Welcome to our webinar entitled A Practical Approach to Coagulation Management in Cirrhosis. We've heard this is the first webinar of the year for ASLD, so we're all the more thrilled to have you join us today. We want to give a special thanks to the clinical practice SIG leadership, Dr. Paul Tulibot and Dr. Chris Cowdley, who gave us the opportunity to speak on this topic today. We have no disclosure. So to start, bleeding in cirrhosis carries high morbidity and mortality and really needs no introduction. The aim of our webinar today is to help one form an approach to managing these difficult and often controversial situations. You'll see today during our talk, we can't give you absolute recommendations and that's strictly because the clinical data on a lot of these issues is quite sparse. But what we're hoping to do is give you a confidence and a comfort level to effectively manage these issues with best practice and perhaps even more importantly, manage these issues with your colleagues in the hospital, be it hospitalists, interventional radiologists, and the such, who all contribute to our patient's total care. We will benefit from ASLD guidance statements on bleeding issues that will be published later this year. We will refer to them during this talk, which will give us further guidance as we seek to manage these patients. Structure of our webinar will be clinical vignette based, which I think will be the most practical to understand these concepts. But before we get into the clinical cases, we do want to go over some background knowledge because I think it's helpful and will help make the clinical cases have more significance. One of those first concepts we want to start with is the difference between portal hypertension related bleeding and non-portal hypertension related bleeding. The former portal hypertension related bleeding, the classic example being esophageal variceal bleeding, that's driven by pressure, carries high morbidity and mortality, and we all know this. But non-portal hypertension related bleeding, such as bleeding that could occur in a cirrhotic after a paracentesis, whether that's a hematoma or peritoneal bleeding, or bleeding that could happen after central line placement or after dental work. This type of bleeding is more unique to hemostatic pathway specific changes that come from cirrhosis as well as unique comorbidities common to cirrhosis. And these types of bleedings also carry very high morbidity and mortality. Another key concept to understand in background is that cirrhotics can clot and bleed. And there are a number of risk factors that can push a patient either way. Some of those risk factors can be dehydration, renal failure, infection, acute alcohol use, inflammation, such as an acute on chronic liver failure. And on the clotting side of things, this can clinically manifest as portal vein thrombosis, which has a number of medical risk factors and can be treated medically and mechanically. Likewise, intrahepatic thrombosis can lead to parenchymal extinction and further hepatic decompensation down the road and may have a role for medical management. Peripheral venous thromboembolism occurs at high rates in cirrhotics and needs the equal amount of prevention and prophylaxis consideration as any other patient. On the bleeding side, of course, portal hypertensive related bleeding, but also as we've mentioned thus far, procedure related bleeding and mucosal and delayed bleeding. And throughout all of these clinical situations, traditional lab testing can really fail us. It has limitations and has major issues with using this to completely guide management. It's helpful also to have a basic understanding of coagulation framework as we approach these topics. Again, the purpose of this talk is more clinical based, so I think for that purpose, we'll look at things under three constructs, one of thrombin generation, platelet vessel wall interaction, and fibrin dissolution. And cirrhosis can affect all three of these constructs. It's important to remember that cirrhosis is no longer considered an absolute hemorrhagic coagulopathy. There's a nuanced pathophysiology and therefore as providers, we need to have a nuanced approach to each patient when attempting to manage hemostasis and coagulation disorders. It's well regarded now that cirrhotic patients have a rebalanced system of hemostasis, and if anything, are likely to lean towards being prothrombotic, and this is termed the mixed and thrombosis prone model. In regard to the construct of thrombin generation, cirrhotic patients have reduced synthesis of pro-coagulant factors, and while certainly and classically this is known, we have to be careful in assuming this strictly means a cirrhotic patient will always bleed. At the same time, there's a reduced synthesis and reduction in plasma of naturally occurring anti-coagulant proteins. I'm going to turn it over to Nick to discuss more specifically about the role of platelets and fibrinolysis. All right, thank you. So the role of platelets is very important in hemostasis in all individuals, primarily in primary hemostasis where the platelets arrive at the site of vessel injury and aggregate to initiate the platelet plug as well as thrombus formation. The platelet surface provides an anionic phospholipid area and substrate for the coagulation cascade to occur on and for thrombin generation to begin and to begin to form the actual thrombus in secondary hemostasis. A very important piece of this process is the multimeric protein found in plasma, von Willebrand factor with factor VIII. These factors serve to assist with platelet aggregation and binding and are very important in hemostasis. Both of these are elevated in cirrhosis. And Tan Lisman's group in 2006 showed this quite nicely in the cartoon pictured at the bottom on the right. They looked at levels of von Willebrand factor in healthy controls and compared them to patients with cirrhosis. And you can see quite clearly that as patients develop progressive hepatic decompensation and more advanced cirrhosis, von Willebrand factor becomes more and more elevated. And this is quite in line with the general concept of the rebalanced coagulation cascade. So remember, thrombocytopenia gets worse as liver disease advances. And as a balancing mechanism, von Willebrand factor then becomes more elevated. And in the photographs above from this experiment, this is demonstrated visually for us quite nicely. On the top left is the baseline level of platelet aggregation you can see with control platelets and control plasma. Now when you introduce cirrhotic platelets to control plasma, you see very similar levels of aggregation. And then it becomes more interesting on the right here when we use cirrhosis plasma with control platelets or cirrhotic platelets, we see significantly increased levels of aggregation indicating that von Willebrand factor plays a significant role in compensating for the thrombocytopenia we see in cirrhosis. Now conversely, another type of study done by Armando Trappotti the same year looked at patients with liver disease with severe thrombocytopenia and found through the use of the thrombin generation assay that the ability to generate thrombin depends on the level of platelets. And they found that if patients with cirrhosis had adjusted platelets to 56,000 or above, they were able to maintain similar thrombin generation as compared to healthy controls. However, when platelet levels were decreased below this level, we saw impaired thrombin generation. So the conclusion was that perhaps thrombocytopenia would be a risk factor for hemostatic related bleeding. Now, I think the caveat that we want to reinforce here for the listeners is that these are in vitro tests and they have limitations. And we know when we take care of these patients, that the situation is much more complicated in hemostatic system and coagulation system are very complicated in vivo. And the clinical evidence really supports this concept and it's very mixed. In fact, when we look at risk of procedural related bleeding and thrombocytopenia, there's really no clear established relationship. So we chose these two studies that are very similar in design, both retrospective studies of patients with cirrhosis in advanced disease, and they tell a little different story each. So the Giannini paper published now 10 years ago, looked at a small group of patients with advanced liver disease undergoing a wide variety of procedures and found that patients they defined as severe thrombocytopenia or platelets less than 75,000 had all of the bleeding events. Now this is in contrast to the Napolitano paper published in 2017, which looked at a much larger group of patients, 852 with approximately 100 of those patients having thrombocytopenia or platelets less than 50. And they found in that group, no bleeding events. And so what this demonstrates is how difficult this problem in this relationship is really to study in real life. We recognize that bleeding events overall are rare in our patients and that thrombocytopenia, we have to remember is also a surrogate of more advanced liver disease. And so when we compare multivariate models, we need a large number of events to study this properly to control for confounders such as that relationship. Furthermore, these studies group procedures together and we know that all procedures are different and they have different baseline bleeding risks. And we think it's important when we take care of our patients to look at the specific literature for each procedure. So we don't have enough time really to go into details on the fibrinolytic system, but we do need to introduce it conceptually and recognize that just like the coagulation cascade and system, the fibrinolytic system is affected in cirrhosis and is rebalanced. And we'll come back to this when we talk about antifibrinolytics. And again, to reiterate, if anything, cirrhotic patients are likely to lean to be prothrombotic. And this is from important translational work done in the last decade, looking at the role of the protein thrombomodulin. So what we're left with is a rebalance where there are changes in prohemostatic drivers, but accompanied by changes in antihemostatic drivers in each of the three general constructs of hemostasis, platelet vessel wall interaction, thrombin generation, and fibrin dissolution. And we're left with a fragile rebalance that's under a constant threat of disturbance. And this threat is what could push a patient either to bleed or to clot. And with that, I think we're ready to look at some clinical cases. So we're going to begin kind of like a typical hepatologist day in the outpatient clinic where we're going to meet Jim P., who's a hepatitis B cirrhotic with the mild of 18. Jim is in liver transplant evaluation, and he's a pretty good candidate. He's asymptomatic. He has no ascites on exam, and CT abdomen done during his transplant workup shows a non-occlusive main portal vein thrombosis. So the first consideration we want to take on today is what is the clinical significance of incidental portal vein thrombosis in a patient with cirrhosis? Portal vein thrombosis carries an incidence from 10 to 15% in patients with advanced cirrhosis. It's more prevalent in advanced child cirrhotics with advanced portal hypertension and lower portal vein velocity. It's thought portal vein velocity is probably the most significant risk factor, but certainly there are contributions from hypercoagulable local factors and endothelial dysfunction. And interestingly, spontaneous recanalization can occur in up to 40% of patients without therapy. Portal vein thrombosis can make liver disease worse, and this goes back on data even 25 years old looking at liver transplant, explant data, which showed focal parenchymal extinction that matched the distribution of small micro obliterative clots in the portal and hepatic venous systems. And portal venous thrombosis has direct and indirect associations with acute decompensation, including variceal bleeding. However, in 2015, a large prospective study showed that portal vein thrombosis was not related to hepatic decompensation, but was a signal of more advanced portal hypertension. That is along for the ride. It wasn't the causative factor, but certainly it was a signal of worse disease to come. Nonetheless, portal vein thrombosis carries implications not just anatomically to complete a liver transplant, but also has implications for worse liver transplant outcomes. So our next question should be, what treatment options should we consider and what are the treatment options? So Nick, what are your thoughts on that? All right. Well, thank you. So we will spend some time here as this is a relatively hot topic in the field, I think, and somewhat controversial. So when we diagnose a portal vein thrombosis, we should certainly begin to consider the use of anticoagulation. And this is really well supported now. You'll see the citation below for the new ASLD guidelines on vascular disorders that are published just this month and now available online. And they note, and we agree, that in a stable patient with cirrhosis, with potentially liver transplant in the future or currently being evaluated, we should really consider anticoagulation as a treatment for portal vein thrombosis. Studies have shown, and we cited here this very nice systematic review, that anticoagulation not only decreases progression of portal vein thrombosis, but also promotes recannulization. And so what we will keep coming back to is the outcome desired or the goal desired from treatment of portal vein thrombosis. And the goal in the liver transplant setting is specifically to help preserve the portal vein for future liver transplant. And the data really support this currently. However, outside of this goal, when the goal is to perhaps try to improve outcomes in patients from a hepatic decompensation standpoint, or to try to improve mortality, the data are simply much less clear. And so understanding what your overall goal is is important as you begin to consider anticoagulation. And so here is the ASLD guidance statement, and we won't go through all of it, but it does recognize that existing clinical trial data are overall weak in determining treatment indications for patients and should be evaluated really on a case-by-case basis. As we alluded to before, there is a clear signal that certain patients, although it's difficult to predict, will develop spontaneous recannulization. And so early portal vein thrombosis that is not extensive and not extending proximally without symptoms, you can observe with serial imaging and evaluate for potential regression. However, in patients with more advanced thrombosis or that has advanced over this time to more proximal regions, including the SMV, you should really begin to consider antithrombotic therapy. So once you've decided whether or not to anticoagulate, the next question in this patient would be what agent are you going to use? And the data up until now have been conducted primarily with traditional anticoagulants. Those include low molecular weight heparin, which is found in carefully selected patients to be safe and effective, however, has clear limitations. Patients don't like it, and it limits compliance with its subcutaneous administration, and our patients develop kidney issues quite frequently, which limits its use in that population. The oral alternative is warfarin, which is old and cheap and has been studied in portal vein thrombosis and found to be safe and effective. However, we know that it is challenging to dose and of course diet dependent. The new agents, direct oral anticoagulants, are used quite frequently and are rapidly replacing our traditional agents. However, we should acknowledge they're not well studied for portal vein thrombosis and cirrhosis, and caution is generally advised, particularly in populations that have advanced child Pu-B and child Pu-C cirrhosis. We'll talk a little bit about the reversal agents, which have some attractive features, and we greatly anticipate the results of the Siroxaban prospective multicenter trial in Spain by Dr. García Pagan, who they are evaluating the use of rivaroxaban to prevent portal vein thrombosis. So we've chosen our agent for this patient, and then these questions generally begin to arise, and I think it's important to take a step back and remember that no matter what you've selected, all anticoagulants have risks and benefits. So when we're deciding on the dose to use, it should be acknowledged that even traditional agents as well as direct oral anticoagulants, really the studies were not conducted in patients with cirrhosis. And so we have very limited pharmacologic data on how these medications are metabolized in cirrhosis, and so we base their dosing essentially on non-cirrhotic populations. There is some emerging in vitro evidence by Tan Lisman's group using thrombogeneration assay that has found that the potency may vary with direct oral anticoagulants in more advanced cirrhosis. One question that people have is, do these need to be monitored? I think we know for sure that warfarin does need to be monitored. And of course, with the INR, which makes management often challenging in our patients and specifically understanding what is a proper INR to use or a proper INR range to aim for. And if our patients bleed, how do we reverse these agents? And we understand that FFP carries associated risks in patients on warfarin or patients with cirrhosis. Prothyramine complex concentrates, which we'll touch on briefly, are expensive. Direct oral anticoagulants do have reversal agents, which are not well studied in cirrhosis and are expensive, but have attractive features in that they are direct reversal agents. And so we don't run the risk of hypercoagulable states. And then the duration, and this is a great question that we would love to have an answer for, but still do not really know how long our patients need to be treated for. So when we look at the ASLD guidance on anticoagulants, they note that really it should be individualized and there is no correct choice of what agent to use per se, and that if you have questions, consultation with experts in hematology is advised. Overall bleeding risk, both non-portal hypertensive related bleeding, as well as portal hypertensive related bleeding, appears unchanged in studies that have had control populations. And direct oral anticoagulants are emerging as the common anticoagulant for all patients. However, the data in portal vein thrombosis remains quite limited, and so caution is advised. Now we should mention tips, is this can be a great alternative or adjunctive treatment for portal vein thrombosis in patients that have not responded to anticoagulant, or with patients that have other indications for portal hypertension related issues such as bleeding esophageal varices or ascites. The group from Northwestern has really pioneered the treatment for chronic portal vein thrombosis with cavernoma in patients who have extensive clot, who otherwise would not be a transplant candidate. They have pioneered strategies for portal vein recannulization via transplenic or transhepatic access, which have served as a nice bridge to get patients to liver transplant. So this can also be an option. ASLD guidance addresses this as well. And note that this is a good option in certain patients. However, of course, multidisciplinary management and specific center expertise is required before pursuing this option. Okay, so let's move on from Jim and meet our next clinic patient, Dolores, who is unfortunately a female we're seeing a lot of, and that's a young female with decompensated alcohol cirrhosis. Dolores is male, matches her age, which is 28, and she has poor dentition with multiple infected teeth. She seeks care from a dentist who indicates multiple extractions are necessary and requests pre-procedure coagulation testing with platelets and INR. Now this blood work comes back, the platelet count is 19,000 and her INR is 2.9. As you can expect, this causes no small upheaval in the outpatient dental clinic, and the patient is promptly sent back for a quote, correction of coagulopathy prior to proceeding with extractions. So the next issue we want to tackle is what should our approach be to procedural coagulation correction? Nick, what are your thoughts on this? So this is a great question. I think we all deal with this patient, whether they're getting dental extractions or some other procedure on a daily basis when we take care of these patients. And I think kind of stepping back and thinking about each scenario broadly is probably the best strategy. And so when we're thinking of hemostasis, bleeding, thrombosis, and cirrhosis, when we're worried about bleeding, we want to think as we approach this patient and what procedure they're doing, are we going to move in with a plan for prophylaxis prior to the procedure, or not give prophylaxis and make a plan in case they do bleed afterwards for rescue? We want to think about potential types of bleeding and recognize that they're all somewhat different. And there's different types of bleeding with cirrhosis. In this case, procedural related bleeding, which we'll talk mainly about, but of course, portal hypertensive related bleeding and spontaneous type of bleeding is a risk. Really what's becoming clear and is very important is what is going on with the patient. And this can change really in each individual patient, even on a daily basis. And so it's important to recognize how decompensated the patient is and what other circumstances are going on. Are they on an anticoagulant? Do they have active infection? Are there other comorbidities such as acute kidney injury? And so what we'll focus on now is what we get asked a lot about platelet correction. And this is a very controversial area. And the whole up until now, the assumption really has been that the goal in our patients is to reach a platelet count above this magical number of 50,000 before elective procedures are performed in that this is based on high quality clinical evidence. The reality, however, is much different. In fact, data are very sparse in this area to support this practice. Nonetheless, of course, we often must oblige proceduralists who believe in this wholeheartedly to get procedures done. So I think it's important to understand the different options we have and the pros and cons of both. And so platelet transfusions, of course, we're all familiar with. These have a short half-life. They are difficult to arrange and have some expense associated with them. One of the problems I think that is not answered is when do you administer these? Often we have outpatient procedures. And is it something that you arrange to give in the morning for a procedure in the afternoon? Does that help? Do you recheck the platelets before? I don't think we have the answer to this question. So to fill this gap, TPO agonists have come along and now have been approved. And these were originally designed to offer a more convenient way to increase platelet levels. The randomized trials have shown quite clearly that they do have a nice durable response and platelets increase for a longer period of time. However, as you can imagine, they require quite a bit of planning ahead of time as the patient needs some time to take the medicine to get the platelets up and they can be expensive. The studies have shown a low risk of thromboembolism with the newer agents. However, you should be cognizant if your patient currently has thrombotic risk factors. This is an example of both Avatrombopag and Lusitrombopag in the dosing scheme. If anyone is interested in this topic specifically, I would refer them here to this review, which is quite excellent and thorough on the subject. The randomized control trials for both Avatrombopag and Lusitrombopag showed quite clearly that both agents are effective at raising platelets. However, they did lack any sort of control arm. And so every patient that entered into the study and underwent a procedure either got a TPO agonist or if the platelets before the procedure were below 50,000 was transfused platelets. And so without a control arm, we're left still with the question, does the patient need platelets to be increased in the first place? And it's difficult to answer that question without a control arm to establish a baseline bleeding risk. Furthermore, in these trials, there was extremely low rates of bleeding overall. And they used a mix of procedures. So both low and high risk procedures. And in fact, the majority of procedures in these studies were considered to be low risk for bleeding in the first place. So back to our patient that's getting dental extractions. As we said in the beginning, it's important to become familiar with the procedural related literature. And so recognizing that a dental extractions are probably higher risk to bleed, but they're in an area that can be locally controlled. And so understanding the literature specific for this procedure is important. And we selected one study here that's a very nice retrospective study in a large number of patients with a wide ranging platelet level. And none of these patients received prophylaxis prior to dental extractions. And the authors found that overall bleeding rates were actually quite low. And in fact, they were able to control bleeding events with local measures. And so applying some algorithm or standard approach to pre-procedural prophylaxis would not necessarily make sense in this patient once we review the literature. So the ASLD guidance has new information here on this, on procedural bleeding risk. And they recognize that it's very complex to sort of determine the risk ahead of time. And that really there's no data-driven specific INR or platelet cutoff in which procedural risk is reliably increased. Instead, they say focus on identifying and correcting modifiable risk factors that may potentially enhance bleeding risk. And that overall, given the low risk of bleeding in general for both low and high risk procedures, that it is reasonable to perform these without prophylactically correcting the platelet count in an individualized approach in patients with severe thrombocytopenia as necessary. So where are we at currently with study in this field? And as we keep alluding to, the literature is quite limited. Viscoelastic testing has really emerged as a potential option and better guide to prophylaxis. We listed three of the prospective randomized trials done in this field over the last four years. And on the right is a picture at the top of a Rotem curve in the bottom, a Teg curve. And these are becoming used more and more widely available in some centers. The issue with these studies, however, is that again, bleeding events are rare and we lack standard restrictive arms again. And so while the parameters for viscoelastic testing certainly reduce prophylaxis transfusions prior to procedures compared to standard cutoffs of INR and platelets, without routine control arms that don't get any prophylaxis, we will fail to understand the true baseline bleeding rate. We currently have a study that we're conducting out of UVA and would invite anyone who's interested. It's an observational prospective multi-center cohort study. And what we're trying to do is get a bunch of patients, as many as possible that undergo procedures in the hospital to help identify risk factors for bleeding. Okay, so let's leave the clinic setting and head over to the medical floor for rounds. We're gonna meet Yvette. Yvette's an overnight admission, well-known to our service. She's 55, she has hepatitis C cirrhosis is what brought her in. Her MELD score is 20. Her platelet counts 60,000 and her INR is 2.2. As you discuss the case with the inpatient hospitalist, you make the recommendation that Yvette needs a diagnostic and therapeutic paracentesis. Before you even finish the statement, the little thought bubbles go off in your head and the hospitalist's head about what correction is needed before an inpatient paracentesis. Well, the simple answer is none. The risk of bleeding following paracentesis in cirrhotic patients is extremely tight. This data here of over a thousand paracenteses done in 628 patients, the majority of them with a high INR and low platelet count showed no bleeding events. When this study was published in 2004, we've since entered an ultrasound era where we have ready access to point of care ultrasound, further making paracentesis quite safe to do without any correction. As we discuss Yvette further, the next question comes up about DVT prophylaxis. Again, wholeheartedly, the answer is yes. Patients with cirrhosis are at risk to develop DVT. It appears safe to prophylax them. And what we would recommend is to use guidelines for medical patients, including a good risk assessment model of your choice to help in decision-making. Certainly the risk of DVT prophylaxis in cirrhotics is couched, but there's not a lot of great efficacy data in the cirrhotic population. And there's always the fear of bleeding, but the benefits we feel are much more. There's obviously an increased risk of VTE in these patients. Prevention is highly desired and better than dealing with downstream complications, particularly in our six cirrhotic patient, many of whom are in the ICU. Speaking of the ICU, let's head over there and meet Bart, who will be our last patient to discuss today, who will give us much to talk about. Bart is in the ICU with acute on chronic liver failure and has alcohol-related cirrhosis. His meld is 40 and he has a severe AKI with the creatinine of 3.5. Bart is not intubated. He's quite miserable. He's on low dose pressers. He's needed multiple central lines. He now has significant oozing from those lines. His platelet counts 35,000. His INR is 3.2. His hemoglobin is six and his fibrinogen is low. So what is our strategy to Bart? So what we would recommend is to focus on platelets and fibrinogen. We need to be mindful of accelerated intravascular coagulation and hyperfibrinolysis with an emphasis on the hyperfibrinolysis. This is seen in particularly six cirrhotics where premature clot dissolution occurs and it's clinically evident by oozing and bleeding from mucosal puncture sites. There are disturbances in clot remodeling and factor VIII can be really helpful to differentiate this from true disseminated intravascular coagulation. As we contemplate Bart over the course of the afternoon, he continues to only do worse, unfortunately. He gets further hemodynamic instability and there are concerns for sepsis. The ICU team is paging and asking whether Bart's in DIC. His platelets are decreasing and his INR is up to 4.8. A diagnostic paracentesis that was done after rounds this morning was performed and now by the afternoon, there's profound oozing afterwards. He's taken the CTA, which shows intraperitoneal hemorrhage and unfortunately, no angiographic target. So the issue we wanna take on next is how to approach catastrophic bleeding in the ICU cirrhotic. Nick, help us with this one. All right, I think these patients are extremely challenging to deal with and we recognize that as the patient gets worse and worse and decompensates more, that coagulation balance becomes more and more precarious and can certainly tilt either way into bleeding or thrombosis. And so in this patient that is visibly bleeding and oozing, but has no surgical or IR related rescue strategy possible, it's really left to try to medically support them and optimize them as much as possible. And in terms of optimization, again, it's very important that we try to correct what we can correct. And it's quickly becoming recognized that comorbidities such as acute and chronic kidney disease can enhance bleeding risk. Studies have shown for procedural related bleeding with paracentesis that acute kidney injury is a risk. To explain this mechanistically, translational data has now shown that AKI is associated with low factor 13 levels as well as impaired platelet function. These patients also have higher rates of anemia, infection and really just get hospitalized more and have more complicated hospitalizations with more procedures, which leads to higher risks of all different types of bleeding. Concurrent infection also can stimulate endothelial heparanoids and enhance risk of bleeding. And then interestingly, data is really emerging here on acute on chronic liver failure and its relationship to the hemostatic system. We think now that these patients have different hemostatic systems that are more fragilely rebalanced. So what agents can we consider? And I think the first thing is to talk about what we probably shouldn't be doing. And most of us know this by now is using the INR and trying to correct it with FFP. This doesn't work and in fact leads likely to worse outcomes as the high volumes increase portal pressure, which can complicate patients in terms of risk of worsening variceal bleeding or increased risk of transfusion related injuries. We recommend and suggest evaluating fibrinogen during active bleeding and replacing with low volume products such as cryoprecipitate. Prothrombin complex concentrate is sometimes considered in our patients. The advantages are it's a very low volume. However, it's not that well studied in patients with cirrhosis. It's expensive and does carry a risk of thrombosis and induction of hypercoagulable state and probably should be reserved for more catastrophic type of bleeding. The agents that are more commonly used in this scenario are antifibrinolytics. There's two available in the United States as listed here. The topical use is relatively common at our center, primarily for oral mucocutaneous bleeding after dental extractions. The evidence to support this use is actually quite limited in cirrhosis, but the practice is common and the risks are probably quite minimal. It is actually well supported and used often in patients without liver disease, getting dental extractions who are on oral anticoagulants. The systemic use of these is controversial and studied in the realm of liver transplantation. But in terms of our use in our patient here, I'll point you to a new trial in the Lancet that was published, it's very interesting. So they did a large randomized control trial using antifibrinolytic infusion or placebo in patients who came in with an acute GI bleed. Now this wasn't designed specifically in cirrhotics, but 41% of a large number of patients did have liver disease. And the investigators found that the use of antifibrinolytics did not reduce death from GI bleeding compared to placebo. But what they did find was that patients that received antifibrinolytics had higher rates of VTE events. And for our purposes, what's interesting is that a subgroup analysis showed that the group with the highest number of VTE events were actually patients with liver disease. So when you're using these agents, select the appropriate scenario, which would be one where you're worried that the patient is developing a hyperfibrinolysis state with oozing, but also remain cautious with their use, particularly in extended period of time. Desmopressin is often considered in patients as an adjunctive therapy to support the hemostatic system. The thought is that it enhances platelet function in uremia and releases von Willebrand factor. It's been considered for use in patients with cirrhosis who have concurrent renal disease in uremia. However, the support for its use is actually quite limited in the literature, although it is a low risk medication. This study pictured here shows this quite nicely. And so what they did is looked at a small group of patients with hemophilia and compared them to patients with cirrhosis. In the top graphic you see here, that administration of desmopressin quite clearly increases von Willebrand factor after one hour. When we look at the group with cirrhosis, they already have very high levels of von Willebrand factor, which we already know, and it does not get increased higher with desmopressin. When they looked at other studies pictured below of platelet aggregation, such as surface coverage and aggregate size, while there was significant improvement in patients with hemophilia, there was no significant difference in patients with cirrhosis. So the evidence is fairly sparse in this area. Okay, so let's try to bring this to a head if we can, as we wrap up the formal part of our webinar. I think just in recap, it's important to remember that cirrhosis is no longer considered an absolute hemorrhagic coagulopathy, and that there's a rebalanced system of hemostasis at play. And if anything, a cirrhotic patient is likely to be more thrombotic. We do need to think about platelets. They're obviously important, but we can't define nor really defend a certain platelet cutoff to provide absolute protection from bleeding in our patients. And above all, there's a very nuanced system at play in cirrhotic patients with hemostasis, and we have to have an equally nuanced approach to those patients. So what about our non-GI colleagues? We in hepatology understand this evidence. We have a respect for the best practice within our field, but how do we get this consistently followed? And just working with Nick as we collaborated on this project, I found very similar experiences at two different centers caring for cirrhotic patients and getting things done with coagulopathy. So it can be a very frustrating thing for liver doctors just to get basic care done on cirrhotic patients when roadblocks come up in the hemostasis world. So we're hopeful that new guidance from ASLD will be very helpful in this area. Just some general collaboration tips. So we think it's important to openly discuss coagulation management with your colleagues. I think this is an issue in the management of your patients that you don't want to leave to chance and you don't want to leave to assumption. One way to do this is to write out specific instructions in your note, including using the good old fashioned if-then situation, which I think is very helpful to delineate what exactly should be done if certain situations happen. We've also found it really important to educate outside of your department's conferences. This usually makes for a very juicy and salient talk at Medicine Grand Rounds or radiology conferences to help us spread some of this knowledge that we've acquired about hemostasis and cirrhosis. And this last point has been a little tougher. It's eluded me more in my short career, but find ways to diplomatically correct other providers after the fact. I think perhaps the gentle question technique might be a way to do this, where you bring up, oh, I noticed on our patient from last night, we gave six units of FFP. I was just asking why did we do that? Sometimes that gentle approach can help open the door to a discussion about best practice and allow us to get to a better level in the care of these patients from a hemostatic point of view. So with that, we'll wrap up the formal part of our webinar, and hopefully we can have a discussion about some issues. We want to thank you again for joining us. Happy New Year. And we'll move into the question section. Okay, so I'll start here on the chat. I got a question from Jessica Davis, who asks if someone has a PVT and needs to be on anticoagulation, but has high risk varices, do you delay starting anticoagulation? Nick, I can answer from my point of view first, if you want, and then- Sure. You can go from there. I generally have been more squeamish about this. I like to at least take a look with endoscopy before committing a patient to anticoagulation, but as I've come along, I've kind of seen situations on both sides where patients have bled, despite not using and vice versa. So I've been more open to starting anticoagulation earlier, but at least when I started out, I was more strict about waiting just to get the varices better treated. Yeah, I mean, I think, you know, it depends on the clinical scenario and recognizing that you do have time, particularly if the thrombus is not extensive and there is a potential that the thrombus will spontaneously recannulize. So I really like to give that opportunity to patients before putting them on anticoagulation in general. But if you need to start anticoagulant, there is data, particularly with low molecular weight heparin that says it is safe even to consider banding on anticoagulation. However, this really hasn't been done in any sort of prospective fashion. And I think it is right to have some caution and having that time period where you try to re-image the patient potentially before starting is a good opportunity to introduce prophylaxis. And so that's generally, I think, my approach. Okay. Next question, was there any role for vitamin K at raising INR and decompensated cirrhosis or maybe use of vitamin K for a raised INR? Yeah, I mean, I think from a standpoint of optimizing them or necessarily improving bleeding risk, I don't think there's a great data necessarily to support that. I think the only use really is to see what component really is nutritional deficiency. And this is quite common in patients that are alcoholics. So I think there's little harm in replacing it, but delaying procedures, I don't think is necessary before that. There's another question in the Q&A box about vitamin K. So I think in general, I think we would say that this is kind of a main part of our plan when we work up these patients, but I don't think it's sort of the pivotal point as far as managing their coagulopathy to use vitamin K. I think it's helpful to see, like you said, if there's a nutritional aspect or if there's any correction that could happen, but I don't think it's a mainstay of treatment. We have a question about any upcoming studies with these recommendations for the pediatric population with end-stage liver disease. That's a good question. I don't know the answer to that. I don't either, I don't. Another question here about managing thrombocytopenia and high INR prior to elective band ligation. I think we hit a little bit on this with our discussion about procedural safety. Actually, one of the cases we were mulling over using, and this was a patient, that first patient, Jim, with the portal vein thrombosis who needed elective band ligation. I think the issue here is we had some time with elective band ligation. So in my practice, the most common thing that has happened is arranging a platelet transfusion, but TPO agonists, if afforded and can be set up, could be an option. High INR, I've been less strict about, but that's a tough one with anesthesia in particular. I don't have an absolute recommendation for how to get an INR down. Usually I've not been as strict about following that, but my colleagues often bring it up and that can be an issue separate from the hepatology point of view. Yeah, I think too, just to add to that, I think the ASLD guidance now helps us, I think, in this area a lot. And clearly says, and there are exceptions, of course, there are always exceptions, but for the most part in both low and high risk procedures, for the most part, correction isn't necessary before. Another question about any role for prophylactic anticoagulation in serotics, meaning they don't have a PVT, they don't have any medical indication for anticoagulation, but given the potential risks. I have not seen that. Nick, have you seen anything? No, I mean, Dr. Vila's study that showed with low molecular weight heparin preventing portal vein thrombosis, reduced mortality and decompensations, I think is very attractive. And I think part of what we're waiting for is more literature in this area and more studies, particularly with the siroxaban study, looking at rivaroxaban for that purpose. So in terms of prophylactic anticoagulation to prevent things like portal vein thrombosis, I just don't think we're there yet. Another question, this patient is after liver transplant 10 years now, and they come with a new diagnosis of portal hypertension as evidenced by moderate varices noositis. So they've transplanted, got varices noositis, normal liver function. Imaging shows a chronic SMV thrombosis. What are the options to offer this patient? So, I mean, I have a handful of patients post liver transplant that develop a portal vein thrombosis. And I think the first thing I typically will do is make sure that they don't have recurrent graft cirrhosis and then consider looking in with hematology, looking in for specific risk factors that they may have developed if they do not have non-cirrhotic portal vein thrombosis, because the underlying etiology would sort of dictate the need for medical management with anticoagulants or if any further workup would be necessary in that circumstance. Yeah, I've seen some use of tips for correcting thrombosis, even after transplant. I would have some caution. I've seen some issues, tips post-transplant where it doesn't quite work as well intended as when the patient had the max portal hypertension as before transplant. But that might be something in a multidisciplinary strategy to discuss too, to see if there's any way. But I would be cautious with revascularization from a mechanical point of view after transplant. Some other questions here. What do you do with patients on warfarin for another indication when they come in for elective banding? Nick, I'll let you go if you want. Sure, I mean, I guess we can generalize this question to what we do with any anticoagulant, whether it's direct oral anticoagulant, which is a much more common scenario for me personally as patients are on these for a wide variety of indications and warfarin to much lesser extent now. But in general, my personal practice, and this varies without particular evidence is to have them hold anticoagulation for banding. It depends on the circumstance, but for the most part, we try to do that. I think the counter argument to that that some of my colleagues have made is that your risk of bleeding is days down the road potentially with banding ulcers. And so, when do they restart it? And are we really helping matters by having them hold for that period of time? I think, unfortunately, we just don't have good answers to these questions. And so it's sort of individualized practice. Yeah, I would agree. I think the general preference for me would be to hold it strictly for that banding ulcer concern, but that also can vary if the patient's coming for their third or fourth session, there's obviously a much less risk of that. And then like you also mentioned, the banding ulcer is a delayed complication. And so holding it really doesn't accomplish anything when we restart it right after the procedure. The next question is, when do you suspect a portal vein thrombosis if decompensation is not an indication? I guess, I think that means if there's not a classic hepatic decompensation, I generally have a very low threshold is to look for this on all serotics. I mean, I think that the concern for thrombosis is high. I think as you were showing me with that Lancet paper too, that even just because a serotic has bled doesn't necessarily mean they are less likely to have thrombosis. So meaning, I think as the sicker the serotic is, the higher the likelihood of thrombosis and why our suspicion needs to be that high. So I would say to look for it on all of your serotic patients. Yeah, I think it's just the chicken or the egg scenario. And what is the controversy? Is the portal vein thrombosis causing that decompensation or are you finding it because they decompensated and it's just sequelae at the portal hypertension? And there is, I think one last question here about ascites and paracentesis, any recommendations to correct coagulopathy? Again, I would say this does not need to be corrected on a traditional general basis for paracentesis. I think paracentesis is a very safe procedure and doesn't require correction of coagulopathy prior. Okay, good. So what about the use of the prothrombin complex concentrate before a procedure such as ERCP in serotic patients? I mean, so the data is actually fairly sparse and specifically when you look at procedure studies, so ERCP studies, the highest risk will be patients who get sphincterotomies. And again, like any procedure where you're cutting, I don't know if that's going to necessarily reduce their risk of bleeding. And certainly there's no evidence to support that practice. I think prothrombin complex concentrate should be reserved for active hemorrhage. I don't think that needs factor replacement. I think there would be a risk of thrombosis or very expensive. I would agree. The times we've used PCC has been in a very sick ICU, serotic with active bleeding. I think that even the logistical and cost of arranging this before any RCP may even prohibit its use. So I think it's not typically used for that type of setting, even just clinically. So the other questions are about HCC, a good topic. So a patient with decompensated cirrhosis, worsening portal hypertension, they have HCC with tumor in vein and bland thrombi is AC beneficial. And then of course the other questions, HCC and DOACs and HCC and PBT. I'll say generally from my practice, if I see tumor in vein and a serotic with HCC, I would recommend anticoagulation in an HCC patient. Yeah, I think I agree with that. Another question here about optimal dosage of low molecular weight heparin for portal vein thrombosis. Would you lower dosage in a child's B or C serotic? That's a good question. I don't think we really have any understanding of what the dosage should be in these patients. So I mean, not only is it hard to dose from that perspective, but also when you're getting into decompensated serotics and you're going by weight-based dosing and the weight fluctuates so much with fluid, I hardly ever use low molecular weight heparin. That's one of the reasons I find it very difficult to dose, but for the most part, simply go by package insert guidance. And I think twice a day dosing, but in child PUC serosis, they would be fairly decompensated and high risk for any anticoagulants probably. And likely to have renal failure concomitantly as well. Last question we can take here, in a patient on the wait list for transplant, if we diagnosed an intrahepatic portal vein thrombosis in the portal branches, do you start anticoagulation or would you wait a little? I would say to discuss this in your committee, but also likely lean to wait if it's intrahepatic and small. Yeah, absolutely. I think, you know, first do no harm. And I think giving that time to see how it's going to develop, because there's a good chance that it could regress and go away without any treatment, but it is important for sure to follow it up. Okay. There is one more question that just popped in. I guess at some point I'll have to put the hard stuff. This is interesting. So intracardiac thrombosis during liver transplantation, obviously underestimated, but a catastrophic event. It's controversial. Some authors stop fibrinolytics, other give heparin, others give TPA. What would our experience recommend to do for intraoperative intracardiac thrombosis during liver transplantation? I know one thing at our center, TEG is used for these patients in the OR, which may be more helpful as far as guiding what to do. But I think beyond that, I don't know if I have enough experience to comment specifically on how to manage that. No, I mean that, you know, the sort of cop-out answer is the multidisciplinary sort of team approach. But in that case, that sounds, you know, like it would require cardiovascular hematology and anesthesia input, but, you know, minimizing antithrombotic or sort of like antifibrinolytics or anything that's going to make it more procoagulant and trying to get by if you can't anticoagulate. But those are really challenging cases. Okay. That was the last of the questions. Thank you everyone. Have a great day.

coagulation management

cirrhosis

bleeding

portal hypertension

hemostasis

platelets

thrombin generation

treatment options

collaboration