Good morning, good afternoon, and good evening to everybody. My name is Christina Rappoll. I am a physician who works in Jena University Hospital in Germany, and it's my pleasure to moderate this fascinating discussion that we will be having regarding endovascular versus endoscopic management of gastric viruses, which is better. I have only a small announcement before we start, and that is that I would like to announce the Portal Hypertension SIG program at DDW. This is an ASLD-AGA combined symposium titled EU's Guided Hepatic Interventions that will be had on Monday, May 23rd. You're all invited to come. It is now my pleasure to introduce Wael Saad, who is currently Senior Physician, Scientist Professor, and Staff Clinician at the NIH Bethesda, Maryland, and is one of the early adopters and advocates for the BRTO procedure in the USA. Wael, we're looking forward to hearing your talk. Thank you, Christina, and thank you for the invitation, ASLD. I'm going to focus on endovascular management of gastric viruses, focusing on BRTO specifically. So what's really at stake here? Gastric viruses have a high mortality. The first-time bleed is about 15%, cumulatively a second bleed is about 30%, and the third bleed, according to Sori and Natal, is about 45%. IR intervention radiology usually kicks in as a secondary prophylaxis after a sentinel first-time bleed. There is some experience with primary prophylaxis, but that's usually amalgamated in retrospective studies with secondary prophylaxis, usually for high-risk gastric viruses. IR is not only BRTO, but it's also TIPS, partial splenic embolization, and other procedures as well that are tailored to the patient's needs and the patient's anatomy. That will not be my focus today on the conversation, but primarily the outcomes of BRTO and the general concepts. So here we go with the concepts. BRTO, conventional BRTO, is approaching the varices from the systemic side of the house. A BATO is approaching the varices from the portal side of the house. And then you have what's referred to as accelerated BRTO, which is still retrograde obliteration and is subclassified into backloading and frontloading. It's not necessarily the acronyms of PARTO and CARTO and FARTO, depending on whether it's plugs, coils, or filters that I use, and we'll uncover that all. Here you have the inflow of the spleen, the splenic vein. You have the inflow to the portal vein is the guts, and this is the portal vein, and the outflow is actually the liver, the hepatic sinusoids. In the presence of a large gastric variceal system, you have reversal of flow in the splenic vein, and sometimes in a very large system, reversal of flow in the portal vein, going feeding through an afferent feeder, at least to the gastric varices. And then in 65 to 85% of the cases, there is some sort of poor systemic shunt, commonly gastro renal shunt. A BRTO is approaching from the systemic venous side of the house, typically the left renal vein going against flow in the gastro renal shunt, blowing up a balloon, and then putting the sclerosants and or gel foam retrograde into the system without overflowing into the portal circulation. And then blowing that balloon up for hours, up to 36 hours, a minimum of 45 hours, and that's conventional BRTO. A BATO is approaching from the portal venous side of the house through the afferent feeder, sclerosing antigrade into the shunt without overflowing into the systemic circulation. And that's through a trans tips or a trans hepatic, percutaneous trans hepatic approach, which the Japanese refer to as a PTO, percutaneous trans hepatic obliteration. And here with the same drawing, what you have is either a posterior gastric vein, a short gastric vein, and those are left dominant gastric variceal feeders, or the left gastric vein, which is a right dominant gastric variceal feeders. And the varices are actually very superficial in the submucosa. All of this is not the varices. This is actually part of the gastro renal shunt. This is the convoluted portion of it, and this is a straight portion of the gastro renal shunt. So the varices are very, at the very tip in the submucosa. If you flatten this whole area out, which is the gastric variceal complex, which we refer to, you have a central hub, which here is defined as a false gastric varices. That's the convoluted portion of the shunt, and this is the straight portion of the gastro renal shunt. Afferent feeders coming in, and then you have a perforator to, through the stomach wall, to the submucosal trans, true gastric varices. And that's what's referred to as the gastric variceal complex. And here you have gastric varices very, at the very top here, the convoluted portion of the shunt, the straight portion, and then it meets with a common stump with a left adrenal vein. And this is an example drawn out. Common stump with the left adrenal vein, the gastro renal shunt proper coming down to it, and then entering the left renal vein. And this is where you need to put your balloon and or your plugs. In the gastro renal shunt proper above the stump, where there's typically a web here, where that's the choke point, where you actually work the system with, with the balloon and or hardware, if you're going to use accelerated BRTO. And here's an example of one going from a femoral approach, finding that stump, getting into the stump, and then getting into the shunt itself proper. Here you have the actual common stump, the left adrenal vein, and this is the gastro renal shunt proper. Going into collaterals, if it's a large collateral, coilets, because you get escape of sclerosant from there, blow up the balloon, the balloon here is full of air, so it's whites. You pull down the balloon to cork and choke the entire system, fill it up with contrast as a rehearsal before you put the sclerosant, and then you put the sclerosant. So to your, to your left is contrast, to your left, to your right is the foam sclerosant. The gastric varices are up high. This is the convoluted portion of the shunt, and this is the straight portion of the shunt. So it's a little background in history. Sclerosis of varices started off with Lunderquist in the 1970s. Prior to the TIPS era, they used alcohol and glue and were not very successful clinically in stopping the bleeding for long. Olsen et al. out of Indiana is the first description for gastric varices using alcohol, and the actual concept is actually, is by Olsen et al., but did not point the term BRTO. Kanagawa and the Japanese in 1991 to 1996 actually perfected this technically and introduced it clinically and coined the term BRTO. The Americans continued kind of the dabble with, with trying to sclerose gastric varices using a different sclerosant. So we did not have 10% ethanol aminolate, we had alcohol and glue with very variable results. However, in 2006, 2015, that's where the real expansion of the practice occurred in the United States. And that's because of the advent of the introduction of Sotrodecol, which is a detergent sclerosant. In Europe and in Asia, the alternative is polydecanol. And with this new sclerosant that we're able to feasibly, technically feasibly actually do the BRTO, we were able to actually expand this practice in the United, in the United States. Here's kind of the outline of the milestones. And in 2012, the American College of Radiology identified BRTO as a viable alternative with just a broad, you know, certain clinical and anatomic scenarios we did not really define and we're still haven't defined it appropriately with, with, with randomized controlled studies. In 2013, a lot of activity accelerated BRTO, one-year outcomes in the United States, two-year outcomes of TIPS and BRTO, 2015 pediatric population. We actually demonstrated that it's feasible to do BRTO in the pediatric population. In fact, to the best of our knowledge, we could actually not find the pathology in the entire literature as far as gastroenteritis with shunts in children. So this, this particular short case series was not only a BRTO first, but I think the first description of gastroenteritis shunts and associated gastroenteritis in children. And in 2017, outcome in liver transplants was also described. 2018 was the first talk we gave to ASLD in Washington, DC, where it started to really become kind of mainstream, mainstream practice in the United States. So what has accelerated BRTO? In the, in the advent of, in two countries, South Korea and the United States, there was an alternative to BRTO, which was TIPS. That was the mainstay. TIPS was done within an, you can do a TIPS within an hour and a half, two hours. BRTO was very taxing. And so we, for advocates of BRTO, we need to accelerate this to actually compete with, with TIPS. So the Koreans and, and the Americans almost at the same time in 2013 came up with the concepts of the accelerated BRTO. In accelerated BRTO, the two real subtypes are backloading the hardware or the plug and front loading the plug. So BRTO, like we said, it was a balloon up, and now we're going to, I'm going to talk to you about backloading the plug first. And the whole idea is to, how do you replace this balloon with hardware, with alloys, coils and or plugs? So backloading the plug is to actually go up against flow, blow up a balloon. There is a balloon. Do the sclerosant or the gel foam, and then coaxially put the, put the hardware, the plugs and or coils, and then deflate the balloon and pull out. You're committing to acceleration of the BRTO at the tail end of the procedure and not at the front end of the procedure. And that's why it's called backloading the plug and needs a coaxial system. I'll show you an example of that. Front loading the plug. Front loading the plug is that there is no balloon. They never will be a balloon. They actually start off with a plug and or coils and have the capture adjacent to the plug and sclerose above the plug, and then detach the plug and or coils at the end and pull out. There is no balloon. And that's backloading. This is the coaxial system. It's a large system, what we call a balloon occlusion guide captor that takes up to a 16 millimeter ampli-ax coaxially. We go in with it. Here's a, the balloon is up. This is contrast. This is posterior gastric vein. We now sclerose it with foam. The balloon still is up and then coaxially, we go in with the plugs and deploy the plugs above the inflated balloon and then deflate the balloon and pull out. Again, like we drew out for you initially, the common stump is actually clear. The choke point is a web right here. The balloon was here. The hardware is all in the gastrointestinal proper above the common stump with the left adrenal vein. Contraindications to accelerated BRTO, no real contraindications. Be careful with patients that have, that are less forgiving when it comes to embolization, either through shunts, like a right to left shunt or patients with poor cardiopulmonary reserve that can't handle PEs, but it is a higher risk for embolization in systemic circulation. Outcomes. This is just an analysis of just adding cases, retrospective cases, unfortunately, that are, but have similar definitions and similar methodology. The technical success is high. It's about 92% in 457 of the RTO procedures. The obliteration rate is also high. It's about 86%. That's by imaging and or EOS. When it comes to complications, this is complications from 25 years of Japanese and Korean literature, mostly using ethanol amino leates, and I'll, I'll, I'll do the variations with Sofuteko at the tail end of the slide. Anaflax is mostly, this is for ethanol amino leates. Also the gross hematuria is ethanol amino leates and renal failure also associated with ethanol amino leates. In the United States, you do get renal failure if it's part of multi-system organ failure. About 3% will get the DIC along with it, and then multi-system organ failure up to 6% of the, of the cohort. Having witnessed and participated in 200 BRTOs, over 60% of them I've done myself, I can tell you that gross hematuria with Sotradekyl still occurs in majority of patients, usually without any clinical significance. Pomeranbulism, one case out of the 200 where a patient transiently desaturated and then, then came back. Cardiac arrhythmias could be PVCs, VTACs, or bradycardia. This case was a bradycardia almost to a systole in one case. DIC about 3% and, you know, pulmonary hepatic failure as part of multi-system organ failure is about 6%. Oral vein thrombosis is about 0.5% with Sotradekyl. To add to this list, unfortunately, stroke where the patient actually decorticated on the table, died within 72 days. Gastric variceal rupture without consequence occurred in maybe three to four cases, although there have been reported, not in the published literature, but there have been reported cases where the patient actually exsanguinated on the table when the balloon is up before they sclerosed. Clinically, grossing material is almost always there, and almost over a third of the patient actually gets some form of embolization into the pulmonary circulation or the portal percussive circulation. When it comes to controlling the bleed, it's about 91 to 100% control of bleeding. That really means nothing in retrospective studies. You'll get that in the same with the tips, because most of these bleeds are intermittent. When it comes to aggravation of esophageal varices, that's very rare, very real, up to 24 months, it's up to 36% and bleeding from these esophageal varices, because you're compressing the system. You're losing a poor systemic shunt as part of the burrito. The bleeding is 24%, in some cases, up to 30%. Portal hypertensive gastropathy is about 5% to 13%. Subclinical ascites, 44%, clinically significant ascites, about 15%. Splenomegaly, one third of the patient will actually grow their spleen more than already large without any effect on thrombocytopenia, which they already usually have. Rebleeds, the common rebleed sites or sources would be portal hypertensive gastropathy and esophageal varices. Looking at the U.S. experience with Sartre-Decaux, there are 100 patients undergoing 103 BRTOs and 18 patients underwent BRTO with coexisting tips. Excluding those 18 patients, focusing on the BRTO cases only, the results are exactly the same, very similar to the Japanese literature of 20, 25 years. Mortality is 90-day mortality, about 9%, mostly DIC and or hepatic failure. Looking at survival, emergent versus elective, those impending bleeds, that's the primary prophylaxis group, 68 patients undergoing elective BRTO after a seminal bleed but controlled, not active bleeding, and 18% with active bleeding, and that's what we refer to as emergent. Statistically significant survival for those undergoing elective BRTO versus emergent BRTO. We looked at multivariant analysis to predict survival for 24 months, looked at the MELD score, synthetic function, presence of ascites, HCC, sites, operators, encephalopathy. The only thing that predicted, the only two things that predicted survival at 24 months or poor survival at 24 months was the diagnosis of HCC at the time of the BRTO, they died from HCC, or a child, B or C patient, or MELD under 13 was considered an indicator of poor survival at 24 months. With intent to treat, life table analysis, the re-bleed rates of BRTO from gastrovarices is 3%, that's a failed BRTO. In other words, with successful BRTO, the re-bleed rate at 24 months is zero, nothing beats this. Esophageal varices re-bleed is 13%. This is a very good result compared to the Japanese literature, and the reason for this is this is not just BRTO and IR boasting this outcome, this is a multidisciplinary approach with GI and with hepatology staying on top of these esophageal varices, preemptively treating the esophageal varices and staying on top of aggravation of esophageal varices. Here is in a case with a wide open tips and bleeding from the gastrovarices with a gastrointestinal shunt, and this is one of the reasons, one of the theories behind why tips is not as effective in managing gastrovarices compared to esophageal varices, and here you have the gastric shunts alive and well, varices are alive and well bleeding, and this is the tips with coiling. And the pet peeve of mine is that if you're going to do a tips with gastrovarices, you just don't only coil, but you have to sclerose the entire system because it will recruit other afferent feeders and continue to survive with this gastrointestinal shunt. And so what happens with tips is that you need to look at these shunts, whether they're man-made shunts or spontaneous shunts, as black holes in space or sink holes. It is not only the size of the shunt, but it's the vicinity and the proximity of the shunt. The closer you are to the shunt, the more you're going to be influenced by this shunt. And so in this particular situation, you have two competing photosystemic shunts, an artificial one, which is the tips and the gastrointestinal shunt, that spontaneous one competing over one another. The theory behind this is that if you look at this run, wide open, wide open tips, the afferent feeder to the esophageal varices compared to the gastric varices is way off for the gastric varices. The distance to decompress EVs is half the distance from the tips compared to double the distance for gastric varices. And in the presence of a gastrointestinal shunt, what you have is a schism, a sphere of influence on the left side from the right side of the portal circulation. And so tips has very little effect in decompressing gastric varices, especially in the presence of a large gastrointestinal shunt, but that's that's the kind of hemodynamic theories behind this. Looking at the need for tips after BRTO, it's about 20 percent at three years, and most of this is varices early on or ascites. Early on, you'll have a peak of ascites and portal hypertensive gastropathy, later on the indication for tips is EVs followed by ascites and hydrothorax. And after 12 to 18 months, it's really progression of disease and it's not really associated with your BRTO itself. Bringing back those 18 patients that had BRTO and TIPS, they had even better results. There's no re-bleeding of any source at 24 months, and there's protection of development and worsening of ascites and or hydrothorax in these patients. Definitely better results with combining BRTO and TIPS or batulin TIPS with statistically improvement when it comes to ascites. This is actually all ascites, not only clinically significant ascites. This redirection of flow that you see after a BRTO increases the pressure, and we noticed that there's actually growth of the liver in 10% of the patients, mostly if you have a MELD of under 14. So if you come in as a child A, there's most likely the improvements in the liver function and more likely to grow your liver. The improvement in the function varies in literature because part of it is depending on the cohort, if your cohort is dominant child B or child C, less likely to improve, but if your cohort is dominantly child A, you're more likely to improve. On top of that, this is a cohort where it shows improvements in the MELD at four months in the same cohort, no improvement in the child Q because it's masked by the ascites. So in general, improvement of hemodynamic effects, the positive things are actually flow driven, the negative things are actually pressure driven, even in the presence of ascites and hydrochlorics, they occur, they're mostly on product because they occur with improved synthetic function. In an analysis, and this is a retrospective meta-analysis and not a prospective randomized trial by Osman et al showed that BRTO had the lowest risk of re-bleeding compared to beta blockers or endoscopic management. Beta blockers had the highest mortality, non-specific beta blockers alone had the highest mortality compared with any other endoscopic and or endovascular management. However, this did not translate into survival. In conclusion, it's a high stakes, high mortality disease process, try to do the best to reduce the risk of that second or third bleed. These are the kind of descriptions of what is a BRTO and a battle and accelerated BRTO as we described. BRTO is very effective with very low re-bleed rates. The need for tips is about at least 20% at three years, mostly due to ascites and hydrothorax early on, it's because it's the tips indication is very sealed bleeding. There are good things and bad things about the redirection of flow. The good things are flow driven, the bad things are on cautic or pressure driven. If you show up as a patient, there's a child A with a melt under 13, you're more likely to grow your liver and more likely to improve your melt. BRTO is superior to tips. There are theories behind that, especially in the presence of a large gastrovenial shunt. It is a high stakes game. So BRTO is definitely better than tips in the retrospective limited studies that we can glean from. However, it doesn't translate into survival. And that includes the randomized control study, which Zach is gonna talk about next. Thank you very much. Thank you very much, Dr. Saad, for this very interesting overview and especially interesting the interventional radiologist perspective that we, or at least we as hepatologists normally don't have. I forgot to say at the beginning that we will do a common discussion for both talks at the end of the second presentation, since we have on one hand, an interventional radiologist and on the other hand, a hepatologist. So that please put your questions in the question and answer section, and we will discuss this afterwards. So it is now my pleasure to introduce Zach Henry, who is a transplant hepatologist at the University of Virginia and has a special interest for portal hypertension, of course, and gastric viruses, and has published articles on the topic, including a clinical practice update from his own lab in the United States. And also a clinical practice update for the AGA regarding the management of gastric viruses. He is going to talk to us about the hepatological or endoscopic management of gastric viruses. Zach. Thank you so much. So yeah, I'm gonna try to focus on the endoscopic side of things, but we'll definitely reference some of the points that Dr. Saad just mentioned. So I have no disclosures. I do have to mention that pretty much every therapy I'm going to discuss is off-label for endoscopic management of gastric viruses. And the way I think about this is kind of the two different scenarios in which you'll encounter this, which the first is just the incidental presentation where you see viruses on a screening endoscopy, and then the second is the actual active bleeding event. Just some definitions of abbreviations I'm gonna include throughout the talk. I think the top one is most important. ECI just stands for endoscopic sinoacrylate injection. I don't like to use the term sclerotherapy just because of ethanolamine and kind of the nasty history of ethanolamine with some of our endoscopic procedures. So that's what I'm getting at with ECI. I think many of the rest of these are kind of more commonly used abbreviations. So the incidental presentation, this is just a patient of mine, 52-year-old with decompensated cirrhosis. The decompensation was ascites, previously had banned ligation of his esophageal varices for primary prophylaxis, and was noted to have this gastric varix kind of on the greater curve of his cardiofundal region there. It was very large, had not had any bleeding from this, had never bled from esophageal varices either. And so when I see this, I often ask myself, what is the next step? And so what we have to figure out is really what is his bleeding risk? And for esophageal varices, we have nice criteria that have been defined for well over 30 years and have been the basis for a lot of our primary prophylaxis studies in that group. For gastric varices, we don't quite have the same data. And so the serine classification seen here on the right is the most commonly used classification system pretty much everywhere in the world. And while it is somewhat descriptive, it doesn't fully convey a risk of bleeding. In this study, most of the varices were along the lesser curve and were contiguous with esophageal varices, what are referred to as GOV1. A much smaller portion were cardiofundal gastric varices, but those did have more severe bleeding when bleeding occurred. Around the same time that Dr. Serin published his kind of anatomical classification, a group from Japan actually looked at endoscopic risk factors for bleeding. And here in the top right on table one, similar to esophageal varices, they found that larger gastric varices in the cardiofundal region, what they called F2 or F3 had a higher risk of bleeding and that a high risk spot or a red colored spot on the varix also had a higher risk of bleeding. Unfortunately, they didn't include any measures of liver function in this study. And so understanding the impact of whether a patient's compensated or decompensated was hard to discern. Now, fortunately, a few years later, a different group did exactly that. They looked at a large cohort of patients with cirrhosis and portal hypertension. And in this particular case, they specifically were looking at cardiofundal gastric varices. They actually were excluding the GOV1 along the lesser curve. And similar to the criteria that we use for risk with esophageal varices, they found that size, child abuse status and a high risk mark or a red spot on the varix were all independent risk factors for bleeding. And actually at the bottom, I really liked this table because through that predictive model, they were able to create some probability tables of putting all of these things together, what someone's risk might be to bleed basically within the first 12 months after diagnosis. Unfortunately, even though this was 25 years ago, a study like this has not been repeated. And so it hasn't been validated outside of this one sample, but it's the best data that we have so far for this. So based on that, this is the only randomized controlled trial of primary prophylaxis for gastric variceal bleeding. This was done in 2011 in patients, again, with only cardiofundal varices and patients that had previously had eradicated esophageal varices or no esophageal varices. And it compared three groups, a group of sinoacrylate injection to a group with propranolol to a group with no treatment. And you can see just very briefly at the top, the sinoacrylate group had the lowest rate of initial GD bleeding and then propranolol and then no treatment and complications were not vastly different. And what we got out of this is that statistically in this graph in the middle for bleeding, sinoacrylate injection was significantly better than both non-selective beta blockers as well as no treatment. However, non-selective beta blocker was not actually superior to no treatment in this group. And then when you look at survival, sinoacrylate injection was better than no treatment but not better than beta blockers. And so what I took from this from a primary prophylaxis perspective is that in patients that have cardiofundal varices that are large, which that's been defined as greater than 10 millimeters in size, that have a high risk mark and or have kind of decompensated child puberty cirrhosis, those patients may be at a higher risk to bleed and may deserve primary prophylaxis. But based on this study, sinoacrylate injection seemingly is the best primary prophylaxis. Though similarly, I have not seen this study repeated either. So for my patient, I have a patient with decompensated cirrhosis with ascites. He was child P class B, had never bled. He did have EVs, but he's got a large gastric varix that's larger than 10 millimeters. It's questionable whether or not, if you see on the right bottom side of that varix, it looks like there's a little red spot. I thought that was potentially a high risk mark. And so I felt like he deserved primary prophylaxis. And the question was whether or not to inject sinoacrylate or to try like a BRTO procedure, because yes, there is data on primary prophylaxis for BRTO, but it is case series, it is uncontrolled data. And so it's kind of difficult to make a determination of which one is the best. In this particular case, although the patient had a gastro renal shunt, he also had a massive splenic vein thrombosis. And we felt like shunting a bunch of blood back through a splenic vein that is predominantly closed down was not a good idea. So we did proceed with injection, which is the picture on the top left. And then the picture on the top right is kind of at the end of the procedure. We did a couple of different injections there. The picture on the bottom left is about a month later with the glue cast coming out of the bottom of the varix. And the picture on the bottom right is a year after the procedure with the variceal bed largely involuted now and the last little bit of glue cast being extruded into the gastric lumen. And this glue cast is very hard. I mean, it's almost cement-like, and it slowly gets pushed out as the vein in the submucosa sclerosis down. And it just mixes with everything else and comes out in the stool. So there are some risks to sinoacrylate injection. Generally, these are all pretty low risk. I've included kind of just four studies with the largest sample sizes of patients. Confirmed infection, kind of one to 6%, but this Caldwell study of 92 patients, no infections were noted. And in our experience here for 25 years, having done about 120 procedures of glue injection, we have not had any infections. And so I'm not sure why there's such a large discrepancy. Embolism is a big concern with glue injection, but actually rates of distal embolism, especially pulmonary embolism is actually very low. And in this top row, the largest study with over 700 patients found incredibly low rates of pulmonary embolism. In older studies, a lot of time pulmonary embolism was defined simply on a radiograph. They would mix the glue with lopidol, which shows up on an X-ray. And if they saw a little dot of lopidol in the lungs, they called it a pulmonary embolism. But in many cases, it was clinically insignificant and it wouldn't have been diagnosed without that. And so it's probably unlikely that there's much risk for a clinically significant embolization, although it does happen. So the acute presentation is very different because this is what I think scares most of us in the middle of the night. In this patient compensated, no esophageal varices. On the left had been an endoscopy from one year, almost one year to the date prior. And on the right is how this patient came in vomiting bright red blood. And we did the endoscopy and see an active stream of blood coming out of the gastric varix. And so initially we all need to remember just the general approach to anyone with variceal bleeding or suspected portal hypertensive bleeding. You know, use of vasoactive drugs like octreotide, antibiotic administration, stabilization, intubation if needed, a very restricted blood transfusion strategy. And then doing endoscopy as soon as we can after the patient stabilized, but at least within 12 hours. And during the endoscopy, we got to kind of define what we see because what we see endoscopically, we're gonna ultimately need to compare to what is seen radiographically. And again, I have the sarin classification here, but in blue boxes are descriptions that personally I prefer where cardiofundal varices are a little more descriptive and lesser curve. Those are the things that I'm trying to distinguish when I do the endoscopy. And here on the picture on the left, the blue area is generally what I would call cardiofundal. And then what's behind the endoscope when you're in retroflexion, generally in this position is the lesser curve. And the reason I segregate these is because the lesser curve as sarin showed with GOV1 are generally contiguous with esophageal varices being fed off of the coronary vein and can be treated like esophageal varices. Whereas cardiofundal varices, like what we have bleeding in this picture are kind of these large bulbous lakes of blood vessels, which Dr. Saad showed with excellent pictures and typically need to be managed differently. So in the moment, we kind of just have to do what we do to stop bleeding because we don't want this patient to bleed out. So band ligation is an option. It has been studied. One of the issues is the gastric varices are very large. And so like the one in the picture on the top there, if we try to ban that, it's gonna end up looking like a snowman. And after a few days, when that band falls off, there's likely to be a large ulceration and re-bleeding. Sclerotherapy using ethanolamine or other alcohol-based substances can definitely get this to stop bleeding, but it causes such severe ulcerations and it actually causes acute early re-bleeding of gastric varices that it is not recommended. And I always include this at the bottom because somebody got creative and decided to clip this. And if you see a spot like that bleeding and you can get a clip on it safely and accurately, as long as you don't stop with the clip, I think you're okay as a temporizing measure. You just gotta make sure you don't lacerate the varix with the clip while you're trying to do it and make the bleed worse. My personal preference though is just to use a gastric balloon. Now, whether it's a Blakemore balloon or a Linton-Nachlis balloon, either way, getting a balloon down and blown up and pulled back up against the GE junction will basically provide tamponade and can stop the bleeding and give you time to do imaging, have a multidisciplinary discussion, and then choose what the best definitive therapy for this particular patient is going to be. I will say this reference at the very bottom of this slide, we do go into more detail about placing these balloons and doing it actually endoscopically at the same time that we're doing the diagnostic endoscopic procedure. And those are what those pictures are from on the right here. So if you wanna read more about that, that reference down there kind of goes into more detail. And then there's always this question, if we're doing this kind of temporizing therapy overnight when the patients present, is it ever definitive? And for lesser curve GVs or GOV1, I think the answer is yes, because if I see this kind of varic bleed, I'm actually just gonna go back down with a bander. And instead of starting banding in the esophagus, I'm gonna go down two or three centimeters below the GE junction and start banding back up into the esophagus. These, like I said before, are generally fed by the coronary vein, just like esophageal varices. And the approach should be very similar. And you can see that down here in this picture from Netter's anatomy atlas of kind of this left gastric or coronary vein feeding into the bottom of the esophagus and into the lesser curve. Once we stop the bleeding though, we're not done. In fact, we're just getting started. And the importance of getting imaging is to try and define all of the shunts and anatomy that Dr. Saad was referencing, because if we're gonna decide between glue injection, which when I say that, I mean sinoacrylate, or BRTO or TIBS, we need to understand, is there a shunt present? What are the inflow vessels? So right, the afferent vessels. What are the efferent vessels taking blood out through the shunt? Is there a thrombosis in the portal vein or the splenic vein? But then also as the hepatologist, I have to have a good idea of what are their esophageal varices like? What is their ascites like? As Dr. Saad mentioned, BRTO can exacerbate other complications of portal hypertension, just like TIBS can exacerbate or create hepatic encephalopathy for some patients. So we have to take all of this into account and really have that discussion between hepatology, interventional radiology, and depending on where you work, interventional gastroenterologists. So at the end of the day, when we decide on a definitive therapy for cyanoacrylate, there have been well over 50 studies over the last 30 plus years looking at cyanoacrylate injection for gastric variceal bleeding. Unfortunately, none of them are at all alike. They're mostly case series and case reports. They use different cyanoacrylate agents and different volumes. Some of them use lopidol. Some of them don't. Some of them are EUS guided where they've added a coil. And so, understanding kind of what the standardized method of doing cyanoacrylate injection really is is kind of unclear at this point. So, on the right, we tried to compare this here. Like I said, we have about 100 plus cases of cyanoacrylate injection over the last 20 years. And we have looked at it with and without lopidol and have actually found out that without lopidol, we inject higher volumes of glue itself, but we do less injections, less total procedures and are trending towards better procedure success, though re-bleeding is really no different. And so, I think in the long run, while I believe in cyanoacrylate injection, we do it here, I do think as a larger community, we need to standardize this a bit better. Generally, one year re-bleeding rates after cyanoacrylate are 4 to 20%. This is highly variable, primarily based on the variability in doing different injection methods. I can tell you our rate has been generally a little bit on the lower end, probably like 5% at a year. And beyond a year, it's hard to know how much is due to the original bleed or if there's now a secondary process, a new shunt, a new varix forming that's causing a new bleed. For acute hemostasis, cyanoacrylate is wonderful. You'll see in a minute that other countries actually use cyanoacrylate in the middle of the night for acute hemostasis. In America, we don't really have like a pre-made kit for this, so drawing up glue in the middle of the night to inject is very cumbersome, which is why I didn't really include it in this temporizing measures, but if you have the availability to do that, that is probably the best temporizing therapy as well. And as I mentioned, there are studies of EUS guidance for glue injection using coils or without coils. I don't know that we're there yet. I don't know that that's necessary. I think rates of embolization are so low that I'm not sure the added expense of EUS is worth it, but again, something that needs to be studied in the future. There is not a lot of comparative literature between cyanoacrylate and band ligation for gastric varices, but there are a few randomized control trials, three that were included in this meta-analysis, and they did find that cyanoacrylate was better to stop the acute bleed and also to prevent re-bleeding the band ligation, but in the long term, there were really no difference in complications or long-term mortality. So how does this compare to TIPS? So if you look at this study, this is from 2007, and this is a small study, but they did randomize these patients, and this looks like TIPS is much better at preventing re-bleeding than cyanoacrylate going out to three years, although on the right, you can see again, no real difference in survival. However, this study combined GOV1 and GOV2, and the problem being GOV1, as I said before, are contiguous with esophageal varices and physiologically are very much the same, and so it makes sense that TIPS would work better for that group, because TIPS works really well for esophageal varices, and if you look, the primary difference between the two groups was found in GOV1, and GOV2, there really wasn't a statistically significant difference, and so I would say that for cardiofundal gastric varices, TIPS and cyanoacrylate at a minimum are equivocal to stop bleeding, and the one thing we always kind of have to worry about with TIPS, as shown here, is there is an increased risk for hepatic encephalopathy, which they did find in their randomized control trial, and on the right is actually a study, now this is a retrospective cohort that we published here from UVA comparing TIPS to cyanoacrylate, and for survival and re-bleeding, they were actually, again, equivocal, but when you look at morbidity, TIPS has significantly more morbidity, and this was almost universally related to hepatic encephalopathy and readmission, just related to the TIPS. So lastly, how does this compare, and let me go back actually real quick, beyond this, there actually is not a randomized control trial comparing these two outcomes. I do think TIPS, because it acutely drops pressure in the system, will stop bleeding, and depending on the underlying vasculature, it may stop and prevent re-bleeding. If it's a feeder off of the left gastric vein, similar to esophageal varices, this could be very helpful, but this is actually another good reason why, once you stop the initial bleed, getting imaging, discussing with interventional radiology, and having that conversation is so important, because you don't want to just TIPS everybody, because if they have this completely separate shunt coming off the spleen, going through a varix, down to the left renal vein, a TIPS is unlikely to prevent that gastric varix from bleeding again. In that case, you know, BRTO, or glue injection, depending on what's available, may be the best therapy. So, for many years, cyanoacrylate was compared to BRTO through large retrospective cohort studies, and I'll be honest, because we've published one from UVA, they generally show that BRTO is superior to cyanoacrylate from a re-bleeding perspective. Last fall, in hepatology, is the first randomized controlled trial published comparing these two for cardiofundal gastric varices with acute bleeding, and the patients were either in the hospital where the study was being done, or they were transferred from an outside hospital, although I have to say, it was transferred, and ultimately treatment was done within four weeks of transfer. The downside of that is that they could have had an acute bleed, and then three to four weeks later, they're actually getting the therapy they were randomized to, but still, like I said, this overall pretty good study. They excluded patients basically that didn't have a shunt, therefore they couldn't undergo a BRTO. Patients that had other complications, so non-steroid portal hypertension, cavernoma, malignancy, renal failure, cardiorespiratory failure, that kind of thing, and I also want to point out that standard therapy for all of these patients, in addition to vasoactive drugs, antibiotics, and a restrictive transfusion pattern that we would use, you know, here in North America, they also all got cyanoacrylate injection to stop the initial bleed, and then they were randomized after that to either get BRTO or to continue getting cyanoacrylate injection every four weeks. The other thing I'm going to point out is they only injected two mls of cyanoacrylate at a time, and again, my personal belief is that you can and sometimes need to inject more than that in a single procedure, depending on the size of the gastric varices in the stomach, but that being said, that's how they did their approach. There were no significant differences in other portal hypertensive complications or kind of serine classifications, you know, presence of EVs or not, and they did find that BRTO had a much lower rate of re-bleeding, which in the top right you can see all of the re-bleeds and where they came from, and there were no re-bleeds from gastric varices in the BRTO group. There were quote-unquote re-bleeds from esophageal varices, and as Dr. Saad explained, probably just exacerbated by the BRTO procedure. In the bottom right, you kind of see, again, free of gastric variceal bleeding, BRTO is superior to cyanoacrylate. One thing I'll point out, the biggest difference actually seems to come after the six to twelve month mark. In the first six months, they're pretty equivocal, and then it starts to differentiate from there, which may mean that BRTO is superior, or it may mean that we need to adjust how we follow up and monitor somebody after cyanoacrylate injection. Understanding endoscopic guidance for how frequently we should be doing endoscopies and monitoring those patients is unclear because we don't have those studies like we do for patients with esophageal varices. On the left, you can see that complications and mortality rates were generally the same between the two groups, and on the right, you can actually see that despite that difference in re-bleeding rates, that overall survival, really not different between groups either. Also, the predictors of survival, I'd say overall, were kind of our common predictors we normally expect. So, liver function in this particular cohort, presence of hepatitis B-related cirrhosis. So, based on this, for cardiofundal varices, where a shunt is present and the bleed has been temporized, BRTO does seem to be superior to prevent re-bleeding, but may not have a long-term effect on survival. So, after treatment, what about secondary prophylaxis? I already said for primary prophylaxis, beta blockers may not work that well. This is kind of the one trial I can find that's randomized on secondary prophylaxis, and in this study, everybody was treated with cyanoacrylate injection for their acute bleed and then also kind of every four weeks thereafter, but one group had the addition of Carvedilol and one group did not. And you can see on the right, statistically, there was really no difference in recurrent gastric variceal bleeding with and without Carvedilol, and this is going out all the way to five years. And unfortunately, on the bottom left in table five, you can see there was a significant increase in adverse events in the patients with Carvedilol, which primarily came down to increased rates of dizziness and dyspnea on exertion. And so, as of right now, at least after cyanoacrylate injection, I wouldn't recommend use of Carvedilol, I can't say that about propranolol or nadolol right now, for secondary prophylaxis. Now, if patients have an alternative reason to be on prophylaxis with a non-selective beta blocker, that's different, but specifically for gastric varices, would not recommend it. So overall, I think just the points I reiterate are, initially, whatever you can do to stop the bleed, you've got to kind of use what's at your hands, what's available at your institution. I would recommend using a gastric balloon to temporize active bleeding, get cross-sectional imaging with CT or MRI, and then plan to have a multidisciplinary discussion and or transfer your patient to a tertiary care center. All right, I will stop it there. I apologize, I know we don't have a whole lot of time for questions, but I'll turn it over to Dr. Ripple. Thank you very much, Dr. Henry, for your interesting presentation. We will now discuss both presentations together. We first have a couple of questions regarding primary prophylaxis. Dr. Saad has already answered in the chat a question regarding BRTO for primary prophylaxis. Dr. Saad, maybe you would like to turn on your camera. It is obviously possible to do BRTO in primary prophylaxis, although I personally would think that the risks of BRTO, at least with the present knowledge, with absolutely no studies in this setting, it is most likely not the first option that we would choose for primary prophylaxis, right? I would agree with that, yeah. And then for Dr. Henry, there are questions regarding beta-blockers in primary prophylaxis. It is clear that since we have the Perdesi study, which included patients with clinically significant portal hypertension and patients with gastric varices obviously have clinically significant portal hypertension, beta-blockers improve survival. And this study that you showed, this is one study, which exactly one study, which has not been reproduced, which did not have survival as a main endpoint, if I recall, right? Correct. Correct. I at least personally would tend always to give beta-blockers, and this is the recommendation of the Babeno 7 conference and will be the recommendation of the portal hypertension clinical practice guidance that is being prepared. But I cannot say that there are patients who do not tolerate beta-blockers, and maybe these patients are those in whom cyanoacrylate is the best option, right? Yeah. And I'm just going to start referring to it as glue because that's easier to say, and it's basically just super glue. So I think that one study, so it showed there was, you know, beta-blockers in this group of cardiofundal gastric varices patients did not improve survival and did not prevent bleeding. I mean, not better than no treatment, but again, small study, one study. That doesn't mean that beta-blockers, that doesn't negate every other beta-blocker study for esophageal varices for other portal hypertensive complications. And so if you want to use them, as I said, for what I would call actually the real indications, I think that's fine. I will say one thing. The presence of a gastric varix may not represent clinically significant portal hypertension the way that it's previously been defined. So previously it's been defined by HVPG. Many people have gastric varices and have an HVPG that's six. We wouldn't call that clinically significant portal hypertension. And so the presence of a gastric varix by itself, without esophageal varices, without ascites, if I did the pressure measurement with an HVPG of six, I don't know. I mean, I don't think beta-blockers, if they tolerate them, is it harmful? Probably not, but I don't think it's going to prevent the bleed from the gastric varix because as Dr. Saad showed, when that forms, you get this segregation of flow and pressure systems between the liver and the spleen that again, extensively for the gastric varix, it's not going to make a difference. This is obviously, obviously in the gastric varices, you see frequently when you have a thrombosis of the splenic vein and left side, what classically is called left-sided portal hypertension, right? And in these cases, HVPG is not, because HVPG only works for sinusoidal portal hypertension. Well, can I make a comment about that? Most of the patients in these studies do not have splenic vein thrombosis. I think they do have an HVPG that's probably 9, 10, 11, 12. Then the shunt develops. The spleen goes through the varix, goes through the shunt, and after they create that shunt, they autotips themselves, right? So then when we go measure HVPG afterwards, the HVPG has come down to where maybe they're in that 6 to 10 range, but they still have a gastric varix. And so that's what I'm saying. It's just a tricky thing to use the same terminology we use for EVs, for GVs, because once that shunt forms, it's a segregated flow and pressure system. And this connects with the second question I have, which is associated to this sort of what you just said, you know, the gastric varix system decompresses. And Dr. Saad showed us these beautiful data of how you can have an improvement in flow, an improvement in, you know, increase in liver volume, and even an improvement in the synthetic ability of the liver, which is fascinating. But then you have all the complications of increase in pressure. And we all know that according to Ohm's law, pressure is depending on flow and resistance. And the problem there is the resistance, because we do have this increase in resistance in the liver. And then when you close it with BRTO, or with glue, then you have more resistance. So this is a question more for Dr. Saad. What do you think, wouldn't you, in the patient who has a bleeding with a gastric varices, when do you, I personally would tend to consider to combine BRTO with tips always. When do you decide with tips or without tips? Well, with tips and without, that's a, that's a more complicated conversation. If it's a, it all depends on whether there's thrombosis or not. That's kind of your over, there's an algorithm, there's a whole algorithm to it. Is there a shunt or not? Is there thrombosis or not? So if there is no thrombosis, like splenic vein thrombosis, and there is a shunt, you can still do a tips as long as the patient is a tips candidate and does not have encephalopathy, basically a tips candidate, and you'll do a BATO. And the key thing is to not just coil, but sclerose the entire system. And there's, there's nothing, in my opinion, there's nothing wrong with that. If there is, you know, a left-sided portal hypertension thrombosis, then, then you're looking at a, at a BRTO for sure, because as you well know, you're decompressing with a splenic vein thrombosis isn't, isn't going to help. And just to note that all these numbers that I've shown you especially with the, the one that I've done, you know, with the American experience, there is no splenic vein thrombosis. This, this portal circulations are all patients, they're all open. So that means that they're all patients, they were all patients with cirrhosis. Yes. Yes. That's interesting. And the, and in the case of a splenic vein thrombosis, would you do, I mean, you do a BRTO and the splenic vein thrombosis and, you know, where, I mean, this is just a, maybe a simple question, where does the spleen then drain? Can I make an argument real quick to not do a BRTO in that patient? Okay. Quick, quick argument, very quick argument. We're at the hour. So chronic pancreatitis causes splenic vein thrombosis and you get non-cirrhotic gastric varices all the time. Those patients need their spleens taken out. Splenectomy is the best therapy long-term. It's difficult when they have, like my patient I presented, right? Had cirrhosis, had a splenic vein thrombosis, had a shunt. And I went back and forth about, should we do a BRTO or glue injection? Wael, what do you think? Well, that goes back to the tailoring. You know, my first slide, the tailoring to the, to the patient, you either recanalize that the, the splenic vein and then do a BRTO. And if you cannot recanalize the splenic vein, you do a BRTO and a splenic embolization. So this is tailoring, and that's a completely different conversation with algorithm treatments that I have that I did not cover today, which is how do you tailor this, all these scenarios and flow thrombosis and patient presentation and needs to, to what we have as an armamentarium to the patient. So it's not just simply BRTO. I just did it for simplicity sake as far as conversation. Well, I think with this, we can conclude a very interesting, I think this just underlines the necessary, that it's necessary to have an interdisciplinary conversation and talking about the patients to see what we can do to, to offer, what we can offer our patients, what can be the best for them. And with this, I would like to conclude. I would like to thank you all for your marvelous presentations, very interesting, and all the attendees for their questions and for, for attending. And I wish you a nice day and a nice evening. Thank you, everybody. Thank you.

endovascular management

gastric varices

BRTO

conventional BRTO

BATO

accelerated BRTO

cyanoacrylate injection

band ligation

sclerotherapy

long-term survival