89 Performance of iADF as an internal marker for evaluating forage digestibility in horses

Abstract

Total fecal collection (TFC) is considered the gold standard for evaluating diet digestibility; however, this method is time consuming, labor-intensive, and requires horses to wear collection harnesses and/or be confined to stalls for prolonged periods. Less frequent sampling of feces can be more practical and less disruptive to the management and routines of certain types of horses (e.g., horses in training, lactating mares, growing horses) but requires the use of a marker to estimate daily fecal output. Indigestible acid detergent fiber (iADF) is routinely used to determine dry matter digestibility (DMD) in cattle but evidence of its effectiveness as an internal marker in horses is limited. This study used 5 mature Quarter Horse geldings to compare fecal output and DMD determined with TFC to that estimated using iADF. Horses were fed alfalfa (Medicago sativa, ALF) or bermudagrass (Cynodon dactylon ‘Coastal’, CBG) hays at 1.6% BW/d in a crossover design. Hays were chosen to represent a large span in fiber concentration (ALF: 38% NDF, 30% ADF; CBG: 70% NDF, 35% ADF). Each period consisted of an 8-d diet adaptation followed by a 3-d TFC. Hays and daily composited fecal samples were analyzed for iADF using 2 methods: in situ incubation in a rumen-cannulated cow and in vitro incubation using equine fecal inoculum from a non-study donor. Samples were incubated for 288 h and sequentially analyzed to determineresidual neutral and acid detergent fibers, with the latter representing iADF. Data were compared using mixed model ANOVA and linear regression. Marker recovery of iADF averaged 98% (SEM 4%) and was similar between iADF methods and hays. Use of fecal inoculum yielded higher estimates of iADF intake (P < 0.01) and iADF excretion in feces (P < 0.01) than the rumen in situ iADF method. Daily fecal output and DMD estimated by rumen in situ iADF (P = 0.27, P = 0.29) and fecal inoculum iADF (P = 0.71, P = 0.79) both agreed with direct determination by TFC. Linear regression of DMD showed a strong relationship between TFC and both iADF methods (P < 0.001), with slightly better prediction when iADF was determined by rumen in situ (y = 0.81x + 7.1; R2 0.85; RMSE 4.8) than equine fecal inoculum (y = 0.79x + 10.0; R2 0.71; RMSE 7.0). Both equations approached a slope of unity and an intercept not different from zero (P > 0.10). Across all methods, DMD of ALF was higher (P < 0.01) than CBG. Findings support use of iADF as a marker to predict DMD of hays in horses. This study measured iADF in samples representing all feces excreted; thus, future work should verify the suitability of iADF as a marker when applied to fecal spot sampling.