Optimizing fecal output estimation in lactating cows: Effects of marker selection, diet, and methodological factors

A study was conducted to evaluate the accuracy, precision, and robustness of fecal output estimates using different external markers (Cr₂O₃ and TiO₂) and internal markers [indigestible dry matter (iDM), neutral detergent fiber (iNDF), and acid detergent fiber (iADF)]. Eight multiparous Holstein cows (144 ± 39 DIM, 34.3 ± 5.10 kg/d milk yield, and 592 ± 57 kg BW) with ruminal cannulas were used in a replicated 4 × 4 Latin square design with 23-d periods. The diets followed a 2 × 2 factorial arrangement: whole raw soybean inclusion (WRS; 0 and 140 g/kg diet DM) and chitosan supplementation (CHI; 0 and 4 g/kg DM). Total fecal collection was performed during 24, 48, or 72 h. External markers were administered daily (15 g/d) in the rumen. Internal marker content was determined after 288-h of incubation in polyester bags with 12 and 25 µm porosity, F57 bags with 25 µm porosity, and non-woven fabric bags with heterogeneous porosity. Cr₂O₃ provided accurate and precise estimates; however, its recovery increased with increasing dietary ether extract. TiO₂ exhibited incomplete recovery, regardless of the length of sampling period, resulting in low accuracy and precision. Internal marker recovery depended on the chitosan inclusion and bag type. Without chitosan, iNDF in polyester bags (12 or 25 µm) achieved complete recovery, producing the most accurate, precise, and robust estimates. The lack of accuracy and precision of TiO₂ and some internal markers (iDM and iADF) affect apparent total-tract digestibility estimates.