LITERATURE UPDATE - ACUTE RENAL FAILURE, ICU NEPHROLOGY

ACUTE RENAL FAILURE: A LITERATURE REVIEW OF NON-DIALYTIC ASPECTS

Norbert Lameire, Ghent, Belgium

Prof Norbert Lameire Renal Division Department of Medicine University Hospital Ghent Ghent, Belgium

---

Slide 1

So we have, of course, communicated with each other so that there will be virtually we hope at least no overlap or very small overlaps because the literature that has been published on these topics is enormous not only devoted to the nephrological literature, you forced us also to read the intensive care literature, pneumology etc. because this is really a multiorgan pathology.

Slide 2

What I would like to discuss is something about the definition and the classification of acute kidney injury. By the way, over the next coming years we are not going to speak anymore about acute renal failure. There is a strong tendency to replace this term by acute kidney injury, which means that acute renal failure will be restricted to patients who really need renal replacement therapy. We have realised over the last years that even small declines in glomerular filtration rate and small increases in serum creatinine have a tremendous and dramatic impact on the outcome of these patients long before the patient actually needs renal replacement therapy and therefore, I think it’s correct that we replace acute renal failure by acute kidney injury. I will say a few words about new insights in the pathophysiology of ischemic acute kidney injury and there I will be very chauvinistic I will restrict myself to two excellent European studies. I will say a few words about the changing epidemiology of acute kidney injury, the prognosis and recovery and if anything new on prevention of acute kidney injury.

Slide 3

This morning in the lunch symposium I spoke about the RIFLE criteria for classification of acute renal failure or acute kidney injury and to my surprise several people came up afterwards and said what does that mean, RIFLE? So, people have never or many people have never heard about RIFLE. RIFLE actually is an acronym, it’s a new form of communication internationally when we speak about patients with acute kidney injury. So it stands for Risk Injury failure Loss and End Stage Renal Disease. So if you take the first letters it is RIFLE. It is based on the intensive care approach of a patient with acute kidney injury and what the intensivist and also the nephrologist is looking at the first time you can think of hundreds of other diagnostic tests but the most important one is first of all what is the GFR doing? We measured that very indirectly and very poorly. We have the changes in serum creatinine and the intensive care people ‘par excellence’ particularly in a post-surgery setting look at urine output. Therefore, by agreement there was at the ATKI group, which is composed of nephrologists and intensive care people agreed that we should call people at risk for acute kidney injury when there is a serum creatinine increase 1.5 times of the basal creatinine if, of course, you know the basal creatinine, the creatinine or the function before injury occurs. That injury we speak when it is doubled the serum creatinine. That we speak about failure when the serum creatinine increases 3 times above the basal value or the creatinine exceeds 3.55 micromoles/L or when there is an acute rise of more than 44 micromoles/L meaning that at this stage many of these patients already are needing renal replacement therapy. Loss is characterised, these are 2 outcome parameters, these are 3 diagnostic parameters. Persistent acute renal failure with complete loss of kidney function for longer than 4 weeks and by definition this is a prolonged form of acute renal failure really. This is injury, this is failure and end-stage renal disease when even an acute patient is maintained for longer than 3 months, we call them by definition and this is also accepted in the chronic renal disease area and end-stage renal disease patient. On the other hand, during output criteria less than 0.5 ml/kg/hour for 6 hours, injury for 12 hours and failure for 24 hours or anuria for 24 hours. If we all would accept these criteria and I’ll show you one study that actually has looked and there are more now over the last years that have used these criteria to determine the outcome of these patients, we at least would have an international common language. You know when we started this criteria there were 32 different definitions of acute renal failure in the literature.

Slide 4

How can we ever compare outcome, renal replacement therapy modalities, therapeutic intervention, even diagnostic criteria? For example, one study which was performed in the U.K. has looked and using this RIFLE criteria in the Sheffield unit what the outcome was in the ICU and what the outcome was of these patients after 6 months. You see indeed when you look at control, the risk, the injury, the failure, there is almost a linear increase in mortality which is, of course, not unexpected but the RIFLE criteria are able to predict that quite accurately.

Slide 5

What is more and what is not always realised and this was one of our fellows when he was with John Kullum, an intensivist in the States, he used a RIFLE class to see what the normal evolution was of critically ill patients when they were admitted, that was a substantial number of patients, when they were admitted in the ICU and see on admission 78% had absolutely no sign of acute kidney injury but 20% during their stay showed an evolution to risk. Some of them went into injury, some of them directly went into injury. 8% were already at risk defined as risk RIFLE at the onset, at admission of the ICU and further had an evolution to injury. 7% had already injury and 6.8% and this is the number, the figure that you find in the literature critically ill patients develop acute renal failure around 5-8%. If you stick to the definition of failure, this is correct but we should not forget this because you see many, many of these patients finally ended up in the failure stage. So we certainly underestimate the prevalence of acute kidney injury in critically ill patients.

Slide 6

What about pathophysiology? Well, this is a slide that is very old and there is a saying that old professors always come with old slides to meetings and occasionally with new wives. This was not the case for me, at least in the second part but you see that we have learned already 20 years ago that there is an interplay between hemodynamic factors, vasoconstriction mostly. We always think of the initiation of acute kidney injury as starting with acute renal vasoconstriction and them some tubular factors, obstruction by casts, back leak and interstitial inflammation and this factor is now very, very hot in the basic research on pathophysiology of acute renal failure.

Slide 7

Just to show that things are not that simple there was a very recent paper coming from the group of Bellomo in Australia where they used in conscious sheep, a model of sepsis induced by intraperitoneal injection of E. Coli and indeed this was a model that is very, very close to what we see in humans developing septic shock.

Slide 8

Indeed, there is a fall in the mean arterial blood pressure, these are the control sheep. There is a rapid rise in heart rate. There is a rapid rise in the animals with shock in cardiac output. There is an increase in peripheral conductance which is actually the --- of systemic vasodilatation. So this is a hemodynamic profile that very closely comes to what we see in patients. So we would expect that at least in the beginning of this phase these animals should show renal vasoconstriction. Well, to their surprise they saw exactly the opposite. Indeed there was an increase and they measured very consciously the renal blood flow in the sheep, an increase in renal blood flow, there was a fall in creatinine clearance. There was a fall in filtration fraction, GFR fell and renal blood flow went up, so the filtration fraction was falling and to their surprise there was and this is very well known also in humans, there was a low fractional excretion of sodium. We always thought that if the kidney is retaining sodium this is a sign of vasoconstriction well in this model and if this can be translated to human pathology this is no longer the case and the attribute and there was tubular damage because the fractional excretion of ureum was also much lower in the septic animals. So for the first time we have a little bit to re-think the hemodynamic aspects of at least septic induced acute kidney injury.

Slide 9

Let’s talk two minutes about the interstitial inflammation. Acute kidney injury is no longer only hemodynamic and no longer only tubular damage, no there is an enormously important component in the interstitium and I’m not going into detail but there’s an activation of polymorphonuclears which translocate across the capillaries, are accumulating in the interstitium by production of oxygen radicals, cytokines are directly toxic to the tubular epithelial cells. But not only polymorphonuclears but also T cells. Immunologically upregulated T cells are also appearing in the interstitium and by release of cytokines and chemokines contribute enormously to the pathology.

Slide 10

In this sense it has been very important that new molecular techniques have been introduced also in this field and there’s a lot of interest in the Toll-like receptors. Toll-like receptors are a sign of the innate immunology in contrast with the acquired immunology and when all these receptors are stimulated like by Heat-shock proteins or necrotic cells, they really contribute to the damage in the kidney.

Slide 11

What is now interesting a group from Amsterdam has published in the JCI, Jan Weening’s group that if you use mice, rodents which are knocked out for this Toll-like receptor, you see immediately a tremendous decrease in the granulocyte accumulation so in the interstitial inflammation in the interstitium and also these animals were protected at least less functionally damaged compared to the controls because the creatinine and the blood urea was much lower. So we will hear more about this in the future about the role and the interaction between endotoxins, Toll-like receptors and kidneys.

Slide 12

For those of you who attended the opening lecture this morning, beautiful lecture on the hypoxia inducible factors HIF by Peter Ratcliff, also he has mentioned very briefly the role of anoxia and of these inducible factors in the ischemia reperfusion model of acute tubular necrosis. Just to briefly summarise and he has done this much more in detail and much more elegantly than I ever can do, you know that in presence of oxygen this HIF factor is completely destroyed by a very complex mechanism.

Slide 13

But in anoxia this HIF 1 factor is translocated into the nucleus of the cell where it binds with some kind of HIF 2 factor and they are upregulating genes and very important genes. Look, the genes for vascular endothelial growth factor, for the glucose transporter and mostly well known to you all of erythropoietin. Now, it could be that by that way we can explain why more and more, at least in animal models erythropoietin is a protective factor against renal ischemia. Indeed as you see in acute ischemic renal injury this HIF may probably be even a protecting factor.

Slide 14

To show that Kai-Uwe Eckardt, who is an expert in this field, has shown that in a model which closely resembles human acute renal failure or acute kidney injury a combination of indomethacin blockade of prostaglandins, a combination of radio contrast agents and blockade of the NO system by L-NAME has seen a nice upregulation.

Slide 15

All these black dots are the upregulation of this hypoxia inducible factor in the animal.

Slide 16

But what is much more important in a paper that also Peter Ratcliff has mentioned this morning that appeared this month so you see Professor Rondeau and I did really our homework we went to the last papers on this subject, he has shown that preconditional activation of these factors ameliorates ischemic function. You see here when you upregulate the HIF by exposing the animals by carboxy monocytes where they produce hypoxia in the kidney, they upregulate HIF and you see when you then plant the renal artery and reperfuse them, the serum creatinine is much lower than in the control animals. We certainly and also with pharmacological means you can block this system, you can block the degradation of this HIF and in the near future we may expect new therapeutic agents working on this pathway trying to prevent the kidney injury. Then last but not least for the pathophysiology there is an enormous interest over the last years on the role of stem cells and the repair of the damaged tubules and some years ago we thought that all these stem cells, which were actually trying to repair the necrotic or the damaged tubular cells were all coming from the bone marrow, we now know since this paper was published in the Journal of Clinical Investigation and that was beautifully discussed in NDT and you know this is one of the leading journals in nephrology that also stem cells which are present in our kidneys mostly in the papilla of the kidney are contributing to this repair by producing growth factors and restoring the integrity of the tubular epithelium.

Slide 17

What about change in epidemiology? The world changes dramatically, also the causes of acute kidney injury. For example, our population gets older, there’s a greater frequency of critical illness and comorbidity. We are now treating these patients in ICUs. When I was young and more beautiful than today deep in the last century I never had to go to the intensive care unit to see an acute renal failure patient, he was on the renal ward. Now, no patient is no longer on the renal ward they’re all in the ICU, there are much more successful cardiovascular resuscitations, surgery is certainly much more aggressive. There are strange infections that we did not know of 20 years ago like for example, imported malaria even in Western Europe, leptospirosis, Hantavirus and then transplantation and I’m not speaking about acute renal failure after kidney transplantation no, surgeons are transplanting 2 and sometimes even 3 organs at the same time, so other non-renal organs and combined to organ transplants are a new risk population for acute kidney injury and of course, the oncology patient since he’s submitted to more aggressive chemotherapy also will be much more prone to acute kidney injury.

Slide 18

Despite this changing negative aspect in the population, high risk population, we see over the years whatever the literature says, don’t believe it. The latest literature shows you on huge databases that actually the prognosis is improving. Mortality both in the non-dialytic requiring and in the dialytic acute renal failure patients is declining slowly but it is declining. So actually, the combination of intensive care and nephrology is doing a great job.

Slide 19

What is worrying me and it should also worry you is the fact that based on a huge database coming from the U.S. and this has not been published it was presented at the latest ASN that when the patient develops acute kidney injury with pre-existing even mild chronic kidney disease, you see 28% after 3 years after discharge from their initial hospitalisation for acute kidney injury are in end-stage renal disease. So, we see now a new type of end-stage renal disease patient that we very rarely saw in the past. This is the one with mild chronic renal failure developing an acute kidney injury and 2-3 years later ending up as a chronic renal failure patient on our dialysis.

Slide 20

To illustrate this that this is very important this paper in KI quite recently looked at the 3-5 year outcomes of children, children who developed acute renal failure in Texas clinic hospital. Just to make a long story short 245 children were observed between these years. 174 survived. 60 immediately never recovered their renal function. 60 were immediately in end-stage renal disease and then they followed the others, the survivors over 3-5 years and they could detect 29 subjects where they could follow with accurate measurements of GFR, proteinuria, etc. renal function.

 

Slide 21

Look what they found. They found that 16 which were on dialysis already and transplanted are immediately afterwards. New cases of after 3-5 years of 13 patients with chronic kidney disease. So don’t forget, acute kidney injury is going to be a contributing factor to chronic kidney disease.

Slide 22

To end up what is new in the therapeutic intervention? You see there are a lot of things that can be done. There is a nice saying that whenever there are too many cures for a disease, it usually means that the disease is incurable and this is indeed true. When you have 25 drugs it proves that none of the drugs is actually working very well. The only one that has really established is the volume restoration and all the rest is very, very doubtful and just to show you there’s been a lot of work going on in acetylcysteine, at least as a protection in radio contrast nephropathy. This is very equivocal, there are pros and cons and it’s absolutely not solved. .

Slide 23

This group in NDT recently has looked at the protective effect of acetylcysteine in the prevention of ARF related to aortic aneurysm repair. If they saw anything, they saw exactly the opposite to what they expected. Actually the group that was receiving in a huge dose acetylcysteine actually did worse, so the serum creatinine went up while the controls improved. So acetylcysteine whatever you read about it for the moment is not shown to be protective.

Slide 24

The only drug that is now in the intensive care unit quite popular and that maybe shows some protective effect is fenoldopam. Forget about dopamine, this is now established. This chapter should be closed, you should no longer use dopamine as prevention or as therapeutic intervention. But fenoldopam, which is a dopamine analogue actually has shown and I’ll show you just one study, in a couple of studies that, for example, in sepsis the survival in the patients with septic shock and acute kidney injury that were treated with fenoldopam was much better in this randomised controlled study than placebo but it requires further studies.

Slide 25

Then one aspect that the nephrologist interested in acute renal failure should be interested in is what is going on outside the kidney in the intensive care unit. You are probably aware of this nice study that was published already some years ago in the New England where intensivists from Belgium this way and this is not chauvinism showed that when you controlled the mean blood glucose at a normal level that actually compared to patients where the blood glucose was allowed to float around and very critically ill patients always develop hyperglycaemia, this is a sign of stress that these where the blood glucose controlled by continuous measurement of the blood glucose an adaptation of a continuous insulin infusion that indeed the mortality was significantly lower and interestingly enough that also the prevalence of acute renal failure in these patients was much better.

Slide 26

This was done in a surgical ICU population. So the obvious question was what happens in the normal medical ICU? So they published a follow up study this year in the New England and to their great disappointment they could not confirm that indeed intensive treatment with insulin whereby the glucose was carefully monitored and normalised compared to conventional treatment had no outcome amelioration in the intensive care unit, in hospital survival or in the intensive care survival. But when they looked at the subanalysis of the group that needed intensive care unit therapy for more than 3 days in this high risk and critically ill patient, they again saw that controlling of the blood glucose with intensive treatment with insulin had an amelioration of the prognosis.

Slide 27

So in conclusion from this very rapid overview of the overwhelming literature data over the last years, the last year has emphasised that testing of the RIFLE classification as outcome measurement and as classification of the patient, the early detection of acute kidney injury by means of biomarkers and Professor Rondeau is going to speak about one of them, the role of interstitial inflammation in the pathogenesis of acute kidney injury, the changing epidemiology of acute kidney injury, the improved immediate survival of acute kidney injury patients and the impact of non-nephrological interventions like glucose control, like activated protein C in sepsis etc, on the incidence of acute kidney injury. But the tendency and the main take home message that I want to convey to you is that to find that acute kidney injury occurring in patients with mild chronic kidney disease is contributing to end-stage renal disease is in my opinion a worrying phenomenon and should stimulate research in this important area of nephrology. Thank you very much for your attention.