By Dr. Robert Swick, NSM poultry nutrition consultant

Broiler chickens consume more soy than any other class of animal. Feed ingredients, including corn, soybean meal (SBM) and edible oil, are variable in terms of anti-nutritional factors and digestible nutrient content. Because feed ingredient specifications are often left vague, nutritionists always try to ensure nutrient requirements are met so that the broilers are being fed feed that allows them to perform to their maximum potential. When there is uncertainty, nutritionists may apply safety margins to critical amino acids, such as digestible lysine, methionine and threonine, and/or metabolizable energy. Variation comes from ingredient origin, processing, shipping and handling, sampling strategy and errors, laboratory analytical errors and application of statistics. To manage variation in the finished feed, an estimate of the variability of each nutrient in each ingredient is required.

Nutritionists must try to get a handle on how much of a safety margin to use. A few strategies include: 1) do nothing and use a published table value of your choice; 2) use the most recently analyzed values from the laboratory; 3) subtract one standard deviation of variability for each nutrient derived from a table or actual analyses over time; 4) use the average of recent laboratory analysis and values existing in the nutrient matrix; 5) adjust five to ten percent below average values obtained in the laboratory; and 6) a combination of the above. Margins may be gradually increased or decreased in two to five percent increments as animal performance continues to be evaluated.

Many nutritionists sample a load of incoming raw material, have it analyzed and formulate based on that information. Common practice is that when a new analysis is obtained, the previous data is tossed, and a new feed is formulated based on the new data because the inherent assumption is that the new data better represents the ingredients than the old data. However, this may or may not be true. When new analytical data is obtained, the nutritionist must ask whether there is a good reason why the composition should be changed. The new number may be no better than the old number and an average of both numbers may have the lowest probability of being wrong. The mean, rather than the new or old number, may be the correct strategy.

The following is a non-exclusive list of cases that call for margins of safety:

• Label guarantees – By definition, in least-cost formulations, the probability for any nutrient to be below target is 50 percent, so formulas may be slightly over-formulated to meet the declared concentrations in feed labels as required by law.
• Nutrient variability – Certain ingredients, such as meat and bone meal or DDGS, can be highly variable in terms of nutrient composition. Agro-industrial products of unknown quality or origin can cause problems if they are not assayed frequently.
• Formulation basis – Diets formulated on total rather than digestible nutrient basis are often over-fortified to account for unknown variability in nutrient digestibility, especially when using alternative ingredients.
• Nutrient losses – Several nutrients, like amino acids and vitamins, are susceptible to destruction during thermal processing (pelleting, expanding and flaking) and during prolonged storage where significant losses of bioavailability may occur.
• The “average” animal – Published nutrient requirements usually refer to the “average” broiler but this definition may leave half of the animals (such as the males) undernourished while the other half (the females) may be overfed.

Fortunately, recent advances in near-infrared reflectance spectroscopy (NIRS) have made the task of estimating critical nutrients such as digestible amino acids and metabolizable energy much faster and easier. The NIRS uses amino acid digestibility calibrations – developed by major multi-national companies like Adisseo, Evonik, Trouw and AB Vista – created from previous measurements using live birds to be able to give answers on digestibility and metabolizable energy.

In the future, the industry may be able to use the data from NIRS calibrations on purchase contracts to value SBM and other commodities based on digestible amino acids and energy which could replace today’s simple lab measurements of moisture, crude protein, crude fiber, crude fat and ash. The beauty of the system is that once an NIRS instrument is set up for the calibration, samples can be scanned in minutes, emailed to the company of choice and the final analysis emailed back to the customer.

Overall, feed ingredients are variable in nutrient content and a strategy should be developed to ensure that the finished feed has a high probability of meeting the requirements for digestible amino acids and metabolizable energy. A major disconnect is that SBM is purchased based on its content of crude protein while animals require digestible essential amino acids to meet requirements for maximum growth. The use of NIRS calibrations to return digestible amino acid values offers a way forward for both formulation use and for purchase contracts. The use of high-quality protein sources such as U.S. SBM with appropriate safety margins applied offers an efficient and economically sound approach to broiler production.