I'm going to be speaking to you about the ICU management of AKI. These are my disclosures. So as a kind of a series of objectives, there will be some overlap, and I hope that this does dovetail on a few of the different themes that have been brought up in the other talks. Ideally, understanding the context of AKI in the context of the critically ill ACLF patient. We know that we've gone through a lot of the newer definitions of AKI and cirrhosis, and the ability to discriminate between different prognoses in ACLF patients, certainly looking at the differences between the presence and the absence of liver transplant. We'll explore the challenges when looking at the indications, timing, and modalities of renal replacement therapy. Just want to review some recent data on the role of interoperative CRRT during liver transplant. And just as a proof of concept, I just want to give you a little bit of an idea about the role of CRRT in acute liver failure. So we're all probably aware of the CLIF-C ACLF score, which is a variation on the original SOFA score. This was refined more recently to look at as the CLIF-C ACLF score as a nomogram out of 100, which includes the six organ failures, and also as a logarithmic equation, also includes age and white count. And as we know previously, this does discriminate patients into different prognostic categories. As you can see here, by looking at patients with a CLIF-C ACLF score of greater than 70, and this was data that we published outside of the original canonic study, this was portending a 90% mortality at 28 days. One of the things that's important, and it kind of highlights the role of hepatology and critical care in the management of these patients, is that ACLF is a dynamic syndrome. And there are patients that with interventional support can actually improve. In this particular study, we demonstrated that almost 40% of patients improved by an ACLF grade, and that their natural history tended to follow their later prognosis, if you will, at day three rather than on admission. So as Dr. Angeli and others have already defined, the definitions of AKI and cirrhosis have become more consistent with the KDGO classification, and similar to some of the other critically ill population. And I know that he already showed this study, but it does highlight that we are able to discriminate patients into very different prognostic categories, which shows their value. One of the challenges that you may find is that there is old dogma, you might call it, of when you're talking to a critical care physician about a cirrhotic patient that potentially may not be a transplant candidate with the advent of AKI, and really a lot of this comes down to some of the older literature on hepatorenal syndrome. We do know that in the absence of liver transplant, based on this very nice study from the Pergenesis Group in Spain, that in comparison with other causes of AKI, that in the absence of liver transplant, we know that hepatorenal syndrome has the worst outcome. But some of the things that are also more important when we look at other etiologies of AKI is that it also has an impact on outcome after transplant, and how we manage these patients also this becomes important. We know that if you look at significant ATN or irreversible AKI, that this not only portends a higher probability of developing CKD post-transplant, but also as showed by Mitra Nadim in this very nice study, that it actually portends a worse outcome with regards to all-cause mortality. So when we talk about therapies, and a lot of these are, particularly in North America where we don't have turlopressin at the moment, a lot of these are based around bringing the patient into the intensive care unit for vasoconstricting therapy. It still goes back to the basic idea that we're trying to increase the effective arterial circulating blood volume, increasing the mean arterial pressure. Often we'll try to target a higher mean arterial pressure of above 75 with the hope to perfuse the kidneys. In terms of the evidence for vasoconstricting therapies, we're all aware of the original Sanyal study that demonstrated that with the use of turlopressin in its bolus form along with albumin compared to albumin alone, that there was a higher rate of HRS reversal. And we know that following these studies that turlopressin has been used widely in Europe. However, one of the challenges in North America was this challenge around the mortality data, which really was an oversight because we realized that a lot of these patients in the absence of liver transplant, it's very difficult to implement or have a mortality benefit with a single therapy. And what was probably lost in this study is the fact that having a patient with significant AKI going into liver transplant becomes a very difficult patient to manage post-liver transplant. However, as also kind of mentioned in the previous study, we know that response to turlopressin is somewhat prognostic. And this follow-up study by Boyer and colleagues in the reverse trial were able to demonstrate that those patients that did actually respond to turlopressin had a better outcome. So if you don't have turlopressin available, what about other alternatives? This was a meta-analysis that was published back in 2014 that showed at least on the studies that were included that norepinephrine is probably non-inferior. So currently in North America, we don't have access to turlopressin. We're often bringing patients into the unit, intensive care unit, and putting them on norepinephrine and along with albumin and maintaining a mean arterial pressure of over 75, which is probably reasonable. It is interesting to note as well that you can also use turlopressin as an infusion. And Professor Angeli and his group published this study in 2016 where they demonstrated that the one benefit of giving turlopressin as an infusion rather in a bolus form is that the overall dose can in fact be lower. So you're exposing the patient to potential lower toxicity possibly or ischemia. There was one other study that came out in 2019 from India that was an opal label study that I just want to highlight briefly that did show that there may be a potential inherent benefit of turlopressin over norepinephrine. But I would highlight that more studies are needed. Where things also continue to be controversial is when to start renal replacement therapy. We don't really have a lot of data on the timing of initiation of renal replacement therapy in ACLF or in the cirrhotic patient. And even if you look at the broader critical care literature, there are a variety of different trials and studies that have looked at the optimal timing of renal replacement therapy and AKI in the general critical care population. And I just highlight one here, the Zarbock study that was in JAMA that demonstrated that there may be a potential benefit if you start patients with stage 2 AKI versus stage 3. When you look at the modality, you've got hemodialysis, you've got hemofiltration, there are combined modes, there are intermittent modes and continuous modes. We still don't have a lot of data with regards to if any has an advantage. From a pragmatic point of view, we tend to use mostly continuous veno-venous hemofiltration partly because it's simple. You have a lower requirement for running any kind of circuit anticoagulation. So often, you can run it with normal saline. You don't have to run any heparin or citrate. Anticoagulation does become a problem because of the concern of potentially coagulopathy and bleeding. And furthermore, we know that patients with liver dysfunction do not metabolize regional anticoagulation with citrate. And I think it's a discussion for another day, but in terms of who do we... In terms of the role of renal support in patients that aren't candidates for liver transplant, this is a very challenging topic. What I would say is that really you have to consider the overall phenotype. So a younger patient with more preserved synthetic hepatic function is a very different patient than somebody who's 60 years old that's profoundly sarcopenic, I would put it that way. So taking it a step further, looking at extracorporeal liver support, and I just want to highlight that probably the most widely studied therapy is MARS, that really there is no good data for this. This was the relief trial that was essentially a negative study. So really at this point, for the patient with ACLF and AKI, the primary modality of management, particularly in the hemodynamically unstable patient, if they fail vasoconstricting therapy is CVVH. So then the question becomes the decision that you have a patient with AKI, that now you're trying to get them through a liver transplant, and we also know that there are a lot of challenges with transplantation, particularly with regards to you're taking an organ that was cryopreserved, that you're going to get a potassium bolus, that there are issues with fluid overload. So one of the questions that Sean Bagshaw was the lead of this pilot study at our center was look at the feasibility of interoperative continuous renal replacement therapy during liver transplant. There is a lot of retrospective uncontrolled data in the literature, but we felt that it behooved us to do a pilot study to look at safety, feasibility, and some physiological endpoints. So just to highlight very briefly, this was a pilot open-label randomized trial that was run between 2012 and 2015. You had to have AKI at least stage one in a MELD of greater than 25. Patients were excluded if they were a live donor liver transplant or they had urgent requirements for dialysis, as you can understand. The standard orders we used was continuous veno-venous hemofiltration. We also used a zero potassium bath, and the reason for this we know is because when you unclamp the aorta post-perfusion, you're going to get this large bolus of potassium. And just to kind of highlight that this study was a cautionary tale, out of 244 patients listed for transplant, there were 60 patients that were consented of which there were patients that died on the wait list or in some cases died waiting for a... If or went to the OR and the graft was unsuitable. We ended up having 32 patients that were actually randomized and went to the OR. The other thing to highlight is that there was actually a large number of crossovers. So of patients randomized to standard of care, there were seven patients that ended up receiving intraoperative CRRT for things like volume overload, et cetera. When you look at the overall outcomes, there were no statistically significant differences in transplant and survival outcomes or in renal recovery. This is as an intention to treat, but as I mentioned before, there were a large number of crossovers. If you look at in a protocol form, it was essentially fairly similar. There was no differences in renal recovery and survival. So in terms of the findings or cautionary tale, while it is feasible, we were able to randomize 53% of eligible patients and we were able to deliver this prescription safely. It's a very challenging thing to do. One of the things that I would highlight is that effectively, if you were starting patients potentially earlier on renal replacement therapy in the ICU setting, you may be able to get somebody through a transplant without necessarily having to do it in the operating theater, which may be easier for your anesthesia group. Finally, I just do want to mention, is there anything that's inherently more beneficial of continuous modes versus intermittent modes? We tend to use the continuous modes more commonly in patients that are hemodynamically unstable. We know that, unfortunately, this does mean that it's more expensive. You're running somebody on a circuit 24 hours a day. Really the only place where we see this more obviously in the literature is actually in acute liver failure and I'm just going to conclude with two quick slides. This was a study that we did from the U.S. Acute Liver Failure Study Group where we were able to demonstrate when we looked at ammonia kinetics that CRRT had a more significant impact on lowering arterial ammonia and we know in the acute liver failure population there's a much stronger correlation with the development of intracranial hypertension and ammonia. The other finding that we found that was interesting was that in patients that are much higher for neurological risks of neurological complications, that intermittent dialysis, which was done in an earlier era, was actually associated with a higher all-cause mortality after adjusting for severity of illness where CRRT was actually protective. Where I can tell you certainly in acute liver failure, the modality of choice is very clearly for CRRT in the acute on chronic liver failure patient, although some of these patients are at a higher neurological risk, there continues to be equipoise. So in summary, ladies and gentlemen, approximately 40% of patients with ACLF patients admitted to the ICU have evidence of improvement with support and the AKI stage based on the K-DIGO classification discriminates ACLF patients with different prognoses. The mainstay of management of HRS-type AKI is vasoconstrictant therapy. If terlopresin is not available, norepinephrine is a reasonable alternative. There is equipoise regarding the timing of initiation of renal replacement therapy, similar to other critically ill population, and there continues to be equipoise regarding the use of intraoperative renal replacement therapy during liver transplant. Thank you very much. Thank you.

ICU management

AKI

critically ill ACLF patients

renal replacement therapy

CRRT