Slide 1

Mr Chairman, Ladies and Gentlemen it’s a pleasure to be here and an honor.

Slide 2

What kind of patients are we talking about? Approximately 20% of the patients in the ICU can have acute kidney injury in our days, with a really high mortality.

Slide 3

Most of these patients have ischemia, with or without nephrotoxicity, as cause of acute renal failure.

Slide 4

And if you have a look at what kind of patients you are treating, you are treating old patients, most of them male, most of them with high, actually very high APACHE scores and more than half of these patients have baseline diseases, not just acute kidney injury.

Slide 5

So, what about nutrition in these patients? Let’s try to frame the problem.

Slide 6

If you have a look in Pubmed or research for these kinds of things, it’s quite strange what we can find. If you look for all years, we have around 700 papers, 25% of them reviews. If you have a look at the last 10 years, we have 150 papers and then 35% of these papers are reviews. If we include “humans” in the research what did we find? We found around 600 papers, 30% reviews and in the last ten years around 100 papers with half of reviews. What’s even worst? If we go and look for randomized controlled trials, we can find only 12 papers and only 5 papers in the last 10 years. So it’s an open field for study, no doubt.

Slide 7

What do we know? We know for sure that in the hospital, patients that are malnourished, they die, as we can see in this population of elderly patients.

Slide 8

We know that in multiple organ failure, bigger the negative balance, higher is the mortality. We know that undernutrition is very frequent in AKI patients.

Slide 9

That’s a paper from Fiaccadori, Italy. We see that more than half of the patients that he studied had moderate or severe undernutrition with crescent mortality.

Slide 10

More recently we did the same kind of research in our ICU and the results were even worst. Almost 2/3 of the patients had moderate or severe undernutrition.

Slide 11

Why do they have that?

Slide 12

Well, we all know that acute renal failure is a hypercatabolic state and many different ways go to that: uraemia, sepsis, inflammation, changes in the metabolism and hormones, acidosis and the use of dialysis.

Slide 13

Actually there is a full bunch of changes in the endocrinous system of acute kidney injury patients. We know they have insulin resistance, an increased release of catabolic hormones like catecholamines, glucagon and glucocorticoids. They have a decrease in the release and a resistance to growth factors and they have hyperparathyroidism.

Slide 14

In fact, if we look at a population of patients with acute kidney injury, we can see that they have all cytokines increased as compared to control patients and this is even worst in the patients that die.

Slide 15

We know that if we infuse in rodents TNF-α or TNF-α/IL-1, that increases catabolism, actually without too many changes in the synthesis of proteins.

Slide 16

More than that, the most usual pattern of acute kidney injury seen currently is from septic patients. We know that these patients have higher resting energy expenditure than the AKI-no sepsis or normal individuals.

Slide 17

But one of the problems, maybe one of the main problems, is how to do early diagnosis of undernutrition in these patients.

Slide 18

The classic anthropometric studies are not useful in these patients because they are fluid dependent, they do not detect acute changes and one thing that is more and more used, the bioimpedance, was not validated in AKI.

Slide 19

If you go to biochemical markers, we always think about albumin. Albumin it’s totally insensitive, it’s an acute phase protein and, more than that, has a very long half life, which makes very difficult to detect changes in the nutritional status of these patients. Transferrin is about the same. Prealbumin has renal excretion and can falsely increase in AKI.

Slide 20

I would like to propose a look to insulin-like growth factor or somatomedin C. This protein is considered a sensitive marker of nutritional status. It’s around a 7649 daltons peptide, it’s produced by the liver. The secretion is constant and has a slow release and has a very short half life of 12-15 hours. The plasma levels in adults are around 200 nanograms/ml.

Slide 21

We already know that it’s a really sensitive marker of nutrition. As we can see, when we feed people, IGF-1 goes high very early, earlier than prealbumin, retinol-binding protein and transferrin.

Slide 22

The same is true when we fast individuals and then we can see that IGF-1 levels are dependent on the protein of the diet. When we give a normal diet, we have a very nice and fast increase in IGF-1. With low protein and normal energy, again we have a very nice increase, but if you don’t have protein and you don’t have energy, IGF-1 stays down.

Slide 23

We have studied IGF-1 in acute kidney injury and actually these results are going to be published this year in Critical Care Medicine.
We have shown that patients that die have significantly lower levels of IGF-1. Actually they have lower levels of cholesterol and albumin too, and the other nutritional parameters were about the same in the group that died and in the group that survived.

Slide 24

It’s amazing the number of patients that die with low IGF-1 levels. The mortality was around 90%

Slide 25

That was about the same with total cholesterol, which was already demonstrated before.

Slide 26

If we do a Kaplan-Meier survival curve, it’s very clear that patients with IGF-1 lower than 50 ng/ml die more, significantly more in the frame periods.

Slide 27

If we plot IGF-1 with total cholesterol, we can see that patients with low cholesterol and low IGF-1 have almost 100% of mortality.

Slide 28

When we move to multivariate analysis, IGF-1, sepsis, oliguria and total cholesterol were the factors independently associated to mortality in these patients.

Slide 29

What can we do about the nutrition in these patients?

Slide 30

Well, everything comes from this very early study published in 1973 in the New England Journal of Medicine. What this paper actually shows was that the use of glucose plus essential amino acids decreased significantly the mortality in acute renal failure of patients, even when they go to dialysis.

Slide 31

The problem is that all these studies, which try to confirm these results later, showed no changes in survival.

Slide 32

They tried to see if a meta-analysis could show something. The results were that there was an inadequate description of randomization in much of these studies, not good statistical analysis and only two in four studies had more than 10 patients. So, it’s totally impossible to have a clear conclusion from this kind of studies.

Slide 33

More than that, we all know that hyperglycemia is associated with increased mortality in ICU patients and mortality is even increased with development of ARF, as an independent factor.

Slide 34

So, it’s not so easy to give high levels of glucose to these patients.

Slide 35

What is recommended and what makes sense is to divide this population in low, moderate or high catabolism. When we go from the left to the right, we have to increase the amount of proteins and the amount of energy and we know that the mortality is going to be very high on the right side. Sure we can use oral or we can use parenteral nutrition.

Slide 36

Do we have data about that? Again we have some data from Fiaccadori, from 2005, that studied different intakes of nitrogen and the effects in these patients.

Slide 37

It is a pilot study in 10 patients. They used total parenteral nutrition, all patients were on dialysis and two regimens, 30 kcal/kg/day and 40 kcal/kg/day, lower calorie and higher calorie. The nitrogen intake was kept constant.

Slide 38

What did he demonstrate? He demonstrated the tendency to better nitrogen balance, lower protein catabolic rate and lower nitrogen rate in the higher calorie regimen. But what he showed was a statistically significant difference in triglycerides, glucose, need for insulin and volume infused in the patients with higher calories.

Slide 39

Graphically it’s easier to see that. We can see the tendency of a better nitrogen balance, a lower catabolic rate, a lower urea generation rate but without significant statistics.
On the other hand, these patients with high intake had higher triglycerides, higher glucose and higher need for insulin.

Slide 40

More recently, Singer tried a high-dose amino-acid infusion

Slide 41

And what he demonstrated was that the patients in the arm with higher amino-acid infusion showed better creatinine clearance and better nutrition balance. Actually, this was associated with a smaller use of diuretics, so that’s a possibility.

Slide 42

About enteral nutrition? We have very, very few studies. Again a study from Fiaccadori, and the conclusion that it is possible to give enteral feed to these patients, but we really don’t have solid data about that.

Slide 43

The only, maybe the better data, came from an experimental study that showed that rats receiving enteral nutrition had a better acute renal failure prognosis than rats receiving parenteral nutrition.

Slide 44

Well, if you go to the European Society of Parenteral and Enteral Nutrition Guidelines, what can we see?

Slide 45

We can see that we’ve just got "Cs" all over

Slide 46

And we got the table 1 that basically repeats what I have shown in the other table too, i.e. that we have to use 20-30 kcal of energy, 35 g/kgBW/d of carbohydrates and so on.

Slide 47

Again, for the indication for the route, for the kind of formulation, we found “Cs” all the way, not even a “B”.

Slide 48

In summary, malnutrition is frequent in ICU AKI patients and that’s a fact. Nutritional therapy in AKI must be individualized, that’s a fact. The degree of catabolism, the associated diseases, the nutritional status and the dialysis modality should be taken into account: that’s probable, but it’s not a fact.-

Slide 49

Nutritional markers for nutritional evaluation in severely ill patients are insensitive or are time-consuming and that’s a fact. IGF-1 might be a good marker and that’s a possibility.

Slide 50

Finally, at the moment there is no clear evidence supporting the concept that nutritional support changes AKI natural history and that’s a fact. If the patient is malnourished or has increased catabolism, the nutritional support should be initiated early. That’s very probable.

Slide 51

Thank you for your attention.

Slide 52

Chairman: Well, thank you very much Doctor Burdmann, an excellent talk and are there any comments or questions from the audience?

Question: If I understand you correctly, you mainly pointed out that malnutrition in patients with AKI especially if in septic patient, the real patient in ICU, is more a state than just something you can do anything about it. You have to wait until the patient gets better and then you have to feed. Is that correct?

Dr Burdmann: I would say that’s totally correct. I agree.

Question: There has been a time, 5-10 years ago, that many people advocated the continuous replacement modality as a way to increase the proteins that you could deliver to these patients. Can you comment on that opinion?

Dr Burdmann: I think you really get the problem. We don’t know, we don’t know how much protein we have to give. Apparently a higher amount of proteins is better, but we have 2 studies and that’s it. We don’t know how much carbohydrate we should give, again apparently more is better but we have studies showing the opposite. We don’t know if we have to go by enteral, parenteral or both. We think both are better, but again, there is no evidence. What’s even worst is that we don’t know what to do with the patients on renal replacement therapy. Should we give oligoelements? Should we give vitamins? Again, we have pros and cons. So, I think we have a really big field to study. What we do know is that malnutrition is very frequent and it is associated with mortality so we have to work on that.

Question: So it’s probably more a marker of disease than something more?

Dr Burdmann: I am betting on the IGF-1. What we are doing just now is that we are trying to modify the nutrition in patients with very low IGF-1 and to see if we can change these levels and to see if these changes could be associated with better survival, but this is just ongoing.

Question: Can I make a comment? I’m very curious about the influence of peritoneal dialysis in critical patients on hyperglycemia, at least in paediatric patients this is very often a big problem. How can you cope with this possible influence of PD on glucose levels and all the indications you get from the glucose level itself?

Dr Burdmann: Well, actually we do a great number of PD in Brazil and South America, but mainly for children and for low catabolic patients. Hyperglycemia is a problem, but when you compare the survival of patients on PD versus other kinds of renal replacement therapies; SLED or even hemofiltration, we don’t observe a higher mortality in PD, but there’s a big bias because of the selection of the patients. But hyperglycemia is a problem and the loss of albumin by day it is a problem too. So, actually we have to give more protein to these patients.

Chairman: Yes please.

Question: We will repeat the question, just one second. Can in the meanwhile people from the technical staff bring us some mobile microphones for questions? So please can you just shortly rephrase the question and give the answer?

Dr Burdmann: If I got the question correctly, we are talking about the propensity to not use a high amount of carbohydrates because produce more CO2. But you see what’s even more confusing, we are talking about nutrition in a nephrology meeting and when we are called to go to the ICU to see a patient because normally they need to do dialysis and you ask these guys what kind of nutrition are they giving, they don’t know. They should be taking care of this and we are worried about that. So, again, even in the ICU it’s not done by the intensivists, this is not very well taken care and I think again that is really an open field to study.

Chairman: Thank you very much Doctor Burdmann for this interesting lecture and I hope that everybody in the room understands that there’s a lot of work still to be done in this area and that maybe the nephrologists might be more suitable for this type of task than we think of ourselves.