Effects of marker concentration errors on digesta kinetic parameters

Authors: DHAKAL, KUNDAN - Oak Ridge Institute For Science And Education (ORISE); Moffet, Corey; Gunter, Stacey

Submitted to: American Society of Animal Science
Publication Type: Proceedings
Publication Acceptance Date: 4/14/2018
Publication Date: 9/28/2018
Citation: Dhakal, K., Moffet, C., Gunter, S.A. 2018. Effects of marker concentration errors on digesta kinetic parameters. Translation Animal Science. 2(Suppl. 1):S121-S124. doi:10.1093/tas/txy030.
DOI: https://doi.org/10.1093/tas/txy030

Interpretive Summary: Modelling the digestive activity in the gastrointestinal tract of cattle is a valuable tool to ruminant nutritionist, allowing them to understand and estimate site and extent of digestion. This modelling activity allows nutritionist to design diets that are more efficient utilizers of natural resources and less polluting to the environment. Nutritionist are concerned with the accuracy of these models because they want the best estimates possible. Because we cannot physically observe the digestion process, a nutritionist questions the validity of the results from these models. A way to test a model, when you cannot see inside the animal system, is to construct the digestion process mathematically so the nutritionist know the true values. Once true values are known, errors of estimation, that we know exist in the animal, can be applied to the dataset. The research can then analyze this new dataset to see if their model will quantify values similar to the true values. In this experiment, we constructed a synthetic dataset with different quantities of error associated with different analytical technics. Based on this analysis, employing technologies that decrease additive errors associated with analysis improved the predictive value of the model; hence, the increasing analytical precision is valuable. With error associated with dosing, multiplicative error, can be managed by the model as long as the nutritionist has developed an accurate and precise method of marker dosing.

Technical Abstract: Use of the 2 compartment, age-dependent (G2G1) model is a valuable tool to ruminant nutritionist to assess the digesta kinetics, and site and extent of digestion. To increase the value of the G2G1 model to scientists, we studied the effects of different types of error on parameter bias and standard deviations. Hence, we constructed a dataset of 307,200 synthetic fecal marker concentrations derived by generating true concentration values at 16 nominal times (0, 4, 8, 12, 18, 24, 30, 36, 42, 48, 60, 72, 84, 96, 108, and 120 hours) then randomly applying 4 different levels of additive (2, 5, 10, and 20 grams of marker/kilogram of fecal dry matter) and multiplicative error (0, 2, 5, and 10% of the fecal marker concentration). The 20 synthetic animals and diet combinations were composed of a factorial combination of 5 body weights (227.0, 340.5, 454.0, 567.5, and 681.0 kilograms) and 4 extents of dry matter digestibility (45, 50, 55, and 60%) which each corresponded to a second order of age-dependency, ', K_2, and t. Bias was computed for ', K_2, t, Fecal dry matter output, gastrointestinal mean retention time, and ruminal retention time then averaged across animal and error treatment and a linear mixed-effects model was fitted to the mean and standard deviation of the difference between the true and estimated parameters. Based on the this analysis, employing technologies that decrease additive errors to less than 5 grams of marker/kilogram of fecal dry matter, will deceased bias in the estimates of ', t, gastrointestinal mean retention time, and ruminal retention time. Further, with an additive error, the bias and standard error will be diminished with increasing dose weight and analytical precision. With multiplicative errors, this analysis shows this types of error can be managed by the G2G1 model as long as the error is minimized to less that approximately 5% and additive errors are less than 10 grams of marker/kilogram of fecal dry matter. With a multiplicative error, the bias and standard deviation will not be effected by dose weight but by inaccurate dose administration. Hence, managing the multiplicative error rate techniques requires the development of accurate and precise dosing methods.