Actually, before I invite him in, I have no financial disclosures. I would also like to thank the AASLD and Naspian for supporting this symposium. So outline for today's talk, we're going to go over definitions of acute on chronic liver failure, review the current understanding of drivers and consequences in acute on chronic liver failure, and then finish up with some of the emerging data in the pediatric literature. So to begin with, definitions. The Asia-Pacific Association for the Study of the Liver set up a working party on acute on chronic liver disease in 2004. They convened a two-day meeting in 2008 with the goal to come up with some consensus statements and guidelines. These were published in 2009, really for the first time introducing consensus definitions for acute on chronic liver failure. The APASL defined ACLF as a previously diagnosed or undiagnosed liver disease affected by a acute hepatic insult manifesting as jaundice or coagulopathy, complicated within four weeks by ascites and or encephalopathy, essentially describing a clinical scenario where one progresses from compensated cirrhosis to a state of decompensated cirrhosis that occurs quickly and as the result of an acute insult. Notably, the association of acute liver failure with high short-term mortality was absent from the 2009 definition. And so in 2014 when they revisited it, the definition was expanded to include this. Notably, this group reconvened in 2019 with the published recommendations. They revisited their definition and decided to leave it unchanged. I think importantly, however, when... Oh. Okay. Importantly, I think when they reconvened in this most recent publication, the authors did look for the first time to define acute on chronic liver failure in children. I'm not going to read the entire recommendations here, but I really just want to highlight these efforts for the first time to define acute on chronic liver failure in the pediatric population. So when looking at acute on chronic liver failure, it's really understood that there are some differing viewpoints and mainly this is in the fact that key terms of acute on chronic liver failure such as acute and chronic and failure have several variations in definitions. This was really underscored by a 2013 review from the group in the Netherlands where they established that at the time there were 13 working definitions of acute on chronic liver failure. I think more recently really two distinct definitions have emerged. The first is the APASL definitions, which we have previously reviewed, and then secondly is the ESIL-AASLD kind of combined definition. And they define acute on chronic liver failure as an acute deterioration of pre-existing chronic liver disease, usually related to a precipitating event associated with increased mortality at three months due to multi-system organ failure and combined with a modified SOFA or sequential organ failure assessment scale to reflect multi-organ failure in the setting of ACLF. Notably there are some key differences between these two definitions, including the duration between insult and the development of ACLF, the duration in which there is higher mortality, the definition itself of chronic, the inclusion of sepsis as an inciting event, and the requirement of extra hepatic organ failure in the definition. With these differences in mind, the group from Philadelphia recently looked to compare incidence and mortality of ACLF using both these definitions. In their study, you know, when applying the different ACLF definitions to over 80,000 patients with cirrhosis in the Veterans Health Administration Registry, they discovered that within their cohort there were 783 who developed ECL and APACL ACLF, almost 4,300 who developed ECL alone ACLF, and 574 who developed APACL alone ACLF. As reflected by these initial findings, the authors reported a higher probability of developing ACLF or developing an ACLF event over time using the ECL definition. And within those, you know, it was notable that those who carried a diagnosis of alcoholic liver disease or a concurrent diagnosis of alcoholic liver disease and hepatitis C were significantly more likely to develop an ACLF event regardless of which definition was used. When looking at biochemical profiles, it was noted that patients with ECL ACLF alone had the lowest AST, ALT, total bilirubin, alkaline phosphatase, and INR levels among all groups. And the authors reported that overall, patients who developed ECL alone ACLF were more similar to those who did not develop ACLF in terms of age, baseline cirrhosis, severity, and comorbidities. Patients who met diagnostic criteria for both ACLF definitions had the highest liver enzymes, total bilirubin, and INR. When assessing mortality, ACLF survival over time was lower using the APACL versus the ECL definitions. And so collectively and somewhat provocatively, the authors state in the discussion that simply put, the ECL and APACL criteria do not measure the same entity. Now this dissimilarity is not new. It's been recognized for some time. Back in 2014, the World Gastroenterology Working Party attempted to resolve some of the differences and published an outline of proposed... or published an outline of a proposed better approach to defining ACLF. However, the proposed unifying definition, which is listed, really did little more than just combine some key aspects of both definitions. And this has not really been rigorously studied in the literature, this definition. I think somewhat importantly though, the authors underscore the thinking that ACLF really wasn't a single disease, but rather a syndrome, which can occur irrespective of the inciting event or underlying etiology. Here, Type B and C ACLF represent the more traditional progression whereby compensated or decompensated cirrhotics experience a precipitating event that leads to hepatic multi-organ failure. This is also expertly reviewed in This Nature Reviews from 2016. I think uniquely, the group also defined the concept of Type A ACLF as a type of liver failure that may be seen in patients with non-cirrhotic chronic liver disease. But with an acute flare resulting in liver failure and that this Type A can be indistinguishable from the clinical presentation of acute liver failure or subacute liver failure. Such patients include those with a reactivation of hepatitis B, autoimmune hepatitis, or other chronic liver disease such as fatty liver with a superimposed drug-induced liver injury. So now that we've discussed the definitions of ACLF, we'll move on to discuss some of the drivers in ACLF. Systemic information really is the hallmark of ACLF, the cause of which is likely multifactorial but really increasingly understood. So when patients with liver disease progress to cirrhosis, they develop well-recognized complications such as portal hypertension, which leads to intestinal dysbiosis, loss of barrier gut function, and bacterial translocation. Now this is also not a new concept, and this is a different concept and this is also not a new concept, and this is demonstrated by this 2005 concise review in hepatology. But what is better understood is how the role of these translocated bacteria or what role they play in exaggerating some of the inflammation and the responses in ACLF and that these bacteria will trigger these pathogen and danger or damage-associated molecular patterns which drive some of this inflammation. And so here the authors from the ESIL Chronic Liver Failure Consortium or ESIL-CLIF present a table demonstrating examples of pattern recognition receptors and their ligands, many of which are bacterial and thought to trigger events which exacerbate systemic inflammation in ACLF. Translocated bacteria may act locally at the lamina propria, increasing the potential for further translocation, and as noted, these bacteria can also contribute to the development of systemic inflammation through various mechanisms. I think importantly, it has been shown that the inflammation in ACLF is due to more than just the presence of underlying liver disease, and in this hepatology paper from that same ESIL-CLIF group, significant upregulation of pro-inflammatory markers were found in patients with ACLF, even when compared to a separate cohort with just chronic liver disease and cirrhosis but without ACLF. So this systemic inflammation can aggravate the underlying liver dysfunction, but also it contributes to microcirculatory dysfunction, mitochondrial and cell dysfunction, as well as cell death. As a consequence from this process, the release of damage-associated molecular patterns and other signals drives further systemic inflammation. This can affect the larger circulatory system and lead to end-organ hyperperfusion. Collectively, circulatory compromise results in end-organ damage and acute decompensation, which again, further compromises circulatory function and promotes end-organ failure. So where are the breaks? You know, lots of green arrows, plus signs, positive feedback. There aren't any apparent countermechanisms noted, and this is because along with this systemic inflammation, there is an additional distinctive feature in ACLF that is immune dysregulation. So to understand this, it's important to look at what happens when a relatively healthy individual undergoes a bacterial challenge. So here, the antigen recognition in red and immune cell activation in green are tightly coordinated to mount an effective immune response against a pathogen. In the setting of cirrhosis, with progression from compensated to decompensated and then to further decompensated disease, such as an acute on chronic liver failure, there's an associated pro-inflammatory phenotype, as we've shown in the previous slides. In addition, key elements of the immune response are impaired in patients with cirrhosis, including antigen presentation capacity of monocytes and decreased phagocytic function of macrophages. That is pivotal for antibacterial immune defense, demonstrated here by the red line. This concept was expertly presented by Martín Mateos in this recent review, and the authors present the concept of immune cell paralysis. So here, as the ACLF progresses, an initial excessive compensatory inflammatory response is followed by the exhaustion of the immune effector cells, combined with some of the metabolic and neuroendocrine disturbances seen in ACLF, such as increased prostaglandin E2, hyperammonemia, and hyponatremia. And collectively, this leads to an overall immunodeficiency state. And so merging these two concepts, in ACLF we have progressive and iterative systemic inflammation that is occurring on the background of a relative immunodeficiency or immune dysfunction. And so what are the consequences? We've already established that ACLF is associated with high short-term mortality, but what contributes to this? Notable is the link between organ dysfunction and outcomes... the link between organ dysfunction and outcomes is well established. So here we see that ACLF with progressive organ failure can associate with mortality rates approaching 80% when three or more organs fail. So this concept has led to a classification system that's used where they grade it from 1 to 3 based on the number of organs that are failing. And more recently, it was shown that those with grade 3 ACLF have greater 14-day waitlist mortality than status 1A patients. And what about prognosis? Well various liver-specific scoring systems have been established, but they are limited in that they do not account for other organ failures. So to improve prognostic capabilities in ACLF, the ESL-CLIF group here looked to develop a better tool. The end result was the CLIF-C ACLF score. The authors were able to demonstrate that in the setting of ACLF, the CLIF-C ACLF score were superior to other prognostic scoring systems in predicting mortality and there's online calculators, like there are for everything, in calculating this for adults. Still, this CLIF-C score is somewhat complicated and convoluted to calculate. And so more recently, the North American Consortium for the Study of End-Stage Liver Disease or... So they looked to validate a more simple scoring system that had previously been established in 2014. Here they defined ACLF as cirrhosis with two or more extrahepatic organ failures. And these were very simply defined as shock, hepatic encephalopathy, grade 3 or grade 4. Renal failure defined as the need for dialysis. And respiratory failure defined as the need for mechanical ventilation. And the authors found that although infectious status remained an important predictor of death, the NAC-celled ACLF can act as a simple and reliable bedside tool to predict 30-day survival in both infected and uninfected patients with cirrhosis. So what about pediatrics? You know, we've already discussed the APASL recommendations. And to date, there really are only four publications that have looked at ACLF in children. So three of these studies come from India, one from the UK. The three Indian studies used the APASL definition where the UK study used the ESIL definition. The underlying etiology for the Indian studies was mainly Wilson's disease and autoimmune hepatitis, whereas in the UK study they really just exclusively looked at biliary atresia patients. You know, underlying disease flare and viral hepatitis were the precipitating events in the Indian studies, while GI bleeding contributed to ACLF development in the UK study. The overall mortality was high and the predictive factors were variable across these four studies. I think unpublished data from the Chicago group was presented at this meeting last year with a manuscript currently in process. And here the primary aim was to characterize and describe the clinical course and triggers and outcomes in children admitted with acute decompensated cirrhosis who develop ACLF. The authors used a modified NAC-celled definition of ACLF for pediatrics to retrospectively identify 27 patients with 31 episodes of ACLF. So no significant differences were seen between the ACLF and non-ACLF groups as it relates to the majority of patient characteristics. However, patients with ACLF were significantly less likely to carry an underlying diagnosis of biliary atresia compared to those with ACLF. The precipitating acute events that resulted in decompensation are listed. The main ones being fever and cholangitis, worsening ascites, and GI bleeding. Notably the numbers were small, but a worsening mental status was seen commonly in ACLF, but not in those who did not develop ACLF. Age at admission was not significant between the ACLF and non-ACLF cohort. And not surprisingly, ICU admissions and ICU length of stay was higher in those with ACLF. Hospital endpoints of transplant and death were higher in those with ACLF, while non-ACLF patients were more likely to be discharged from the hospital. Mirroring the data from adults, increasing the number of organ failures substantially increased mortality rates in those with ACLF, with respiratory failure being the most common extrahepatic organ failure to develop. Admission labs reflected a more severe hepatic injury from their precipitating event, as well as some multi-organ dysfunction with elevated creatinine. Infection was more commonly seen in those who developed ACLF and similar to the adult NAC-celled study, was predictive of ACLF development and poorer outcomes along with admission creatinine, white blood cell count, and AST. The conclusions from the study were that ACLF developed in 15% of hospitalizations for decompensated cirrhosis. Mortality was 26% and increased with extrahepatic organ failure. Children with biliary atresia were less likely to develop ACLF. ACLF was associated with longer length of stay, longer ICU length of stay and significantly different hospital endpoints. And admission creatinine, white blood cell count, AST, and having a positive blood culture predicted the development of ACLF. Future collaborative efforts are underway to better understand some of the systemic inflammatory cytokine profiles in these kids, as well as biomarkers that can predict outcomes. So key takeaway points. ACLF is a frequent syndrome in patients with cirrhosis characterized by acute decompensation, multi-organ failure, and high mortality. Drivers of ACLF development include systemic inflammation in the setting of immune system deficiency. Emerging data in pediatrics demonstrates different underlying disease precipitant factors, but similar morbidity and mortality. And more standardized definitions will enable improved understanding and drive future research efforts. Thank you. I'd like to invite Dr. Wainwright and Dr. Behrens back up to the podium. And we will invite questions for the group. So anyone who has a question, come forward to the microphones. And we'll get started. Yes? So by perfor negative, can you clarify, So, I don't think we had enough patients to... The vast majority of patients that we looked at were perforin-expressing CDAT cells. It was really the minority of indeterminate PALF patients that did not stain for perforin, and it might have been one or two cases, so I don't think it's enough to kind of make like a general statement about that population to be able to say that they are somehow unique or different. So, what I will say is I've had the opportunity recently to have one case where we had three different biopsies so that we were actually able to follow over time, and what was interesting is that in the first untreated biopsy, the CDAT cells were triple positive, CD8, CD103, perforin positive, and then that patient received steroids and some other medicines, and subsequent biopsies, the cells remained 103 positive, but the perforin became negative, so it's also possible that patients that are seeing some kind of immunomodulatory therapy are actually down-regulating their perforin in response to the drug. And then with respect to why you might not see classical markers of inflammation elevated, that's a complicated one. If it is a purely interferon-gamma-driven disease, then I can imagine it might be possible that it doesn't drive things like a CRP, which is more of an IL-6 disease, and the SED rate, I think, is very confusing to follow in patients that don't have normal fibrinogens because that's the major acute phase reaction that drives the SED rate, so all bets are off on that, too, I think. So, you know, I think perhaps these more specific markers of immunosuggestion might be better. Lorenzo Dantiga, Italy. Question for Dr. Squires. I think acute and chronic, I think, is very different in children compared to adults, and I think we need to find a classification. For instance, the acute and chronic would be patients with chronic disease who are looked after, they have a trigger, and they develop acute liver failure, whereas there are patients presented for the first time acutely having an underlying chronic disease, and the easel classifies them as chronic liver disease presenting as acute liver failure. For instance, autoimmune or Wilson presented with acute liver failure. So I think they're very different. What do you think? I think it is confusing. To me, acute and chronic are very few in our centers, at least in centers where there is a liver transplant, whereas chronic disease presenting as acute are probably many. I agree. You know, I mean, I think that 2014 proposal, you know, where they tried to classify the type A acute and chronic, I mean, I think that's what you're speaking to, which is those patients who are yet to be diagnosed or have a disease that isn't really chronic but can act or present in an acute manner. You know, I think back to your first point, I think better definitions, I think, are clearly needed, and in pediatrics, you know, I think, again, the APASA group has proposed their first recommendations. I think those are worth reviewing, but I think a more refined definition absolutely in this case is going to be important for the pediatric population more than adults. I think, you know, there's a lot of, you know, when you look at those of the adults who do present with acute and chronic, you know, a lot of it's alcoholic hepatitis. There's a lot of, you know, data that we didn't present that talks about microbiome and the effect of the microbiome that is different in alcoholics that contributes to this process, and so I think all of that will clearly be different in kids. So I mean, I agree with your points, and I think, you know, better definitions will allow us to define this and then study it. We have used them but there are no published data to show that they are actually neuroprotective. The rationale for both would relate to decreasing the systemic inflammatory response and therefore decreasing the exposure to all the cocktail of pro-inflammatory cytokines. I'd say based on anecdotal experience that it's impossible to say whether they are directly helpful because when they're used, at least when there's someone like me advocating for their use, it's in the context of other escalations of care like escalating the intensity of temperature control, for example. I would say that they make excellent mechanistic sense based on the likely, almost certain, contributions of neuroinflammation to brain injury, brain edema, and acute liver failure in children. But there's no evidence to help show what they actually contribute individually to improved outcomes. In fact, this whole idea of neuroprotection in this field is confounded by what are you actually using as a measure of outcome. Well, is it the need for transplant, survival, long-term functional outcome? As you can see, even in the studies that we published just looking at the utility of EEG as a more sensitive biomarker, we rolled together death and liver transplant, which is, you know, there's significant issues with doing that. So anyway, that's a very circuitous answer to say I think those therapies make sense, neither based on my anecdotal experience or certainly anything I'm aware of in the published literature could one actually say that they have definitively shown... Yeah, it makes excellent sense, and it's a similar rationale for therapeutic hypothermia and a reminder that when we're thinking about neuroprotection in this context, we're trying to maintain the brain in a relatively, a state that's compromised, but a state that's reversibly compromised, and that reversibility may occur when the healthy liver, either the liver recovers or there's a healthy liver transplanted. Thank you. Thank you. Stasios Gramatikopoulos from King's College Hospital. It's a question for Dr. Squires. We obviously described the first British biliary atresia population from Britain, and obviously the cause of ACLF was completely different than it was from other parts of the world reported. Now, we reported that GI bleeding was one of the major factors precipitating an ACLF episode. I just wanted to see what are your thoughts about GI bleeding, because you, I don't know, you went too fast on your slides to be able to appreciate that. And also, you know, since we know GI bleeding is associated directly with infection, how do you see that being prevented, and have you got any thoughts about surveillance endoscopy on these children? Wow. So I think your first point, and I apologize for going fast. Yes, in the data from the Chicago group, you know, GI bleeding was, I believe, their second highest decompensating event leading to the ACLF other than fever and sepsis. And so I think it is a significant problem in kids. Obviously compared, you know, your study with the Indian study, I think that, you know, their population was very different. And I think, you know, reactivation of hepatitis and hepatitis A was their leading decompensating events. I think GI bleeding, you know, should be... I think it's important. You know, whether or not it should drive our decisions to do primary prophylaxis and screening in these kids, I think is a more difficult question to answer. I'm not so sure I would go that far at this stage in the game given all the other data about, you know, first GI bleeds in kids with portal hypertension. But I think, again, as this body of literature grows, I think that there could be a potential role for screening for this, but I wouldn't say that I'm ready to recommend that at this time. notes that it's biliary atresia which possibly is not causing pediatric ACLF, and I agree with you that we haven't found in our cohort biliary atresia being one of the common etiologies. One reason that we felt was it was so because even the adult apastel definition doesn't talk about cholestatic liver diseases because they have their bilirubin already high, so that further increase in bilirubin is really not appreciated. Those two things, that coagulopathy and jaundice to begin with, is not there. Is that the only reason that you feel even the easel cohort doesn't really have biliary atresia, or do you think differently about the easel cohort? The other thing that within those two days of review that we had of the apastel definition of ACLF in 2019, at least three of us are sitting there who were there in the part of that review, and we noticed two more things about pediatric ACLF. One, that biliary atresia, of course, is not there, and the second was the metabolic liver disease that we find mostly in our population because we don't have a very strong neonatal screening program, so even the metabolic liver disease never presented to us as pediatric ACLF. What you're right about is that our cohort is a little different because a large percentage of our ACLF are those type A who come to us for the first time as ACLF. They're mostly undiagnosed CLD who have possibly a viral hepatitis as a precipitating event, and then they come to us. So my question here is that what is your take on these two findings? Why is metabolic liver disease not really coming as, and of course, biliary atresia? Again, I think common themes in my answer is we need more data, and so I think that studying this more will lead to more information. I think your first question about is biliary atresia really protective? Again, I don't know if I'd use the word protective. I think for reasons that are to be determined, children in these very small cohorts that have been studied with biliary atresia seem to develop it less. Now whether or not that's partly because of the way they're being managed or if it's inherent to the underlying disease or the way that the disease is treated, I don't know. I think it relates to your second question in metabolic liver disease. I guess one question would be what do you mean by metabolic? If it's the metabolics who don't have intrinsic liver damage and don't have portal hypertension and don't have some of the other things that we think are contributing to the process, then that may explain why some of the metabolic kids don't have these types of phenotypes. But I think that, again, we just need larger numbers and better understanding, and I think there's a lot of people who are looking at this now in the pediatric field, and so I think that's a good thing. And hopefully the next time we have a symposium about this, we'll be able to present better data about what some of these things are. Deirdre Kelly, Birmingham, England. My question is to Dr. Wainwright about neuroprotection. I think you're right. We're not very good at getting this as accurate as we should do. One of the scenarios that is very similar to this is primary graft dysfunction, when essentially you've got an acute liver following a graft dysfunction. What about the role of cooling? Do you know, have you any data on whether cooling is helpful in that situation? And perhaps I missed it, but could you also comment on the use of sodium benzoate to manage hyperammonemia? Yeah, I didn't address sodium benzoate, and I simply didn't review that. show that in the acute liver failure from the acute liver failure group that there may be a benefit for under the age of 25. For pediatrics, there's no published studies that I'm aware of showing about experience with hypothermia in pediatric acute liver failure under any circumstances, including the scenario you mentioned. The field is, I think, moving towards targeted temperature management, taking the lessons from the use of targeted temperature management in the management of resuscitation after cardiac arrest, but has also taken the lessons from the management of severe traumatic brain injury that the early use of therapeutic hypothermia temperatures down to 33.5 may carry complications including unrecognized volume depletion, and certainly in the case of severe TBI, increased mortality. So what I see across the field from my colleagues is that for children with acute liver failure, there's much more attention paid to very tight control of temperature, and that would include using the arctic sun or its equivalent, so within a much narrower range. There's acceptance that, as children acceptance, there are centers at which there's varying degrees of willingness to use therapeutic hypothermia for the children who are progressing to stage 3 or stage 4 coma, and to give them a bridge until a liver can be made available. But there's also recognition that despite the neuroprotective efficacy of therapeutic hypothermia, that that has to be done with caution, because we know from the data from severe TBI that carries an increased risk of mortality for reasons that really are not entirely clear. For sodium benzoate and the clearance of ammonia, I think one of the themes I was indirectly alluding to in addressing neuroinflammation was that the role of ammonia, clearly one can't have encephalopathy without ammonia or cerebral edema without hyperammonemia, but it's not the ammonia alone. The effect of ammonia on astrocyte function, the neurovascular bundle, the integrity of the blood-brain barrier, is not just through the swelling of astrocytes and producing edema. It's actually more directly likely related to direct toxicity to the sodium chloride transporter and dysregulation of extracellular potassium, and thereby compromising astrocyte function in general, which is to say that sodium benzoate on its own is fine, but it has to be done in the context of all these other approaches to try to regulate cerebral metabolism and maintain astrocyte health. I mean, it's very effective, obviously, in urea cycle defects and also in exopatic portal vein obstruction and hyperammonemia. I just wondered if it was useful in addition to all the other therapeutic aspects. We just don't think of it often enough. I think I glossed over it. I'm thinking more in terms of the inflammatory mechanisms, but yeah. I'm the one from London. I just wanted to clarify the data we were talking about with the atresia. That study was just done on biliary atresia patients with the motive of looking at natural history of a disease that we know and the outcome we know. What we found, there were 26% higher mortality in that group who developed acute and chronic by the definition we used, highlighting that they need better points on the waiting list. We don't have a waiting list system in the UK like the points you have. So the purpose of our study was very different. It wasn't looking at epidemiology or what are the conditions that we disposed to it. The motive was very different. In a very defined cohort of patients, it's a long-term, very close follow-up. That's what was that what predicts it and how we can avoid it. Thank you.

acute on chronic liver failure

ACLF

Asia-Pacific Association for the Study of Liver

EASL/AASLD definitions

pediatric ACLF

neuroprotection approaches

hyperammonemia management