Thank you, Dr. Begins. The last talk of this module will be presented by the professor Dean Carvelas about critical care for acute and chronic liver failure. Dr. Carvelas is a professor of medicine, critical care medicine and gastroenterology and hepatology at the University of Albert. Thank you, doctor. Chairs, ladies and gentlemen, I want to thank again the organizing committee for this wonderful opportunity to speak to you today about the critical care management of acute and chronic liver failure and the patient you're considering for liver transplant. So as an outline for what I'll discuss today, I'll review the indications for ICU transfer for patients with cirrhosis in ACLF, describe volume status assessment and management in ACLF patients with sepsis and other forms of shock, review respiratory failure and ventilatory strategies in ACLF, review renal replacement therapy strategies in ACLF and finally understand prognosis with and without liver transplant. So we know that acute and chronic liver failure is a multi-organ dysfunction syndrome. It occurs with a diverse set of precipitating events and tends to follow a predictable pathway irrespective of the inciting event. And what I do want to mention is that for some of these organ failures, we have ACLF or cirrhosis specific data that guides our management. But for some of these other organ failures, we rely on the general critical care literature because there is a lack of data in the specific cirrhosis or ACLF population. So just to remind us that this was the canonic study that Richard Moreau and colleagues published back in 2013. And this was one of the first studies to kind of identify the use of multi-organ failure scores which in some sense is a variation of the original ICU SOFA score that looked at discriminating patients with different prognoses with different numbers of organ failure. And as you can see here, patients with three or more organ failures that met the CLIF definition for ACLF had a 90-day mortality of almost 80%. So just to highlight, there are different definitions of ACLF across the world. And part of the reasons for this is the etiology of cirrhosis and chronic liver disease is different in different parts of the world. So in Asia, we know that hepatitis B is very prevalent. In European and North American populations, we tend to see more alcohol-induced liver disease and MASL. So as such, there are slightly different definitions. So the question is, when would one consult your ICU colleagues to potentially bring a patient to the ICU? Primarily, it's for these reasons. It's for circulatory support, respiratory support, hematologic support, for example, in the setting of a variceal bleed, and renal support in the setting of acute kidney injury. So I'll talk firstly about circulatory failure and shock. And I note that the first two speakers did a lovely job of elucidating the circulatory abnormalities of cirrhosis and the hyperdynamic circulation. And really, where these become a concern to us as critical care providers is when patients do develop relative arterial hypovolemia due to this hyperdynamic circulation, which in one sense leads to upregulation of the secondary neurohormonal pathways, such as the renin-angiotensin system and the sympathetic nervous system, which can lead to hyponatremia, which generally is an early warning sign for HRS-type AKI. But as mentioned before, because of this decrease in systemic vascular resistance, often this can mask significant underlying cardiac issues, of which cirrhotic cardiomyopathy is just one of them. So we heard in Dr. Biggin's talk as well that volume overload can be associated with deleterious outcomes in cirrhosis patients. And I just highlight the cautionary tale of the Attire study, where these were patients in Britain that were randomized either to albumin transfusions in the hospital to a target of an albumin level of greater than 30 versus standard of care. And we've now seen this tale over and over again, where albumin infusions weren't associated in the prophylactic setting with improved outcomes, and in fact were more associated with respiratory complications and specifically pulmonary edema. And we've also seen this in the setting of the recent confirmed study with terlopresin, where one of the potential reasons for respiratory complications with terlopresin might have been the amount of albumin patients were given prior to entering the study. So there are several different methods to assess volume status, and it is important. One of the benefits that we have in the ICU setting is we have, you know, more devices with probably the most widely used one now being point-of-care ultrasound. And the benefit of this is that it's dynamic, and we can get information with regards to both left ventricular and right ventricular function. We can look at pulmonary, indirectly at least at pulmonary pressures or right ventricular systolic pressures. We can also look at fluid responsiveness and assess filling by looking at the inferior vena cava diameter and whether or not it's compressible. Many centers are still using CVP monitoring, but as mentioned, there are some limitations that are related to this. And because of the widespread use of point-of-care ultrasound or bedside echocardiography, some of the more, the thermodilution techniques have not been as widely adopted. So in terms of kind of the basics in terms of management of shock and ACLF, perform early baseline assessment of volume and perfusion status and cardiovascular function. As mentioned previously, balanced crystalloid solutions like plasmolyte and Ringer's lactate are preferable to avoid hyperchloremic acidosis and primarily use albumin if there's evidence to support this, classically being SBP, HRS-AKI, and large volume paracentesis, but not as a primary resuscitation fluid. In choosing a vasoconstricting agent, norepinephrine remains our first-line agent, but certainly we consider terlopressin or vasopressin in the setting of HRS-AKI. And while in the general critical care population, the target mean arterial pressure is 65, we'll target a higher mean arterial pressure in the setting of HRS-AKI where you're hoping for some potential reversibility. And it should be noted that relative adrenal insufficiency is common in cirrhosis and ACLF, so always consider the possibility of supplementing with hydrocortisone. We've learned more today, particularly in the previous session, that cirrhosis patients are at high risk of developing bacterial infection through translocation. We know that there's a loss of the tight gap junctions, which leads to bacterial translocation. And because of these portosystemic shunts, you get microorganisms that bypass the reticular endothelial system and can end up seeding sterile spaces, such as the peritoneum developing SBP, and also the bloodstream developing bacteremia. And while there's been a few changes with regards to some of the published recommendations on surviving sepsis, we used to talk about the sepsis six-pack. We're unfortunately left now with early appropriate antimicrobial therapy and resuscitation to perfusion targets. But what we do know is certainly, based on this sub-study of the CADS database, that each hour of delay to early appropriate antimicrobial therapy in patients with SBP and septic shock increases your odds of death by almost 80%. So talking about respiratory failure and acute lung injury, as Dr. Fallon went through in eloquent detail, there are causes of respiratory failure that are specific to cirrhosis, such as hydrothorax, hepatopulmonary syndrome, alpha-1 antitrypsin, and portopulmonary hypertension. Then there are also nonspecific etiologies, such as pneumonia, thromboembolic disease, volume overload, and atelectasis. Once again, this is borrowed literature from the general critical care literature, where we advocate for a lung protective strategy. And what this generally means is a low tidal volume with a high positive end expiratory pressure. And the purpose of this is to prevent atelectasis, and at the same time, with a low driving pressure to prevent volume trauma and barrel trauma. One of the challenges with this type of a strategy in an ACLF patient is that they tend to be more fluid dependent. So if you go too high on the positive end expiratory pressure, you can decrease venous return to the right heart, which can lead to hypotension. So that's really the only caveat. When looking at patients with altered mental status, as this is another reason why patients might end up intubated for airway protection, as Dr. Bajaj talked about before, certainly on initial presentation, you need to consider other causes besides hepatic encephalopathy. And I just kind of highlight a few of them here. But one of the things to remember, though, is that generally if somebody's had repeat episodes of HE, certainly in the same admission, you don't need to kind of go through everything, including the imaging to rule out structural causes, as if you've already clearly identified this as hepatic encephalopathy. In terms of what are some of the specific differences between the ward and the ICU, generally there's a lot of the same themes. We will often transfer patients to the ICU if they've got a coma grade of 3 or greater, or a Glasgow coma score of less than 9, because this is generally an indication to intubate somebody for airway protection. We will work them up on the first case for other causes of delirium, and certainly if there's a lack of response to hepatic encephalopathy therapy. One thing I will mention is that we do tend to see a lot of patients, a lot of ACLF patients in the ICU with abdominal distension, and there's always a concern of abdominal compartment syndrome. And this is one of the reasons why we sometimes will substitute polyethylene glycol as opposed to lactulose, targeting the same 2-4 loose bowel movements per day. And as you can see from the study from 2014, it showed that there appeared to be efficacy with using PEG in treating HE. We try to use medications or sedating medications with short half-lives. We like propofol and dexmedetomidine, and certainly we avoid drugs like benzodiazepines. And a further thing, we know that ammonia is certainly a neurotoxin, but generally we kind of advise against the use of serial ammonia testing once you've identified somebody as having HE. So I want to talk in a little bit more detail about renal replacement therapy. As Dr. Biggins mentioned, there were the International Club of Societies recommendations, and then recent harmonizations with some of the adkeys with the K-Digo classification. So one of the good things now is generally when we're talking about different stages of acute kidney injury, what people are talking about in the critical care literature and the hepatology literature, people are now talking the same language. Renal replacement therapy is a bit challenging, and we do know about the recent START-AKI study, which was a general critical care study that did not demonstrate a benefit to accelerated renal replacement therapy versus standard initiation of renal replacement therapy. And one of the interesting things that did come out of the general data of this study is that there was potentially a benefit to the initiation of continuous renal replacement therapy in the critically ill as opposed to intermittent hemodialysis. And this likely has significant relevance in ACLF because a lot of our patients are hemodynamically unstable and would likely tolerate CRRT better than intermittent hemodialysis. So if you are going to start renal replacement therapy, this is certainly something to consider. As mentioned, timing is still an area of equipoise. Potentially where you might consider starting early is somebody that is listed for transplant that you are trying to optimize metabolically prior to going to the OR for a procedure, either bringing their potassium down or drying them out. Other practice-related factors such as circuit anticoagulation, we generally start certainly with patients on CRRT to run them without anticoagulation. And often what you can do is run in hemofiltration the replacement fluid pre-filter. So you are effectively diluting the blood to prevent clotting. In the setting where you have a hypercoagulable cirrhosis patient or ACLF patient, then you can potentially use citrate, a regional anticoagulation over the short term, which appears to be safe, but generally you would not be doing it for several weeks. And in those patients that that's not an option, you could also consider something like prostacycline. Taking it a step further to looking at other liver support devices, and as you may all be aware, there hasn't been a study to date that's shown a mortality or a spontaneous survival benefit with the use of a liver support device. But I think one of the challenges clear here is it's actually the end point of these studies. And the primary end point of a study looking at an extracorporeal liver support device should primarily be looking at the bridge to transplant because we know that these patients have a natural mortality that's very high in the absence of transplant. So just talking briefly about prognosis, I find often when I speak to my ICU colleagues that it can be fairly pessimistic to talk about ACLF. But one thing I do want to mention is that there is a subset of patients with ACLF that have some reversibility. And in the study that we did of both North American and European data back in 2018, we found that almost 40% of patients with ACLF improved by at least one organ failure. And certainly that can have a significant impact either potentially if you're looking to bridge somebody to transplant or somebody that you're essentially trying to buy time if there is something in their history that you could reverse that might make them a transplant candidate. And certainly in the setting of the patient on the transplant list, I just want to highlight this study that the late Vinay Sundaram published that demonstrated that patients that came in with grade three ACLF or three or more organ failures, that following support in the ICU improved by at least one organ failure, their natural history after transplant was actually non-inferior to those patients that came into the ICU at the beginning without ACLF3. So in summary, acute on chronic liver failure is cirrhosis with multi-organ failure. With regards to the management of sepsis, early appropriate antimicrobial therapy improves outcome. With regards to circulatory failure, perform early baseline assessment of volume and perfusion status with dynamic modalities, including point of care ultrasound, and avoid fluid overload. In the setting of respiratory failure, employ a lung protective strategy, but be cautious of the use of high positive end expiratory pressure. And in renal failure, continuous modes of renal replacement therapy are preferable to intermittent modes such as intermittent hemodialysis. And other factors such as timing of initiation of renal replacement therapy and circuit or anticoagulation warrant further study. And finally, I just want to highlight that after a great deal of work by our co-members and by my co-chair, Dr. S. Rennie, who's sitting beside me, the ASLD clinical practice guideline on acute on chronic liver failure and the management of critically ill patients with cirrhosis finally came out this week. So thank you very much. Thank you everybody for outstanding talks. I'll start with some questions both that were posed online and then also if folks can make it to several of the microphones that are throughout the room. So one question to Dr. Iwakiri, a question for you is about lymphatic drainage. Do you think it changes in the setting of heart failure? So lymphatic drainage in the liver, does heart failure cause any issues with that? That's a very interesting question. I don't have any answer. This is an unexplored area of the research. Definitely, there are some possibility of some role of lymphatics in this heart failure. But I think this is something need to be carefully explored. But it's an interesting area to be exploring in the future. So whoever asked that in the audience, that's a research project with Dr. Iwakiri for you. And then the second question also to you was, does the disruption of the lymphatic system predispose to infection? You talked about ascites. But what about infection? Yes, infection plays an important role as well, too. I think this is also another new area. And it's not been known, actually. But infection definitely may cause the lymphatic endothelial dysfunction. And definitely, it could also lead to some thrombosis within the lymphatic vascular system, and also some inflammation within the lymphatic endothelial cell. Again, this is not yet being explored. But it's a great possibility. Andres? Hi. Thank you, everybody. Like beautiful talks, this is a great session. So one question that is mostly directed to Dr. Biggins, but I will appreciate, actually, everybody chiming in. So in the care of post-AKI, once your patient recovers from AKI, regardless of the etiology, what do you do with the diuretics? And let's assume the patient has some fluid overload. Do you send the patient home off diuretics? Or do you do half dose? What do you do? What do you do? Yeah, I don't know if I have the right answer. I think reducing the diuretics to make sure they don't have an immediate post-hospitalization AKI that bounces them right back in. I think if you have a severe injury, it is very important to include nephrology. And they should be seeing us in the hepatology. They should be seeing us. They need very close lab follow-ups. And a really medical home type of approach to them. Microphone right here. Hello, I'm Luis Rosales from Mexico. I have a question for Dr. Corvallis regarding, or rather, whether he could speak a little bit of the role of interleukin-6 in the pathophysiology of acute and chronic liver failure, and whether he could consider it a potential therapeutic target. A recent research article published last year showed that great acidic concentrations of IL-6 greater than 4,200 correlated with a greater mortality. Therefore, maybe IL-6 inhibitors could be a potential therapeutic window or target. So I think that's a very good question. And as you are probably aware, during the COVID pandemic, we got very interested in IL-6 with tocolizumab and other anti-IL-6 inhibitors. I think being above board and honest, the challenge is that ACLF presents with multiple different phenotypes. And it's a multi-organ, multi-mediator condition. And I think the challenge is that inhibiting a single pathway likely would probably not give us a robust therapy, unless it was for specific presentations or specific organ failures. And I guess that's my main concern with a directed therapy like that. Thank you. Microphone over there. Mohamed El Fiki, Sioux Falls, South Dakota. My question for Dr. Biggins, would you please elaborate more on how tips may improve GFR that you mentioned in your talk? Because as far as I know, we need a good kidney function for tips to kick in and help with portal hypertension. I didn't get the full question. Is it related to the improvement of GFR with tips or the improvement in GFR with paracentesis? With tips. With tips. Yeah. So I think the application of tips for hepatorenal syndrome is increasingly of interest. If the portal hypertension improves and the ascites, the tense ascites, improves, I think we can see an improvement in GFR. Thank you. Let me follow up with Dr. Fallon. A quick question, then you. Same thing with tips. Could you comment on tips and implications in patients with cirrhotic cardiomyopathy? OK. Yeah, so as you know, acutely after tips, the hyperdynamic circulation worsens. It doesn't improve. So you increase finished return to the right side of the heart. So tips is likely, as I showed in the Toulouse study, probably 13%, 15%, 18% of people, depending on how you choose them, will have cardiac dysfunction after tips. The long-term effect of tips on cardiac function is not clear. But I don't think tips is probably the way to go for cirrhotic cardiomyopathy. I think it's probably a risk factor rather than a treatment. Right, as in terms of should we be screening for that prior to tips? Absolutely. I think we should. I mean, everybody should get an echo before tips. I think when you have the luxury and it's not acute varicose bleeding. So I think that's very important. And outside of the Toulouse study, there are not any great studies in the acute setting. But I think in the chronic setting, we should be looking for particularly the diastolic markers, because those are the ones that probably predict the risk. If you have systolic dysfunction, you're probably not going to do the tips. OK, question there? Yeah, I'm Dr. Virender Singh from Chandigarh, India. This is regarding that post-recovery AKI. If diuretics are not the cause, and we have treated the etiology, we have treated, suppose it is because the infection, then we will definitely put these patients after the discharge on low dose of diuretic, maybe 5 and 50, and then monitor closely. OK. Hello, Sabata from Chile. I would like to ask Dr. Fallon about carbidolol in the complications, in terms of we're using many patients after the predatory study carbidolol. What are the limits on patients on myocardiopathy? And what are the effects on the pulmonary patients with hepato or portopulmonary syndrome? Yeah. So good questions. So yeah, the question becomes really where in the sequence of events would you want to use carbidolol. I think the study I threw out there from Shiv's group is really an interesting one, in that if you have diastolic dysfunction, it may be that carbidolol actually improves that. And that may be one of the mechanisms through which it may actually have its beneficial effects that aren't being realized. I think in those with systolic dysfunction or severe diastolic dysfunction, it's probably a higher risk. I don't know of any data that tells us exactly when to use it, but I would not use it in systolic dysfunction, low EF or low strain. But I think in the others, we need to know better. But we are probably already using it in people with diastolic dysfunction. We just haven't checked. And so it would behoove us, I think, to check. In terms of hepatopulmonary syndrome, there's not any good data around whether beta blockers are effective or not effective in that scenario. The only data that's really out there is it looks like tips or decompression of the portal system using that is probably not, in people that need a tips other than hepatopulmonary syndrome, it's reasonable to do a tips. And those that have hepatopulmonary syndrome, doing a tips doesn't seem to make it better. So I don't know that it makes a difference there. And then in terms of pulmonary hypertension, obviously, in the general pulmonary hypertension literature, particularly with moderate or severe pulmonary arterial hypertension, the concern about beta blockers is decreasing RV function as well as LV function. So they're generally contraindicated. Again, I don't know of any specific data that says when exactly you need to stop it in the setting of portal pulmonary hypertension. But I think most people would hold beta blockers in that setting, use band ligation if needed for a varices, and then treat the pulmonary hypertension before thinking about restarting. Then question for Dean. Just to comment on POCUS, one is, is the use of POCUS different in cirrhosis patients as compared to the non-cirrhosis patients that you see in the ICU? So that's a good question. Generally not. I do take Scott's point that I agree there is inter-observer variability. And there's a difference between doing a structured course over an extended period of time, which we've actually done at our institution, versus people going to a conference and doing a two-day course in POCUS. I think really the focus is, the second thing is the number of scans that you have done, where generally, certainly if you've done more than 100 scans, then you are likely going to be getting very reliable or reproducible results that can be confirmed by others. And I understand that dabbling in things always has a risk. The one benefit, I think, is that actually cirrhosis is one of the populations where we can probably take advantage of this more, just because it's so difficult to assess volume status in somebody that can have all this third-space fluid. And furthermore, even in terms of not just cardiac ultrasound, but chest ultrasound, looking for polio effusions, abdominal ultrasound for ascites, that I see a lot of advantages with POCUS in cirrhosis patients. And a question up there. Ashutosh Barve from University of Louisville. This is a question for Dr. Fallon. What do we know about the reversibility of the cirrhotic cardiomyopathy after liver transplantation? That would have implications for post-transplant survival. Yeah. Well, I don't know very much about the reversibility. We do know that there's concern about cirrhotic cardiomyopathy and cardiac dysfunction post-transplant, particularly in those with evidence of diastolic dysfunction. Most of the people with systolic dysfunction don't want to get transplant. So I think there's a long-term concern about cardiac dysfunction becoming evident. And so, again, it's not clear to me that there's clear-cut guidelines about when not to do it or when to do it. And I think it really behooves this question of, how are we going to use this new data to specifically characterize a broader array of people so we can make good decisions? But it's clear that transplant's not a cure reliably for cirrhotic cardiomyopathy. Thanks. And then a question for Scott, two related questions. We went through an entire session, but octreotide and midrin was not brought up. That's one. And the second thing is, if we're not saying the O&M word, if patients start getting placed on terlipressin, your thoughts on freezing the MELD score? Or you can use this. I think this one's working now. So I'll tackle the second one first. And in my role as the chair of the liver and intestine committee, we've been actually looking at this very closely. There is clearly a detriment to individuals who are on the transplant list who have an improvement in their numbers, but not an improvement in their mortality. And one consideration that we've looked at is not freezing the creatinine, but just lengthening the intervals in which the MELD update is required. So particularly the higher MELDs, over 25, it's an update every seven days. Should we just increase that to 14 days? We're learning from what's happening in Europe and that octreotide mineral. So we've been doing this for a while. It's been something that we've done on the way to having terlipressin available, I think. Does it really help? In my experience, it's uncommon. OK. Did you have any questions? OK. And then I'll, yeah, Dr. Fallon, question. So I wanted to ask Scott a question, because we have been battling our pharmacy relentlessly about terlipressin and being able to use it because of the cost. They keep saying it's going to bankrupt the whole hospital. And so the question we've had is, should we spend the first 12 or 15 or 16 hours actually checking the creatinine with albumin? Because it should happen quickly. I don't know why we picked 24 and 48 hours, quite frankly. And then the other question is, can we do a day? We're getting asked to do a day of octreotide, mitadrine, and albumin to see if there's improvement before we go to terlipressin, perfect just from a cost perspective. Does that seem totally crazy? I'll say the title of my talk was something beyond terlipressin. But the cost is really important here. And my prior institution, the University of Washington, we had a very difficult time getting in. And we used a lot of norepinephrine. And the Stanford group has had a study that came out about using norepinephrine in a step-down unit. So I think if you're running into cost concerns, norepinephrine would be something. We've also used, at my prior institution, we used oral droxedopa. So I'll just put that out there and found some reasonable effects there. And then one last question to end with Dr. Fallon is an interesting question. So how do we balance the risk of progressive pulmonary remodeling in patients that have low MELD and poor pulmonary hypertension? If we're deferring transplant, are we just not risking irreversibility of the lungs? Yeah. Another really good question. So there's a lot of questions there. The first question is, only half the people get better after transplant anyway, right? So it's not clear that everybody gets better. At four years, 50% of people are off drug. So whatever is driving the remodeling isn't reliably resolved by transplant. The second issue is we already mandate that people have a response, right? So that response may just be tone-related. But in other forms, it appears to have an impact on remodeling. So we're already asking for remodeling to be there before we will transplant them. So I do worry about disadvantaging people. But I think our criteria already sort of address that. But it's a really good question. I don't know the answer to it. All right. With that, we'll take a 20-minute break. And then we'll be here for the final session at 4 o'clock. Thank you again.

critical care management

liver failure

ICU transfer

renal replacement therapy

liver transplant

cirrhotic cardiomyopathy

terlipressin