CME Slides Forum - D. Fouque

A joint Congress by ERA-EDTA and ISN

DIAGNOSTIC CRITERIA FOR PROTEIN-ENERGY WASTING

Denis Fouque, Lyon, France

Chair: Christian Combe, Bordeaux, France

Joel D. Kopple, Santa Monica, USA

fouque

Dr Denis Fouque
Department of Nephrology
Hospital Edouard Herriot
University Lyon 1
Lyon, France

Thank you very much. As you know, it’s still very difficult to make a diagnosis of malnutrition. I’ll try to show you some of the results of discussion of the International Nutrition Society and also if we have gained new tools to diagnose and to predict morbidity and mortality linked to malnutrition.

So the first point is how to define malnutrition and is it possible to predict survival using new tools?

As you know, in every chronic disease there is a particular thing with body weight that is every time you get a complication over time you lose a slight part of your body weight and then over time you’re not able to regain this weight and then at the end of a too big loss of body cells and active body cells you eventually face the death of your patient over time.

So the problem we face particularly in dialysis is to try to maintain this well being body weight before the patient was sick. We have to aggressively treat this nutritional defect. So in 2006 the International Society for Renal Nutrition and Metabolism organized an expert panel and wanted to clarify some of this complex terminology. We have all heard about malnutrition, sarcopenia, uremic cachexia, protein-energy malnutrition, malnutrition inflammation atherosclerosis, MIA syndrome etc.

Then we came out with a proposed new nomenclature on ‘protein energy wasting’. You all this complex flow chart and in fact, in dialysis in CKD, in overall CKD we face standard problems such as anorexia, anaemia, diabetes mellitus, infection but we also have specific problems in response to kidney disease such as volume overload, inflammatory cytokines, endocrine problems like hyperparathyroidism, IgF insulin problems and we eventually face 4 different and 4 independent problems. The first one is malnutrition which is in fact, under nutrition, not enough intake of the patients. The second thing is uremic toxins. We all know about a big number maybe 200, 300 uremic toxins and most of them have not yet been identified even. The third one is inflammation and specifically in CKD we have this inflammation problem which is not well understood but as you know, half of the dialysis patients have a CRP greater than 5 mg/l and you know that CRP is strongly linked to morbidity and mortality. Last we have hyper catabolism which may be for example loss of nutrients during the dialysis technique.

Every patient can face 1, 2, 3 or even 4 of these problems and to reach this protein energy wasting syndrome and eventually if this is very strong, final cachexia. So we defined 4 items, four groups of items. Biology, body mass, muscle and dietary intake. Diagnosis of protein energy wasting is made if at least the patient has one criterion in at least 1 or 4 groups.

So I will go through these different groups and not highlight everyone but just the yellow ones. The first group of biologic criteria which are very simple to measure and to assess contains serum albumin lower than 38 g/l using bromoresol green you have to be careful because this is not always mentioned you have to go back to your lab and even though in the publications this may fulfil some results because it’s not always mentioned. For example, nephelometry gives values lower than 3 g/l lower than BCG for example. So the second one was serum albumin less than 300 mg/ and serum cholesterol less than 1 g/l. The second group is body composition with BMI, weight loss and body fat and I will go over this. First to mention that in fact serum albumin is very low as you know but it is still low and it hasn’t changed since 1995, we have the exact same levels.

This is current data in July 2008 in France over 10.000 patients. The mean serum albumin is 36 g/l. if you put this threshold at 38 g/l you see that at least 75% of our patients enter the first criterion for protein energy wasting.

The second was Prealbumin. Prealbumin is very robust indicator of nutritional status and this is the FineS study which was run 5 years ago and published 2 years ago. Using oral and IDPN nutrition in hemodialysis patients. These patients were malnourished and you can see that the prealbumin at the start of the study was very low and after you start the re-nutrition you see how fast and how sensitive prealbumin is and very quickly after 3 months it almost regains normal value.

When you re-nourish your patients, you can gain at least 50 mg/l which is a quite important variation in this indicator. So, prealbumin is an exquisite nutritional marker and can be used safely and with many indications.

Moreover, if you can measure the prealbumin increase during the first 3 months we did here, if this prealbumin increases by 30 mg, you are able to reduce the mortality of your patient by 60%. So this is the first time we were able to show that manipulating nutritional status could in fact, increase survival in your patient. So an increase in prealbumin by 30 mg/l after you start re-nutrition is an indicator of increased survival.

The second group is body mass and body composition and this is the reason why we picked a BMI less than 23 kg/m². You know that the WHO speaks about malnutrition when the BMI is lower than 18.5 kg/m².. In fact unfortunately, in dialysis we have already malnutrition and risk for mortality with BMI larger than that. This paper from Professor Coppell’s team by Kalantar et al. showed that in fact when the BMI is lower than 23 kg/m² you already increase your risk of mortality.

The same occurs for weight loss. Weight is a very sensitive indicator and you have to measure weight and to pool your weight to see if you have changed. This is of course the dry body weight and you see that as soon as you lose 1 kg this is every 3 months but the loss of 1-2 kg or 2-3kg or 3-4 kg every 3 months increases rapidly the mortality of your patients. This is not a very important variation if you lose 2 kg over 3 months it’s not always easy to discover that.

The third thing is fat mass. Fat mass may not be as evident to measure but right now with have anthropometry, DEXA and there are new tools such as body composition monitor by Fresenius for example which in a very short time, IN less than 1 minute you’re able to derive your body composition quite reliably and you get your fat mass also. This gives YOU the fat mass of the patient. You see this very evident result that when the fat mass of your patient is less than 12-10%, there is a sharp increase in mortality.

The third and fourth groups refer to muscle mass here and dietary intakes. So the muscle mass is less than evident to measure in fact and again we need anthropometry or body composition monitors but still a loss of muscle mass of about 5% over 3 months or by over 10% over 6 months is very indicative of a wasting process underway.

I would like to give more information on serum creatinine because I don’t think it’s used enough. In fact, serum creatinine allows you by creatinine kinetics, to estimate muscle mass. In this paper from Eric Desmeules in 2004 you see that the ratio lean body mass over body weight is a strong determinant for morbimortality in these 500 cohorts of patients over 5 years.

We also wanted to assess if serum creatinine could help us to better define body composition. So we have a small study running in the Rhone-Alpes region not far from here with 1400 patients and we tried to apply this criterion.

What we observed in fact over 1 year mortality was the risk of mortality and you know that the red line is no difference. If you go to the left of this red line, you have survival, improvement in survival. So you see for example that BMI, if you have low BMI you are on the wrong side. If your BMI is increasing, you get an overall protection. But look at serum creatinine here and every time your serum creatinine is increased by groups, you get a strong survival advantage here. For example, here for creatinine over 860 µmol which is about 90 mg you get a 50% decrease in mortality.

So the problem is how does this go with BMI? You all know that BMI is protective. If you have a higher BMI, you get lower mortality. In fact, it’s probably not absolutely true because BMI does not completely measure lean body mass and we did this small analysis grouping BMI over or lower than 25 and serum creatinine over or lower than 800 µmol. So of course, you see that the patient with high BMI and high serum creatinine almost doesn’t die during the first year. By contrast if you take this one which is of course you guess low BMI under 25 and low serum creatinine under that 800 µmol you have this strong mortality. But look at the third group here which is high BMI greater than 25 but low serum creatinine less than 800 µmol. You see that this group clearly is not protective. So I think it is time now to get more information when we speak about BMI. Of course BMI is protective but not all BMIs may be protective it depends on lean body mass. BMI should be clarified using lean body mass index and serum creatinine, we all measure that and probably we don’t use it enough.

In fact you all know that we have an appetite defect and anorexia during dialysis. This is a nice study again from Doctor Coppell’s team 5 years ago and Doctor Kalantar et al. just asked a very simple question, do you have a very good appetite? Good, fair or poor appetite? So 4 groups of responses and then he derived subgroup analysis according to age, dialysis vintage, serum albumin, Crp, protein intake and mortality. You can see very clearly that according to the group of appetite filling the mortality was very different with 30% mortality if the patient declares having a poor appetite or 5% if the patient declares having a very good appetite. That was independent of dialysis and age. Of course, you can see that patients with good appetite had a large protein intake 1.1 nPNA and the poor appetite had a low protein intake.

More interestingly you can see that the very good appetite group was not inflamed whereas the poor appetite group of patients had more inflammation and quality of life was inversely related also. Then he derived a survival curve just based on this question and you see after one year here 12 months I think there is no comment here, very good and good appetite don’t die and very poor appetite here have a very strong 30% mortality. So again anorexia and appetite are of concern in these patients.

So protein intake should be assessed monthly and either by nPNA which is very convenient to measure. Almost all dialysis machines are giving you these numbers now every session or food reports which are more time consuming and need expert dieticians. The problem of energy intake of course, there is no machine to measure energy intake and the only way is to get these food reports every 3-6 months but this is more difficult and time consuming.

So the levels we picked were protein intake lower than 0.8 g/kg/day and a dietary energy intake less than 25 kcal/kg.

So where do we stand right now? This is again the French observatory results over almost 10.000 patients. This was June-July 2008 and it’s reassuring to see that nPNA is greater than expected in fact and if you take the minimal requirement which is 1.0 g of protein/kg/day using nPNA, you see that here maybe 70% of patients have enough protein intake. The threshold for protein energy wasting here fortunately for these patients doesn’t deal with many patients.

So in order to go forward we wanted to assess these criteria to see if we could define a score, a very simple score to be calculated every month for example, for each patient. We wanted something very small, accessible and including lean body mass. So the targets we picked, we picked in fact 4 different items: serum albumin, nPNA, BMI and serum creatinine to see that in each group of the nomenclature we picked one parameter. So if the patient is over this parameter with an albumin greater than 38 g/l he has 0. If he’s under he gets 1. It’s the same for nPNA, 0 if he has over 0.8 g/l BMI 0 if he’s greater than 23 and serum creatinine 0 if he has a greater serum creatinine than 665 µmols.

So 0 is the best, no malnutrition. 4 is severe protein energy wasting. We made a survival curve in our patients for 1 year and you can see easily that this score appears to be very discriminant if you have a 0 score, you don’t die after one year and if you have a score of 3-4, you have at least a 20-25% mortality. So again this is mixed. You can get this score, the blue score here whenever you have low protein intake or low albumin or low body mass.
So I think it’s something convenient, easy to measure and if your patient is changing groups, it probably can indicate a better survival or worse survival and I think this score should now be validated in larger prospective studies.

So the next thing is that is nutrition still a strong determinant for mortality? We wanted to assess among many different parameters for morbimortality one year survival with almost 5000 patients. We started in the Observatoire French study. So this started in 2007 and these are the results of June 2008

and these are recent data and I have to insist on the fact that we have to generate new data because treatments are changing everyday and we need fresh studies. Of course we expected to see things change in fact. Unfortunately, you see that in these modern treated patients with no restriction in drugs or time on dialysis or anything you could imagine still the strongest determinant for mortality is nutrition. These were adjusted with lots of comorbidities, classical comorbidities and you see that in fact age of course you can change, you always find this increased mortality but if you go to a phosphocalcic metabolism, for example, see that for high serum phosphate you have absolutely no overt mortality. So in these kinds of results even if we expect to wait 5 years we won’t see any impact of high serum phosphate in this group of patients. PTH doesn’t go out, you can’t see anything with PTH and low serum calcium but still we find BMI, low serum phosphate which is linked with under nutrition. You know that if you don’t eat, you have a low serum phosphate. Albumin of course and haemoglobin also. All these parameters confirm that in fact nutrition is still a very prominent state of danger in these patients.

So in summary, I just wanted to share with you a potential new terminology which might be protein energy wasting. Fortunately, in the literature since last year people are going on with our idea and now use this term in many publications. We wanted also to provide a pragmatic assessment tool for everyone to be used easily everyday in the dialysis unit. We just started this preliminary validation in patients with a potentially useful predictive score to be largely validated now. This data and recent data still confirm that the impact of nutritional disorder are the most powerful determinants of patient survival even in 2009.

Chairman: Thank you very much we have time for perhaps one question Bill? Doctor Bill Mitch?

Question: Thank you very much, very interesting. I wanted to ask two brief questions. Firstly do you correct the serum creatinine for age? The second is that I was surprised at your model that there was no indication of inflammation when, in fact serum albumin and prealbumin really are the opposite of inflammation and have you considered that?

Dr. Fouque: Well, interestingly we didn’t pick inflammation because in our model it didn’t go. It was rejected and therapy was less strong as a determinant of mortality and came out of the model. So we didn’t consider CRP that was a curiosity but we didn’t keep CRP in this analysis. For creatinine we did not correct by age and maybe I think it’s a good suggestion to think about that. I don’t know how to do that but we can think about that.

Question: One small comment thank you for your talk. You had 75% in the albumin group which was below the standard and in the nPCR as I like to call it protein nitrogen appearance was 75% in the group. I think I have understood why this is because the mainstream nephrology dialysis world ten years ago realised there is a rebound after dialysis and you have to take this into account and we got away from the same pool. Whereas the nutritional people were all discussing these esoteric things at that time and they never paid attention that this has to be revised too because the post-dialysis on which you base your protein appearance is artificially low. Therefore, you over estimate in the current way that you do nPCR. You alluded that now there is perhaps a way with a machine with the clearance to get to such a value. This may get better.

Dr. Fouque: Yes I think you’re absolutely right. We all know that inflammation is very important in albumin. So albumin is not always protein intake and nutrition intake and there is a very nice paper from René -- very recently showing the important interaction between inflammation and albumin.

Chairman: I think we’re going to have to go on I’m afraid. Thank you very much Denis.