Slide 1 Hi there. I'm Hans H. Stein, a professor at the University of Illinois, and I'm here today to talk to you about standardized total tract digestibility of phosphorus by pigs. And I will go over some of the principles that are involved in calculating these values. Slide 2 So we'll talk a little bit about how to determine phosphorus digestibility in different feed ingredients, and we will talk about the different expressions for the digestibility of phosphorus that we may use. And this includes apparent digestibility, true digestibility, and standardized digestibility. And we will talk about the basal and total endogenous losses of phosphorus, because these values are needed if we are to calculate values for the true or the standardized digestibility of phosphorus. We will also talk about the effects of fermentation of feed ingredients on phosphorus digestibility, and I will show you some data on how to formulate diets based on values for the standardized total tract digestibility of phosphorus, including which matrix values we should use for phytase when we formulate these diets. And finally, I'll draw a few conclusions and give you a few perspectives for future research. Slide 3 A few of the overall questions about phosphorus digestibility include: whether we should determine ileal digstibility or total tract digestibility. And that basically determines whether we're using pigs that are cannulated in the distal ileum, or if we simply use pigs that are inserted into a metabolism cage. So we need to make sure that we understand if ileal digestibility or total tract digestibility give us the best values to estimate phosphorus digestibility in feed ingredients. We also need to know if there is an effect of the level of phosphorus in the diets on the digestibility of phosphorus. And we need to know if calcium has an influence on phosphorus digestibility. And finally, we need to know if it's most accurate to formulate diets based on the apparent digestibility, the standardized digestibility, or the true digestibility of phosphorus. So, we'll try to answer all of these questions to make sure we understand how best to formulate diets fed to pigs. Slide 4 The first question was whether to use apparent ileal digestibility or apparent total tract digestibility, also called AID and ATTD. And to answer that question, we have some data here from an experiment that was conducted in 2005. There were three feed ingredients: the low phytate corn, the yellow dent corn, and soybean meal. And both apparent ileal digestibility and apparent total tract digestibility values were determined. And we can see here for all three ingredients there is no difference between the apparent ileal digestibility values and the apparent total tract digestibility values. Slide 5 And in the second experiment, that was published in 2007, three different diets were used. We had a control diet, which is shown in the blue bars, that's a straight corn/soybean meal diet, then we have that same diet with 500 units of phytase; that's the yellow bars. And finally, we had the control diet, with 1000 units of phytase; that's the red bars. And the phytase that was used in the diets with 500 and 1000 units of phytase was OptiPhos. We can see here that there's very little disappearance of phosphorus prior to the duodenum -- less than 10% -- and there's no difference between the three diets. However, when we get to the end of the ileum, there is a big difference between the control diet and the two diets containing phytase, and we can see that the digestibility increased from about 32% up to about 60%. However, there was no difference between the two diets that contained phytase. We also note here that there is no difference between values for the ileal digestibility and values for the total tract digestibility regardless of the diets that were fed. So, in agreement with the previous experiment, we see here that values for the ileal digestibility are no different from values for the total tract digestibility. That means that we can use either values for ileal digestibility or values for total tract digestibility, because there is no difference between these values. However, it is easier and less tedious and less expensive to determine values for total tract digestibility rather than values for ileal digestibility. So, in most cases, we decide to determine values for total tract digestibility. Slide 6 The next question, then, is: Do we need to determine apparent total tract digestibility values, standardized total tract digestibility values, or true total tract digestibility values? And values for the apparent total tract digestibility of phosphorus are calculated very easily from pigs inserted into metabolism cages, and we simply measure the total intake of phosphorus and we subtract the fecal output of phosphorus. And when we do that, we have the apparent total tract digestibility values. If we are to calculate values for the true total tract digestibility of phosphorus, then we need to determine the intake again, which is easily done by weighing out the feed that we supply every day. We also need to have the output of phosphorus in the feces; that's also relatively easily determined by collecting the feces and analyzing for phosphorus. However, we then need to subtract the total endogenous loss of phosphorus from the fecal output of phosphorus, and by doing that, we can calculate the true total tract digestibility of phosphorus. Likewise, if we are to calculate the standardized total tract digestibility of phosphorus, we need to have the intake, the output, and the endogenous loss. However, in this case, it's not the total endogenous loss of phosphorus we subtract from the fecal output of phosphorus. In this case, instead, it is the basal endogenous loss. So, it appears here that if we are to determine true total tract digestibility or standardized total tract digestibility, then we need to determine either the total endogenous loss of phosphorus, or the basal endogenous loss of phosphorus. Slide 7 If we look at the apparent total tract digestibility values, then we can see that there are some challenges with these values. And this is illustrated in an experiment that was conducted at the University of Guelph about ten years ago. And in this case, four different levels of soybean meal were included in diets, and soybean meal supplied all the phosphorus in these diets. So there was 13.3, 27.3, 40.80, or 54.6% soybean meal in the diets. And what we see here is that as the concentration of soybean meal increased, and therefore the concentration of phosphorus increased in the diets, the apparent total tract digestibility values increased. So if we only had 13.3% soybean meal in the diet, the apparent total tract digestibility value was less than 20%, but as the inclusion of soybean meal increased, that value increased to more than 40%. So the problem here is, we have one feed ingredient -- in this case soybean meal -- but we have four different digestibility values for phosphorus. So, based on this, it is not possible to decide which value is the correct one to use in diet formulation. And the reason we get these different digestibility values is that at relatively low inclusion rates of phosphorus, we get a relatively high contribution of the endogenous loss into the total fecal output of phosphorus. And therefore, we get a low calculated value for apparent total tract digestibility of phosphorus. So this experiment illustrates that values for the apparent total tract digestibility of phosphorus are not standardized across different inclusion levels of phosphorus in the diets and therefore, it is difficult to use these values in practical diet formulation. Slide 8 So if we cannot use apparent total tract digestibility, we have to use either true total tract digestibility or standardized total tract digestibility. And as shown before, if we use true total tract digestibility, then we need to determine the total endogenous losses of phosphorus. That has been accomplished in six recent experiments that have been conducted in North America, and the estimates for the total endogenous losses of phosphorus, determined in mg/kg dry matter intake, are shown on this slide. And we can tell here that there is a variation in these estimates from 8 mg/kg dry matter intake, and all the way up to 670 mg/kg dry matter intake. So it is clear that our ability to accurately determine the total endogenous loss of phosphorus is not very good. And that is a big limitation to this procedure, and it makes it very difficult to calculate values for the true total tract digestibility of phosphorus. Slide 9 Alternatively, we can calculate standardized total tract digestibility of phosphorus, and to do that we need to determine the basal endogenous losses. And here are data from ten different experiments that have been conducted, and we can see here that there is much less variability among different experiments, and we can relatively accurately determine the basal endogenous losses of phosphorus. And on average, over ten different experiments, we have total endogenous losses of 199 mg/kg dry matter intake. So, we can conclude from this that we can calculate values for standardized total tract digestibility by correcting apparent total tract digestibility values for basal endogenous losses, and we can use a value for basal endogenous losses of about 199 or 200 mg/kg dry matter intake. Slide 10 So, by using that value, we can correct the apparent total tract digestibility values for the basal endogenous losses and by doing that, we can calculate values for the standardized total tract digestibility of phosphorus. Slide 11 To illustrate the importance of using values for standardized total tract digestibility, we will look at an experiment that was conducted here at the University of Illinois a few years ago. In this case, we had three different whey products: whey powder, whey permeate, and a low-ash whey permeate. And we can see that the phosphorus concentration in these three ingredients was 0.63%, 0.57%, and 0.10% in the low-ash permeate. Slide 12 We determined the apparent total tract digestibility of phosphorus in each of these three ingredients, and we can see here that there was no difference in the apparent total tract digestibility of phosphorus between the whey powder and whey permeate; however, the low-ash permeate had a much lower value for the apparent total tract digestibility of phosphorus than had the other two ingredients. Slide 13 However, we also determined the basal endogenous losses of phosphorus in this experiment, and when we corrected the apparent total tract digestibility values for the basal endogenous losses, we could calculate standardized total tract digestibility of phosphorus. And we have that represented here in the blue bars, and we see that there is no difference among the three ingredients in the standardized total tract digestibility of phosphorus. So, this clearly illustrates that values for the apparent total tract digestibility -- which are the orange bars here -- they are not accurately estimating the digestibility of phosphorus in the low-ash permeate because the apparent total tract digestibility value underestimated the standardized digestibility in this ingredient. So, again we see here that apparent total tract digestibility values are inaccurate and should not be used in diet formulation. In contrast, values for the standardized total tract digestibility apear to be more constant across different ingredients. Slide 14 We can also get to this conclusion by looking at the data that were presented a little while ago by Fan et al., where he had four different levels of soybean meal included in diets fed to pigs. We already looked at the apparent total tract digestibility values and concluded that there was great differences among these four different diets depending on what the level of inclusion of soybean meal in the diet was. However, if we correct the apparent total tract digestibility values for basal endogenous losses and calculate standardized total tract digestibility values -- which are shown here in the blue bars -- then we will see that there is much less variability among diets. And again it appears that we have much more accurate values when we use the standardized total tract digestibility values because these values are not influenced by the level of phosphorus in the diets. So therefore, we can have one value for the total tract digestibility of phosphorus in soybean meal and we can use this value in practical diet formulations. Slide 15 So, to conclude on this, we have seen that values for the basal endogenous phosphorus losses are less variable than values for total endogenous losses. That means that we can calculate values for the standardized total tract digestibility of phosphorus in our feed ingredients, whereas it is difficult to calculate values for the true total tract digestibility of phosphorus because values for total endogenous losses are more variable. When we calculate values for the standardized total tract digestibility, we can formulate diets that are independent of the phosphorus concentration in each ingredient, and therefore the values that we have for each ingredient are additive in mixed diets. And that, of course, is the objective when we formulate mixed diets and feed them to pigs. So based on this, we conclude that values for standardized total tract digestibility of phosphorus are the values we should use when we formulate diets fed to pigs. Slide 16 However, it is possible that the concentration of calcium in the diets can affect phosphorus digestibility. So, we have conducted a few experiments to investigate this topic. Slide 17 Here's an experiment in which we included six different levels of calcium carbonate, and therefore we had six different levels of calcium in diets that were fed to pigs. And we can see here that when we calculate the apparent total tract digestibility of calcium for each of these six diets, there is actually no difference in the calculated values, and that means that the level of calcium in the diet does not affect the apparent total tract digestibility of calcium. Slide 18 However, when we calculate the apparent total tract digestibility of phosphorus in these same six diets, then we see that there is a linear decrease in the apparent total tract digestibility values of phosphorus as we increase the level of calcium carbonate in the diets and therefore increase the level of calcium in the diets. So what this means is that our values for phosphorus digestibility are influenced by the concentration of calcium in the diet. So we need to make sure that when we formulate diets that are used to determine phosphorus digestibility, then we have calcium levels in these diets that are not in excess of the requirement, because if we have excess calcium, then we will reduce phosphorus digestibility. Slide 19 We also conducted an experiment in which we looked at the effect of phosphorus concentration on the homeostasis of calcium. In this case, we had two diets. One diet contained phosphorus, and the pigs were eating 12.21 grams over the five-day experimental period. That's the orange bars here. And then we had a second diet that contained no phosphorus, depicted in the blue bars. We can see here that the intake of calcium was slightly different between the two diets, and therefore the absorption of calcium was also slightly different. However, when calculated on a percentage basis, the absorption of calcium was not different between the two diets. So the pigs that were fed the diet without any phosphorus, they absorbed as much calcium from the diet as the pigs that were fed the phosphorus-containing diet. However, when pigs were fed no phosphorus, they were not able to synthesize bone tissue, and therefore they were not able to utilize the calcium that they had absorbed. So the majority of the absorbed calcium was actually excreted in the urine, which we can see here in the last two columns. If there was phosphorus in the diets, there was very little calcium excreted in the urine; however, if no phosphorus was in the diet and therefore no phosphorus was available for bone tissue synthesis, then most of the calcium that was absorbed was excreted in the urine. What this means is that the regulation of calcium homeostasis mainly takes place at the renal level and excesses are excreted in the urine, as we can see here. In contrast, there seems to be very little regulation at the intestinal level because the pigs will simply absorb as much calcium as they can, regardless of whether they need that calcium or not. Slide 20 As we're starting to use more alternative feed ingredients, we often use ingredients that have been steeped or fermented during production. And it is therefore relevant to look at the effect of fermentation on the standardized total tract digestibility of phosphorus. Slide 21 We have, therefore, compared the standardized total tract digestibility of phosphorus in four different feed ingredients: corn, distillers dried grains with solubles, high protein distillers dried grains, and corn germ. And of these four ingredients, corn and corn germ -- which are the red bars and the orange bars, respectively -- were not fermented, whereas DDGS and HP DDG -- the yellow bars and the blue bars -- have gone through fermentation in the ethanol plants. And we can see here that the digestibility of phosphorus in the two fermented products -- DDGS and HP DDG -- is much greater than the digestibility of corn and corn germ, which were the two unfermented ingredients. So based on this, it appears that fermentation, as such, increases the digestibility of phosphorus in feed ingredients. Slide 22 We also added phytase to these same feed ingredients, and when we added 500 units of phytase to the ingredients, we see that the differences among the ingredients are much less than they were before. And the reason for that is that the effect of phytase is much greater in corn and corn germ than it is in DDGS and high protein DDG. So by increasing the digestibility of corn and corn germ, and not changing the digestility of DDGS and high protein DDG, we get values for the standardized total tract digestibility that are much more equal among the four ingredients. Slide 23 Another ingredient that is sometimes fermented is soybean meal, and we have determined the standardized total tract digestibility of phosphorus in both unfermented soybean meal and fermented soybean meal. And we can see, we have both of these diets without phytase -- that's the orange bars -- and we have them with phytase, which is the blue bars. And we can see that for conventional soybean meal, shown on the left here, the digestibility of phosphorus is much less than in fermented soybean meal. So, in agreement with the previous data, we see that fermentation of soybean meal increases the digestibility if there is no phytase in the diet. However, when we add phytase to the diets, then we see a relatively big increase in the digestibility of phosphorus in conventional soybean meal but not in the digestibility of the fermented soybean meal. And these data are also in agreement with the previous data where we saw that corn and corn germ had a larger increase in digestibility when we added phytase compared with DDGS and HP DDG. So it appears that fermentation of feed ingredients basically has the same effect as addition of phytase. So when we ferment a feed ingredient, we make more phosphorus available to the pig, and they don't need phytase to be able to absorb that phosphorus. However, if we don't ferment the ingredients, we can get a similar effect by adding phytase to the ingredients. Slide 24 We have now seen that we can determine values for standardized total tract digestibility of phosphorus. So, we will now focus on how to formulate diets based on values for standardized total tract digestibility, and I'll show a few values that were included in a manuscript that was published in the Journal of Animal Science in 2010. Slide 25 In this work, we first determined the effect of phytase on the digestibility of phosphorus in corn, soybean meal, and DDGS. And we have each ingredient without phytase in the orange bars, and we have each ingredient with addition of 500 units of phytase in the blue bars. The values we are showing here are the standardized total tract digestibility of phosphorus. And we can see if there's no phytase in the diet, we have the lowest digestibility of phosphorus in corn, second lowest digestibility in soybean meal, and actually a pretty high digestibility of phosphorus in DDGS. However, when we add phytase to the diets, we see that now the digestibility in corn is more than 60%, the digestibility of phosphorus in soybean meal is over 70%, which is similar to the digestibility of phosphorus in DDGS because the digestibility of phosphorus in DDGS was not increased when we added phytase to this ingredient. So, in agreement with our previous data, we again see that phytase does not increase the digestibility of phosphorus in DDGS, but it works very well in both corn and soybean meals. Slide 26 So, having determined these values, we formulated four different diets that all were supposed to contain 0.32% standardized total tract digestible phosphorus. In the first diet, we used corn and soybean meal, no phytase, and no DDGS. To get enough phosphorus into this diet, we had to add 1.15% dicalcium phosphate. By doing that, we got up to our 0.32% standardized total tract digestible phosphorus. The second diet was similar to the first diet with the exception that we added 500 units of phytase to this diet. So, we still had the corn and soybean meal, but we also had 500 units of phytase, and because the digestibility of phosphorus is greater when we add phytase to corn and soybean meal, then we could reduce the inclusion of dicalcium phosphate to 0.35% instead of 1.15% as we had in Diet 1. The third diet was formulated by using corn, soybean meal, and DDGS, and there was no phytase in this diet. Here, we needed 0.65% dicalcium phosphate to get up to our 0.32% standardized total tract digestible phosphorus. However, in Diet 4, we used again DDGS, soybean meal, and corn, but we also added 500 units of phytase to this diet. And by doing that, we could remove all the dicalcium phosphate from the diet because we had enough standardized total tract digestible phosphorus in this diet simply by using corn, soybean meal, DDGS, and 500 units of phytase. And all of these diets were formulated based on the actual digestibility values of phosphorus that we had determined in the previous experiment. Slide 27 We then fed these four diets to four different groups of pigs from 11 to about 21 kg of body weight. We can see there is no difference in the starting weight between the four different diets, and there is no significant difference in final body weight among diets, although there was a tendency for pigs that were fed the two DDGS-containing diets (which is Diet 3 and Diet 4), they had slightly greater final body weight and also a slightly greater average daily gain and average daily feed intake than pigs fed the corn/soybean meal diets, which were Diets 1 and 2. However, except for these few differences, there were no differences among diets, and we can see that even for pigs fed Diet 4, which contained no dicalcium phosphate, we had very good performance that was not different from that of pigs fed the other three diets. We can also see that the retention of phosphorus was actually increased a little bit in the diets that contained DDGS compared with the diets without DDGS, and that may indicate that we actually had underestimated the digestibility of phosphorus in DDGS by a little bit. However, when we look at phosphorus excretion, we can see that pigs fed the control diet -- the corn/soybean meal diet without any phytase, which was Diet 1 -- they excreted 1.68 g of phosphorus per day in the feces and urine combined. However, when we added phytase to that diet, we could cut the excretion to 0.82 g/day. Pigs fed Diet 3 excreted 1.43 g/day; however, when we added phytase to the DDGS-containing diet, we got down to 0.82 g/day again. What we see here is that by adding phytase to the diets, we could basically reduce phosphorus excretion by 50% from these pigs, and there was a significant effect of both DDGS and of phytase on the excretion. And that means that we can include DDGS or phytase or both DDGS and phytase to the diets and we will reduce excretion of phosphorus from pigs. So these data show that it is possible to formulate diets based on the standardized total tract digestibility of phosphorus without creating any problems for the pigs, and the pigs will perform as expected when they are fed these diets. Slide 28 As we have seen, phytase inclusion in diets can reduce the excretion of phosphorus. So it is important we know how to best formulate diets that also contain phytase. Slide 29 So to do that, we have conducted an experiment in which we had four different levels of phytase: either 0, 420, 720, or 1100 units of phytase, as shown here. And we can see, as we added phytase to the diet, we increased the digestibility of phosphorus as we have seen before. And there was not much change between the 420 and the 1100 units of phytase in these diets. We can, however, describe the response to phytase added to corn by using the equation shown at the bottom of this slide. So the standardized total tract digestibility of phosphorus can be calculated as 42.3 plus 0.059 times the units of phytase that are included in the diet, and then we subtract 0.000028 times the phytase included, squared. And we can see this equation has an r-squared of 0.63, and the model is significant. So what this means is that for every 100 phytase units, we basically increase the digestibility of phosphorus by about 5.9%. Slide 30 We also included corn germ in this experiment, and here we included 0, 390, 910, and 1400 units of phytase and added that to corn germ. In this case, we see again that the digestibility increases as we add phytase to the diet, and we can also see that there was an increase from 390 and up to 910 units of phytase. This response can be described by the equation shown at the bottom of this slide. And here we see that standardized total tract digestibility of phosphorus can be calculated as 35.5 plus 0.067 times the phytase units, and then we subtract 0.000034 times the phytase units, squared. And the r-squared for this equation is 0.79, and the model is also significant. So, this is how we can calculate the response of phytase added to corn germ. Slide 31 We did this same thing for DDGS. And in this case, we can see that the digestibility of DDGS was not significantly improved when we added phytase to the diet. This result is in agreement with our previous data that I have shown, where we also saw that there is no effect of phytase on the digestibility of phosphorus in DDGS. So, the model here is not significant. Although we could calculate an equation, this equation is not significant, so there is no reason to use this equation. Slide 32 When we did the same thing for high protein DDG, we see we did get a significant increase in digestibility. It was, however, a relatively small increase in digestibility when we added phytase to the diets. The model was significant; however, the r-squared here is only 0.36, so it's not a very strong prediction. So what we can see here, basically, is that there is very little effect of phytase to the digestibility of phosphorus in high protein DDG, as we also saw in the previous experiment. Slide 33 If we add this all up, we can see the four different equations that I've shown. And we can see here, based on the blue numbers, that there is a different response to inclusion of phytase in diets containing corn, corn germ, DDGS, and high protein DDG. And what this means is that we will not get the same response every time we add phytase. So the response to phytase will be different among diets that contain different feed ingredients. And that means we cannot use a standard value for the response to phytase when we formulate diets. Instead, we have to use different digestibility values for each ingredient depending on how much phytase we have in the diet. So if we add 500 units of phytase to diets, then we will need to use the digestibility values for each ingredient that are based on inclusion of 500 units of phytase. If we use 250 units of phytase, we need to have a different digestibility value. And that is the way we have to formulate diets fed to pigs. It is not correct to use a fixed matrix value for phytase across all diets because the response is different among diets depending on the formulation of the diets and the ingredients that are included into these diets. Slide 34 So in conclusion, we have seen that values for the standardized total tract digestibility of phosphorus are independent of diet phosphorus concentrations, and therefore these values are additive in mixed diets. And this is important because we don't feed pigs on individual feed ingredients; we always feed them mixed diets, so we need to be able to predict the concentration of digestible phosphorus in these mixed diets. And we can do that if we use values for the standardized total tract digestibility of phosphorus. We have also seen that values for standardized total tract digestibility of phosphorus support pig performance as expected, so we can actually use these values in diet formulation. And we have seen that if we do that, then we can reduce the excretion of phosphorus from the pigs. Slide 35 To calculate values for the standardized total tract digestibility of phosphorus, we need to determine the basal endogenous losses of phosphorus. However, we can use a value of approximately 200 mg/kg dry matter intake across different experiments, and therefore, we do not necessarily have to determine basal endogenous losses in every single experiment. We have also seen that values for the standardized total tract digestibility of phosphorus may better capture the value of phytase than values for apparent total tract digestibility because these values are not influenced by the level of phosphorus in the diets as we saw is the case with apparent total tract digestibility values. It is, therefore, important that values for the standardized total tract digestibility of phosphorus is used in practical diet formuation and also in modeling of phosphorus metabolism in pigs. Slide 36 Finally, we have seen the values for the standardized total tract digestibility of phosphorus may be influenced by phytase; however, that depends on which ingredient we are talking about. As we saw, corn and corn germ and soybean meal, they will have an increased digestibility of phosphorus when phytase is added to the diet, whereas values for the standardized total tract digestibility of phosphorus in fermented soybean meal and DDGS are not influenced by phytase in the diets. It is, therefore, important that matrix values for phytase are added to each ingredient, and not to the phytase premix. So we may need to have three or four or five different values for the standardized total tract digestibility of phosphorus in each ingredient depending on how much phytase we have in the diets. So, we have one value for each ingredient wihtout phytase, we have another values with 250 units of phytase, another one with 500 units of phytase, and so forth. And by formulating diets this way, we can capture the value of phytase and accurately formulate our diets. Slide 37 And with that, I want to acknowledge the people who have conducted the research I have discusssed here. Those are students, post-docs, and technicians in my research laboratory, and without these young people and their hard work, we would not have been able to complete this research.