digestibility (NDFD) was employed; it uses continuously CO2-gassed Erlenmeyer flasks in water baths and requires manual swirling of vessels for mixing. This method was compared to a system that uses glass tubes in 50-mL tube racks, are crown-capped to seal, not continuously gassed, incubated in a shaking incubator, and do not require manipulations during fermentation. Four fibrous feed samples (alfalfa silage, corn silage, soyhulls, and ryegrass) were incubated for 24, 30, and 48 h in duplicate at each hour in 3 separate fermentation runs. Rumen inoculum from 2 to 3 cows was blended for each fermentation. The ratios of sample, Goering and Van Soest medium, reducing solution, and inoculum were identical for both methods, with the tube system using half the amounts as in the flasks. NDFD differed between the methods by the effect of sample x vessel, with alfalfa silage, soyhulls, and ryegrass giving similar values in the two systems. However, the tube system gave lower values for corn silage. Standard deviation values calculated from mean NDFD of each sample in each fermentation were greater for the flask system (57 g/kg) than the tube system (23 g/kg). In the second experiment, 2 silages and 4 fresh pasture grasses were used to test the forage preparation method on in vitro NDFD results. Forages mixed with dry ice were ground through a meat grinder to pass a plate with 4.5-mm openings. The ground forage was subsampled with portions retained as frozen (FW), dried in a forced-air oven at 55 degrees C (OVD), or freeze-dried (FD). Samples were fermented in duplicate in the tube system described above for 24, 30, and 48 h in duplicate fermentation runs with inoculum blended from 3 to 4 cows. Both neutral detergent fiber (aNDFOM) as a proportion of dry matter and NDFD differed among forages by processing method. For 4 forages, FW, OVD, or FD gave the greatest aNDFOM value, but maximal differences between methods (25 to 43 g/kg) were small. For NDFD, two forages gave 51 to 64 g/kg greater values for OVD than FD, one gave a 26 g/kg greater value for FD over OVD, and the other forages showed no detected effect of processing method. FW showed no advantage over OVD or FD in NDFD determinations. In summary, the tube in vitro fermentation system gave smaller values for corn silage, similar NDFD values for all samples except corn silage, and was more precise than the NDFD measurements in the flask system. Effects of forage sample processing method varied by forage sample, but use of a wet ground frozen sample showed no advantage over the dried samples, even for fresh forage grasses.