SCCM Resource Library-Acute Kidney Injury: Soup to Nuts

Acute Kidney Injury: Soup to Nuts

Video Transcription

Good morning, everyone, and welcome to this session on critical care nephrology, which is called Acute Kidney Injury, Soup to Nuts. My name is Akash Deep.  I work at King's College Hospital in London as a pediatric intensivist, and I chair the critical care nephrology section of the European Society of Pediatric and Neonatal Intensive  Care, and I have the honor of chairing this session. So we've got a total of four talks in this session.  Each talk will be about talk time, 25, 27 minutes, and at the end we'll have all the questions from the speakers.  Without much ado, I want to invite our first speaker, Dr. Ashita Talwani, who is a professor  of medicine and nephrology at the University of Alabama in Birmingham, and she'll be talking on what the ICU team needs to know about renal replacement therapy. Ashita.  Thank you for that kind introduction, and I'm happy to be here today. All right. This is my disclosure information. Good morning, everybody.  So when looking at acute renal replacement therapy in the ICU for AKI, these are the key issues. One, when should acute renal replacement therapy be initiated?  Two, if you're going to initiate the therapy, what are the critical elements for the delivery  of successful acute renal replacement therapy in terms of modality, dose, and anticoagulation? So let's look at some of the literature regarding these issues.  Let's start with timing. So if you look at the KDGO guidelines from an international expert, these were published  now almost 10 years ago, and basically they say to initiate renal replacement therapy emergently when life-threatening changes occur, as we all are aware of, but they also indicate  to consider the broader clinical context and look at not just a single lab value, but trends.  So since these guidelines have been published, which are still kind of vague, do we have any more data supporting timing? Well, let's look at that.  So if you look at the definitions in the literature, how is timing defined? So conventional is considered to be late initiation of renal replacement therapy, and again, these  are for the indications of life-threatening emergencies. In the literature, if you look at the trials that have been done, early is considered to  be preemptive, and you start at an earlier stage of AKI based on a staging system and hopefully to prevent complications of AKI.  So let's look at this theoretical benefits of early initiation. So these are the advantages. Uremic metabolic control, improved volume management, acid-based homeostasis, prevention  of other organ dysfunctions, such as cardiac dysfunction, congestion of the liver, et cetera. But the reason that we do not provide renal replacement therapy early without some of  the conventional indications is because we're all concerned about the theoretical drawbacks.  Complications of the procedure itself, infectious mechanical complications of catheter placement,  incompatibility with the membranes or dialysis devices, bleeding from anticoagulation or blood loss from circuit clotting, risk of hypotension or arrhythmias with further delay  of renal recovery. And depending on the type of therapy you use, loss of nutrients, drugs, increased cost.  And if we start renal replacement therapy on patients earlier than the conventional indications, what about the fact that they may spontaneously recover renal function and  may not need the therapy? So based on that, what do we know? Well, I can tell you that observational studies, data that I've not presented today, indicate  that there seems to be a signal that earlier is better than later. However, if you look at how earlier is defined, many of these studies use either a BUN level,  staging criteria, timing to admission to the ICU. But what I'm going to focus on is the three major randomized trials done over the last couple of years and what they showed.  So these were three major randomized trials in the ICU setting looking at patients who were critically ill, either on vasopressors or on mechanical ventilation.  And all of these three studies used criteria of K-DIGO or staging AKI to define what meant  earlier versus conventional. So of the three trials, the AKI trial, which was done in France, and the IDEAL trial, also  done in France, when defined by what they meant by earlier, it was by a K-DIGO stage  three definition or a creatinine three times baseline or a urine output less than 0.3 milliliters per kilogram per hour over a 24-hour period.  And so if you look at the ELANE trial, which was done in Germany, these patients in the early arm were started at a staging earlier than the rest of the two other trials.  The delayed arm for all these three trials was conventional indications that we know in terms of starting renal replacement therapy.  If you can look at this table that I've presented for you, you can see that the patient characteristics varied between the trials. There is difference in SOFA scores.  There is clearly more sepsis in the AKI trial and the IDEAL trial, where 100% had septic shock, as compared to the ELANE trial. The ELANE trial also had primarily surgical patients.  The modality for renal replacement therapy differed between all three trials. Both the AKI trial and the IDEAL trial use a combination of dialysis modalities for patients  in both arms, while the ELANE trial used specifically 100% CRT at the start. So what did these three trials show? There was no difference in outcome between AKI and IDEAL.  But in the ELANE trial, there was a statistic difference in patients in the early arm had improved outcomes.  So you may argue, why did one out of these three trials show improved outcomes? You may argue it's because of the maybe starting earlier than the other trials, maybe surgical  patients, the modality. But probably what's more correct is that all these trials were probably truly underpowered to really detect an outcome. So where do we go now?  Well, there's one more trial that should be presented soon called the START-AKI trial, which again is a ICU trial looking at timing of renal placement therapy.  But how this trial differs from other trials is that it's a much larger trial encompassing an international amount of institutions.  And essentially, about 3,000 patients have been enrolled in the studies being analyzed. In this trial, patients were eligible for enrollment when they reached K-DIGO stage  2 criteria, very similar to the ELANE trial. Those who met the criteria were randomized to the accelerated arm or the early arm within  12 hours, while those in the other arm were for standard indications, except if they emergently needed.  So stay tuned for those results, and perhaps we'll have more information to guide us in terms of timing.  But one thing I want to point out is that there's other factors we consider when we look at these patients on a daily basis that goes beyond just looking at a staging of AKI.  So we consider the severity of AKI. We follow the trajectories of the BUN and creatinine. We look at the level of acid base. We look at the severity of critical illness.  What other organs are affected? The degree of volume overload. Again, it's trajectory that we're looking at.  And we have to think about the potential risk of renal placement therapy. If patients have severe coagulopathy, is it safe to place the line?  Are there complications with line or inability to place the line? And finally, reality is there's also other issues that we have to consider.  Availability of machines, staff, what day of the week it is, what time the consult occurs at night, and also whether we think the therapy is going to be futile.  So because we have to think about all these things, the ADKI, basically another international group, had a consensus conference on what's the next step to think of timing for renal  placement therapy, specifically for patients in the critical care setting. And their consensus statement was that acute renal placement therapy should be considered  when the metabolic and fluid demands exceed the total kidney capacity. So not based on a sole function or AKI stage, which makes sense. Can the kidney meet the current demand?  And if you look at it that way, essentially, we're looking as per an individual patient where that spot is where benefit is more than harm.  And this degree and impact of fluid and metabolic derangements may vary from patients, and what is considered early for one patient may be too late.  So hopefully there'll be further information in terms of looking at timing per this way. But this is probably what we do realistically on a day-to-day basis, more than looking at  staging criteria. So that is where the evidence stands for timing. Now, if we're going to start renal placement therapy, what about these other aspects?  Is there an optimal modality, an optimal dose, or anticoagulation in terms of technical aspects of the therapy? So let's talk about modality.  So looking in the ICU setting, these are the major modalities that are available. This is our whole armamentarium.  Intermittent therapies consist of either conventional hemodialysis or prolonged intermittent renal placement therapy, also known as sustained low-efficiency dialysis or SLED.  And you have the continuous therapies, of which CRT is a big part. But CRT itself has different types of therapies, depending on how solutes are removed.  And the million-dollar question is, which modality is best in the ICU? OK, let's again look at the data.  So K-Digo, again, almost 10 years old now, basically stated to use them as a complement based on the studies we have so far.  However, they stated there may be areas where CRT or conditions where CRT is better, specifically  the hemodynamically unstable patient and in patients with acute brain injury who are at risk of increased intracranial pressure. How about the studies?  If you look at the studies comparing intermittent hemodialysis to CRT, guess what? From the randomized trials done, in terms of mortality, there has not been shown to  be a benefit between one study versus the other. So what about now PERT or SLED? And what is PERT or prolonged intermittent renal placement therapy?  Well, essentially, it's a hybrid between CRT and intermittent hemodialysis. It basically is longer hours and a slower therapy than dialysis, so you still have the  benefit of the hemodynamic effects by decreased volume removal and solute removal per unit time. But at the same time, it's not a continuous therapy.  So essentially, it allows patients to go for procedures and also to increase mobility. So it's a hybrid between the two. What are the studies looking at PERT?  Again, not surprisingly, if you look at meta-analysis of randomized trials comparing PERT to CRT in mortality, no difference. So people ask, what about renal recovery?  Does renal recovery, is it hastened by using CRT? And again, this is one meta-analysis looking at the first type of renal placement therapy patients were placed on.  And what their outcomes were in terms of renal recovery. And if you look at all these observational studies, there's a signal that essentially renal recovery was better with CRT.  Looking at the randomized trials, though, the signal was not there. But the overall meta-analysis suggested a preference for CRT.  A more recent analysis comparing CRT to both intermittent hemodialysis and to PERT showed no difference in terms of renal recovery. So where do we go now?  First of all, again, many of these studies were underpowered. But there are clear indications of why we use CRT and some evidence that supports it. What are these indications?  It is a slower therapy. You remove less volume and less solutes per unit time. It has been developed specifically for ICU patients. So it is better for hemodynamic instability.  Because of the continuous solute removal, there are patients that could benefit from it who have a continuous catabolic state.  For example, severe tuber lysis syndrome or severe rhabdomyolysis, where they continue to have problems with acid base, hyperkalemia, hyperphosphatemia, et cetera.  Continuous volume control, this is important in many of our ICU patients who are all volume overloaded and for ARDS patients.  And we talked about the issues of the increased intracranial pressure. One thing that many don't realize is that it can be used to treat severe dysnatremias in patients with AKI.  A standard hemodialysis machine, the lowest sodium is 130. And so if you have someone with severe hyponatremia, even if you slow down the therapy, you can overcorrect them rapidly.  So this would be another indication to use this type of therapy. What are the disadvantages? Of course, it requires an ICU level of care and is probably more complicated than the  others. It is not good for slower removal of toxins. You have issues of patient immobility because they're tied to a machine 24 hours a day.  It often requires anticoagulations, and there's risks with that. Also, it increases the labor and complexity.  Hypothermia, nutritional losses, drug clearance, and of course, increased costs. I must say, though, even with PERT, these issues exist because we do not know how to  dose antibiotics effectively also with that therapy. So let's look at some of the data supporting CRT.  This graph on the left is from a randomized trial looking at intermittent hemodialysis compared to CRT. It was done at Cleveland Clinic looking at volume removal.  And what it showed is within a 24-hour period, patients were more successfully volume negative within that period compared to patients with intermittent hemodialysis.  It was statistically significant, and the reason being is that in the patients of the intermittent hemodialysis group, they had decreased mean arterial pressure and increased  vasopressors. Similarly, on the right side, this is based on a prospective observational studies of patients with AKI done in five institutions in the United States called PICARD.  What they showed is looking at the patients specifically who required dialysis, that volume removal was much more successful with CRT as compared to intermittent hemodialysis.  As far as the data to support the intracranial pressures with CRT, this is a study looking at patients with cirrhosis who had intracranial pressure monitors.  And what this showed is that the intracranial pressure increased substantially with intermittent hemodialysis and not from CRT.  And the reason for the increase was a rapid fall in serum osmolality from intermittent hemodialysis. All right.  So now that we talked about some of the benefits of CRT, what type of CRT do you use? So the different modalities of CRT either use convection to remove solutes or diffusion.  What's the difference? Diffusion is like standard hemodialysis. You have a dialysate that runs countercurrent to the blood and toxins or solutes to remove  down a concentration gradient. Only small molecules are effectively removed, molecules less than 1,000 Daltons. All are electrolytes.  Diffusion on the other hand, which is depicted here, is essentially what happens is a hydrostatic  pressure is placed across the filter and large amounts of plasma water are pushed or dragged through the filter, dragging with it solutes.  And then the rate limiting step for what are the size of the solutes that can be removed is the size of the CRT membrane, which technically most CRT members are 30,000 Daltons.  So the benefit for potentially convective clearance or the thought processes is that perhaps by removing cytokines or middle molecular weight molecules, it can portend a better  outcome. So I'm not going to go through this in detail, but these are the four CRT modalities. SCUF, slow continuous ultrafiltration, is purely a modality of volume removal.  There's not enough ultrafiltration to really cause a lot of solute removal. And it was used in the past of heart failure patients.  The other three modalities depend on how they remove solutes. Continuous veno-venous hemofiltration uses convection.  So high rates of plasma water are pushed or dragged through the filter. And as a result, the volume and the electrolytes have been replaced back to the patient because  there's such high rates. CVVHD is continuous veno-veno hemodialysis. And basically, you have a dialysate that runs countercurrent to the blood.  It's a lot slower than conventional hemodialysis, but solutes are removed by diffusion. And CVVHDF is a combination of both.  So you have a dialysate, diffusive clearance, and you have convective clearance that requires a replacement fluid. Now, are convective therapies better than diffusion?  Is one modality of CRT better than the others? Well, again, let's look at the data. If you look at this meta-analysis of hemofiltration or convective therapy compared to diffusive  therapy, 19 randomized contrials were looked at, 16 used CRT, no difference in outcome. Those patients who were on convective therapy did have losses of middle molecular weight  molecules. They also had increased clotting of the filter. So to this date, there's no role for convective therapy, and I can't say one modality is better than the other.  So how do we look at renal support in the ICU? This is essentially the summary. Most practitioners will use CRT for cardiovascular instability, cardiogenic shock, septic shock,  acute liver failure, for volume control, or cerebral edema issues. However, intermittent hemodialysis or PERT would be better for patients with severe hyperkalemia,  drug poisonings, anticoagulation or access issues with CRT to increase patient mobility. But the big issue is to understand this is a spectrum, and you're going to tailor your  therapy according to the changing clinical parameters of the patient. And often, you may use these different therapies in the same patient, depending on the course  in their ICU setting. All right, so I'm going to end with, because of time, dose. OK, just keep me on my toes.  All right, so now that we talked about the difference in modalities, is there a certain amount of clearance we aim for, or there's evidence for, in terms of targeting for these  renal replacement therapy modalities? If you look at the CADEGO guidelines, they do mention that you need to prescribe a dose  for a patient and deliver what you prescribe by measuring what you are clearing. And you need to continuously reevaluate if you're meeting your targets. So what is dose?  So what's conventionally been looked at for renal replacement therapy by nephrologists is urea clearance. That's how we target it.  But you can also look at clearance of larger molecules, like we talked about, and time on the machine, and volume management.  All these things can be considered part of your whole dose of therapy. When you look at intermittent hemodialysis, conventionally, as nephrologists, we use a  concept called KTRV, which is basically clearance of urea, or urea reduction ratio, and the frequency of hemodialysis sessions. For CRT, it's very different.  How to calculate urea clearance is very different, because the mechanism is very different in terms of how solutes are removed.  In summary, in CRT, whatever comes out of the filter, you're total effluent to ultrafiltrate, your spent dialysis, your replacement fluid amount, plasma water that's removed.  For most solutes, the concentration of that approximates that of plasma water. So with CRT, different from intermittent hemodialysis, because the dialysate flow, or this effluent  flow, is so much slower than the blood flow, you have time for complete equilibration of small solutes. So what I love about CRT, the calculations are completely simple.  The concentration of small solutes in the effluent is that close to the plasma water. So guess what? Solute clearance, or urea clearance, strictly approximates the effluent flow.  It can't be easier than that. So what do I mean by the effluent flow? It's basically all the removal of the fluids and solutions that come out of the filter. It's basically urine.  It's what you see. We call it the effluent bag. So for CVVH, it's your total ultrafiltration, which means whatever you're pulling through  the filter and replacing back, in addition to extra fluid removal. That's UF. For CVVHD, it's the spent or dirty dialysate at the end of the filter, plus whatever ultrafiltration  you're using for patient fluid removal.  And for CVVHDF, it's a combination of both. And what's really nice is the CRT devices tell you what the effluent rate is per hour. So it's very easy to calculate.  The way we express dose, however, for CRT is in milliliters per kilogram per hour. And this is also displayed on many of the CRT devices.  So given that, what is a target dose we strive for? There have been two Sentinel studies, one done in the United States looking at dose, and the other one done in Australia.  In the ATN trial, which was done here in the United States, basically, patients were randomized  to a less intensive strategy, which was conventionally what was done at that time, to intensive management.  If they were stable, they were randomized to intermittent hemodialysis at three times a week versus six times a week, aiming for that target KTRV of 1.2 to 1.4 is what we  aim for in the outpatient setting. If they're hemodynamically unstable, they're either randomized to CRT at a dose of 20 versus 35 or SLED, depending on the investigator's choice.  The renal trial, different from the ATN trial, was purely a CVHDF CRT study, looking at a higher dose of CRT, 25 was a lower dose, versus 40.  Both of these trials had over 1,000 patients. What were the results? In both these trials, there was no difference in providing a higher intensity of dose.  However, even in the conventional dose arm, the results were better than what was historically done in the literature or basically published in the literature.  So based on this, the standard of care currently is that for CRT, you should aim for an affluent  rate of 20 to 25 milliliters per kilogram per hour to ensure that's sufficient as long as you measure it and make sure you're delivering that.  And many guidelines will state to go a little bit higher to account for time off. For hemodialysis, there's no need to provide treatments more than three times a week for  a target KTRV of greater than 1.2, but additional treatment may be needed for fluid management. So that's currently what we stand for or where we go.  All right, finally, septic patients. We still want to believe in the cytokine removal with high volume hemofiltration using convective therapy.  Long story short is if you look at the randomized trials, high volume is considered greater than 50 milliliters per kilogram per hour, mainly convective therapy.  Compared to standard, no study has been shown that there's a benefit in terms of outcomes by randomized trials.  Even more recently, there was a high volume hemofiltration study looking at patients post-cardiac  surgery who had severe shock and high dose catecholamines randomized to either high volume hemofiltration at a higher rate or standard care, which is starting CRT at conventional  indications, and there is no difference. Similarly, another septic AKI study looking at high dose CVVHD versus conventional showed  no difference, although in the high dose arm cytokines were removed. So at this point, there is no data to support convective or high dose convective therapy of presepsis.  And currently, what's thought to be is that there's a best practice window for what we consider to be the optimal dose of therapy, understanding, though, that sometimes we may  need to increase the dose to manage acid-based or electrolyte issues temporarily. So because of time, I'm not going to go into anticoagulation except to say that for CRT,  the recommended anticoagulant is now citrate. There are multiple studies to support that.  And in order for you to understand what citrate does, you have to understand the complications to be able to manage it.  I'm happy to deliver you these slides, but I just want to let you know that that is what's up and coming.  So in summary, there's no benefit to mortality compared to one dialysis modality to others,  but CRT is preferred in the hemodynamically unstable patient and those at risk of elevated cerebral edema.  There's no benefit to high-dose therapy, but there is a standard of what's considered to be minimal dose, and it's important to measure that and achieve it.  And finally, if you're going to use CRT, citrate's probably the way to go, which I can do a whole another talk on.  So in summary, I borrowed some of these slides from Dr. Conner, who's sitting in the front here.  What are the steps for successful RRT, close collaboration between critical care medicine and nephrology? We're able to achieve all these things we talked about.  So we can discuss goals of therapy every day, volume removal, clearance goals, how to transition between therapy, what are we trying to achieve?  And if you're going to use CRT, how to keep it running. And finally, we can address with them on a daily basis medication, nutrition support, and avoid complications.  So it's a multi-team effort. Thank you very much.  Thank you. Thank you, Ashutosh. We'll have questions at the end. Thank you. Can I now invite our second speaker, Dr. Sandra Gill, who is a professor of pharmacy at the  University of Michigan.  University of Pittsburgh School of Pharmacy and faculty in the Center for Critical Care Nephrology. And she'll be talking about biomarkers in acute kidney injury. Hello.  For those of you who may be expecting Azra Biorak, who is intended to speak today, I don't look like Azra, and I certainly don't have Azra's accent, for those of you who know her.  But I can offer you an out and about, my Canadian accent. So Azra's talk was going to be about prevention and protection of the kidney in a surgical patient.  And so I'm going to try to pay homage to her theme, but talk about biomarkers for acute  kidney injury and the applications for prevention or protection of the kidney with consideration for biomarkers. I have no conflict of interest related to today's talk.  My objectives are really to discuss structural biomarkers as an early warning signal for the prevention of acute kidney injury.  And then give some examples of how structural biomarkers can be used in clinical practice.  So let's think about what I'm going to call the older biomarkers, the traditional biomarkers, the ones we're all familiar with, and start thinking about urine output.  And just remember the advantages and disadvantages, right? The advantages to urine output is, first of all, we're all familiar with it, been using it for multiple years.  It's easy to measure. It's sensitive. And it does suggest as an early functional marker if there's decline.  The disadvantages being that it can be difficult to collect, that there is potential human  error associated with collection, and of course can be easily influenced by diuretic therapy and also fluid status of the patient.  The biomarker we rely most heavily on is serum creatinine. The advantages, again, we're very familiar with it. It's easy to measure.  It's measured frequently so we can monitor trends and see changes in a patient's kidney function over time. The disadvantages is really that it is not a real-time biomarker, right?  The key is is that after insult, a rise in serum creatinine can lag behind somewhere between 24 to 48 hours.  Also, it requires a substantial amount of kidney function loss before there's an increase  in the serum creatinine, and also can be affected by things such as age, gender, and of course muscle mass.  So when we think about the ideal characteristics of these newer biomarkers that have arisen,  I'm going to give some examples from the drug-associated acute kidney injury literature to support some of these advantages.  But the major advantage is that they can be early warning markers, right? I just suggested that serum creatinine has this lag, and so if we can identify a patient  who is at risk for developing acute kidney injury before the serum creatinine, that can be very advantageous.  So an example would be NGAL as the biomarker, which is neutrophil gelatinase-associated lipocalin.  And in this particular example, the NGAL concentrations rose much more quickly for amphotericid-induced acute kidney injury.  It actually rose 1.7 to 3.2 days sooner than serum creatinine. Consistent with our older, more traditional biomarkers, the newer biomarkers are readily  available and they can be measured in the urine as well. So we have things such as the NGAL, NAG, and then TINP2, IGFBP7, which is a tissue inhibitor  of metalloproteinase-2, and in combination with insulin-like refactor-binding protein-7. So this is one study that is suggestive of some benefit, which is that potentially, as  we learn more about these biomarkers, maybe a rise, depending on what the concentration of the rise in these particular biomarkers are, it could suggest maybe a type of damage,  right? What the etiology of the damage is. There is one study that's available that suggests that NSAID-induced acute kidney injury,  the NGAL levels are higher than if the etiology was hypovolemia or type 1 hepatorenal syndrome.  However, NSAID-induced AKI, the NGAL concentrations were actually lower than if the etiology was ATN.  These newer biomarkers can be differentiated based on their levels and suggestive of what the risk might be.  KIM1, which is kidney injury molecule 1, higher concentrations suggest more damage. There's been evaluations for cut points for NGAL, a level of 150 to 300 is moderate risk.  Greater than 300 is high risk for progression to acute kidney injury. TINP2-IGFBP7, greater than .3 to 2 is actually suggestive of moderate risk for progression  to acute kidney injury. And greater than 2 is high risk for progression. So I'm talking about structural biomarkers. Another way to think about this is damaged biomarkers.  So when there's some type of injury, that's when you're going to see this increase in the concentration of these biomarkers.  It can be for various reasons that could be because the synthesis of the biomarker or protein is upregulated during AKI.  It could be because the molecule is released when the cell is distressed or injured. Or it could just be that this protein is a byproduct and the filtration, reabsorption  and secretion are altered by kidney damage. So let's think about that in context of some of the more common biomarkers.  KIM1 is a transmembrane protein that's upregulated when there's damage to the kidney. For NGAL, again, it's upregulated during renal ischemia.  And then for TINP2, IGF-BP7, these are released during cell cycle arrest when the kidney is stressed.  So I'm going to talk a little bit about TINP2, IGF-BP7, and I'm going to explain a little bit of why. There was a study that was done in 2013, which was a SAFIRE study.  And it was a discovery and validation study. In this study, it was around 35 institutions across North America and Europe that collaborated  and looked at 340 different proteins to figure out which protein was most commonly predicted acute kidney injury, stage 2, stage 3, 12 hours after insult.  And so after the evaluation of 340 proteins, TINP2 and IGF-BP7 rose to the top. And specifically, when you use them in combination, the AUC to ROC, the ROC curve was 0.8.  After identifying that these particular biomarkers maybe performed better than some of the other  This was then proceeded to do a validation of these, including serum creatinine and clinical  variables for evaluation of how well the model would work at predicting AKI stage 2, stage 3. And the AUC was 0.81. And then substituting in TINP2, IGF-BP7, it rose to 0.87.  The next study was the OPAL study. And the OPAL study was really an evaluation of cut points for the TINP2, IGF-BP7.  And if the value was less than 0.3, this had a high negative predictive value of 97%, a very low risk for progression to AKI.  And then for moderate risk to high risk, you can see here, just as I mentioned previously, the cut points were 0.3 to 2 for moderate risk, and then greater than 2 for high risk.  The next study that was required by the FDA before approving TINP2, IGF-BP7 was the TOPAS study.  They actually required the investigators to do a clinical adjudication of acute kidney  injury and then to see how the biomarker compared to the clinical adjudication and TINP2, IGF-BP7 performed better than the serum creatinine.  So this particular TINP2, IGF-BP7 is FDA approved in the United States for use of assessment of risk in hospitalized critically ill patients.  I'm also going to talk about it because it's what we use at our institution. So I'll show you a couple examples of how we're using it.  Just to reiterate what a cell cycle arrest marker is, when a patient receives some type of insult, in this particular case, I'll suggest that it's nephrotoxin, a chemotherapeutic  agent.  There's insult to the kidney and if we remember back to our cell cycle or reflect back on  mitosis and the stages when a cell is replicating between G1 and the S phase, this is at the point in time where the TINP2, IGF-BP7 is released.  Because the cell has received some insult, it stops, pauses or goes under cell cycle arrest at that point in time, says, okay, should I replicate, should I repair myself  or should I undergo programmed cell death, it's at that point in time where these biomarkers are released into the urine.  So again, paying a little homage to Dr. Burek's original talk, protecting the kidneys during  surgery, I'm just going to go through a couple of studies that use biomarkers for this purpose. And the first one is the BIGPACK study.  The BIGPACK study is a randomized controlled trial that was occurred at a single center. It included 121 patients who underwent elective non-cardiac surgery.  And after surgery, the patients were evaluated for the risk factors for progression to acute kidney injury.  If a patient had at least one risk factor, then a TINP2, IGF-BP7 was measured. If that value is greater than 0.3, remember I said that the cut points, 0.3, suggest at  least moderate risk for progression to acute kidney injury, then those patients were randomized into the control group or the intervention group.  In the intervention group, the patients were, those individuals were actually made sure that they followed the CADEGO bundle or the suggestions or recommendations for caring  for a patient that's developing acute kidney injury. And that included things such as avoiding nephrotoxins, optimizing fluid status, maintaining  perfusion pressure, or in this particular study, it was required that these individuals had a nephrology consult. The primary endpoint for this study was any AKI at 7 days.  And as you can see here, there was a decrease in the incidence of AKI, but that was not statistically significant. Although the stage 2, stage 3 reduction in HKI was from 20% to 7%.  Looking at the drugs in a little bit more detail, in this particular case, the nephrologist made recommendations for how to manage the nephrotoxins.  And in 35% of the patients, they made a recommendation for drug management. And about 76% of those recommendations were accepted.  The next study was also done in, this particular study was done in cardiac surgery patients, also a randomized controlled trial at a single center.  And in this particular case, post-surgery patients had a TINP2-IGFPP7 measured. And in these patients, if their value was greater than 0.3, suggesting moderate risk  for progression, then they were either randomized to the control group or the intervention group. Again, the intervention group received the Cadego bundle.  In this particular study, they specifically mentioned avoiding nephrotoxins, withholding ACE inhibitors or ARBs, monitoring serum creatinine urine output,  assuring that the glucose was less than 150, limiting IV contrast, or optimizing volume status. The endpoint for this particular study was NE-AKI at 72 hours.  And there was a significant reduction in NE-AKI and also stage 2, stage 3 AKI for this particular study. Looking at the drug management in a little bit more detail,  first of all, just want to indicate the ones here in red are the interventions that were significantly, or that were different between the control group and the intervention group.  So looking at these drugs in a little more detail, there was significantly less ACE ARBs use in the intervention group. And although the vancomycin diuretics in contrast  were not significantly different between the two groups. So I keep highlighting the drug portion of this. And there's a reason. Yes, I'm a pharmacist. And so I like drugs.  There's that. But in addition, the fact of the matter is that drugs are one of our modifiable risk factors, or modifiable exposures. If we could optimize the use of medication,  there is a potential that we can decrease AKI severity or potentially decrease the occurrence of AKI, depending on the stage at where the patient is at.  So I'm going to go into that in a little bit more detail, which is thinking about the use of these biomarkers for drug management.  So this particular figure has kidney function on the y-axis. And you can see here, this is the time of insult. And so this is before, where the patient's kidney function is  declining, but they're not at AKI yet. This is an ideal time where you want to think about doing that biomarker assessment. Now, there's other biomarkers after a patient  has acute kidney injury that may be beneficial in determining prognosis and recovery. That's not our emphasis for today, because we're thinking about protecting the kidney  and prevention. So this is really what I'm talking about right here. And so the drug management that can occur when a patient may be at risk for progressing  to acute kidney injury is you can individualize the care plan for these patients. You can say, OK, this patient's at risk for progressing. Let me make sure that the drugs are  renally adjusted appropriately. Let me make sure that I don't give them a nephrotoxin unless I really have to. Or maybe you want to monitor that vancomycin level  more closely in this patient who's apt to be progressing. So this is just a figure that indicates where some of these biomarkers are located in the nephron,  and also where damage from some drugs occur. We're not at this stage yet, but I like to think into the future and say, OK, well, let's see. We know that a rise in NAG would be  associated with some damage in the distal tubule. And we know that amphotericin B causes damage in the distal tubule. So if you measured different biomarkers,  maybe it can eventually get us to the point where we can say, OK, you know what? This injury is due to the drug and not due to something else.  Again, a futuristic approach, but hopefully at one point in time we go in this direction. So I did want to present to you some data. This is the kinetics of urinary biomarkers,  both CHEM1 and NGAL, after vancomycin exposure. So this was an evaluation that was done in 85 patients who had been exposed to vancomycin. And 11 of those patients actually  developed acute kidney injury. And so this was an evaluation of patients who developed AKI versus those who didn't. And as you can see here, the AUCROC curves  for serum creatinine compared to CHEM1 and urinary NGAL, you can see on day zero and day one the serum creatinine is not rising as quickly.  And so the AUCROC in predicting the actual patients with AKI is lower than CHEM1 and also NGAL. Again, going back to the point, these  could be early biomarkers compared to serum creatinine, helping us to understand those patients who might be progressing, allowing us to maybe manage them slightly differently.  In this particular patient population, when urinary CHEM1 and NGAL were evaluated, the AUCROC actually increased above the individual ones.  Here's another evaluation that was published in Critical Care Medicine by Osterman and colleagues. And this is an evaluation looking at TINP2-IGFBP7  after the exposure to vancomycin. They evaluated some other insults as well in the study, including surgery. They evaluated NSAID use. They evaluated exposure to contrast.  I particularly highlighted the vancomycin because of the previous study that I had just shown you. So you can see the comparison between these different biomarkers.  In this particular case, you can see here this is the day of exposure to vancomycin. Here's the median TINP2-IGFBP7 concentrations.  And you can see on day one, after exposure to vancomycin, for those patients who developed stage 2, stage 3 AKI, the concentrations of TINP2-IGFBP7  are substantially higher than those patients who did not go on to develop stage 2, stage 3 AKI. So this could be suggested if you give a dose of vancomycin,  you check the TINP2 level. If it's high, then maybe that's a person you need to really be cautious in with using the vancomycin. If you don't dose it on a scheduled basis,  you could actually maybe dose it based on trough levels or something like that. It's a person you want to be more cautious in. I'm going to briefly switch from biomarkers  and steal Dr. Goldstein's thunder a little bit since this is his work. But I'm going to just mention it briefly, some of the work that he's doing at his institution.  He's developed an alert that indicates when a patient may be having some nephrotoxic burden or they have a long duration of exposure to a nephrotoxin.  So when a patient is receiving three or more nephrotoxins or they've been on aminoglycoside for greater than or equal to three days or vancomycin greater than or equal  to three days, the pharmacist receives an alert. The pharmacist receives an alert, evaluates the patient, goes and talks to the team.  And their intervention is monitoring serum creatinine. This is in pediatric non-ICU patients. Dr. Goldstein's evaluated the outcomes associated  with this program in doing this intervention and highlighting hypervigilance and monitoring of these patients who are at risk for developing acute kidney injury.  And they've had very successful outcomes. At one year, they've decreased AKI intensity and reduction of days of AKI. They evaluated also three years later  and again quite successful. I'll let him talk about his program in detail. But I'm highlighting this because it is very consistent with what we're doing at our institution.  So I wanted to make sure that there's an understanding between this and what I'm going to present next. So Dr. Goldstein's program is called NINJA.  I hope my NINJA there does you justice, Dr. Goldstein. It's called NINJA because it is the nephrotoxic injury negated by just-in-time action. And so that's why it's called NINJA.  So at our institution, related to the biomarkers, you know, we had seen the success in which they were having at Cincinnati Children's Hospital and with the NINJA program.  And there was some question, okay, well that particular work was done in pediatric non-ICU patients. Would those data uphold in adults? Would it uphold in an ICU patient?  We monitor serum creatinine all the time, right? If the intervention, at least part of the intervention is monitoring serum creatinine, or like more frequently,  would those results still be upheld in this population? And the answer is I'm not sure. We've never tested it. And maybe just because you monitor serum creatinine  doesn't actually mean you're looking at the serum creatinine and doing anything with it, where Dr. Goldstein's program does. So maybe you could get the same results.  But at our institution, we said, okay, well, we check monitor serum creatinine regularly. What happens if we look at TINP2 IGFBP7 instead?  So pharmacists in the MICU receive an alert when a patient has serum or nephrotoxins. And they evaluate to see if the patient already has AKI.  Because if the patient already has AKI, remember, we're trying to predict progression to acute kidney injury. If they already have AKI, there's not a need to go ahead  and test that TINP2 IGFBP7. So if they don't have the acute kidney injury already, then the pharmacist order the biomarker test. And they inform the physician that they did.  So why do we go with a pharmacist-driven program? The reason is because clearly time is of the issue here. Time is of the essence here. We need to do something quickly.  By the time the pharmacist would contact the physician, then get a verbal order to put that in, it would take time. And so we have approval at our institution  so that the pharmacist can actually place that order. Of course, with continuous communication with the team and especially the attending.  When the results come back, the pharmacist works with the team to interpret the results and makes drug recommendations accordingly. This is a QI project and it's under evaluation  so we don't have outcomes for this project. But I did want to explain to you what we are doing.  One of the feedback I get when this program, even when introducing it to the medical team, was again, we monitor serum creatinine already. Getting an additional biomarker  is not going to provide me any better information for managing this patient. I don't even know that I would do anything with that.  Continuously, I hear that, not just from there, from other places who are trying to figure out how to use this. So one of the things I did was looked at the first  105 patients that received or that had a biomarker test, the TINP2 test, TINP2 IGF-BP7 test at our institution. So the first 105 tests, this was in 100 patients.  Then divided those groups into those patients who had a value of greater than .3 or less than .3, suggesting the moderate to severe risk  for those individuals who had a value of greater than .3. And I wanted to see, similar to the PRVAKI study and similar to the BGPAK study,  who said, look, we're going to follow the Cadego bundle to see if we can improve outcomes. Well, I wanted to see when the attending or the physician was provided this information,  did they follow any of the Cadego bundles? Did they apply anything? And so that was the evaluation completed here. These are the components of the Cadego bundle looked at in detail.  The first part is explicit notes from the physician where the physician actually said, look, it got, it did an AKI risk assessment, measured the TINP2 IGF-BP7,  and here's how I responded to that information. Didn't ask them to do it. Remember, this is them doing it on their own. It's just a retrospective evaluation  to see how they responded. And in these particular notes, they said things like, I'm avoiding the initiation of an efertoxin in this patient, or I'm avoiding contrast,  or watching volume status. They were very explicit about what they did. For some people who didn't write or document exactly a response to that particular information,  we did this implicit evaluation where we said, okay, they got this information. In the next 24 hours, did they adjust the meds? Did they discontinue any drugs,  or was there additional therapeutic monitoring that was done? And so combining those explicit and implicit evaluations,  in summary, you can see that patients who had moderate to severe risk, 76% of patients had some form of the Kadego bundle applied.  And those patients who were at lower risk, they did not. So this information did guide therapy in some capacity. Again, I don't have an outcome evaluation of this.  It was really just to see the responsiveness to this information.  So in summary, using biomarkers for prevention or protecting the kidney, these newer damage stress biomarkers may be an early warning signal for acute kidney injury.  I suggest that because of the information that's been presented in the Pravaki and the Big Pack study. Combining the biomarkers with the application  of the Kadego bundle, there was improvements in outcomes with regards to acute kidney injury. And then not sure where it's going yet for using early warning biomarkers  for medication management, but maybe it's useful. In our institution, you can see the response for 100 different patients, so it could be useful.  I'm not sure how it's gonna relate to outcomes yet, but maybe there's potential in this area. Thank you. Thank you. Thank you, Sandra, we'll have questions at the end.  So next is me.  So I'm gonna take you to an entirely different world of critical care, the world of liver ITU. I come from King, so liver is the most important organ, without a doubt.  So I'm gonna take you through kidney injury in liver disease. So before we do that, no conflicts of interest, and I'm gonna take you through some myths we have,  the myths of liver, the myths of kidney. The first myth which we all think is chronic liver disease equals cirrhosis  and kidney involvement takes place only in patients who've got cirrhosis. Then we say every kidney injury in liver disease is hepatorenal.  AKI in liver disease mimics acute kidney injury in non-liver failure patients, and therefore the criteria for non-liver failure patients  can be used in patients with liver disease as well. We think HRS, which is hepatorenal syndrome, is purely functional with no structural changes,  which means once you give this patient's liver transplantation, you should be able  to reverse completely the damage to the kidneys. We also think that the pathophysiology of HRS is purely based on hemodynamic alterations, and vasopressin analogs will be able  to reverse hepatorenal syndrome. So what I'm gonna do in the next 20, 25 minutes is to bust this myth. I'm gonna talk about liver failure as a concept of inflammation,  a concept of oxidative stress in various organ failures. I'm gonna talk about and try and convince you that renal dysfunction in liver disease is not entirely hepatorenal syndrome.  I'm gonna introduce concepts of HRS-AKI and non-HRS-AKI, and go through the new nomenclature definitions and diagnostic criteria, and finally talk about  some recent advances in acute kidney injury. Now this is a slide I always begin talking about, any liver disorder. This is a patient who's got a structurally normal liver,  but with a genetic defect. This is a liver of acute liver failure where there are areas of regeneration and areas of collapse, and there's a completely collapsed architecture here.  So in each of these patients, plus in patients post-transplantation, there's always a risk that these patients will and can develop acute kidney injury.  Now before we go into acute kidney injury, it's very important that we distinguish two distinct concepts. The first concept of acute decompensation of liver disease,  and second concept of acute and chronic liver failure syndrome. Now if you look at this patient, this is a cirrhotic patient sitting on the ward,  and this patient's liver disease is gradually worsening. The liver disease worsens, the liver numbers go off, the portal hypertension, ascites, everything gets worse,  and this patient's kidney disease parallels the deterioration in your liver disease. On the other hand, here is a patient sitting on the ward, stable, cirrhotic, have some ascites,  but something happens here. So there's an insult in the form of bleeding, infection, massive parasynthesis, and you convert a ward-level patient  into a patient with multi-organ failure, and this is ACLF. You've got multi-organs involved. However, if you intervene, this patient has the potential  to go back to the pre-insult stage. So the mortality is higher, the short-term mortality is higher. However, timely intervention can lead  to the patient going back to the pre-insult stage. So therefore, the two concepts are different because the prognosis is different. And this slide is gonna form the basis  of a lot of my further slides, the concept of inflammation, the concept of cytokine storm, and the concept of an imbalance between pro- and anti-inflammatory cytokines here.  This is a patient with cirrhosis who's got a chronic systemic inflammatory state going on. There's translocation of bacteria which takes place continuously,  and that's why there is systemic inflammation. What happens in ACLF is there's a sudden burst of this translocation, sudden burst of this inflammation,  and you get multi-organ failure. So the concept of inflammation remains the same, whether it's acute decompensation or ACLF, but it's about how severe it is  and how many organs get involved. And as you see, kidney here is absolutely central to ACLF. And you can see here as well, when there is a pro-inflammatory phenotype,  you've got the Kuffer cells getting activated, but mainly it is your activation of the monocytes. Once the activation of monocytes take place,  you've got an inflammation-related AKI.  So again, inflammation, SERS, DAMPS, PAMPS, all these immunological features take place in patients who've got ACLF. And this is how the pathophysiology is.  You can see that your portal hypertension is the starting factor, but as you go along, you've got release of reactive oxygen species, you've got splancting vasodilatation,  but if you have super added insults in the form of acute liver damage or nephrotoxic drugs, you can get all the multi-organ failures of adrenal, hepatic, hepatopulmonary syndrome,  and renal dysfunction. Now, we know whether it is the PEL scores, the MEL scores, or any of the ICU scores, they're not very good in predicting mortality. Therefore, the CLIF,  which is the Chronic Liver Failures Consortia, came up with this scoring system of diagnosis of acute liver failure. And you can see that kidney forms an absolute pivotal role  in determining how severe the liver damage is, in determining where this patient should be listed for transplantation, and how early or late should the patient be listed.  This is how the online scoring looks like, and if you put these various numbers, you'll get to categorize your patients in ACLF one, two, or three.  This is the pediatric SOFAR scores, and as you can see, this is based on centiles. It's based on more than one, more than two upper limits of normal, as opposed to adults,  where we have got standard fixed numbers. So, in summary, renal injury in acute on chronic liver failure will determine the prognosis, which is mortality.  It will determine whether you should be listed for transplant or not, and it will also determine whether there is a priority for transplantation and survival.  This is unpublished data from Europe, from King's, Paris, and Netherlands, where we actually looked at 130 patients with cirrhosis, the primary outcome being combined,  which means liver transplantation and or death at day 60, and the secondary outcome was survival at 28 days. And you can see, as the category of liver failure becomes worse,  your survival becomes worse. So, worse the grade of acute kidney injury, worse was the survival. Just to give a perspective to how the kidney and livers have been interacting,  it's not new, 1863's Austin Flint had first described that there's an association with cirrhosis and fulminant renal failure. It was Gordon Haid in the 1920s to 1930s  where he described a clinical syndrome referred to as the Flint syndrome, again, associated with appendicectomy and renal failure.  And finally, the hepatorenal syndrome came out in 1960s. A number of consensus conferences, number of consensus meetings took place to actually define this syndrome.  But the question is, you get acute kidney injury, you get renal failure, is everything hepatorenal? We know that 20 to 50% of cirrhotic patients admitted to the hospital  will develop acute kidney injury. However, there is a new concept, a new concept with International Club of Ascites where they said,  we are gonna divide patients with cirrhosis and ascites into two distinct categories, non-hepatorenal syndrome AKI and HRS AKI. And when you talk of non-HRS,  they talked about hypovolemias, they talk about gastrointestinal bleeding, excessive use of diuretics, loss of gastrointestinal losses from gastrointestinal infection.  They also include in that acute tubular necrosis, but very importantly is bile acid nephropathy and contribution from intra-abdominal hypertension.  And of course, HRS AKI was a functional and that is categorized as a separate entity. Bile acid nephropathy, I think it's important because as bilirubin levels start to go up,  there's accumulation of bile acids which have got a direct toxic effect as well as tubular obstruction. And the bilirubin casts have been found in approximately 80% of patients  if you do renal biopsies, which seem to be pathogenic rather than just simply bystanders. And in addition to HRS AKI, non-HRS AKI, you've got something like the renal diseases  that can either occur as a result of primary liver disease or they occur together with the liver disease as a part of the whole syndromic process.  If I have to summarize what all things can go wrong in a patient with liver disease when they have a kidney dysfunction, it could be pre-renal,  which could be hypervolemia and HRS AKI. It could be intrinsic renal disease, which includes your ATN, acute tubular globular nephritis,  bile acid nephropathy, very, very rarely post-renal, and of course, the renal diseases which occur as a part of the liver disease process. Now coming to epidemiology,  we know that 50% of patients with cirrhosis and the situs will develop AKI. But HRS will constitute a very, very small fraction of them.  In one of the series, 129 cirrhotics with AKIs, it was only 7.6% of them who had HRS as a cause of deterioration. Similarly, in another multicenter trial,  423 patients with AKI, 35% had ATN, 32% had pre-renal azotemia, and there was only 6.6% who had HRS too. Now can we compare the adult incidents with the pediatric incidents?  Now if you look at the pediatric cause of chronic liver disease, it's most of the times is biliary atresia. So these patients undergo a CASI operation,  and it's said that if you don't clear your jaundice within two months of doing a CASI, the incidence, the risk of getting a liver transplantation  or death by two years is almost 90%. So once they develop portal hypertension, they develop ascites, they're kept under a closed vigil, the waiting lists are not that long.  However, in adults, you've got very long waiting lists. Their etiologies are varied. Hence, they undergo all the complications of liver disease, including hepatorenal syndrome.  So their hepatorenal syndrome, though low, our pediatric incidence is even lower. Now this is what forms the basis of what the new classifications are all about.  So you talk about rifle criteria, you talk about AKN, you talk about KDIGO. You just, there are two main parameters, serum creatinine and your urine output.  And if you look at what was the conventional criteria for AKI in cirrhosis, it says a percentage increase in serum creatinine of 50% or more  to a final value of more than 1.5 milligram per deciliter. So it's talking about absolute value now.  But the paradigm is changing. The paradigm is changing because we've got problems with serum creatinine and urine output.  Of course, in general, but in liver disease, you've got liver synthetic function, which is decreased. You've got muscle mass in the form of malnutrition.  Therefore, you have a very low baseline serum creatinine. There's increased tubular secretion of creatinine in cirrhotic patients, and of course, they've got ascites.  So everything is saying creatinine might not be great in kind of diagnosing AKI in liver disease. And if you look at urine output, these patients could be frequently oliguric  with average sodium retentions, but yet might have a normal GFR. On the other hand, they might be having polyuria because the patients are on diuretics.  So therefore, there was a consensus meeting in 2015. The International Club of Ascites met and said, these criteria which are used to diagnose AKI  in non-liver failure patients are not good enough. So what did they say? Let's take out the urine output criteria. So urine output criteria is out,  and they talk about an increase in serum creatinine more than 0.3 milligram per deciliter within 48 hours, or a percentage increase above the baseline,  which is supposed to have occurred in the last seven days. They talk about progression. They talk about regression. They also talk about response or no response.  So in summary, what they said was, we would want to diagnose acute kidney injury in liver failure patients the way we do in non-liver failure patients, but with certain changes.  And those changes are no urine output and a percentage increase in the criteria. But these again were not a hepatorenal syndrome diagnosis. This was AKI.  But a diagnosis of HRS is one of exclusion, and it is renal failure in a patient with advanced liver disease in the absence of an identifiable cause.  So these are the traditional 2007 diagnostic criteria. And what these criteria were, there are four different things. Number one, these criteria said,  we have got HRS-1 and we have got HRS-2. HRS-1 was acute. It was doubling of serum creatinine to a value of more than 2.5 milligram per deciliter in less than two weeks.  So this, I'm gonna tell you why this was not right. It also said that type II HRS is a gradual deterioration serum creatinine to a value between 1.5 to 2.5.  But very, very importantly, this criteria had that there should be no improvement in serum creatinine once you withdraw your diuretics and you give a fluid challenge.  And that fluid challenge was in the form of albumin, one gram per kilo up to a maximum of 100 grams per day. So there should be no response. You shouldn't be on nephrotoxic.  You should not be on any inotropes or septic shock. And you should not have any parengymal injury. So in summary, 2007 guidelines were  that you need to have patient with cirrhosis and ascites. You need to have HRS-1 if your serum creatinine is more than 2.5 in two weeks time. And the problem with these things was  that the baseline creatinine is low. So for you to mount a response to go up to 2.5 is gonna take ages. By that time, your GFR would be really, really low  and you would not be able to start your vasoconstrictor therapy if you keep waiting to 2.5. Therefore, there was a change. And the change was people wanted now to make HRS  as a part of the phenotype of renal dysfunction. And they said that since it gets precipitated by illnesses, we are gonna propose this.  The new definition of hepatorenal syndrome 2015 is this. Diagnosis of cirrhosis and ascites, great. The second thing is diagnosis of AKI  according to the ICA AKI criteria which I just mentioned, i.e. rise of 0.3 milligram per deciliter in 48 hours. There is no hepatorenal syndrome one  or hepatorenal syndrome two in the new classification. The HRS one is now called HRS AKI and the HRS two is now a part of CKD and it's called HRS CKD.  So therefore, everything else remains the same. You need to have cirrhosis, ascites. You need to have diagnosis of AKI according to HRS AKI criteria  and rest everything of 2007 has remained the same. So time limit of two weeks to diagnose HRS is removed. Limiting threshold for serum creatinine  to reach 2.5 milligram per deciliter is removed and the advantages are that there's earlier treatment of patients with HRS one leading to better outcomes  and HRS two now is included in the CKD. You can see here, this is HRS one which is now equivalent to HRS AKI and HRS two is now HRS CKD or AKD.  So in summary, this is how I'm gonna define HRS for you according to the latest definition. Cirrhotic patient with ascites would develop AKI  by detecting a change in the serum creatinine value of more than 0.3 milligram per deciliter or an increase in serum creatinine more than 50% from baseline within 48 hours  with a lack of volume responsiveness with albumin without evidence of shock, recent exposure to nephrotoxic drugs or preexisting structural renal disease.  And this is the study from Pediatrics. Very, very little literature but this is a study from All India Institute of Medical Sciences in Delhi looking at the profile risk factors  and outcomes of acute kidney injury. They had about 600 patients with chronic liver disease. They excluded the compensated and decompensated ones  and they had about 84 patients with ACLF and the incidence of AKI in these patients was 22.6% of which 47% was at admission and the remaining developed during the hospital stay.  The median duration from acute kidney injury onset was about four weeks and if you look at what their causes were, one third had hepatorenal, one third was sepsis  and about one fifth had nephrotoxic injury induced. So you can see that acute kidney injury in these patients is not all about hepatorenal.  It's about different causes of acute kidney injury. And if you look at the various risk factors, high bilirubin, high SERS, high PELSCOLS, high INRs  and spontaneous bacterial peritonitis were the univariate analysis and they actually found that if you have a bilirubin of more than 17 at admission  and the presence of systemic inflammatory response syndrome, your hazard ratio would go up to about seven for death in case you have these features.  And this is the data from our institute just analyzed about last week. We looked at patients with ACLF who were admitted to our PICU from 2012 to 2019.  We excluded all our compensated and decompensated liver disease patients. So exclusive ACLF patients, total 62 of them and the mean age was about 6.4 years.  And as you can see, a variety of etiologies for these patients with ACLF. Majority of them had biliary atresia and Wilson's disease. And you can see that 54.8% is a huge number.  54.8% of these ACLF patients had acute kidney injury and of that, of these patients, about 70, 80% of them had severe acute kidney injury pre-renal in 26, 80 and in 40  and 14% of my patient population had hepatorenal syndrome. And if you look at what was wrong depending upon the different stages of ACLF, as the ACLF stage is going higher,  your baseline serum creatinine, your peak serum creatinine is going higher. The need for transplantation is going higher and very importantly, higher the grade of ACLF,  higher the grade of acute kidney injury and more is the mortality. The mortality in acute kidney injury in ACLF3 is 48%.  Therefore, higher grades of ACLF equates to higher kidney injury and higher mortality. Now coming to pathophysiology of hepatorenal syndrome,  we know portal hypertension is the big killer. When you have portal hypertension, lots of bacterial translocation, lots of bacterial products, cytokines get released  and you've got a splanchnic vasodilatation. Once you get splanchnic vasodilatation, you get reduced effective blood volume and then your compensatory mechanisms kick in.  You've got renin-angiotensin-aldosterone system kicking in, you've got sympathetic nervous system and all these lead to renal vasoconstriction and reduced GFR and in addition,  you get an insult because of sepsis or any other thing you lead to hepatorenal syndrome. So in summary, you have splanchnic vasodilatation  as one of the precipitating factors of pathophysiology which leads to hepatorenal syndrome. And if you look at the pathophysiology of other causes of acute kidney injury,  in addition to what I've just said, it is the surge, the infection, the bleeding leading to tubular damage and you can have functional renal failure  which can be pre-renal or hepatorenal and you can have tubular damage and lead to acute tubular damage. So why do we want to know the etiology?  Acute kidney injury is acute kidney injury. I can just take off the nephrotoxic drugs, put them on renal replacement therapy. So why is it important specifically in a patient  who has got liver disease to know what has caused kidney injury? I think A, the treatment is different. You will want to give volume resuscitation pre-renal.  You might want to put renal replacement therapy for ATN and might want to use some vasopressors and albumin for hepatorenal. Two, there are certain treatments  which might be contraindicated in one as compared to the other if you know the etiology. And very importantly, you have a patient with liver disease who's gonna get transplanted.  If you know right in the beginning, my patient has got hepatorenal, what is the chance that after transplant, the kidney function will get back? So renal recovery.  It also depends upon the survival post-transplantation as well. And this is what I was talking about, survivals. This was a study where looked at different causes of AKI  and you see that the survival with hepatorenal syndrome was the lowest at 15% and the odds ratio was 6.88. And if you look at the impact of AKI etiology  on mortality following liver transplantation, you have got ATN here and you've got HRS here. You can see that if a patient with liver disease has ATN  and that patient goes into liver transplantation, that the chance of this patient dying is higher than a patient who goes into transplantation with hepatorenal syndrome.  Similarly, this is a chance, accumulative incidence of CKD. You can see if you've got ATN, the chance of you getting CKD is much higher than if you had HRS or any other etiology  of acute kidney injury. And this is the EGFR. If you have got no AKI, this is the chance of your GFR. But if you've got ATN, you can see the GFR does not recover  and this is the time after transplantation. So you can see that the etiology of acute kidney injury dictates what happens to survival, what happens to the incidence of CKD  and what happens to recovery of GFR. This is what you find in the various urinary findings in renal dysfunction. Fractional excretion of sodium,  fractional excretion of urea can be used, but these patients are on diuretics, these patients have got renal vasoconstriction, so not very reliable in patients with liver disease.  So are there biomarkers which are better than serum creatinine and better than urine output? And we need them because we want to differentiate the etiology of AKI and we need them  because we want to prognosticate about what's gonna happen to them post-transplantation. So this is a study of biomarkers done in 2014, published in Journal of Hepatology,  and you can see here whether it is pre-renal azotemia, whether it's chronic kidney disease, it's hepato-renal syndrome, and this is the acute tubular necrosis.  So if you have got tubules which are damaged, you'll have higher urinary engel. This is all good, but if you look at the difference between pre-renal and HRS,  and HRS and acute tubular necrosis, the values are in between.  So is it trying to tell me that there is some structural damage in patients with hepato-renal syndrome as well? Which is not clear. And is that the reason that a lot of patients  with hepato-renal syndrome, when they're put on terdepressin or any vasoconstrictor, they do not respond? So biomarkers are good,  but biomarkers does not, in this particular scenario, give me a clear-cut answer that I am dealing with ATN and I'm not dealing with hepato-renal syndrome.  And this is a study which actually told us that if you have engel more than 365, that was discriminatory of acute tubular necrosis. What are the drawbacks of these biomarkers?  A, not universally available throughout the globe. Though biomarkers of structural damage are characteristically elevated, as I've said, they are low in pre-renal,  but they're in between between pre-renal and acute tubular necrosis. So we have to take it with a pinch of salt  and say it cannot be discriminatory. And this is the classic renal biopsy of patients, very clearly telling us that there's inflammation, there's inflammatory infiltrates  in acute tubular necrosis, as compared to this. If you look at this, this is calprotectin. In patients with acute tubular necrosis, almost nothing in HRS and nothing in controls.  So there is inflammation, that inflammation plays a major role.  So what is the treatment? I think we need to treat the precipitating factors first,  whether it's a GI bleed, whether it's hypovolemia. And again, my favorite thing is please, please make sure that intra-abdominal pressure is taken care of.  You measure your intra-abdominal pressure, if that's pressing on your renal vessels, you're not gonna win. So do parasynthesis and make sure it's not a massive parasynthesis  and you need to replace it back with 20% albumin. And if you've got high risk factors like high serum bilirubin or systemic inflammatory response,  start doing the basics first again. Take care of nephrotoxic drugs, stop the diuretics, treat your sepsis with antibiotics, which are kidney friendly  and use albumin as a volume expander. And the management of HRS, because we talked about splancting vasodilatation, we can vasoconstrict using vasopressors  like your terlepressin and albumin, which is gonna increase your blood flow, is gonna decrease your renal vasoconstriction and gonna reverse your hepatorenal syndrome.  So this is quite evidence-based about use of terlepressin. Unfortunately, not available in the United States, so people can use norepinephrine or they can use vasopressin.  So this is how terlepressin acts on the V1 receptor, causes splancting vasoconstriction and that it reverse the whole vicious cycle. But the management dilemmas are,  should we use vasoconstrictors alone or should I use it with albumin? Which vasoconstrictor to choose from? Should I use it as a bolus or as an infusion?  And how do I define a terlepressin responder  and will treating inflammation or the cytokine storm in ACLF help the associated AKI?  Do you know what time?  Five, five, yeah, okay, thank you. So this was a study of terliprasin by Arun Sanyal from Virginia. Two groups. It was terliprasin and placebo with albumin in both the groups.  And you can see that in the placebo groups, the terliprasin is going up, the creatinine is going up, whereas the terliprasin-treated group, the creatinine is nicely coming down.  And the best predictor of HRS response was a low serum creatinine at the beginning of terliprasin. So you start terliprasin at an early stage where the  creatinine is high but not too high, you are at a higher chance of getting a response. And this was a study done in 2016, same group, and they said actually you don't need to have a  complete hepatorenal syndrome reversal. Even a decrease in serum creatinine from the baseline of 30% can actually predict that this patient will have a better prognosis in terms of  HRS reversal and in terms of mortality. But the question is, do all patients treated with terliprasin and albumin respond? And the meta-analysis shows only about 52%.  And the reasons could be, A, either we are starting it too late because we say early onset will respond better, or B, there is something else going on  besides just the splanchnic vasotilization, the inflammation. We are talking about structural changes. We have not taken that into account.  And can we identify those patients who will respond? Because this drug is great, but this drug has got side effects. So we want to restrict the exposure of this drug  only to that group which is supposed to respond. And the way we should know when to start terliprasin, I think we should also know when to stop terliprasin.  And this was a study looking at association between different grades of ACLF and the response to terliprasin. You can see where your ACLF grade one, two, or three.  It is the lower the serum creatinine at presentation, higher the response. So it did not matter what grade of ACLF you were. The response depended on that.  And these are the responders versus the non-responders. You can see that whether you are in ACLF one, two, or three, it depends upon the number of organ failures.  More the numbers fail in ACLF, lesser will be the response of terliprasin and albumin. So the question is, if you are in grade three ACLF  and the response rate is 28%, would you even consider starting these patients on terliprasin? And if you look at the 28-day mortality, it's  actually slightly better in responders than the non-responders. So what my practice is, I do not care what grade of ACLF are they in. I would give the bog standard treatment first  and then start these patients on terliprasin. But, of course, cost is a big issue. If you look at the cost of terliprasin versus vasopressin and noradrenaline, noradrenaline,  of course, is cheaper. We in intensive care are more used to using noradrenaline. And, therefore, noradrenaline sometimes and in most of the units form actually the first line  of treatment for these patients. Now, I don't have those slides, but just talking about whether should I use bolus or infusion, we have now said bolus dose is associated with more  side effects, and we start on a terliprasin infusion and we use noradrenaline as our first line of treatment. In addition to that, we have to drain the ascites.  Transjugular intrahepatic portosystemic shunting can be done. Renal replacement. Controversial in these patients, but only for those patients who are listed for transplantation.  If they're not listed for transplantation, unless they have got hyperkalemia and fluid overload, we do not start them on renal replacement therapy.  And, of course, liver transplantation is considered to be the gold standard. But if you look at the reversibility, it is shorter the duration of HRS-AKI.  You're associated with better outcomes and greater renal recovery. And if the AKI persists for more than six weeks, it's less likely, and the patients might then  need to be considered for a combined liver kidney transplantation. So if you look at the different eras, this is the survival in the 1990s, and in the 2008, you  can see whether it's HRS-1 or HRS-2. I'm not supposed to use these terms anymore, but this is what it was that time. The survival has become much better.  Just summarizing the acute kidney injury in patients with liver disease, you need to know whether you are in stage 1, stage 2, or stage 3. Take care of nephrotoxics.  Make sure that you've taken care of gastrointestinal infections. Once you've done that and your AKI progresses, you need to give fluid resuscitation with albumin.  There's no response to albumin. Now you know you probably are dealing here with a patient who's got cirrhosis, who's got ascites, who's not responding to your volume resuscitation,  probably is hepatorenal syndrome. You might want to consider starting vasoconstrictor therapy at this point in time. And this is just kind of a graphic presentation  of what I just talked about. And this is what I do at King's. Noradrenaline first, max of 1 microgram per kilo per minute. I add in terlepressin infusion at 10 micrograms  per kilo per hour, very carefully monitoring the ischemic side effects. Double the terlepressin if there's no response in 48 hours. And if there is no response, by day 4,  I actually stop terlepressin. The maximum I've used terlepressin is about two, two and a half weeks. But of course, very, very carefully monitoring the ischemic side effects here.  What do I have in the future? Do I have any drugs which can be used? But before I come to drugs, I would like to reiterate is the correct differentiation of acute tubular necrosis  from hepatorenal syndrome. Because if you start vasoconstrictors in acute tubular necrosis, you're not going to get a response, and you will label this child  or a patient as a non-responder. You've got partial vasopressin agonist. You've got serolexin recombinant form of human peptide 2, which will selectively dilate the renal vessels,  not causing any effect on mean arterial pressure. And if we can use them in patients, that would be great. But very importantly, can we reduce the inflammation  and organ failures by using plasmapheresis, mesenchymal stem cell transplantation, or GCSF? Because this is the multi-organ failure, and there are these biomarkers coming from  these multi-organ failures which affect the kidney as well. So treat the inflammation. So in summary, patients with cirrhosis prone to developing AKI.  The new ICA AKI criteria provide a simple but prognostically relevant staging system. Potential triggers of AKI should be recognized and removed.  Differential diagnosis of HRS AKI includes ATN as well. And biomarkers have a role, though not very clear-cut. And vasopressors such as terlipressin and noradrenaline  can be used in combination with albumin. And this is just our invitation from Stuart Goldstein and myself. We are organizing the first international conference  of critical care nephrology in children in London, 27th to 29th, and I hope to see you guys there. Thank you. And now I invite my dear friend Stuart Goldstein,  who is the professor of pediatrics and the director of Critical Care Nephrology Center for Acute Care Nephrology at Cincinnati Children's to talk about the past, present, and future  of pediatric AKI in CRRT. Thank you, Akash. And actually, this all worked out quite well. Dr. Cain-Gill and Dr. Talwani set up the past pretty well.  I'm going to reiterate a few things and build upon what you've heard today. Here are my disclosures, and we'll go from there. So AKI is difficult.  There is absolutely no magic bullet, but there are things that we can do today. The progress in AKI has really moved forward since we developed a standardized definition.  Prior in the past, we had 30 different definitions. Acute renal failure led to RRT, which was often late, as Dr. Talwani showed us. Most of the data came from single centers.  We didn't even think about neonates, and we were not even thinking that AKI would lead to chronic kidney disease. So what I'm going to talk to you about now  is how we've moved forward since we have a standardized definition. First definition that was an empiric definition was the rifle criteria, which you've heard about.  And what's really important about this, it was multidimensional. It isn't that patients had AKI or they didn't. That's what we thought about before.  Patients had acute renal failure or they didn't. There was no staging. There was no grading.  And it was multidimensional in that there are urine output criteria and GFR criteria. Then the akin criteria built upon this because of the 0.3 milligram per deciliter rise  in serum creatinine, or a 50% rise, which has been shown to be independently associated with mortality. And you can see from Dr. Deep's talk that this has now been incorporated  in the hepatorenal syndrome guidelines. And then KDGO ended up harmonizing rifle, akin, and pediatric rifle into a single definition. So as of today, February 18, 2020,  this is the definition of acute kidney injury. There will be another KDGO consensus conference in the next year or two where we will look at the KDGO criteria.  But when we think about this,  these are very small changes in kidney function, something that when I was a fellow and a resident, we would have thought were part of the error  of the laboratory measurement itself. And this should actually make sense to us because those of us who are nephrologists and participate in kidney transplantation,  when we remove the one half of the nephron mass from an adult and put it into another patient, that donor's creatinine does not change. There may be long-term effects,  but the creatinine doesn't change. The urine output doesn't change. And we've taken one half of the nephron mass away. So a 0.3 rise per deciliter or milligram per deciliter  or a short period of oliguria tells us there's something going on systemically with the patient. Scott Sutherland looked at this, looked at a number of different criteria and found  that mortality rates, irrespective of the criteria, increased in children in the ICU with each stage. And for that reason, Scott and those of us  who were on this paper said we really should use KDGO just so we can harmonize with our adult colleagues. Length of stay also, while there are different associations,  they're only slightly different. And so we've agreed that KDGO should be the law of the land. So does this matter? So we participated in a study called the Assessment  of Worldwide AKI, Renal Angioepidemiology in Children. And this was a study of over 5,000 children admitted to one of 32 ICUs, nine countries, four continents.  And you can see 5,200 patients, slight male predominance, and you can see the primary ICU diagnoses here, shock, cardiovascular, respiratory, surgical, et cetera.  And when we looked at 16 variables that were associated in bivariate analysis, including KDGO stage 2 or 3 AKI, and put them into a multivariable model,  we found that severe AKI, stage 2 or stage 3, was independently associated with mortality in these critically ill kids. Odds ratio was 1.77. So this association is strong.  It's strong in children. And children often don't have the other comorbidities that, chronic comorbidities that adults have. So that's why this is particularly interesting.  And when you look at the Kaplan-Meier curve, you can see 28-day mortality increases with each stage of AKI. We do not need to do another epidemiologic study  to prove that AKI is bad. We just don't. In the adult world, similar thing. The AKI epi study, looking at about 1,800 critically ill  adults, multinational study, looking at KDGO criteria,  and again, 1,800 patients looking across the globe. And irrespective of the model, stage 1, stage 2, stage 3 AKI, independently associated with mortality.  So we couldn't even get here until we had a standardized definition. And now that we have a standardized definition, it doesn't matter what the underlying cause is. AKI is bad.  And so we could stop right now and say AKI is bad. But OK, Stu, we get it. AKI is associated with mortality and morbidity in hospitalized children and adults.  So what are you and your AKI friends going to do about it? Because it's not enough to continue to do these large epidemiologic studies.  We need to get on with it and figure out what to do. So I'm going to talk to you about NINJA and Dr. Gill did a little bit. So NINJA stands for nephrotoxic injury negated  by just-in-time action. And what I'm going to show you in terms of what I think is the present or present future is a couple of studies that if you just pay attention  and are systematic in how you pay attention, you may be able to improve some outcomes in patients who are at risk for acute kidney injury.  Right now, there are only two things really that we can do for patients who have acute kidney injury or are at risk for it. We can remove unnecessary nephrotoxins.  And if it gets severe, as Dr. Tolwani was discussing, we can start to think about when will we start renal replacement therapy. So things we can do today, nephrotoxin avoidance,  educate about AKI, and I'm going to show you a little more data about how we may be able to incorporate biomarkers. So the NINJA vision statement is children  should only get the nephrotoxic medications they need for the duration they need them. We are not saying that patients should not get nephrotoxins.  These medications are lifesaving and are important for critically ill and non-critically ill patients. But we should be very vigilant about the ones  they're getting, making sure that they get them, and not just say multiple nephrotoxin exposure is the cost of doing business in a tertiary health care center.  So the objectives of NINJA were to help physicians and providers become better providers at the bedside, to do all of the work of screening  and give them the information that this patient was at risk for AKI, that being on three or more nephrotoxic medications at the same time,  or an IV aminoglycoside for more than three days is putting a patient at risk for acute kidney injury. When we looked at the data that I'm not going to show you right now, but we  looked at kids who were getting prolonged aminoglycosides, depending on the definition, 20% to 33% of those children developed acute kidney injury, and these  are non-critically ill kids. Do we tell families, I'm giving your child a medication that can cause acute kidney injury, one out of five to one out of three times  when they're getting gentamicin? No, we don't. When we give somebody rituximab, we have to spend about 40 hours going over all the side effects of rituximab.  Rituximab does not have an adverse event rate of one in five to one in three. It doesn't. And we have just taken and thought about nephrotoxic exposure as part  of the cost of doing business. I don't tell a family that their patient, their child, has a one out of five to one out of three risk of a major organ injury that could  lead to chronic kidney disease. So here are the NINJA criteria, and we developed these criteria based on the epidemiologic work that I just alluded to.  And we came up with a number of outcome measures. We looked at the high nephrotoxic medication exposure prevalence rate per 1,000 patient days.  And again, these are in children who are not in an ICU because we know that patients in the ICU have AKI that's multifactorial. We looked at the AKI prevalence rate  per 1,000 non-ICU patient days. We looked at the percent of patients who were exposed in one of those two ways who developed AKI, and then the AKI intensity rate, the days of AKI  per 100 days of exposure that a patient had. And we thought this might be the metric we'd move first because we found that when patients were exposed  in this way in our previous studies, serum creatinine was only being measured at least every four days 50% of the time. And since nephrotoxic AKI is non-oliguric,  if you're not measuring creatinine, you're not going to detect AKI. So what NINJA did was when patients were exposed in one of these two ways,  we just asked the pharmacist to recommend to the team to get a daily creatinine. That is the intervention. Then we would continue surveillance for as long as the patient was exposed  and then two days after the exposure ended. If a patient developed AKI, we would follow them until their creatinine returned to baseline, or at 30 days we would end the surveillance  and close the record. And Dr. Gill showed this to you, or alluded to this earlier. Over three and a half years, we observed a 34% reduction in high nephrotoxic medication exposure.  I didn't tell anyone they couldn't use vancomycin. I didn't tell anyone they couldn't use piptazobactam. But we just gave people the information  and then they became providers and said, maybe I don't need to use this combination of medications. More importantly, we had a 62% reduction  in acute kidney injury over that period of time. Again, just giving people the information about high nephrotoxic medication burden. And if you look over that period of time  at Cincinnati Children's, had NINJA never been put in place, we would have had 633 additional exposures and we would have had almost 400 additional AKI cases.  Just by giving people information and telling them to pay attention to what they were doing. So, it's great that the nephrons in Cincinnati kids are being protected,  but a quality improvement project isn't helpful if one, you can't sustain it, which we have, but two, if it doesn't work elsewhere. So, in the last three years,  we've disseminated NINJA at nine pediatric hospitals through a couple of grants. We measured the impact of NINJA on nephrotoxic AKI in these hospitals. But more importantly,  we wanted to look at contextual measures. I am very fortunate to be at Cincinnati Children's where we have a massive safety program, a massive QI program,  but other institutions may have a lone NINJA, a lone nephrologist or pharmacist who's doing this on her or his own. And so, we need to understand the contextual measures  and factors that accelerate or create barriers to success in NINJA. And these data have been accepted for publication and I think they're coming out next week  in Kidney International. But here's what we found. Again, we found about a 24% reduction in acute kidney injury in the nine NINJA hospitals. So, this does work elsewhere.  And what's really interesting is all of the rates that I showed you before, the exposure rates, the AKI rates, not the intensity rates, but the ultimate rates that were achieved  in the NINJA hospitals were identical to what we saw at Cincinnati Children's. So, quality improvement can get you to this point, but it can't get you any better.  And we've been looking at this now for eight years. So, we need in the future, and I'm going to discuss how we're going to do that later in the talk,  how we might be able to improve this. NINJA is now being rolled out to about 30 pediatric institutions, the Solutions for Patient Safety, and ultimately, every pediatric institution  hopefully will have some sort of NINJA program. Again, interestingly now, in the number of patients who had AKI who were exposed, one out of six, you can say confidently,  one out of six children who are on an IV aminoglycoside for three or more days or on three or more nephrotoxins, they are going to develop acute kidney injury.  So, let's look at some adult data about let's have people pay attention when a patient has acute kidney injury. And so, this is from Dr. Kellum and the Pittsburgh group  and looking at clinical decision support for in-hospital AKI. And in their system, now their team gets an alert, warning of possible acute kidney injury.  This result tells you that the patient has had a 0.3 milligram per deciliter rise based on this reference. This is KDGO Stage 2. This is a clinical diagnosis  and clinical evaluation is recommended before management is altered. Consider consultation or if the patient may be critically ill, consult an intensivist. And what did they find?  Well, just by having people pay attention, and this is an interrupted time series chart and here's where the intervention went in. The upper line are patients with AKI.  The lower line are patients without AKI. So, there wasn't some other hospital-wide initiative with the lower graph that was decreasing mortality,  but there was a 10% decrease in mortality at UPMC just by having patients or physicians pay attention and alerting them to what was going on. No bundle here,  just giving people the information. That's what we can do today. Okay, so you've heard some about Land of Biomarkers and I'll go, we call this troponin worship.  But why do we call it troponin worship, right? Troponin by itself, while it's an important biomarker, the reason it's important is because it tells you something's going on  and we want you to do something about it. Take the patient to the cath lab. Just having a biomarker out there itself is not going to improve outcomes.  And so, you heard a lot about NGAL and just the issue here, again, as Dr. Gill noted, here in kids after cardiopulmonary bypass where there's a high rate of AKI, about 40%.  NGAL came up earlier, two days prior to a change in serum creatinine. So maybe there's something that we can do in this period before the creatinine changes  because I've already shown you that once the creatinine has changed, the horse is out of the barn. The risk for mortality is significantly increased.  TIM-2, IGP-7, again, the stress cell cycle arrest markers. And again, this is from the SAFIRE study showing the prediction of acute kidney injury.  So where I'm thinking that we're living now is we need to think about these biomarkers and how they can direct our therapy. And so Dr. Cain-Gill showed this to you as well,  but this is really the future, what we need to do, or the present future. Once you know that there's stress or once you know there's injury, that's who you go ahead  and put your very exhaustive, extensive interventions. And you can see all of this. This would exhaust us to do all of these things for all patients.  We should direct this for patients who are at risk, avoidance of nephrotoxin, discontinuase, et cetera, et cetera. And again, you can see a very simple, elegant bundle in PREV-AKI.  And what they found was dramatic. A massive reduction in all AKI, a massive reduction in stage 2 to stage 3 AKI. Small study, but this is how we should move forward.  Why is this important? Because every drug trial in AKI has failed because we've been waiting for urine output and creatinine to change.  But now we might have markers where we can randomize only those patients who are truly at risk for AKI to the drug study. I want to step back even further.  I want to know in whom I should get the biomarker. And so Mink Chawla and I developed this concept called the renal angina index, or renal angina. Now, cardiologists hate that term  because there's no squeezing pain on the kidneys. But what we want to understand is in whom should I get the biomarker? If you go to an adult ICU,  a significant percentage of patients are going to have troponin leak. They'll have an elevated troponin, but they're not having the acute coronary syndrome.  What we wanted to do is say, look, I want to figure out who's at risk for AKI, get a biomarker in them, and enrich the sample for whom I need to think about the intervention.  So based on a number of pediatric studies,  Mink and a number of us developed this renal angina index. And patients rule in if this product, based on are they in the ICU, they get a point. Do they have some transplantation?  They get three points. Are they intubated on a presser? Those are your Framingham risk factors. And then we look at changes in creatinine or fluid accumulation  at 12 hours of ICU admission. And it's a very good rule-out test. If it's less than 8, patients rule out. If it's greater than 8, that's who we think about giving  or getting a biomarker in. And in this single center study from Dr. Shine and Menon, we looked at PRISM-3, sepsis, transplant history, and renal angina at 12 hours  to predict K-D go stage 2 or 3 at day 3 of ICU admission. If we could predict that, that helps the intensivist. Should I give a lot of fluid? Should I not give a lot of fluid?  Should I get a biomarker? Should I not get a biomarker? When do I start thinking about renal replacement therapy? And you can see the odds ratio for renal angina  was significantly higher than all of these other risk factors. And then when we added a biomarker, so we added NGAL, we had almost perfect prediction of stage 2 to stage 3 AKI.  Now, this has not held true because this was a small study. But again, look at the negative predictive value for renal angina index.  If the patient rules out, their risk of AKI is very low. If they rule in, their risk of AKI is almost a flip of a coin. That's in whom you get a biomarker.  This is Framingham risk factors plus chest pain. Now let's get a troponin and do something about it. So we're trying to recapitulate the acute coronary syndrome paradigm.  And this is what we're doing. So this is the present, but really the future. So we've been doing this study called Taking Focus 2 for the last two years.  So we have programmed our electronic health record, just like we did for NINJA, to put out the renal angina index at 12 hours. It gets put in Epic. It's right in the Results tab.  And based on that, if the renal angina index is less than 8,  patients get standard of care. If it's greater than or equal to 8, then NGAL, which is the biomarker we have available, is reflexively ordered.  A research nurse doesn't have to look at it. A resident doesn't have to order it. It happens automatically. And then based on that, if it's less than 150,  patients get standard management. If it's greater than 500, now this is where we start to address some of the issues that Dr. Tawani spoke about.  Okay, now I know I have a patient who's at risk. They have a damaged biomarker. That's very high. That's going to... Our intensive care group is now going to treat this patient  a little bit differently and fluid restrict or start renal replacement therapy at 10% to 15% fluid overload because we know higher fluid overload percentages  are associated with mortality. This gives the intensivists more confidence about putting a catheter in and starting renal replacement therapy because we've taken demographics  and then a damaged biomarker to tell us this patient's going to have acute kidney injury for quite some time. If we have a middling number between 150 and 500,  then we institute the furosemide stress test. So a patient will get one milligram per kilo of furosemide and we'll look at the urine output over the next two to six hours.  And we have data in children to show that if the urine output does not rise to greater than three mils per kilo per hour, their risk of progression  to worsening acute kidney injury is greater. And it's kind of interesting because in the adult world, there was a 200 mil increase in urine output  and the average adult is supposed to be 70 kilos. So 70 times three is 210. So it actually worked out the same way. And if it does not increase again,  then we know this is a diagnostic, it's not a therapeutic, it's a diagnostic dose of furosemide. Again, our intensivists will get on a more fluid restrictive strategy  or start renal replacement therapy sooner. And when we've done this, when we've added NGAL and we've added the FST, our positive predictive value  to the renal angina index goes up significantly from about 30% up to about 60%. Specificity and sensitivity also improve. These will need to be calibrated and they're not perfect,  but this is giving people information at the bedside to help them predict the future. Looking at NGAL in NINJA, David Askenazi and I, one of the problems with NINJA,  especially for children, is that you've got to get the daily creatinine, have to put and do venipuncture. So we looked at a biomarker NGAL  to see if we could rule out who is going to have AKI. And you can see in terms of stage two and stage three AKI that NGAL does a pretty good job  to predict who's not going to have AKI. And so we're now working on supplanting daily creatinines with NGAL and then only getting the creatinines  more periodically or less frequently until the patient has an elevated NGAL. And then finally, this is something, again, we're working with Sander Vinks  and we've put in a grant for this. As I showed you or I told you about NINJA, we have not improved the AKI rate after the initial institution  of the NINJA Quality Improvement Program. And so now we're going to start to look at aminoglycosides and vancomycin and start to look at how biomarkers  and other physiological factors can tell us who's going to have acute kidney injury. So first we're going to do standard PKPD over the entire spectrum of pediatric disease  from neonates infants all the way to adolescents. And then using modeling and simulation, we're going to add a biomarker and we're going to get the biomarker  before the patient is actually exposed to those medications and based on that, determine if we can use dose adjustment nephron sparing strategies  based on the biomarker to try to limit acute kidney injury. And then finally, we will incorporate this into a physiologically based PK model that will take all the information  from the electronic health record, including demographics, biomarker inputs, to give physicians at the bedside through an app, a clinical decision support tool to say  in that patient with cystic fibrosis who has this creatinine, this NGAL level, this set of demographic factors, here is the dose of tobramycin that you should give.  And this is what a PKPB model looks like and again, all of these components go into it, but our goal is that we would do this for all nephrotoxic medications,  but we're starting with ones that we can measure in our clinical lab, aminoglycosides and vancomycin. So that's the future to, I think, where we need to go.  What Pittsburgh and a number of others have shown is what we need to do is give people the information at the bedside to be better physicians at that point in time.  Leverage the electronic health record, leverage demographics, tell them who's truly at risk, and then we can go ahead and do the studies, hopefully with interventional strategies  that will prevent acute kidney injury. So in my mind, we have a bright future ahead. Thank you very much.  Can I now ask Dr. Talwani and Dr. Gill to please join me on the podium? The floor is open for questions now. Please come to the mics there on the...  Those are very good questions.  Those are very nice presentations, Samir Kadri, NIH. Question for all of you.  Although most of us have clinical experience with this, and I think we know what our threshold should be in terms of dosing for diuretics,  is there any guidance for the critical care community in terms of what the sweet spot should be  to make sure we give a high enough dose to elicit a clinical response and not miss it and not cause toxicity? Yeah, sure. Is it okay if I handle that?  Yeah, I think that what Mink did with the furosemide stress test is really helpful. So in adults, one milligram per kilo, in a patient who has not had exposure to diuretic,  1.5 milligram per kilo, and then look at what happens over the next two hours. And when you look at those data, performs as well or better  than any of the novel biomarkers we're talking about. And that will tell you it's a diagnostic dose if a patient is going to respond to a diuretic.  And so then don't go down the path if the patient doesn't respond, well, now I'm going to give three per kilo, or now I'm going to do a continuous infusion,  or now I'm going to go to Bumex because I'm a cardiologist and I think that works better. You will understand if that patient  is going to respond with AUCs in the 0.85 to 0.88 range. I think a more critical, but then how do I manage the patient? To me, the diuretic should give you enough volume space  so that your patient can get all the blood products, all the medications, all nutrition that he or she needs without developing worse fluid overload.  Because if they can't, then you need to think about either fluid restricting, which we don't generally want to do in the critically ill, or starting renal replacement therapy.  Looking at a number of response, either in kids in mLs per kilo per hour, or even KDGO mLs per kilo per hour, isn't helpful in isolation.  You have to determine what do I need as the intensivist, or if I'm the nephrologist working with the intensivist, what do you want to give that is optimal for that patient?  And if the patient cannot tolerate those volumes, and diuretics do not give you that volume space, then you have to progress on to renal replacement therapy.  People view renal replacement therapy and diuretics as competing therapies that are complementary. It's just escalating the intensity of care. Anyone else? Great talks.  Thank you very much. We tried to bring in the biomarkers at our facility, and encountered unbelievable resistance due to the cost. So I was really fascinated about the renal angina,  and then I got a little bit confused, because you thought maybe you'd replace routine creatinines with NGALs. So anyway, I've got a couple questions.  The first one, did you do any comparison or make a choice between NGAL versus nephrocheck? But perhaps I should ask the other question first. If we did the renal angina, do you  think that would lead to doing less renal biomarkers? And I could present that case to our pathology people, say, I'm being smart. Yeah, absolutely.  So about 10% of kids rule in for renal angina. So it gives you a directed way of doing that. So then 90%, and that's what you want a screening system  to do on population health, is you want to rule out patients you don't need to pay attention to. So yes, for us, renal angina hasn't been looked at as extensively in adults.  But again, I think if you are talking to your pathology group or your clinical lab and saying, we're only going to get it in patients who are truly at risk,  I think that would go a long way. And then the second issue is that NGAL works really well in kids. Full disclosure, I do consult for the company. I've also consulted for Astute.  And NGAL was discovered by my boss. But it's Prasad Devarajan. But NGAL works very, very well in kids. And so that's why we chose it. Cell cycle arrest markers work well as well.  But they haven't been studied as extensively in children. And NGAL may have some limitation in adults because it is associated with inflammation.  So it could be rise in some chronic diseases and sepsis. So that's one of the reasons why, in addition to the fact that it's FDA approved for us, was  why we went in that direction for the TINP2IJFBP7. But I also agree with the fact that, look, there's a lot of things we're not doing to assess risk for individuals.  Before we even get to the biomarkers, right? Do you know how much AKI occurs in your ICU? Do you know ways in which you could assess those individuals  and already be preventative just based on a risk score before we even get to the biomarkers? People aren't even doing that yet. So that's a good first step. Great.  I think we've reached the end of the session. So thank you all for coming. And thank you to all the speakers for the wonderful talks. Thank you.

Video Summary

The video discusses the relationship between liver disease and acute kidney injury (AKI) in cirrhotic patients. Around 20-50% of patients with cirrhosis develop AKI, which has a significant impact on kidney function. Differentiating between non-hepatorenal syndrome AKI and hepatorenal syndrome (HRS) AKI is crucial for appropriate treatment and prognosis. Treatments for HRS include vasoconstrictors like terlipressin and liver transplantation. The video also highlights the need for improved biomarkers to diagnose and manage AKI in liver disease patients. Overall, the discussion emphasizes the importance of recognizing and treating AKI in patients with liver disease, while understanding the different types and causes of AKI in this population.

Asset Subtitle

Renal, 2020

Asset Caption

The Society of Critical Care Medicine's Critical Care Congress features internationally renowned faculty and content sessions highlighting the most up-to-date, evidence-based developments in critical care medicine. This is a presentation from the 2020 Critical Care Congress held February 16-19, 2020, in Orlando, Florida, USA.

Learning Objectives

Examine key mechanisms of acute kidney injury (AKI) in preclinical models and explore their potential relevance to clinical AKI syndromes
Discuss the evidence on renal replacement therapy (RRT) indications and timing, dose, and modalities as well as logistics in order to apply an effective strategy for managing RRT for severe AKI
Identify differences in diagnostic and management approaches that are warranted in pediatric and postoperative patient populations as well as less common but potentially treatable AKI syndromes

What the ICU Team Should Know About RRT
Ahita Tolwani

Use Biomarkers for AKI
Sandra L. Kane-Gill

AKI in Liver Disease: Is Everything Hepatorenal?

Akash Deep

The Past, Present and Future of Pediatric AKI & CRRT Studies

Stuart L. Goldstein

Meta Tag

Content Type Presentation

Knowledge Area Renal

Knowledge Level Foundational

Learning Pathway Introduction to the ICU

Membership Level Select

Membership Level Professional

Membership Level Associate

Tag Renal Replacement Therapy

Tag Kidney

Year 2020

Keywords

liver disease

acute kidney injury

AKI

cirrhotic patients

non-hepatorenal syndrome AKI

hepatorenal syndrome

HRS

treatment

prognosis

vasoconstrictors

terlipressin

liver transplantation

biomarkers

diagnosis

management

	Society of Critical Care Medicine 500 Midway Drive Mount Prospect, IL 60056 USA Phone: +1 847 827-6888 Fax: +1 847 439-7226 Email: support@sccm.org	Contact Us About SCCM Newsroom Advertising & Sponsorship DONATE	MySCCM LearnICU Patients & Families Surviving Sepsis Campaign Critical Care Societies Collaborative	GET OUR NEWSLETTER
© Society of Critical Care Medicine. All rights reserved. \| Privacy Statement \| Terms & Conditions The Society of Critical Care Medicine, SCCM, and Critical Care Congress are registered trademarks of the Society of Critical Care Medicine.