Starch digestion by feedlot cattle: predictions from analysis of feed and fecal starch and nitrogen

To evaluate the utility of N as a digestion marker to predict total tract starch digestion, data from 32 metabolism trials involving 147 steers and 637 individual starch digestibility measurements were compiled. All trials were conducted at the University of California Desert Research and Extension Center. Total tract starch digestibility was determined from concentrations of starch and chromic oxide in feed and feces. In all trials, the steers were adapted to diets for 10 d followed by 4 d for collection of samples of feces. During collection, fecal samples (approximately 200 g, wet basis) were obtained twice daily. Samples from each steer within each collection period were composited for analysis. Diets contained 46.5 +/- 7.4% starch and 1.85 +/- 0.20% N. Apparently digestible N as a percentage of diet DM was closely associated (r(2) = 0.73; P < 0.001) with dietary N concentration. Fecal N concentration (FN, % of DM) explained 35% of the variation in fecal DM excretion (S(y.x) = 4.3; P < 0.001). Incorporating FN into the model, starch digestion was estimated as follows: starch digestion, % of intake = 100 {1 - [(0.938 -0.497FN + 0.0853FN(2)) FS/DS]}, where FS is fecal starch concentration (% of DM) and DS is dietary starch concentration (% of DM; r(2) = 0.94; S(y.x) = 0.68; P < 0.001). Fecal starch concentration alone explained 96% of the variation (S(y.x) = 0.45; P < 0.001) in total tract starch digestion: starch digestion, % = 99.9 - 0.413FS -0.0104FS(2). Omitting cases in our data set in which observed total tract starch digestion was less than 95%, the r(2) between FS and starch digestibility decreased to 0.82 (S(y.x) = 0.26; n = 529). However, estimated starch digestion using the equation incorporating FN remained closely associated with the observed starch digestion (r(2) = 0.90; S(y.x) = 0.22; P < 0.001; n = 529). Equations also were developed to predict NE(m) and NE(g) concentrations of common feed grains based on starch digestibility and FS. Starch digestion can be accurately predicted based on FS. However, incorporation of FN into the model markedly enhanced the estimates of grain quality and the efficacy of processing when total tract starch digestion exceeded 95%.