Evaluating Feed Quality

Feed is considered as the most expensive component (it can be up to 75 per cent depending on the animal species) of the total production cost in the animal industries. Volatility in feed ingredient prices, particularly over the past few years, has been very challenging for the livestock industry. Expansion of the biofuel industry and increasing demand for human food are among the important contributing factors to the current upward trend in feed costs. It is becoming much more important to develop pragmatic approaches in order to alleviate some of the negative impact of these challenges on the industry. 

Considering the limitations associated with the supply of feedstuffs on a global basis, it is very important for the industry to be able to obtain the most nutrients possible out of a feedstuff. A critical step in this process is to know the actual digestible nutrients available to the animal from feed ingredients such as wheat, corn, barley, soybean, peas, and Dried Distillers Grain with Solubles (DDGS), so that the diet formulation can meet the animal’s requirements more closely. 

The nutrient content of feedstuffs is normally governed by genetics and environment. Feedstuffs are usually obtained from different geographical locations and, as a result, their nutrient content can vary substantially. We not only need to have a good understanding of these variations, but we also need to use practical approaches to predict these variations for diet formulation. Increasing the accuracy of diet formulation should improve animal production performance and reduce nutrient excretion due to over-formulation.

What can we do?
There are a few approaches that can be taken when it comes to the prediction of nutritive value of feed ingredients.  However, it must be noted the accuracy of findings resulting from these approaches can be quite different.

Physical measurements
Physical measurements such as bushel weight have been used to assess feed quality. However, it has been shown that, in general, these measurements are not good indicators of digestible nutrient content. 

Chemical analyses
Analyzing samples in the laboratory can certainly provide good information on the nutrient or proximate content (e.g., protein, fat, fibre) of a feedstuff or ration, but these analyses are time consuming and expensive and, as a result, cannot be of direct help when an immediate answer is required. In addition, these analyses do not provide information as to the digestibility of the nutrients.

In vivo digestibility technique
The digestibility of nutrients and energy in a feedstuff can be determined by feeding animals (e.g., broiler chickens, laying hens, broiler breeders, turkeys) in an in vivo study. Basically, we measure the difference in nutrient content fed to the animal with that excreted by the animal. The difference is assumed to have been digested. An in vivo experiment is the most accurate approach, but it is a long (e.g., two-week), labour-intensive and expensive procedure and, therefore, has very limited applications for routine feed quality evaluation.  

In vitro digestibility technique
A faster and less expensive approach is to simulate the digestive system of the animal using specialized solutions of the feedstuff with specific enzymes. This technique is called an in vitro assay. In vitro techniques need to be validated using the in vivo assay to ensure that there is a solid relationship between the two methods.  As soon as the in vitro method is validated, it provides the opportunity to analyze large quantities of samples relatively inexpensively.
 
Near Infrared Reflectance Spectroscopy (NIRS) technology
NIRS technology is currently available to the animal feed industry as it is used in many commercial laboratories around the world. By using this technology, the nutrient content of different feedstuffs can be predicted quickly (within a few minutes). “Real-time” analysis supports real-time decision making for both the sellers and the buyers of the feedstuffs. In other words, payment for a load of feedstuff arriving at the feed mill could be based on the actual feeding value of that specific load.

NIRS technology is a secondary method in that the sample is not altered nor destroyed and no chemicals are required for analysis. The technology relates the amount of light energy absorbed or transmitted by hydrogen-containing molecules in the feedstuff to its chemical composition or digestibility. NIRS needs to be specifically calibrated to estimate nutrient content or digestibility of different feedstuffs such as wheat, corn, barley, soybean meal, peas, and DDGS for each of poultry, pigs and cattle.  Developing the calibration models relies on a reference method such as the in vivo or in vitro techniques discussed previously. Creation, maintenance, and updating of the calibration databases are very important and these are done using the data provided through laboratory analyses of unknown samples. 

Another interesting avenue that NIRS may move toward in the near future is prediction of animal performance, which can be of significant benefit to the animal industries. This gives the industry an opportunity to relate performance of the animals to NIRS-predicted feeding value of feedstuffs.  

Where do we go from here?
Researchers at the University of Alberta, Agriculture and Agri-Food Canada (Lethbridge) and Alberta Agriculture and Rural Development (Lacombe) with support of industry and funding agencies such as the Alberta Crop Industry Development Fund (ACIDF) have been working on linking the above-mentioned approaches (i.e., testing in animals or in vivo technique, in vitro digestibility techniques, and NIIRS technology) to help the crop and livestock producers to have access to a fast and accurate tool for the evaluation of nutritional quality of feed ingredients in poultry, pigs and cattle. Wheat is one of the main cereal grains included in poultry rations, especially in Western Canada. Accurate and fast prediction of energy content of different batches of wheat for poultry diet formulation is of critical importance to the industry. It is hoped that current efforts will lead to the creation of robust NIRS calibration databases for the prediction of nutritive value of wheat in poultry. n

1. University of Alberta
2. Alberta Agriculture and Rural Development, Alberta, Canada

Acknowledgments
The authors would like to thank all project supporters, including ACIDF, Alberta Agriculture and Rural Development, and University of Alberta, for their continuous support.  

Contact information
Dr. Mary Lou Swift: 403-782-8693; mary-lou.swift@gov.ab.ca
Dr. Doug Korver: 780-492-3990; doug.korver@ualberta.ca

References