Effects of sample size on NDF digestibility of triticale forages using the ANKOM daisy II incubator system

Authors: COBLENTZ, WAYNE; AKINS, MATTHEW - UNIVERSITY OF WISCONSIN; Ogden, Robin; BAUMAN, LISA - UNIVERSITY OF WISCONSIN; STAMMER, ANDREW - UNIVERSITY OF WISCONSIN

Submitted to: Journal of Dairy Science
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 4/24/2019
Publication Date: 7/12/2019
Citation: Coblentz, W.K., Akins, M.S., Ogden, R.K., Bauman, L.M., Stammer, A.J. 2019. Effects of sample size on NDF digestibility of triticale forages using the ANKOM daisy II incubator system. Journal of Dairy Science. 102(8):6987-6999.

Interpretive Summary: Accurate and precise determinations of in-vitro neutral-detergent fiber digestibility (NDFD) are critical to proper evaluation of forage nutritive value, and essential for proper diet formulation for dairy cows. Currently, NDFD is determined with a variety of methodologies that include traditional incubation of an unrestrained sample within a sealed dedicated tube or flask (STD), as well as the batch procedures of the ANKOM Daisy II Incubator System (ANKOM). Our objectives were three-fold: i) relate NDFD values from 48 triticale forages determined by ANKOM procedures at multiple endpoints ranging from 12 to 240 h using 0.25- or 0.50-g sample sizes with concentrations of fiber-related analytes or growth stage; ii) directly compare NDFD values determined with 0.25- or 0.50-g sample sizes by ANKOM procedures after 12-, 24-, 30-, 48-, 144-, or 240-h incubations; and iii) compare NDFD values determined with 0.25- or 0.50-g sample sizes at 30 and 48 h with those determined by STD procedures obtained from a commercial laboratory. For the 48 triticale samples selected for this evaluation, plant growth stage at harvest proved to be a far better predictor variable for NDFD than various nutritional analytes; this occurred in part because growth stage was not impacted by the physiological process of grain fill, while concentrations of nutritional analytes (particularly NDF) were diluted by this process. Comparisons of 0.25- and 0.50-g sample sizes sealed within fiber bags indicated that greater fiber disappearance occurred from 0.25-g samples than 0.50-g samples, especially after 24-, 30-, and 48-h incubations. For the 24- and 30-h incubations, the discrepancies between sample sizes were exacerbated as fiber became more digestible. By comparison, agreement between sample sizes was much improved with 12-h, as well as extended (144- and 240-h) incubations. Comparisons of ANKOM and STD methodologies suggest that the best agreement for 24-, 30- and 48-h incubations occurs when 0.25-g samples are sealed within fiber bags, but these efforts need to be expanded to a broader array of forage species before broad-based recommendations can be made.

Technical Abstract: Accurate and precise determinations of fiber digestibility are essential for proper diet formulation for dairy cows. Our objectives were three-fold: i) regress in-vitro NDF digestibility (NDFD) values from 48 triticale forages determined at multiple endpoints ranging from 12 to 240 h with ANKOM DaisyII Incubator System (ANKOM) methods using 0.25 or 0.50-g sample sizes on concentrations of fiber-related analytes or growth stage; ii) directly compare NDFD values determined with 0.25 or 0.50-g sample sizes by ANKOM methods after 12, 24, 30, 48, 144, or 240-h incubations; and iii) compare NDFD values determined by ANKOM methods after 30 and 48 h of incubation with those determined by traditional sealed-tube procedures (STD) obtained from a commercial laboratory. Generally, plant growth stage, which was quantified with a linear model suitable for serving as an independent regression variable, proved to be a better predictor variable for NDFD than NDF or ADL. For direct comparisons of 0.25 and 0.50-g sample sizes using ANKOM methods, the regression relationship for a 30-h incubation was explained by a linear model (Y = 1.206 x – 1.1; R2 = 0.933), in which the slope differed from unity, but the intercept did not differ from 0. After a 48-h incubation, a linear model (Y = 1.014 x + 7.1; R2 = 0.964) indicated that the slope did not differ from unity, but the intercept was > 0. A linear regression (Y = 1.040 x – 1.8; R2 = 0.861) of the 30-h incubation results obtained by ANKOM methods using the 0.25-g sample size on those from the commercial laboratory indicated the slope and intercept did not differ from unity or 0, respectively. A similar relationship was obtained from the 48-h incubation (Y = 1.021 x – 3.4; R2 = 0.866). Relationships were poorer when the 0.50-g sample size was used by ANKOM methods, particularly for the 30-h incubation, where the slope (0.824) was less than unity. Generally, NDFD values were greater for the 0.25-g sample size by ANKOM methods, especially with 24, 30, and 48-h incubation times, and agreement with traditional STD methods was improved with the smaller sample size. Synchronization of results between ANKOM and traditional methods needs to be further verified across a wider range of forages and harvest/preservation methods before definitive recommendations can be made.