Methods of ytterbium analysis for predicting fecal output and flow rate constants in cattle

Abstract

Continuous or pulse doses of Yb-labeled feedstuffs with subsequent fecal sampling can be used to estimate digesta passage rates and fecal output in ruminants. However, the validity of such estimates is affected by mineral elements in fecal samples that interfere with atomic absorption analysis of Yb. A procedure was developed involving co-precipitation (CoP) of Yb with lanthanum (La) oxalate at pH 1.0 to separate Yb from interfering elements present in the fecal matrix. The procedure was tested for accuracy of Yb determination, repeatability, and for validity of predicting fecal output. Repetitive analysis of the same sample resulted in a coefficient of variation of 2.2% for the CoP technique. An experiment using 12 mature Angus cows offered 1 of 4 diets tested the accuracy of predicting fecal output using one- and two-compartment models. Cows were pulse-dosed with Yb-marked orchardgrass neutral detergent fiber, and fecal samples were collected from the rectum at 9, 12, 15, 18, 24, 32, 40, 48, 60, 72, 84, and 96 h after dosing. Ytterbium content of fecal samples was determined by neutron activation (NA) or atomic absorption spectrophotometry after slow oscillation of the fecal ash for 12 h in 3 M nitric and 3 M hydrochloric acid (acid leaching; AL) or CoP of Yb with La oxalate. For fecal Yb concentrations fit to the one-compartment model, the k0 parameter (scaling factor related to initial marker in the age-dependent compartment) was greater (P<.05) for CoP than for AL or NA. Likewise, calculated first appearance of marker (Tau) and the age-dependent rate constant (k1) were greater (P<.05) for CoP than for NA. For the two-compartment model, the initial marker concentration estimate (lambda0) was greater (P<.05) for CoP than for NA or AL, and Tau was less (P<.10) for NA than for CoP. Rate constant estimates (lambda1, lambda2) were not affected by method of analysis. For both models, fill and retention time estimates differed (P<.05) between CoP and NA. Fecal output estimated from both models was similar to actual fecal output for CoP, but the one-compartment model estimate of fecal output for AL and NA over-estimated (P<.05) actual fecal output. Likewise, the two-compartment model estimate of fecal output for NA was greater (P<.05) than actual fecal output. Co-precipitation of Yb with La oxalate appears to be a valid analytical procedure that may yield more accurate estimates of fecal output than other reported procedures.