Articles and Resources

Formulation Of Diets On A Digestible Amino Acid Basis And Factors Affecting Digestibility Of Feedstu

Introduction
Over the past several decades, the poultry industry has moved from formulation of diets based on a crude protein basis to a least-cost and total amino acid (AA) basis, in part due to the introduction of affordable computing power. More recently, the industry has begun to switch to newer concepts that may prove cost effective. These include digestible formulation, precision feeding, ideal proteins, modeling. All concepts that have been presented to the poultry industry and have been adopted world wide to some extent. As we look at the industry, we see the entire gambit of methods to provide a nutritional regimen to poultry. In some parts of the world, nutritionists will still use the old Pearson Square as a diet formulation tool and most will formulate on a crude protein basis. In the undergraduate Monogastric Nutrition class I teach, we use the hand calculation method at the beginning of the semester to give them an appreciation for the professional computer formulation package they will use during the rest of the class to formulate far more complex rations based on total amino acids, digestible amino acids and ideal protein ratios. While the industry in the U.S. and much of the world has made the switch to computer formulation, the switch to what may be referred to as more advanced formulation methods has been a bit spottier. A solid percentage of the birds now being fed are using digestible amino acid formulation. We¡ãOll look at some of the arguments for making the switch to mor advanced formulation methods, show several ways to make the switch and look at factors that affect digestibility and how to maintain a database on feed ingredients.

Digestible Formulation Saves Money
Very few nutritionists would tell anyone that they do a poor job with their diets and that they need help desperately. In fact, most folks are doing an adequate job in terms of bird performance. The real reason to look at newer formulation methods is strictly an issue of money. Providing maximum growth for birds is generally fairly easy, just overfeed all of the nutrients that are required and you will probably come close to maximal growth. But at what cost? When we factor the economic side into the equation we find that overfeeding can be expensive. Thus, least cost diet formulation. Unfortunately, least cost diets are not necessarily the most cost effective diets. A cheap diet that reduces bird performance is not very cheap. We should be looking at least cost per pound of gain or even better least cost per pound of salable meat. To achieve optimal cost efficiency, we need to feed as close to the true requirement as possible to get the performance we seek without overfeeding nutrients, what folks in the U.S. are now terming precision feeding. One of the main ways to feed to the true requirements is by using digestible values for AA. A number of people over the years have questioned this or stated that it wasn¡ãOt terribly valuable and that what we are currently doing is fine both from a bird standpoin and an economic standpoint. In turn, I have then asked how many of those folks feed birds based on gross energy values or total phosphorus levels in the diet? I¡ãOm sure that at some point in the pas we did that as well, but no one does it now because we all agree that use of some measure of availability for energy and phosphorus improves accuracy of the formulation and will save us money. A similar argument holds for digestible AA formulations and could be argued to be the next logical step in poultry nutrition.

But is it worth it? Switching to a different method of formulation will take some time to get set up and some work continuously to keep digestible nutrient levels updated. Some of the other aspects of precision feeding will also increase the management costs as well. Let¡ãOs look at digestible formulatio first and see how money can be saved. Digestible formulation is really about two things. First, feeding the correct requirement and second, proper pricing of ingredients based on the available nutrients. For example, if we compare a soybean meal sample with a meat meal sample on a total AA basis, we may find that lysine content of both products is about 3.00%. If we look at these same products on a digestible lysine basis, we will note that the soybean meal is generally higher in digestibility, in some cases as much as 10%. If our soy sample is 92% versus 82% digestible lysine, the actual lysine content of each product available for bird use is 2.76% versus 2.46%, a substantial difference. While we are looking only at a simplistic view of a single amino acid, the point is that the value of the product should be based on its usable content rather than its total content with respect to AA. Once again, we wouldn¡ãOt think of using gross energy values when data is available for metabolizable energy. I used correctly it thus appears that digestible AA formulation will save some money. When checking with US industry nutritionists, it appears that $1-2/ton (USD) are saved

Digestible Formulation Results In More Accurate Requirements
Digestible amino acids result in the requirement being expressed on a more accurate basis. We have shown this in our research trials when using higher levels of by-product additions formulated on a digestible versus total amino acid basis. However, let me provide a simplistic calculation to make the point. Let¡ãOs say we wish to provide 1.00% total lysine or 0.90% digestible lysine to meet a requiremen for our birds from a feed ingredient that contains 3.00% lysine that is 90% digestible or contains 2.70% digestible lysine (such as soybean meal). To meet this lysine requirement, we can feed 33.3% soybean meal (3.00 x .333 = 1.00% total or .90% digestible). We could also feed a by-product (such as meat meal) at the same level which also happens to contain 3.00% lysine. The problem occurs if the digestibility of lysine is lower, say 80% or 2.40 digestible lysine versus 2.70%. On a total lysine basis, we are meeting the requirement (3.00 x .333 = 1.00%) with either ingredient. However, with the lower digestibility ingredient we are only providing 2.40 x .333 = .80% digestible lysine whereas the higher digestibility of lysine in the first ingredient met the requirement. This is taken into account when formulating on a digestible basis, but not when formulating on a total basis. The result: lower lysine levels as the amount of by-product is increased. This is not a problem if formulating on a digestible basis, although the higher digestibility ingredients may be included at higher rates when formulating on a digestible basis.

Digestible Formulation Allows For Reduced Protein And Increased Energy Without Fat Additions
One of our nutrition professors asked me one time why we feed such low energy rations to growing turkeys in the early stages of growth. If you formulate turkey rations you know this is due to the high protein contents of the starter rations which necessitates high levels of high protein, low energy feeds. The only solution to this is the addition of substantial quantities of fat to the ration. The same is true when formulating broiler rations, less fat may be needed to maintain the energy content of the diet which can now be provided by grains rather than fat.

Digestible Formulation Allows For More Precise Determination Of Amino Acid Requirements
When a researcher attempts to determine the requirement for an expensive nutrient such as the amino acids, precision is important. Formulation on a digestible basis is more precise in terms of amino acid delivery across the intestinal lumen. As an example, two turkey diets are shown below (Table 1). Both diets have similar total amino acid contents, but when compared on a digestible basis they differ by .07% on lysine, .08% on methionine and .02% on threonine. We see this also when we attempt to do titrations of an amino acid such as lysine. Generally, what is done is that a reduced protein diet is formulated and crystalline amino acids are added back. If we were determining the requirement for lysine, we might add lysine back in small increments until we found the level of lysine that would support optimal growth. The problem comes in when we then utilize this requirement data for a regular diet formulation. Let¡ãOs say we added back 0.30% lysine to our basal ration to get to th lysine requirement. This 0.30% lysine was added as crystalline lysine which is considered to be 100% digestible. When we formulate a standard ration, this is replaced with ingredients such as soybean meal or meat and bone meal which are not 100% digestible. If we did this on a total lysine basis rather than digestible, we have just inadvertently come up with the wrong requirement.

Digestible Formulation Is Useful Even If You Just Feed Corn And Soybean Meal
A number of people think that they don¡ãOt need to formulate on a digestible basis because they primaril feed corn and soybean meal. While I would certainly agree that it is more important if one feeds a complex diet, corn-soybean meal diets alone do not change my opinion. Just the precision argument noted above is reason enough to make the switch. The other question I always come back to is why don¡ãOt we put more ingredients into our matrix to cheapen the ration? Certainly corn-soy alone ma provide an adequate ration, but the benefits of least cost ration balancing mostly disappear if there are no feedstuff choices for the computer to balance with.

Time To Make The Switch?
All in all it appears that the benefits of digestible formulation are significant. The downside is that you will need to set things up on your computer to do this and may need to do some routine feedstuff assays to get digestible values. The potential savings are quite real however and many companies have made the change. Let¡ãOs next look at what¡ãOs involved in making the change both from t standpoint of the computer changes needed, databases to utilize, assays that can be done and what factors affect digestibility of feedstuffs.

Data Needed For Making The Switch
There are basically two parts to the data needed to make the switch. These include a database of the feedstuffs that will be commonly used and digestible requirement data. Let¡ãOs look first at th feedstuffs data needed and input into the computer.

Feedstuffs Database
There are a number of locations to access feedstuff digestibility values and there are several things that should be looked at in regards to this. One of these is the methodology with which the data was collected, the number of samples that have been tested and if the data was then used for requirement determination. In other words, if I am using data from cecectomized roosters for the digestibility values, is the requirement that has been determined based on this same type of data? A number of sources are available for the database needs. These will include your amino acid manufacturers, some published literature and in some cases computerized databases. I have included some of our data using the cecectomized rooster model for a starting point (Table 2).

Collection Methods
There are a variety of methods for collecting digestibility values, but all of the in vivo methods revolve around some form of feeding the animal and collection of excreta. In the U.S., a common method is to utilize cecectomized roosters (roosters with the ceca surgically removed). The rooster is removed from feed to clear the gut, force fed a known quantity of feed and its excreta collected for a period of time. Another bird is either left off feed or fed a non-nitrogenous feedstuff and excreta collected for adjustment to account for endogenous loss or loss that occurs regardless of the feed being fed. Other methods involve feeding with a undigestible marker that allows quantitation of feed in the gut and collection of feces from the small intestine after killing the bird. This tends to be more expensive and time consuming. Either method will provide a reasonable representation of what portion of the amino acids are being absorbed by the bird.

Factors Affecting Feedstuff Digestibility
A number of factors can affect the digestibility of feedstuffs both from the bird standpoint, diet standpoint and the from the feedstuff standpoint. Looking first at the bird, variation may occur due to ages, gender, species or strain, environmental temperature, gut length, gut conditions and level of feed intake. While there is little hard and fast data on these affects, it appears the very young bird does not digest feed as well as after several weeks of life and probably the older bird may depress digestibility somewhat as well. Dietary factors can also affect feedstuff digestibility. We have shown that high levels of dietary fat can improve digestibility of feeds with lower digestibility coefficients. High levels of dietary fiber can reduce digestibility of feeds as well. There is little data on interactions of different feedstuffs in the gut either. While these factors have been shown to occur, for the most part they are of little practical significance. Feedstuff digestibility may also occur with less differential in the grains and well controlled processed feeds such as U.S. soybean meal. More problems occur when one looks at feeds such as by-product meals where there may be differential inputs of fat, ash (from bone), whole carcass, etc. as well as different cooking conditions due to these changes. This can result in overcooking in some cases and leads to changes in digestibility values. One group of research samples of a meat product changed digestibilities each day of product output. Similar effects can be had in processed grain products where different levels of hull may be left in the product causing changes in fiber levels and thus potential changes in digestibilities of the product. The bottom line here is that one should attempt to get solid suppliers that provide a similar and consistent product.

Computer Input
There are several ways to input data into the computer. First, one can split things up based on new nutrients versus new ingredients. Using new nutrients is probably the method of choice if there are experienced operators doing the formulation. Basically, one adds new nutrients on to the database for each ingredient such that we now have lysine (or total lysine) and diglysine (or digestible lysine). If doing this from scratch, one would be best served by putting each amino acid and its digestible counterpart following for ease of comparison. Having taught computer formulation to novices (undergraduate students) for many years, we tend towards the other method which is to make separate feedstuffs for use of digestible values or corn (corn on total AA basis) versus digcorn (corn on a digestible AA basis). I have found there are fewer errors with this method as there is no question that they have chosen an ingredient based on digestible values.

The next data input question is how the digestibility values are calculated and entered. Two basic methods can be used: 1) Do the calculation first followed by input into the database. Basically, this involves taking the level of the total AA such as lysine multiplied by the digestibility coefficient in percent (%). So if lysine is 1.00% total basis multiplied by a digestibility coefficient of 85% the digestible lysine is .85%. In some cases, the coefficients are not expressed as percent and must be converted. 2) The other method is to input an equation into the database with the digestibility coefficient multiplied by the total AA such that the computer does the calculation. Again, the second method may be preferred by the experienced user, whereas the less experienced may do the calculations by hand or on a spreadsheet to keep things simpler.

Requirement Data
While there appears to be a good number of sources of data for digestibility coefficients, there is less data available on requirements for broilers and while data for layers is coming soon, I found nothing published in journals as yet. There are several ways to get values on a digestible basis. Before good data was out, many people used the NRC requirements multiplied by some averaged digestibility value (85-90%). Another way to come up with requirement data is to do what I refer to as back calculation. With this method, we will use a diet formulation based on a company¡ãOs current total AA requirement and back calculate to determine the currently fed digestible requirements. This is easily done by using the method above to get the digestible AA into the database and adding new nutrients to the formulation (digestible lysine for instance). One then formulates on a total basis and then looks at the values obtained on a digestible basis to see what current requirements are being fed based on a digestible AA basis. These values are then used for the digestible requirements. The advantages of this are twofold. First, one will basically feed the same AA levels as are currently being fed and that one is comfortable with. Second, this allows us to compare the currently fed digestible requirements with the book values found through research on digestible values. Below are values for broilers from the group at Illinois based on their Illinois Ideal Protein Ratio (Tables 3 and 4). The ideal protein is the exact ratio of amino acids needed to provide optimal performance without excess and is based on lysine at 100% with all other AA in a ratio to lysine. Theoretically, as the requirement for lysine changes (for example from strain x to strain y), all other AA will change related to that.

A variety of factors potentially affect the requirements of the AA including gender, strain, age, temperature, energy content of the feed, etc. These factors have not been well researched to date.

Future Formulation Technologies
While making the switch to digestible formulation is the first step in moving towards more precise feeding, there are other steps that one can take in that direction. One of the next steps is formulation on an AA basis, where the protein constraint is totally removed. This requires excellent information on the digestible AA requirements such that the order of limitation is known to one AA beyond what is being provided as crystalline in the diet. This allows the protein to be as low as possible with intact protein providing the first limiting AA after the crystallines. Another area of interest is the lysine (in an ideal protein relationship) to energy ratio. This will help balance energy needs and AA needs more precisely as well as allowing for calorie cost formulation to be more effective. Ultimately, we will model all of the data and the computer will help with a great deal of the decision making needed in the future.

References
Baker, D.H., and Y. Han, 1994. Ideal amino acid profile for chicks during the first three weeks posthatching. Poultry Sci. 73:1441-1447.
Emmert, J.L., and D.H. Baker. 1997. Use of the ideal protein concept for precision formulation of amino acid levels in broiler diets. Journal Appl. Poultry Res. 6:462-470.
Firman, J. D. and J. C. Remus, 1993. Amino acid digestibilities of feedstuffs in female turkeys. J. Applied Poultry Research. 2:171-176.
Firman, J.D. and J.C. Remus, 1994. Fat additions increase digestibility of meat and bone meal. J. Applied Poultry Research 3:80-83.

Tags · Feed formulation · Diet · Digestibility · Amino Acid · Feedstuffs · Protein

17.02.2008. 09:08