Slide 1

Thank you Mr Chairman. Good afternoon Ladies and Gentlemen. I fear that I will not give you much more clearances but nevertheless, we will discuss it. Before going to renal replacement therapy in ARF, let me make some additional general comments on ARF because I think it’s very important for the nephrology community and not only for the ICU community.

Slide 2

This is certainly the organ dysfunction which has the perception, the view on which we have fundamentally changed during the last years. 10 years ago ARF was a very simple organ dysfunction, it could easily be replaced by renal replacement therapy. So it has not attracted the interest it deserves. I think there are many here in this room who believe that the patient doesn’t die from but with ARF because renal function can easily be replaced. But we have learnt during the last 10 years that this is certainly not the case that ARF is a very serious complication in our patients which has a fundamental and probably the most fundamental impact of all organ dysfunctions on the prognosis of the patient.

Slide 3

A couple of years ago we made a large multicentre study in Austria and we investigated the impact of acute renal failure requiring renal replacement therapy on outcome and the gross mortality for instance was 63 versus 16%. This was a very large study in 17.000 patients and 839 patients with acute renal failure and from this large database it was possible to perform a case control study. So for each patient with ARF we identified patients with the same disease severity, the same age, the same sex and these were also corrected for treatment centre factor because a university hospital has probably other conditions than a smaller hospital but even then there was a highly significant difference in the prognosis of these patients.

Slide 4

So acute renal failure has an attributable, a specific effect on the prognosis. You could say probably if he has high acute renal failure, the severity of disease is higher but in these investigations we have corrected for the disease severity, for each patient with ARF there was a patient with the same severity of disease but then he had been more ill in other organ systems etc.

Slide 5

This is a difference and this is true for all age groups and this fact in the meantime has been confirmed by several other groups, for instance here De Cheyron from France. These are patients with liver cirrhosis again corrected for severity of disease and you see the huge difference, the huge impact ARF has on the cause of disease.

Slide 6

If you perform an animal experiment, you can show that already 24 hours after induction of an ARF there are massive consequences for the pulmonary system. Pulmonary oedema, microhaemorrhage etc This was a dog experiment from ---- in Baltimore and you can see the massive differences only by the fact that in this animal an ARF has been induced, a distant organ injury induced by ARF.

Slide 7

So we have to accept and I think this is a very important message that ARF is not a problem only of the kidneys but it’s a problem of the whole organism. affects all organs systems of the body and all biological functions. Thus, only some of those effects are here listed; cardiovascular system, cardiomiopathy, pericarditis, pulmonary system, oedema formation, alveolitis, haemorrhage, pneumonia, gastrointestinal etc. I think the most important consequences are immunologic and metabolic consequences and they are tightly interconnected because if you look at your patients what is a patient with ARF dieing of? Mostly it’s infection, it’s multiple organ dysfunction.

Slide 8

So ARF is much more than a kidney disease and this just to illustrate these effects a study from Brazil last year and they compared patients with ARF and non-renal dysfunction in mechanically ventilated critically ill patients and only ARF was defined by an increase of serum creatinine of more than 85%. So a minimal increase as you can see but you can see that even these minimal increases had a profound effect on the extubation and the length of ventilation etc.

Slide 9

So patients with AKI the new name for ARF had identical initial severity of disease as measured by APACHE scoring but had an increased occurrence of severe sepsis and septic shock, had a higher need of antibiotics, a higher use of vasoactive drugs and increased duration of mechanical ventilation. Increased duration of weaning, increased lengths of stay and massive impact on mortality. So ARF is certainly one of the most important, most influential complications a patient can achieve in the hospital.

Slide 10

So what are the clinical consequences of this recognition of the importance of AKI? The first must be we must prevent ARF in any stage where we can. I don’t want to go too much into detail. The timing is the next point. We certainly have to start extracorporeal renal replacement therapy earlier than we previously thought. The second point but this is not the subject of this talk, is then we have to adapt the doses of dialysis or CRRT to these effects and the dose is certainly higher than we previously thought.

Slide 11

Coming to timing, again we have to say that there are not very many data available. There are some animal experiments from already nearly 10 years ago where an ARF was induced by pneumonia and resulted in sepsis in dogs. The hemofiltration was immediately started after this induction of this septic process or it was retarded to 24-48 hours. You can see that if there was a late institution of renal replacement therapies in the septic ARF animals, there was a massive decrease in the blood pressure. So they developed septic shock and the animals died. But if the hemofiltration therapy was instituted earlier, immediately after the challenge with bacteria, then the hemodynamics of these animals were maintained.

Slide 12

In the clinical situations there are not so many studies available. The first study is a retrospective analysis of Gettings again nearly 10 years ago and what he noted is that if patients with post-traumatic ARF were started earlier and here are the initial BUN concentrations, if they were started earlier as compared to late, so this means BUN was 100 in the later group and this is about what we recommended 10 years ago and there was a 30% difference in survival. So the timing had a profound effect on the prognosis of the patient but this is a retrospective analysis.

Slide 13

But what was also interesting in this study is that this phenomenon of starting earlier or later is confined to the first days. Yes you see that here the days of treatment and here is the BUN concentration so what was achieved by this earlier starting is not something in the further run of the situation but that obviously the treatment within these 3-4 days in the early renal replacement had a massive impact on the further cause of disease and prognosis.
So the uremic intoxications, avoidance of uremic intoxication in the first days was a very relevant phenomenon in this study.

Slide 14

Something similar was also suggested by Ravi Mehta from San Diego a couple of years ago where he didn’t investigate the timing of initiation of renal replacement therapy but of the consulting by a nephrologist and he assumed that this would result in an earlier treatment.

He noted in these studies that the number of organs failing and the further cause of disease was much higher in those patients where the nephrologist had not been consulted earlier but later.
So we do not have many studies available but also in other studies the problem of timing had been addressed as a secondary endpoint.

Slide 15

Everybody knows this study by Claudio Ronco, the most influential study concerning the dose of therapy to treat ARF. As you know, he has investigated in a prospective randomised controlled study 3 doses of therapy, of CRRT; 20, 35 and 45 ml/kg/hour filtration volume in a continuous hemofiltration process and he noted that in the lowest group there was a higher mortality than in the other two groups.

Slide 16

But as a secondary element he also investigated the timing of the therapy and the BUN at the start of continuous hemofiltration was a significant factor for the survival of the patient. The earlier he started the treatment in these patients, the higher was the probability that the patient would survive.

Slide 17

From the same group unfortunately a study with only historic controls and these historic controls have not been treated adequately with modern standards. As you know, it’s 35 and not 20 and he compared these historic controls with an early treatment group where he started less than 12 hours after admission or below 24 hours after developing oliguria and in these patients he made a high volume hemofiltration followed by a low volume hemofiltration again in this early status there was some advantage concerning survival.

Slide 18

We have more of those not very well controlled data. One of these is of Elahi in cardiothoracic patients as opposed to cardiotomic patients and he investigated also that the impact of the timing on the evolution to organ failure and mortality and you can see that the early starters had a lower development of organ function and had a much better prognosis but again, it’s not optimally controlled this study.

Slide 19

Some of this data we have available is from Wu from China and you can see they investigated patients with acute liver failure and ARF. This acute liver failure actually was not very well defined in this group but nevertheless, they had an early starter group and a late starter group and you can again see that early starters had some advantages concerning the outcome.

Slide 20

Unfortunately we have only one really well controlled randomised study and this is from Amsterdam from Mrs Bouman and this is something disappointing. What I have said until now would clearly favour an early start of renal replacement therapy in ARF but she had 3 groups; early high volume hemofiltration, early low volume hemofiltration and late low volume hemofiltration and you can see that there was not much difference between all these groups. One has to say that these patients were not so sick as we usually see in the ICU because the survival here was more than 60%. The early low volume had a somewhat poor prognosis but there was no statistical difference but it means there was a survival rate of about 70% which is very high for the ARF in the ICU as we have heard from Doctor Burdman in these really critically ill the survival is certainly usually between 20-30%.

Slide 21

So coming to an end, so when should we start our renal replacement therapy in ARF? One must say the conventional retention parameters of BUN or creatinine certainly are poor indicators because usually we start too late when we stick to these indicators.
If we have other indicators of uremic intoxication such as pericarditis, acidosis, so that we must say then the patient has already developed an acute uremic state, an acute uremic intoxication and this certainly should be avoided.
So it’s my belief but it’s more belief that the volume state, the fluid retention , the response to diuretics, the cardiovascular system, the pulmonary system are much more better and much more important indicators of when we should start renal replacement therapy in a patient with ARF. Certainly these decisions are very dependent on the individual patient whether he’s or she’s a stable patient or presents additional organ dysfunctions etc.
The optimal situation is certainly to avoid the evolution of an acute uremic state and in general we have to start earlier than we previously assumed but again it must be an individualised decision from patient to patient. A patient who is not hypercatabolic and is hemodynamically stable you can certainly wait longer than in a patient who has a high severity of disease.

Slide 22

So there is some data clinically available but we don’t have very good controlled, prospectively controlled trials and we don’t have many hard data but I think for the clinical routine we have some basis to make quite strong decisions. Thank you for your attention.

Slide 23

Chairman: Thank you Doctor Druml for this very nice overview. In short summary I would say that the major problem is that we really don’t know exactly what we are doing that the measures we use for defining ARF status are not adequate, urea creatinine and so on and in that regard I think it’s of utmost importance that especially for our non-nephrology colleagues, for surgeons and anaesthesiologists and so forth that we make them understand that urine production is not a sign of kidney function. That’s probably the reason why in that study of Ravi Mehta the use of diuretics was related to a higher mortality because indeed if you give the patients diuretics, they start peeing again and they think that the kidney is safe whereas for the filtration of course it doesn’t matter. Can you give a statement on that?

Dr Druml: I can only agree because in transplantations they have 10 L of urine but have a very low renal function, so I think you have to look at the patient as a total to identify the necessity.

Chairman: Questions, remarks from the floor? I think there must be many people who have questions.

Question: Hello now I have a question you know a practical problem in the ICU is low blood pressure. Patients on vasopressors, on dopamine, dobutamine. Now the patient has a mean arterial pressure of 70 or 60 mmHg and the patient is in severe ARF, is there any cut-off you would suggest of systolic blood pressure where we should not initiate renal replacement therapy? And if we don’t what do we do? Do we do intermittent peritoneal dialysis? Or how do we manage the situation?

Dr Drumlo: You have to stabilise hemodynamics before you start renal replacement therapy. If you do intermittent as is mandatory, you should not dialyse a patient with such low blood pressures. You have to give vasopressors, you have to lower volume in septic shock or in patients with cardiogenic shock and if they have persistent low pressures we would advocate the performance of continuous renal replacement therapies and these continuous renal replacement therapies have a much lower impact on the hemodynamics or even have a stabilising effect on the hemodynamics.

Chairman: I would not completely agree on that but that’s the discussion of the next lecture, so we will have the discussion after the next lecture. I think if you have hemodynamically unstable patients especially if they’re acidotic then you should dialyse because correction of acidosis might make them more prone again to listen to their vasopressors agents.

Dr Druml: Yes but you have to stabilise with vasoconstrictors or etc to modulate the hemodynamics before you start your therapy.

Chairman: Thank you very much Doctor Druml