Awesome. Thank you, Dr. Williams. And we'll move to the [INAUDIBLE] announced portion now. And we're really excited to have an interdisciplinary group today. So Dr. Brendan Bowman, from the Division of Nephrology, Dr. Kyle Enfield, from the Division of Pulmonary and Critical Care Medicine; and then Dr. Zach Henry, from the Division of GI and Hepatology. They've asked me to keep introductions short, even to just say that they were three guys that have trained here, and stayed here, and never got a job anywhere else but here. [LAUGHTER] But I'll say just a few more words. So today's presentation is obviously on hepatorenal syndrome, a very complex topic and interdisciplinary. And this group is very uniquely qualified to cover this topic. So Dr. Bowman is an expert in renal physiology, and teaches across the spectrum, from medical student course directorship to resident teaching, didatactics on the wards, to running the fellowship program here, as well as teaching nationally. Dr. Enfield serves as the Medical Director for the MICU and is one of the strongest clinical educators for our residents and for the fellows. And then Dr. Henry is an expert across several areas of hematology and is really the most active with researchers and educators within the division there. And all have earned several both clinical and teaching awards over the years. So that being said, we are very excited to have them here. They want to make this interactive and interdisciplinary. So I'll leave it at that. And, without further ado, please join me in welcoming them. [APPLAUSE] So this is going to be a giant experiment in three guys presenting at the same time. And we've gone through this not once. [LAUGHTER] So with-- how kind of you. So when we were originally approached to do a pro-con debate, we thought about lots of topics. I really advocated for Marvel versus DC Comics. Unfortunately, we all wanted to pick the Marvel side and, except for Zach, who was going with Calvin and Hobbes. [LAUGHTER] And so we really couldn't rely on this. If this happens to come up on MOC question, I'm very sorry. It's my fault. None of us have any financial disclosures, which makes this a much easier topic to go over. So we're going to go over several things. And, hopefully, we will make this fun and exciting for all of us. We're going to try to review the criterion and controversies in the diagnosis up Hepatorenal Syndrome, talk about some of the ways that we approach this as three guys from UVA. And doing that through a Case Presentation, then I'm going to hand things over to Zach to talk about the Physiologic Changes in Cirrhosis. We will then go into Brendan who who's going to talk about AKI Challenges. We'll switch under the ICU Management, which is actually the shortest of section of this. And then we'll talk about Transplant and what we hope you guys will take home from this talk and you can go back and practice. So it's a typical day in the ICU right now. I originally said, I don't know if I can talk about liver disease for an hour. And then I rounded today, and I spent most of an hour talking about liver disease. So this patient is a 48-year-old gentleman who comes in, no history of cirrhosis. Initially diagnosed around three years prior from a GI bleed. Was found to have hemochromatosis mutations. He's had to have a pretty typical course of large volume paracentesis every two to three weeks, and is on a fairly typical home regiment of Lasix, in the morning, and the evening, and spironolactone daily. And this is a fairly average patient for a day in the MICU, particularly one that comes in with the Emergency Department with lower extremity edema. Had an LVP 10 days prior with an appropriate albumin response to that. He complains of increasing abdominal girth and weight. He denies any dietary indiscretions, and has been very faithful about following his low sodium diets. So when he presents, his blood pressure is 77 over 40. He's afebrile. His heart rate's 96. He's sat at 86% on room air. He's got the constellation of elevated JVP, tachycardia with a murmur, some distended protuberant abdomen with umbilical hernia, a lower extremity edema, scleral icterus, and spider angiomas. This actually describes 25% of the MICU patients currently in today. The labs have been notable for a rise in his creatinine to 3.8 from a baseline of 1.3. tbili is 22, sodium of 130, bicarbonate 18, and albumin of 2.2. And his urine sodium is less than 20 millimoles per liter with a bland sediment on examination. So this prompts us to ask several really good questions. So what treatment in the inpatient unit should we use to manage this patient? What etiologies describe what has happened and why this patient has gotten worse over the last 10 days? What next steps should we take? And what is a short-term and long-term prognosis for this patient? So as we go through this talk, there's going to be some terms that come up that I've just noticed, actually, for a long time that we sometimes use interchangeably. But they have some very specific definitions. And then a new one, or relatively new to most people, so compensated cirrhosis is basically cirrhosis without complications. But really it's cirrhosis without significant portal hypertension that causes these clinical decompensations. So ascites, encephalopathy, actually acute variceal bleeding, those are decompensating events. Just having varices though is not a decompensation. And that's important from a mortality perspective because compensation and decompensation have much different one, three, five year mortalities. So when we use these terms, we need to make sure we're using them correctly. MELD score, we use all the time, right? Everybody who presents a cirrhosis patient tells us their MELD score. Remember just a couple of things. MELD was originally described in TIPS to try to predict mortality after TIPS. Its main mortality prediction is really over about a three month period. There are some studies that have looked at longer timelines, but that's its main predictive power. It is used for transplant for-- not for listing people but for moving them up and down the list as a suggestion of severity. The newer term though is acute on chronic liver failure, and this is an important thing you're going to see a lot now. This started a few years ago in Europe. And, essentially, it's basically decompensated cirrhosis with progressive organ failure. And the reason it's important is because if you just look at patients with cirrhosis or decompensated cirrhosis versus those that start to knock off organs, and you can see here just knocking off one, two, three more organs, similar to what you've seen with, like, SOFA scores and other ICU scoring systems, their mortality increases significantly. But, as this is coming into prominence, more and more studies are associating not just this with mortality but with infectious risk, risk of developing AKI, et cetera. And so this is actually a really powerful prognostic, tool which we'll come back to in the end, that you can use especially in the ICU setting to help guide some family discussions in decision making. So from a physiologic perspective, I promise I won't spend a ton of time on this. But I do think it's important to understand. Portal hypertension is vascular, OK? So just think blood vessels. It starts with the sinusoids and the portal veins, the portal system in the liver. But it ends up affecting the whole rest of the body, right? So if you think about things like hepatopulmonary, hepatorenal, cardiomyopathy related to cirrhosis, all of these things are really vascular complications from portal hypertension. And those are their downstream effects. And so I think most people understand the basic physiology of blood flowing into the gut through the portal system to the liver and back to the heart. So I don't begrudge this. So-- but what changes in cirrhosis is when this guy on a microscopic level gets scarred down and shrunk, you get elevated portal pressures, which tend to back up into the splenic bed. You get release of vasodilatory substances, like nitric oxide and endothelin. And it starts to trigger a cascade of events that, again, is all vascular. So you'll start to trigger the renin-aldosterone system because those vasodilatory substances get out into the systemic circulation. And so systemic vasculature will dilate. And what happens when it dilates? Blood pressure goes down. Our body has natural mechanisms for that to come back to vasoconstrict again. And, over time, our body kind of maxes out. One of the things that we'll show you-- so when that happens you develop ascites. And then here, hepatic hydrothorax, which is just ascites leaking back up to the abdomen. This is one of the first things that typically happens in patients with cirrhosis. One thing I would like to correct-- I didn't make this slide. I'm going to blame that on Dr. Bowman. [LAUGHTER] I don't-- You forgot angiotensin, though. So I was going to-- We'll get there. We'll get there. That's fine. Ascites really doesn't leach out from around the bowels, which is kind of a common phenomenon that we think about. More likely, it leaches out from the liver itself. All of this at a microscopic level. There's some older studies that suggest that you max out the lymphatic drainage, as well. And once that occurs, you start to get a ascites development. And one of the typical things we see in early cirrhosis is the very hepatic ascites. So it's just a rim of fluid around the liver. They don't really have distension elsewhere, and it's probably because that's where it's starting from and building from the liver directly. So the physiologic changes are everywhere. So you start to get underperfusion because this-- the aorta is clearly in systemic circulation. That's the only vessel that matters. So the fluid level goes down when things vasodilate, right? And so you can't perfuse the kidneys. So that can lead-- simple, right? Just leads to kidney failure. However, our body vasoconstricts, this goes back up. But these things still get hurt. And that's what we're going to talk about today. Your heart tries to pump harder, and pump more, but it can't keep up. That's why eventually people get cardiomyopathy. And, yes, Brendan, angiotensin-aldosterone, and eventually ADH. So when all these things start to max out, norepinephrine or noradrenaline, the RAS system, kind of the last thing that really kicks in is that ADH system because if this tank is low and we can't vasoconstrict anymore, we just have to fill the tank back up. And the easiest way to do that for the body is to hold onto more water. So we talked about this a little bit. This is important because, from a cardiac perspective, this was kind of an older mantra where you can kind of see here compensation, decompensation, starting to get hyponatremia. This is where we start to see more of the ADH, and then progressive kind of long-term renal dysfunction. And the original thought was that if you measured cardiac output, it would just stay up over time. It just couldn't keep up. And what we know is actually true is really as this starts to get worse, your heart your heart cannot keep up. And it's actually a cirrhotic cardiomyopathy where the contractility of the heart just can't keep up with what's going on in the liver, and what's going on systemically. I won't belabor this, I promise. But if you just look, these are all, right? Measures of vasculature, basically. If you look at that, basically you have kind of cirrhosis, decompensated cirrhosis, and at the time of HRS. And if you will get things like renin levels, so early on decompensated cirrhosis, they're elevated chronically. And at HRS, they're almost doubled. Same with noradrenaline levels. And this is because the body is trying to respond as much as possible to increase blood flow. But you can also see, regardless of this MAPs are still significantly lower as you progress through this vascular phenomenon. Does that kind of make sense? OK. No one's overtly shaking their head. And no one's gone to sleep yet. OK. [LAUGHTER] I'm just going to point this out. This is a great review article in The New England Journal of Medicine by the king of nephrology. Correct me if I'm wrong, Dr. Schrier from-- I knew we were thinking the same thing. Yeah, not Dr. Bowman. And then-- [LAUGHTER] --one of our pre-eminent hematologists from Barcelona, Pere Gines, they spent most of their lives researching this. And this is a wonderful review that explains that kind of vascular physiology. So if you're going to read one thing to try to help understand this, I would recommend this one New England Journal article. And it's literally called Renal Failure in Cirrhosis. If you just search it, it will come up. Yeah. Yeah. Sure, sounds good. Yeah, we actually had Dr. Schrier here. He came as one of our invited lectures a few years ago and gave just a wonderful lecture, too, as well. I don't think we recorded them back then. All right. So we'll spend a little bit of time about acute kidney injury, and the kind of perils and pitfalls with diagnosis of kidney injury and HRS and those things as we go through there. We did say we wanted to make it kind of question and answer too. If you have questions that come up during the presentation, I think we're on time. We've been trying these kind of integrated, multidisciplinary cases in the School of Medicine. It seems to work OK. At least they don't complain too much. Less than the usual level. So GFR estimation in cirrhosis. So this is tricky. So a couple of things to know that happened in cirrhosis. So, first of all, if you think of creatinine as kind of your pro-drug for creatinine, that's synthesized in the liver, so as liver function fails, creatinine production is going to go down. So just by that alone is an issue. Secondly, muscle mass is generally lower in our patients with liver disease. And so you may produce less. But, in addition to that, you also see changes in the site of creatinine removal. This happens in chronic kidney disease, too, by the way. So if we're carrying about filtration at the glomerulus, right? The glomerular filtration rate. That if creatinine is being removed kind of under the radar down here in the proximal tubule by secretion, so we're kind of artificially reducing the creatinine total in the circulation. And this leads to overestimation if we just use a creatinine-based equation for this. So that hasn't changed a lot. That's been a problem. So things that we typically have used historically-- Cockcroft-Gault, MDRD, and what we use at our hospital now, the most accurate creatinine-based equation, CKD-EPI-- still underestimate kidney function and GFR and cirrhosis. So this has been difficult. There's been a lot of move for other biomarkers other than creatinine-- and some places now have Cr-Cystatin C. But there are new equations, things like creatinine Cystain C for cirrhosis where those were actually validated against the gold standard to see what the GFR truly was in a population of patients with cirrhosis. But none of this stuff is available for you. The point being is that if you see a creatinine of 1.4 or 1.5 in a patient with cirrhosis, that indicates a significant impairment in their kidney function, and the CKD-EPI equation at the bottom that you're reading off Epic doesn't know about that. OK? So you just have to keep that in the back your mind that your dosing may not work for those folks. And so one thing, if you think about the patients we see-- especially, like, the alcoholics that are malnourished, that have-- you look at their arms and legs and they have no muscles. You'll see their baseline. If you look back, always look for their baseline. They'll have a baseline tracking of 0.3 or 0.4. So even coming in with the creatinine of 1, which on the surface may look normal, or 1.1, or 1.2, may be a significant AKI. And so you need to pick that up right at the get-go so that you don't just start trying to diurese their ascites. And then you come in the next day and their creatinine's 2. That was a short interruption. That was-- that's good. I'm trying. [LAUGHTER] We said we would interrupt each other and not take offense. So we're going to keep going with that theme. All right. So whenever you do a talk about HRS you're contractually obligated to show the only person that ever did angiograms in patients with acute kidney injury and HRS. So there's never been another picture. So if you see a grainy picture from 1970 from the American Journal of Medicine, and you don't ever see-- you guys say, well, where's the update? It's never been done again. So-- And it will never be [INAUDIBLE]. We'll show it to you. This is not IRB approvable. Yeah. This is not good. So this is kind of-- the '60s and '70s were kind of an exciting time in discovery for HRS, right? So the first transplant was done from a patient with HRS taking the kidney out. And that's a '69 paper from The New England Journal of Medicine where the kidney did well. And so this kind of brought the idea that this was a functional change in kidney function, and not a structural. In other words, the supply was bad to the kidney. The kidney itself is OK. So then these guys-- Epstein did angiograms pre and post on patients with a AKI. So just to orient ourselves here real fast, so kidney. And we're there. So on the outside of the kidney is where the action is. This is the cortex. This is the filtration area. This is where all their nephrons are. These are the medullary pyramids, and this is where we do all of our concentration. All right? So this is where we store all of the sodium that-- when that urine sodium is less than 20 in the urine, this is where it's going, to allow you to retain water, OK? So our blood supply down here though-- so it branches out just as we do anywhere else. So I say that because what we're going to see here are the renal angiograms of these patients at the time of the acute kidney injury. And this should be a nice fill of all the renal vasculature here. And this is the edge of the kidney. That's with the arrow marks down here. And so you can see, again, this fills the renal artery, maybe some of the lobars, but you're not getting any cortical blood flow. Same kidney, post-mortem, fills beautifully, OK? So this is where the validation study could be at least suggested that renal blood flow is decreasing tremendously. So this doesn't make a lot of sense. And I see Dr. Carey here, so I got to be careful talking about angiotensin II, and those things. So in the afferent arteriole, there's not a lot of angiotensin receptors that are going to lead to constriction, right? This makes sense. We don't want to constrict the afferent arteriole. However, that really revved up renin-angiotensin-aldosterone system. However, it will still affect the large vessels. And so the suspicion or the theory of the classical underfill theory that Schrier propagated was that you'd have a low blood pressure, low MAP state, but you also have high renin-angiotensin-aldosterone. And, in fact, you can check ultrasound resistive indices in these arteries and find that they're quite high. Much higher than normal. And so the constriction probably goes from hilum out to cortex. So it's moving from the large vessels to the small vessels. So it's a bit unusual. So what about it? So we hear a lot about HRS. This is the famous cause of acute kidney disease, and the one we think about first. But all AKI is not HRS. If we think that our response rates to typical HRS therapy aren't 100%, well, why might that be? So this is-- I'm going to make a pun here-- a pretty gutsy study. Sorry. We're NGI so I had to do that. So we-- [LAUGHTER] I know. When you have your second kid, you have to make puns all the time. This is part of being a father. So these guys actually-- is a French study. And they did about 10 years of biopsy studies. Patients that had worsening creatinine, proteininuria, microscopic hematuria. And they did transvenous biopsies of these patients. This is the only study I think to anyone's knowledge that that's ever been biopsies in patients with later stage cirrhosis. But you can see there's quite a bit of glomerular disease in here underlying. 80% of patients had glomerular disease. 90% had some evidence of other chronic interstitial diseases. And 75% had evidence of acute. So I guess what I'm saying is that there's a lot of underlying pathology, probably by the time the patients get to us. How much of the disease process that we're seeing is potentially due to HRS versus due to some of these underlying problems. And the most common would be IgA. That's pretty known association down here, too, as well. So there's a lot happening in the kidney in these patients that we don't know about. So, I mean, none of us remember the time we biopsied that patient with the decompensating cirrhosis in the MICU. We're not going to do that. That's probably not going to change any time soon because the kidney is essentially a bag of blood vessels. And it's pretty high risk. But we just have to keep in mind that not everyone is going to respond. And part of this may be because of that. All right. So the diagnostic criteria. So when I was a resident, I first heard of the International Club for Ascites on rounds. It might have been with Dr. Caldwell. And he was a pretty fun person around with, I'll tell you. So I just assumed that the International Club of Ascites was like this. It's not. [LAUGHTER] But they came up and recognized this was a special case in '96, and had their first diagnostic criteria. These have been revised multiple times. And so the Acute Kidney Injury Network and multiple AKI working groups got together with the International Club for Ascites and came up with the revised diagnostic criteria. And this was changed back in 2015. So a couple of key points. Creatinine as a threshold marker was taken out. And we looked at exactly what Zach was talking about, changes from your baseline creatinine. And that's important. I mean, we think about differences not just in disease, but in sex. So if you're using a diagnosis of acute kidney injury and it's a female versus a male, the creatinine levels might be lower and that person might end up with even a lower MELD score, for instance, if the kidney injury is-- so there's a lot of implications for that. So by changing it to the patient's baseline, that put a lot more people in the bucket to start getting them on your radar for you need to treat this person up front. And these are the typical ones. They didn't approve with a trial of volume expansion, and they don't have any other reason to have kidney failure. So the problem is, is that, yes, it's a diagnosis of exclusion. But these are pretty-- we've just said before that a lot of patients are coming to the table with a lot of underlying renal pathology. And so by looking at these and following them strictly, we may be leaving out some patients that really do have HRS superimposed on a pre-existing process. So keep this in mind. And they're good for research. Well, and another thing. So NASH is quickly taking over, right? As the number one liver disease, not only in the United States, but in the world. It's associated with metabolic syndrome. So diabetes, hypertension, things that are going to cause CKD. In the future, more and more of our patients are actually going to come to us with a baseline with some amount of CKD. And so they're going to have some underlying kidney pathology. And we're going to have to capture this on top of that. So it's an excellent point. It's that last little part, if it's acute and it's obvious. But, otherwise, you kind of have to take it with a grain of salt and try to figure out, is this really an acute change? Or is this more related to their chronic kidney. And just to add onto that, you'll notice the criteria of absence of shock. And most these patients are in the ICU, and distinguishing someone with shock without shock in the ICU can be very difficult because we remember shock is a state of poor oxygen delivery to oxygen demand, which we usually diagnose by either development of lactic acidosis, which how many liver patients come in with the baseline lactic acid? So there's lots of confounders here that make using these very strict criteria difficult, and maybe not always clinically applicable, even though they're great for our research trial. All right. So other problems with HRS, so-- yeah? [INAUDIBLE] He wants the punchline now. [LAUGHTER] It's real bad. It's not nearly as good as he was doing. Sorta bad. Nah, this is really good info. I'm enjoying it. In the histology study, I'm surprised it's so [INAUDIBLE] but the thing with alcoholic related [INAUDIBLE] Yeah. The largest group is about 40%, 50% alcoholic. It was the-- [INAUDIBLE] explain, because I don't think IgA [INAUDIBLE]. Yeah. Well, so I guess one of the things I would say is I think that that doesn't mean that that was the primary driver of the reason they were coming to the table for a kidney. I think a lot of those were over calls because the IgA, it just shows up in the mesangium. And then you see that a lot. And so it gets diagnosed. That doesn't mean it was really causing any of the renal problems. But I think its under-diagnosed in general. So it could be. But I don't know that it was the driver. I just know it was there. So this is also a slide for Dr. Caldwell, too, as well, because we used to talk about this when I was a resident, is that if eschema is on the same-- it causes ATN eventually. And HRS is a state of chronic eschema in the kidney, then it stands to reason that HRS and ATN maybe on the same continuum of disease. And so one of the problems with having a diagnosis that you need a bland sediment is that you don't know where you're catching HRS in the time frame, right? And so if we start with HRS and we give it enough time, we eventually move from functional to structural kidney disease, you may not have a bland sediment at that time. OK? So this makes the conversion of HRS to ATN very possible. And they shouldn't be mutually exclusive. You can't look at a urine sediment and say, I see casts. And this accounts for X% of the kidney failure. In other words, you could still have a smoldering HRS component to your AKI. And you won't know that. And there's no way to diagnose that. So this begs the question of, well, what have we got in the hopper? How can we differentiate these things? So one thing also to keep in mind is in studies they define patients. They say, well, this was the HRS group. This was the ATN group. Well, how did you know the difference? Oh, we all got together and said they met the criteria. We had a meeting about it and we said this guy has HRS. So what test did you use that's a gold standard? We don't have one. OK. So that's the punchline to this. When you look at all this data, we can't validate a biomarker versus something else, because we still remain without a gold standard. It's still a clinical diagnosis. So I just say this because there are options out there that are thought to potentially differentiate things like HRS from ATN. NGAL is one example. This is a tubular protein that gets expressed during tubular necrosis at high levels. So a cut point has been theorized to about 194. So we've always been up here as the kidney doc saying, biomarkers are coming. Biomarkers are coming. But I will tell you, finally, there is a-- we actually have a biomarker pilot study going on post-surgical patients, too, as well. So we're not always lying you. OK. Come back in 2020, biomarkers will still be coming. That's right. [LAUGHTER] And we'll talk about steroids that year, too. [LAUGHTER] All right. So just another example of other ones. KIM-1, IL18, and all these [INAUDIBLE]. But I would also say that they don't provide always a tremendous amount of value over good old-fashioned albuminuria. So a lot of the simple stuff still works just fine. OK. I'll let Kyle have a chance. This is the longest you've ever had to go without being at the podium, so-- I know, and-- [LAUGHTER] I was trying to be really quiet, too. So now we made it completely clear how easy it is to diagnose HRS, we're going to talk about how incredibly clear it is to manage HRS, both kind of on the floors and in the ICU, primarily in the ICU. And what I'm going to call is, is that the best part about critical care when it comes to hepatology is, is that there are almost no patients included in any randomized controlled trial. And when they are, we have trials of the size of 50, which is very close to, in my opinion, so-- not quite. But pretty close. So we know that albumin versus placebo has shown a benefit in giving a albumin load. And there's been shown a significant mortality benefit. That's been around for-- since my residency, which was only a decade ago, or a little more. And so that's kind of the mainstay of our treatment. And AMO therapy really came out from that. We want to address this splanhnic vasodilation. And typically in the US we use octreaotide. And that has the combination of albumin plus octreaotide plus midodrine has shown benefit. In Europe there is a higher use of terlipressin. And the great part about us talking about terlipressin now is that in the next couple slides we'll talk about terlipressin more and maybe where we're going with that. When we look at sort of the classic therapy on the floor with AMO and terlipressin on the floor, what we really see is-- this is a small study. And I'm going to get how many patients were in it. But what we see is that when we give terlipressin over AMO, we see a benefit in the MAP response rate. And we're going to talk a little bit more about why the MAP response rate might be important in a few minutes. But Levophed may be safer. And why do I say that? So this is the original Levophed study. And essentially-- so I'm going to just tell you right now. Zach threw some slides in here at the last minutes. [LAUGHTER] And I was on rounds this morning, trying to-- as soon as we got done with rounds, I had to read a whole bunch of paper. So-- but this-- this, and this-- so if I stumble a little bit, that's why. But this is what-- so the early Levophed studies with patients with type 1 HRS and basically showing that over time you see an improvement in creatinine clearance. You see an improvement in serum creatinine. And this sort of drove home the point that Levophed may be a useful agent in treating these patients. And that's important to remember. Because if you remember back to the physiology we talked about before, these patients come in with a very high norepinephrine levels at baseline. So we're giving them norepinephrine on top of the norepinephrine that they're already systemically producing. And we do show some improvement in many of the markers of renal function. And you can see though, even on top of that, they still get an improvement. If you go look-- go back. If you look at their MAP response from day zero to day 10, even already maxed out with their endogenous noradrenaline levels, giving Levophed on top of that still has a benefit. So if you look-- this is-- these are meta-analysis of four head to head trials combining-- comparing norepinephrine to terlipressin for HRS. And this is sort of what drives the use of norepinephrine right now in the United States over terlipressin. And I'll let Zach talk about why we may shift that in the future. But really when you look at outcomes of HRS reversal, the four studies really show no difference between the two agents. The 30 day mortality shows no difference. And then there's this maybe trend towards an overall safer, but if you look at all the studies, they all crossed over the line of indemnity. So what that means is that while pooled the means show that there is some benefit. It's based on a pooled systematic review and that that benefit may be somewhat overstated. And what I'll say is right now this debate's a little bit moot because terlipressin is not FDA approved. But what I will say, Levophed is. Has been for a long time. We use it all the time, especially in septic shock patients. But one thing Kyle and I were talking about earlier this week, when we have patients in the ICU with hepatorenal syndrome, we still see them on AMO therapy. And if you're already in the ICU, should probably just use Levophed. So Levophed and albumin, as was previously shown, has had a significant benefit. And in those trials with terlipressin, from a clinical outcome perspective was equal. If you can give it in the ICU, it's going to be better than AMO. So when you have that patient in the right setting, you should put them on Levophed and hold their midodrine and octreaotide until you need to transition them back to the floor. In those prior studies, and just all comers with decompensated cirrhosis and HRS. Levophed terlipressin were even. However, a more recent trial in acute on chronic liver failure has actually shown that terlipressin is substantially better than Levophed. I can't completely explain the physiology as they're both vasoconstrictors. There may be a little bit of a benefit to terlipressin because of its mechanism of action in the kidney as opposed to the general systemic just vasoconstriction. But you can see it affects. Survival. You can also see just the response rates for reversal of HRS, terlipressin is better. And just to make a plug, we are still-- for those of you that didn't know or didn't know. We're actually part of a trial right now in the United States for terlipressin for hepatic renal syndrome. And I think this-- the trial is kind of winding down. We're probably six months from the end. But the early data suggests what Europe has known for now 15 years, and that's the terlipressin is beneficial and it's better than AMO therapy. So we may actually-- hopefully, within the next one to two years, we'll have it approved and have it ready for use. Yes, sir? [INAUDIBLE] Don't-- you got to-- let's make the conclusion. We're going to get to that at-- Yeah. Opinion-based-- [AUDIO OUT] Yeah. We'll get to that, but-- [INAUDIBLE] [LAUGHTER] [INAUDIBLE] First of all, Levophed and [INAUDIBLE]. [AUDIO OUT] The other thing we were seeing-- [AUDIO OUT] --good advice and we-- [AUDIO OUT] --automatically huge savings. [AUDIO OUT] Right. And is-- whereas Levophed's an infusion, the terlipressin in the trial is an injection three times a day. And much easier, safer from a monitoring perspective. However, I don't know if-- it's been a while since I was a resident. I don't know if they still do the renally dosed dopamine. I don't think so, right? That's not a thing anymore? OK. Good. We learned beyond that. So however-- That's so 1998. It never came from us. In the record, it never came from us. If you look at the trials for Levophed and HRS, the actual dosing is pretty low. It's usually like two to four micrograms. It's not high doses. And so if you had telemetry, if you had appropriate monitoring, there's actually probably a setting where you could do it on the floor. It just requires writing a protocol and having appropriate monitoring of it. [AUDIO OUT] Just generically? No, in this-- in this graph. So it's for HRS. It's reversal. I think-- oh, oh, yeah. No. It's creatinine. I think it was 1.5 to 2, a reduction of 1.5 or-- he's at 1.5 grams per deciliter or 2, I can't remember exactly. But yeah, it's a creatinine [INAUDIBLE]. So really the upshot of this slide is in a few years we'll be practicing medicine that they're practicing today 15 years ago in Europe. [LAUGHTER] So-- [AUDIO OUT] So a number of years ago they tried blocking nitric oxide-- [AUDIO OUT] No, yeah. They gave-- they gave-- I believe-- systemically, it didn't work. And then they actually tried it and have had a pulmonary syndrome because they felt like that was where maybe the major effect was. And they used [INAUDIBLE]. I can't remember exactly what the lung thing is. And I think it's because nitric oxide is just one of many. So when we talk about the vasodilation that happens here. And you get this release from the abdomen basically from portal hypertension of these vasodilatory substances. The things we can identify are like nitric oxide and endothelin. They're kind of the [INAUDIBLE] substances that you can measure. But there are clearly other factors involved, as well. So that brings the question that comes up often is, are there people who respond to these therapies versus non-responders. And this is really looking at a very small group of patients that showed some degree of response in their MAP, their mean arterial pressure. And what were the outcomes associated with that? And I'm just going to highlight the creatinine because we keep coming back to this creatinine and I want to try the creatinine to the MAP to survival just a little bit. And what we see is that people who when started on therapy do not have an improvement in their MAP. Their creatinine after treatment hasn't worsened or stayed about the same as opposed to those people who are responders who had some degree of improvement, the statistical significance at 0.045. It's maybe marginal. And this has really led to several studies that were pulled together. This is an article that was published in the American Journal of Kidney Disease in 2012. And it's a pooled meta-analysis of several studies. And what they were trying to figure out is there a MAP to creatine ratio. They wanted to draw that line. Now, I wanted to bring this article to this, because if the residents look at the up to date section on hepatorenal syndrome, it says raise the MAP by 10. And where does that number come from? It probably comes from this article, even though they don't cite it. It shows that-- what they showed with this analysis, is that if you raise the MAP by around 8.6, that with treatment, you've got an associated creatinine decrease of about 1. And we're going to talk about why that may be important in a second. And I think it brings up-- to some of our opinions that we're going to state here in a little bit. But one of the things that we struggle with, almost on a daily basis, and patients ICU is-- what is the appropriate MAP level for this patient? So you'll hear us talking, and you know, there's been several studies that looked at higher MAPS versus lower MAPS, showing that MAPS of 65 are probably good in all comers. But again, the nice thing about critical care and the liver is, there's not a lot of studies involving liver patients in the ICU. And so, we sometimes will say, oh, we'll tolerate a MAP of 60, because this patient has cirrhosis and their baseline MAP is very low. But it may be that even if you start at that lower level, you really should be targeting a bump of around 10 millimeters of mercury in the MAP. Or if they come in at 65, maybe we should be trying to push that to the 70-72 range-- maybe even 75. So I think that's controversial. There's not going to be great studies to do that. But there is some suggestion, at least in this meta analysis. So what does success look like, Brendan? It doesn't look like Brendan, I don't think So-- [INTERPOSING VOICES] He's the one in the tie, though, so he must be successful. I thought we agreed on the tie. We were going to talk about this. You said Alexander Hamilton garb originally, so. This is [INAUDIBLE] So, yeah. So the question always comes up, right? It's like, how long do I do this? This person is doing well on norepinephrine, or they're not doing well, or whatever the case may be. So I only say this to say, well, where's the slope of the line, right? So these are your non-responders. This is the [INAUDIBLE] trial, but they all look very similar. This is creatinine change over day, right? So the slope of this line is most severe at the creatinine drop in this first time, day 0 out to day 6, day 7-- you're seeing response. So I remember being on a case with somebody else-- one of our colleagues from hepatology. And we sat down-- it was about a week. And when do we kind of sit down and have that meeting about this? We need to take next steps of those things. But really, at 3-5 days, if you're not seeing a response in patients down there, it's unlikely that you're going to see a sudden response, OK? For whatever reason that may be-- and we don't know whether this is converted to a structural kidney injury, whether something else is happening-- a bile cast nephropathy, if you believe in such a thing. Whatever the case is, it's not likely they're responding to your therapy, if you've done all of these things. So I think maximal care at this point is all those things we've described, and driving that MAP at least 10 points higher from the baseline from which they'd be compensated at. I think if you've done all those things, then you can tell yourself you've really done what you can do. So this is a nephrologist's perspective on when to dialyze. So I'm going to tell you it's a nephrologist's perspective because clearly, every patient knows, right there at the decision point, who's a transplant candidate and who's not a transplant candidate, right? That's why. So this gets to be a murky decision making one. All of the algorithms for each HRS or these things kind of end with this kind of box, that says, and usual indications for dialysis. So I think, to be on the less cheerful side, we have to say that this is a really bad disease, with a bad outcome for kidney failure. Both mortality and recovery-- these are bad. We're going to come back to that in a second, and show you that the more recent data that we have about the outcomes. But because the overall prognosis is quite poor in nature, as we know that these patients have a very high certainty of death within a very short time, the decision to initiate dialysis isn't just usual indications. It really needs to be thoughtful. Is this patient a potential transplant candidate, or are they potential transplant candidate on the sooner side? That's really how we think about them. If there is something reversible, it's probably reasonable to think about a limited time-- limited trial of dialysis. But same thing-- I'll show you the average time from presentation to death. And it's pretty fast. So a time limited trial is the reason when patients need to know this when they come in, because we don't do a great job about having that expectations discussion. OK, so quick point-- I'm going to skip through this, just because I think 45 minutes is our-- We're running out of time. Just know that dialysis is hard. OK. There you go. CRRT versus dialysis. Hemodialysis-- I would just say that hemodialysis, taking a lot of fluid off a very sick patient, is very difficult to do. We tend to opt for CRRT. There are not good trials in this. We're probably not going to see those things. This is one of the things we wanted to get through to you, though. This is finally the nicest-- probably review. This is Mass General Brigham women's combined data over about 10 years. They looked at outcomes in patients with cirrhosis. And the kicker to this is that-- in dialysis dependent-- so this is not just all AKI, though that has its own bad prognosis. But dialysis dependent AKI-- the median time to death was 12 days in ATN-- slightly longer for HRS. So we don't have a lot of time to work on this. The people that survived up to-- were all transplant patients. At the end of six months, only about 10% of people were still alive at the end of this. So this is a very, very difficult disease. And because it's so dangerous and so deadly, you have to think about your therapeutic interventions and when you're going to do things. OK, so back to our case. So what happened with this patient? So we had lots of discussions about risk and benefit, and we decided to pursue renal replacement therapy, while hoping that the patient would go to a liver transplant. We went with CRRT. And there's lots of reasons for doing that. But I think that you will see that at UVA, the mainstay for these patients is CRRT. They're on norepi, without an ability to really pull fluid, due to recurrent hypotension. The patient remains intubated hypoxia. And we-- the CRRT, really-- the benefit of it is there's not more fluid accumulation, not that there's fluid removal. The-- this is always a great topic with the transplant nephrologists. It used to be high debate when Dr. Keith was here. Who do we give a kidney to? So we have patients with cirrhosis. They get renal failure all the time. They may or may not be on dialysis. Some of them have CKD. Some of them it's all acute. And we're stealing kidneys, when we give them to our patients, from the whole list of kidney transplant recipients that are waiting 5+ plus years for an organ. So there has to be some definition of who is going to get benefit. And so we have to talk about this study, because this is patent worth of study. I believe our eminent chair is on here as well. And so they looked at liver recipients-- this is, I believe, from the UNOS database-- and found that if you look at those that recovered, the mean duration-- this is after transplant-- was about two weeks for people who needed renal replacement therapy, versus about a little over a month. And they excluded the simultaneous-- so if you see this? That means simultaneous liver kidney. So this is kind of liver alone. And renal recovery actually was also associated with longer survival for the patients. One other thing that's important to look at, though, is off of the transplant. So days of pre transplant renal therapy required, versus post transplant. So what I'm saying is, basically-- how long were you on dialysis before? And then how did you do after, as far as your kidney function? And this seems pretty logical, right? So basically, the longer you're on dialysis before transplant, the higher risk you have of not recovering afterwards without a kidney transplant. And so what it came down to on this was 60 days or greater. And this has actually, in our literature, been roughly weaned down to about six weeks or greater. If you're on dialysis pre transplant, you're more likely to need a kidney with your transplant. So in the acute setting, in these HRS patients where we do start dialysis-- if one of them actually survives six weeks on dialysis-- those are the patients that probably need a kidney, versus if it's 60 days or less-- or we actually roughly use six to eight weeks or less-- they may get away with a liver alone. In the past, we weren't super excited to take that risk. And so we were more apt to push for the kidney with the liver. However, recently changes have been made to how kidneys are distributed, so that for our liver patients, if we're in that gray area, or if we're pretty confident it's HRS and not underlying CKD, we can transplant them without a kidney now. And then when we put them on their calcineurin inhibitor afterwards, and we give them numerous reasons to have kidney failure with all of their rejection meds, and their antibiotics, and everything else-- at one year, if they're still on dialysis and still requiring therapy, then we can list them for a kidney. And this is probably a better system-- not just for our patients, but obviously for the kidney patients, because it hopefully prevents us from misusing kidneys in patients that probably would have recovered anyway. Because as we said before, this is all hepatorenal syndrome, when you replace the liver and you no longer have portal hypertension, it's vascular hemodynamics resolved and the kidney should recover. So we're going to wrap up with managing expectations in these patients. And this is important. And this is multidisciplinary. So this is where-- this is why three of us are giving this talk. We're not all so dumb we can only give a third of a talk. But it requires significant communication-- well-- [LAUGHTER] So we were trying to help Brendan. So in reality though, this requires a multidisciplinary discussion. You know, we all talk about the family meeting in the ICU, with the sick cirrhotic patient. And it really comes down to the intensivist, and nephrologist, and the hepatologist sitting down and really talking about this. And talking about the data that's available, and whether or not we really think someone is going to survive. And just because we've presented some of these studies as black and white-- it shows this MAP, or this therapy, or this outcome after dialysis-- we all know in reality it's not always that simple. But we have to use the data we have. So most of these patients in the ICU don't just have HRS. They have ACLF. They have that acute chronic liver failure I was talking about at the beginning. And as we said before, this is-- remember, grade 1, 2, and 3 is just sequential organs getting knocked off. So 1, 2, 3 or more-- and as you can obviously guess, that means mortality rates go up at 30, 90, and 180 days. And so this is an algorithm that you can use. It's been recently published together as a team, to outline what do I really think someone's prognosis and survival are going to be? What are our options for them, as far as dialysis or not? And what are their likelihood of making it to transplant? Are they a transplant candidate or not? And you can use this to support your decision making. And so this kind of-- I like this, because instead of having to use a mathematical scoring system, all you really need to know is that slide from the beginning, about how to define these things. And then just kind of follow it down. However, there is an ACLF CLIF score. That's Dr. Bowman's Clif bar he had this morning. The ACLF CLIF score-- if you Google ACLF CLIF, the first website that comes up is the score. Go to that website. It's a nice calculator. It's from EASL. Put all the numbers in. It'll give you a score. And it'll even give you an interpretation at the bottom about the 30 and 90 day mortality. It can change. So obviously, as they progress, sequential organ failure, you can go back and keep putting the score in, and see how they're doing. But both of these are reasonable to use, especially when you're going into that family meeting, or even just that multidisciplinary conversation about what's the best therapy. How hard do we push? Do we keep going for transplant, or is it time to stop and pull back and have some discussions? So hopefully this can help you guys in the future. I know we're at the end here, but-- because Anthony had a very embellished introduction for the three of us-- I know that's probably it. But you can't tell the family, when you're sitting down with them, we're not going to offer you dialysis. So families will make a good decision when they have information. Not always, but I mean-- usually they'll come around to it. But the way to discuss this is, we're not going to offer you a therapy that isn't going to work, and has the potential of harm. It's extremely difficult to dialyze these patients, even if you're using CRRT mechanically, it's difficult. There's a high number of risks. It's overwhelmingly more likely in these cases that you're going to do harm. So if you have the discussion a little different than that, then the reception is a little different. They don't take it as, you're denying me a therapy. But you wouldn't offer them know chemotherapy for a HRS, right? No chance of that working, right? That doesn't make any sense. And you shouldn't be offering things that don't have a chance of working. But you need to have that discussion in a way as you're not taking something off the table, because people get deeply distrustful about that. So that's one of the things to just think about how you're having your conversations with families. And that may help. And please be patient with us, when you're asking the question about are they a transplant candidate or not. That is not always a simple question. And even sometimes when we think they are, and we start their work up-- three days later, they're not, because we just found something that excludes them. So just-- I promise we're not hiding it. But every time we answer that question, we're trying to use the data we have available. So to sort of kind of wrap up here, before everyone gets paged and has to leave. So this patient actually does well. I know that's a surprise-- we're presenting an ICU case. But yes, critically ill patients do survive, often. And in this case, this patient went to liver transplant. We're really, always excited when one of our patients goes up for a liver transplant. The [INAUDIBLE] usually. That's a lot of high fives, because we feel like we've successfully got someone to where they need to be. And this patient went on to stabilize and come off dialysis in their creatinine-- despite all the drugs they were on-- stabilized at 1.5. And they were able to do well. So what are we trying to tell you, in a few take home points? It's a complex disease. That's why you needed three people to talk about it-- that, and we're all very specific to our one area. Renal function is often overestimated by using the GFR traditional calculations. AKI, especially when dialysis is needed, is exceedingly a poor marker of prognosis unless transplant occurs. And this is the one that I think that all the residents-- nephrology fellows, hepatology fellows, and critical care fellows-- know, is that when that liver patient comes in and develops renal dysfunction, we know that things are going to go poorly. We need new tech tools. And I'm going to take Brendan up on that. There's a biomarker coming. And dialysis modalities really aren't different, but we prefer CRRT at UVA. I think that we need to think about HRS earlier, and really look at that baseline creatinine, and how much it's changed, because I think that we are missing patients that are having renal dysfunction as they come into the ICU. And I see that more and more frequently. And that we need to use the baseline MAP as a place to start, and then target an improvement in that baseline MAP for our therapies, rather than just putting an arbitrary number of 60 or 62.5. Post transplant AKI prognosis is good for those who have been on dialysis less than 30 days. And there's options for liver-kidney transplants, which is great. And with that we'll stop, with questions, at 1:23. [APPLAUSE] So I wanted to make a comment that [INAUDIBLE]. First of all, it was a great presentation. And I wanted to metion although it wasn't [INAUDIBLE] and I want to-- There's a Clif bar up there for you. [LAUGHTER] Going back to the [INAUDIBLE] of hepatorenal syndrome, talking about [INAUDIBLE] inject dye, there's nothing [INAUDIBLE] ICU and critical care [INAUDIBLE] always talking about tissue perfusion, when all know one great marker of tissue perfusion [INAUDIBLE], but in [INAUDIBLE] putting in the diagnostic criteria [INAUDIBLE] and I have problems with that. Why is that [INAUDIBLE], because maybe some people have thought, you know, there are people with hepatorenal and still making urine. But for me, it's hard to understand, how you have no urine output, no renal perfusion, [INAUDIBLE] And then there is no monitoring? [INAUDIBLE] is not an outcome. I don't want to stop you, but so-- when they did the revised consensus ones, they kind of pulled in the AKIN RIFLE type stuff. So the urine outputs now-- now in HRS criteria and from that standpoint. And I read that in the papers and those kind of things. We didn't talk about it, because it doesn't fit the narrative we wanted to share with you, which is that, to drive the MAPS higher, the creatinine went down. But interestingly, as well as they could track urine output in that pulled analysis in AJKD, there was not a change in urine output, at least charted. Again that's-- it's tough. We all know how tough it is to get Is and Os in the hospital. So I wasn't sure what to make of that. But the urine output didn't change in that. So it still begs the question of, when that great trial is done-- that's something we need to get good control of. And to maybe your underlying point-- I'm not aware of any studies that prove that urine output was not significant, that led to that changing the criteria. You know, for years, it was always-- you had to have all of -- uric AKI was one of the criteria for HRS, because you do almost always see their urine output go down. I think why they probably waffled is because of Brendan's slide on HRS and ATN. We see these overlaps between these syndromes. And he talked a lot about that kind of scheming of progressing ATN. I know we've also seen the septic patient that's either pre renal or developed ATN. They kind of get a little bit better, and then they develop HRS after that. So you have that constant moving target. And I do think that urine output's important. I don't think we have a great idea of exactly what urine output means what, though. One other thing [INAUDIBLE] Right. No difference. Right. Why do you have to do so much to the patient. We don't have clear criteria for [INAUDIBLE]. As soon-- I think-- so I don't disagree with that, actually. Yeah, we would favor treating everyone the same. That's the-- you should treat the one reversible cause you have, because I don't believe that you can differentiate HRS from ATN. I mean, listen. Somebody comes in, they get-- bam. The hammer gets dropped on them. They have some horrible nephro doctor-- yes. If you have a perfect time-- that never happens. And when it says the absence of nephro toxic drugs-- I mean, everyone's on a PPI. I mean, you can might make anything a nephro toxic drug. I'd say it's very low, in terms of side effect profile, for HRS treatment. So take a chance. If we're talking 3-5 day trial, at most, and the disease is horribly deadly, that's the only component that you can reverse with an intervention. ATN is going to get better or not on its own. But also, too, as Brendan mentioned, all of the studies-- all of them-- for differentiating ATN and HRS-- the gold standard for diagnosing and differentiating is agreement. Either between-- most of them are two nephrologists. Some of whom even throw a heptatologist in there-- but it's just agreement between a certain number of physicians who are part of the study, right? So even sometimes what we call ATN or HRS, in the end, may not have been. And I think, for treatment, you're right. I mean, if you're going to basically-- what are we doing for HRS? We're vasoconstricting a volume, resuscitating to perfuse the kidney, right? That's going to work for ATN in theory, or it might, so it's probably not going to hurt them. So I agree with you. I think the reason for us to, in the future, be able to differentiate them, is prognostic value, because theoretically-- and this is theoretical, because again, how we define HRS in studies is based on agreement. But theoretically, you would think ATN has a higher chance of renal recovery without transplant than just HRS. Our thought for type 1 HRS is that once it happens, their kidneys are gone. And if you can dialyze them to transplant, great. But if you can't, the chance of recovery is low-- to the point where when we round, and someone's renal function does recover, some of our first thoughts are, was this really HRS? AND it's anecdotal. I mean, that's my opinion, because the studies are subjective. Thank you for being patient. Last question. Where [INAUDIBLE] important [INAUDIBLE] is that when you're doing [AUDIO OUT] Yeah, so I think-- I think going in that direction, yeah-- So-- go ahead. So basically, right-- so in liver disease, there's severe systemic inflammation. And when you transplant someone and their renal function gets better, they're on a ton of immunosuppression. And is it really a vascular flow mediated benefit to resolving portal hypertension? Or could there be a component, or more of an inflammatory benefit for giving immunosuppression, and as the inflammatory cascade dies down after transplant, the kidney can recover? And I think going in that direction, that's a hard thing to prove, because you can't necessarily differentiate. However, those-- I've wondered about this. And that's when I go back to those studies from the late 60s and early 70s, where I guess it's the same, but they took the kidney out of the liver-- out of the cirrhotic environment-- and transplanted that. And then it did really well. Now, you can make the same argument. They're getting immunosuppressed after their kidney transplant. You could also argue that the immunosuppression in the late 60s and 70s wasn't nearly as good as what we have now, and they still did pretty darn well. But I think you're not going to really do a transplant without immunosuppression, so that's going to be a hard timeline, differentiating. [INTERPOSING VOICES] [INAUDIBLE] But although there's one study that shows that TIPS can improve renal function, we also have plenty of studies that show that it doesn't make a difference. We didn't address it. We gave you the classical underfill theory from trial, but there's a whole-- just like in cardio renal, that it's not that simple. That there's also an inflammatory component, too. So that's the other area of research around it-- we did have time to go into that. So I don't think that's a wrong theory to think about. Looks like we're starting to have lots of people run away. So we-- as you probably should at this point-- but we'll be around if you guys want to ask some more questions. Thank you very much for coming. Thank you all. [APPLAUSE]