Estimation of indigestible neutral detergent fiber in forages from cell wall components

Abstract

Ruminants can utilize forage fibre material due to an unique adaptation of the stomach that is called the selective retention mechanism. The fibre cell wall is composed mostly of cellulose, hemicellulose and lignin. The first three components are measured as neutral detergent fibre (NDF). The NDF is either slowly digestible or indigestible (iNDF). Even if indigestible iNDF is unavailable to microbes, it is a critical component affecting the nutritional value of feeds. Organic matter digestibility is influenced by iNDF, therefore iNDF will influence the amount of energy available to the ruminant. Feed intake is also influenced by iNDF content which can be estimated by several methods. These methods are time consuming, expensive and not all laboratories have the necessary equipment in order to implement these methods. There is an urgent need to develop a cost effective method that can accurately predict iNDF and could be easily implemented in feed analysis laboratories. The aim of this trial was to develop accurate and precise prediction equations for the estimation of iNDF across selected groups of forages. One hundred and two milled grain and forage samples were received from Afgri (Pty) Ltd, including oats, sorghum, lucerne, ryegrass and Eragrostis curvula hay. Samples were analysed in duplicate for various chemical components as well as incubated in-vitro for 240h to estimate iNDF. A simple ANOVA was used for comparison between the different groups, as well as simple linear regression analysis and stepwise multiple linear regression. Akaike`s information criterion and R2 values were used to evaluate the models and to establish the best fit models. Indigestible NDF was predicted by generating power functions. The independent variables included NDF, acid detergent fibre(ADF), acid detergent lignin (ADL), ADF/NDF, ADL/NDF, iNDF, iNDF/NDF, hemicellulose, cellulose, hemicellulose/NDF and cellulose/NDF. Significant differences were found within groups for the different variables used in the regressions. The R2 values for simple linear regression analysis for all the groups combined (“All”) ranged between 0.03 and 0.60. The R2 values for the individual feedstuff within the groups ranged from 0.64 to 0.99. Therefore individual species had higher iNDF prediction accuracy than the combined groups. As can be seen from the data when specific groups are considered, the value 2.4 is not appropriate. The R2 values for the multiple linear regression analysis for the combined groups of forages in dry matter (“All”) was 0.75 where the R2 values for the specific groups ranged between 0.73 and 0.98. The R2 values for the multiple linear regression analysis for the combined groups of forages in NDF (“All”) was 0.72 where the R2 values for the specific groups ranged between 0.21 and 0.98. It was concluded that it is possible to accurately estimate iNDF from prediction equations. Indigestible NDF can be predicted when only one variable is taken into account. The most accurate results can be obtained from using simple linear regression analysis within specific species.