Well, it's nice to be here, and thanks to Ahmed and Linda for inviting me to give this talk. I'll be talking about liver biopsy, of course, and as well as treatment of portal hypertension, portal pressure measurement, and other vascular therapy. My disclosures. So therapeutic endoscopy, you know, early on we were doing amazing things with EGD, and then ERCP took the stage, and amazing things were being done with ERCP, and now really it's EOS has taken over as the main player, and it seems like most of the advances in the last few years and going forward will probably be EOS-based. This concept of endohepatology has emerged, and basically it's liver biopsy, EOS liver biopsy is I think the cornerstone of that, and the idea is that you can have a one-stop shop. You get a referral for someone with chronic liver disease. You can do an EGD to rule out varices, portal hypertensive gastropathy. You can do an EOS to look at the bile duct, the gallbladder, the liver, potentially even do a shear wave assessment of the liver, then can do an EOS-guided portal pressure evaluation, finally a liver biopsy. So all these can be done in one setting and without having to go to IR, and I think this is the trend that we're, you know, aspirational, you might say. As we go forward, hepatology hopefully is referring to us in the interventional GI realm and less referring to interventional radiology. So if you think what, you know, IR, we obviously still need them for tips, for the very rare failed cannulation, you know, Amy Tyberg was talking about how we get into the ducts, but every once in a while we still need them for that, and if there's a coagulopathy and the person needs a liver biopsy, fine, we'll send them for a transjugular. But we want to work more closely with our hepatologists. I refer you to the Tide's journal. I edited, it was a two-part series, and this is a nice review on all things endohepatology, so check that out. First I'll start with the EOS-guided liver biopsy. So I don't always do liver biopsies, but if I do, they're EOS-guided, obviously. So among the three ways of doing it, again, transjugular obviously are for those people maybe with a coagulopathy, and in the past it was for those people maybe you want a portal pressure measurement at the same time. And then the battle then is really between percutaneous and EOS-guided, and it depends on one's referral patterns. I mean, the fact is some hepatologists are very old-fashioned and still insist on a percutaneous liver biopsy. You know, GI fellows aren't even trained in doing liver biopsy anymore, so those hepatologists who insist on doing their own liver biopsies, you know, they're retiring bit by bit. And so I think a lot of it's education to our referrers, telling them what we do and why it's better. And my hepatologists, they still order a lot of biopsies, you know, they have transelastography and everything else, and they're still ordering biopsies, so biopsies are not going away. And we did talk to them. We had a PA that actually was doing all the liver biopsies. This was, he retired like 10 years ago. And after that, they all went to IR, but I just talked to the hepatologist and I said, you know, we can do them. And they were like, oh, okay, fine. And it was very easy, I have to say, and I know it's not easy at a lot of institutions. We went back and looked at trends in the technique and volume. So we looked at this 10-year period, 2007 and 2017. And back in 2007, there was 166 biopsies done that year at Geisinger and Danville. And you can see most of them were IR image guided. And 30% were, I say gastro, that's that PA that was doing them. And a few were transjugular. Now we've captured almost, you know, a huge part of that market, IR. The transjugulars are the same, and IR gets a few, but not that many. And you can see our volume. I mean, in 2017, we did 427 liver biopsies. I mean, these are referrals from a hepatologist. So I will say that I think that's a little unusual. I think a lot of places, the hepatologists aren't doing biopsies as much, but our hepatologists still feel the need for those. The history of EOS-guided liver biopsies is interesting. There was the first true-cut liver biopsy was done by Moritz Worsema and Mike Levy. And if you look at that paper, what's interesting, they got like one portal tract. So it was more of a proof that you can stick a large needle into the liver and get away with it. And then Abraham Matthew, who's at Penn State, he had a, it was like a letter to the editor in the Red Journal where he did parenchymal biopsy with a true-cut needle. And then Stavros Stavropoulos wrote a paper that had a great impact on me. He just used a regular 19-gauge FNA needle in 22 cases, and he showed specimen adequacy in 20 out of 22 cases. And before that, we were trying to use this spring-loaded true-cut, it was called a quick core needle, and it was very frustrating because I can never make it work. But after I saw that, it changed my approach, and we started using a 19-gauge needle. And I organized this multi-center study, and it was very high specimen adequacy. And again, this was just a regular FNA needle. Oops. I don't know why that happened. Oh. This one, David. Okay. This to go forward, this to go back. All right. It's a little more stable. This one is a little jumpy. I see. Okay. Okay, and this was a study that we did showing we compared tissue yields of percutaneous, transjugular, and EUS-guided, both lobes. And this, again, this was with a regular 19-gauge needle, and this study showed that the tissue yields were comparable or better using EUS guidance compared to percutaneous or transjugular. And I will mention that I excluded cirrhotic patients, because, you know, they'll have lower yields, and it was still better. And again, that was with a regular 19-gauge needle, and I think with the 19-gauge core needles we're using now, there's no comparison. So you know, we get these beautiful tours, and if anybody's on Twitter, people, I enjoy seeing people posting their beautiful core specimens. But you can see the slides, and these nice, long cores that the tech in the lab lines them up and makes it easy for the pathologist to look at when they do that. So it's good to talk to your pathologist about this effort. You know, and it's a 19-gauge needle, it's not like a 16-gauge needle, which they used to use. But the width is very good, you get good portal triad counts, and there's still people out there that insist it has to be a 16-gauge needle, and you know, I don't think that's true. And in fact, a lot of IR docs use an 18-gauge needle, so an 18 versus 19 is pretty close. These are, so you know, if you look at those, this is a cirrhotic liver, and because you can see the individual nodules, and using a core needle, you get these beautiful specimens. The advantages of EUSLB, well, first of all, you can do, if the person needs an endoscopy or an EUS or a colonoscopy, in addition to a liver biopsy, you can do them at the same setting. So it's definitely cost-saving there. The sample yields are very good. You can biopsy both lobes very easily, which is something that's relatively hard to do percutaneously. It's improved patient experience, and I, when I'm consenting the patient, I congratulate them that they're going to have the liver biopsy done in the most, you know, humane way possible. It keeps care within the GI department, which I think is important. And it has an excellent safety profile. Something I'm going to touch on a couple times, you have to avoid inverting splenic biopsy. That can be a problem. So if you identify the spleen and then exclude it and go for the left lobe, then it's rare to have a problem. So we've done a lot of studies trying to figure out the best technique, and we've learned a few things. We learned that wet suction is better than dry suction, and I'll show you a short video how to do that. We learned that a 19-gauge FNB is better than a 19-gauge FNA. We figured out that one stick and three to-and-fros with the needle, three actuations, is better than one pass and one actuation. A 19-gauge FNB is better than a 22-gauge FNB, and I really, I don't recommend a 22-gauge FNB. If you're going to do a liver biopsy, you should probably use the largest possible needle, and it'll be a 19. There are studies on 22, but what we found is the specimen looks pretty good coming out, and then it goes to the lab, they turn it into those slides like you saw, and it comes back very fragmented because the specimens are half the diameter. So I don't recommend a 22-gauge FNB. And if you're looking for NAFLD, bilobar is better than unilobar. People have talked about the various kinds of needles available. We tend to prefer the core needle. You can use a regular non-core needle, but if the liver, if you're worried about a high degree of fibrosis or preserosis, it's better off using a core needle. So as far as the technique, I have a video I'll play, and I'll narrate it as it goes. This is, that cartoon is showing how we access the right lobe. We go to the duodenal bulb, and you can see, go to the duodenal bulb and rotate counterclockwise, and you see a lot of liver. And you would think that the left lobe, which you get through the stomach, would be easier because the scope is straighter, but I find the right lobe a bit easier, technically. These are the regular Manghini-type needles, and again, they can work, but as a general rule, we're using a FNB needle. This is the wet suction preparation. So pull the stylet, and we use a heparin flush, and I'll talk more about that. It prevents blood clots in the needle. So wet could be heparin or saline, but these heparin flushes, they're everywhere in the hospital. They're easy to come by. So you flush until you get a couple drops out. Then you set the syringe, just drop a little bit of fluid, and then cock the plunger. So now there's vacuum there, and we use full vacuum. And so now the needle's ready for use. So here's this. I'm going to harp on this. Confirm it's the left lobe. So this is liver, hyperechoic. This is spleen. It's the same echo texture. And sometimes the spleen's quite large, particularly if the portal pressure is high. So you have to look very carefully and just make sure you're looking at liver and not spleen. Now with a radial scope, you can see the liver's on the left, and you can see the spleen on the right, and they're almost the same size target, same echo texture. So there have been bleeds described by inadvertent splenic puncture, and I think it's not the spleen so much, it's the vessels, the splenic vessels that'll cause bleeding, because we do rarely have to do a splenic biopsy. So I can't underscore this enough. Just make sure you're looking at the liver and not the spleen. That's, again, for the left lobe. So then you find a trajectory. It doesn't have to be super long. I mean, it could be three centimeters, sometimes it's two and a half, sometimes it's more. A quick punch to get it through the gastric wall or duodenal wall, and then you can, here's where we do three and two and froze. Turn the suction on once you're in, and turn the suction off before you come out again. Again, the left lobe is transgastric. That first punch can be very hard sometimes, particularly with the FMB needle. Sometimes it's hard to pierce the gastric wall, but you can see full travel. If you don't have vessels, you can do full travel, and you'll find that your specimens are longer the longer you can travel. But if you can't do full travel, that's all right. So here's the right lobe from the duodenal bulb. And we like to see the needle the entire way, but particularly if it's a very hyper-echoic liver like here, it's kind of hard to see the needle. Sometimes you can rotate a little, but you get a sense of where the needle's going. So three, two, and froze is what we're currently doing. The specimen, you don't want to handle it too much because you don't want it to fragment. So you can either put it straight in the formalin, but I came up with this, it's like a filter, it's like a little funnel with a piece of nylon mesh on it. And if you've heparinized the needle, the blood won't clot, which is nice, generally speaking. So you have blood, but it's liquid, it'll run through the mesh, and then you can also just take a saline syringe. Oh, importantly, if there's blood in the syringe, not a big deal, but know that you could have liver tissue there. And sometimes the liver tissue is at the end of that stopcock. So look carefully, because you could have a nice core sitting there. So don't forget about that. And then you can wash it. And the advantages are, first of all, you have an immediate assessment of adequacy. You can see it's good. And this peels off, and you can just drop it in the formalin. You don't have to handle it or touch it or anything. And those pieces sort of float off by the time it ends up in the lab. And then it makes it easier on the lab, too, because they don't have to pick through the blood and stuff. One point about blood in the syringe. It's very common to have blood. It's no reason to panic. Sometimes the fellows freak out, or the nurses freak out, and you stick, you're doing it, and then the syringe is filling with blood. It's not a big deal. It doesn't correlate with post-procedure bleeding. You don't have to necessarily, OK, I'm just going to turn off the suction. Finish your 3, 2, and froze, turn it off. But just understand that in that blood, there could be tissue. So you want to put it, I like putting it through the filter. I mean, you can put it all in formalin and let the lab deal with it, but then it takes them longer. They have to pour it out and pick through it. But you notice that the blood is liquid. It generally doesn't, you're not getting these blood noodles. Sometimes on Twitter, people, when they show how beautiful their specimen is, I look at it, and it's like, wow, it's like there's a blood noodle there, and maybe there's some tissue there. But I'm wondering if they're thinking that their blood noodle is the specimen. And I will refer you to this article I was invited to write for GIE. It's open access, so feel free to download its top tips regarding U.S. guided liver biopsy in February. And it summarizes everything I've talked about. Next, I want to move on to portal pressure measurement. There's a station out in the lab showing this device, this AcroTip Insight. And the station is a great way to learn how to set it up. It's very important to set it up properly without having bubbles in the system. This is the transjugular method using a wedge balloon. And I know that hepatologists are often there, the purists really want to see this tracing of the wedge pressure and the free. And we don't get a tracing, as you know. We get a little number on the dial, which the purists may not like it as much. But I've talked to Cook about, can they somehow work it out where it wirelessly transmits and creates this? Because this is kind of the standard way of doing the measurement. And you're familiar with the degree of hepatic venous pressures and the risk of variceal bleeding and so forth. I think with the PPG, it sort of draws attention to this one group, this sort of people in the six to nine millimeter range, who might have asymptomatic portal hypertension or there's something going on in the liver, but there's no varices. So it is hoped that expansion of the EUS-guided PPG will shed more light on this subgroup. We're taking a 25-gauge needle. We're sticking it into the liver, sticking it into a vessel, but it is a 25-gauge needle. And we're already doing a 19-gauge needle biopsy, so I consider this very safe. Indications, these are generally coming from our hepatologists. So depending on where you are, you may get referrals or not. There are these occasional patients that you think they have some sort of portal hypertension, but they're not obviously cirrhotic. They would be a good patient for that. To clarify the cause of portal hypertension, occasionally, if it's a known cirrhotic, a patient needs a high-risk surgery, there may be some prognostic value in having that. And some programs that have busy transplant centers, the transplant surgeons are asking for this to see if they need to do a kidney-liver combined transplant. And the advantages are similar to liver biopsy. It unifies care. It stays within the department. It can all be done at the same time. It's not an unsedated transjugular approach, so it's more acceptable to the patient. And actually, adverse effects are rare. You find these vessels, and there's a short video that I got from Ken Chang's group, which I'll show. But if you're in the hepatic vein, it has this triphasic Doppler signal. And if you're in the portal vein, it has a monophasic. So that's one good way to know if you're in the right vessel. The other thing, the portal structures have a white halo around them, so a white wall. So those are means to figure out if you're in the right area. One of the most important things is to make sure you go through parenchyma to get into the vessel. So this is a pretty large vein. I mean, you can really almost see the IVC there. But you can see the needle is traversing parenchyma, and that will probably prevent bleeding. This is a portal structure with the white wall. I will say that usually the vessels that we target are not quite as big as that. They're smaller. They're a little bit more peripheral. But the concept of going through parenchyma to get to the vessel is the same. This is from Ken's work. Ken has done most of the work in this area. And here, in a pig, they actually did EOS-guided with simultaneously transjugular. And there was very good agreement there in measurement. And then they followed up with this study with these patient subgroups to see if the portal measurements made sense. And they did make sense. So it was good correlation of the EOS PPG with clinical parameters of portal hypertension. So the technique, first you set everything up. So you have to flush the system and eliminate bubbles. And Michael from Cook is doing a good job of demonstrating how to do that. So I encourage you to have him show you that. The manometer has to be at the mid-axillary line and not move. So you have to figure out a way to just keep it there. And your assistants will be keeping track of that and actually reading the numbers. You will not be. You'll be looking at your EOS monitor, making sure the needle's in place. So you're not actually looking at that little device. I found sitting for this makes a lot of sense. Because if you sit for it, you can brace your arm while you're holding the scope. And they can be, you know, it could be like 15 minutes or 20 minutes to do it, just because it's all the multiple measurements. And if you're just trying to stand there like a statue, it's hard. But if you sit and brace your arm, I highly recommend trying that. We generally have been using endotracheal intubation in general. It can also be done with MAC. Certainly earlier in your experience, it's probably better to have under general so that you don't want the patient to move at all. I did mention checking the Doppler signal. You'll get a different signal from the HV or PB. And you can check close to the probe to make sure you're not going through a varix to get through the liver. So that's the getting ready part. And once you're there, again, plunge the needle in through parenchyma into the vessel. If you get a funny measurement on the first measure, you can disregard it and repeat it. The problem can be like when you go through the liver, there could be a little bit of tissue at the end of the needle, which interferes with the manometer. You do flush it, but sometimes there's a little piece stuck there. And it's not that, I would say it's not that technically difficult. If you have good aim, if you're used to hitting small lesions with an EOS needle, then you can do this. But you have to be patient. You measure for 60 seconds, you flush, you wait for the thing to come down. So you can't rush doing it, but you have to make sure that that gauge is absolutely in the same position for the whole procedure. There is what's been called the elephant in the room about this. It's known that transjugular is generally done without anesthesia. So it's known that propofol does affect the wedged hepatic vein pressure. So it seems to decrease the gradient by about 10%. So that's known. But what about what we're doing? We're doing with propofol or general, and we don't really know fully the effect of the means of anesthesia on these measurements. And this is something that has to be worked out, I think, if there's going to be widespread acceptance of this. There was a recent report, which was published just in July. They used three patients, and they were sedated with, I would say, ultra-low dose midazolam. It was like a milligram. And ketamine, kind of high-dose ketamine. And with the thought being it will have less of an effect on the portal pressure. That may be so, but it may be challenging to sedate someone like that. And obviously, you have to be very comfortable with the technique so you can do it quickly. I should say as quickly as possible. Next I want to move on to EOS-guided gastric variceal glue and coiling. So everyone's familiar with the sarin classification, and we're talking about GOV-2 and IGV-1 mainly. The GOV-1's current thinking is that can be managed with banding. Now I don't want to overly complicate things and go through every aspect of this, but do know that the venous drainage of the stomach is very complex. And the important points are you can see the inflow is that sort of light blue. And the inflow can be through the left gastric vein, the posterior gastric vein, the short gastric veins, a combination, all three, two out of the three. So it's very complicated. And I guess my point is, I'll make this point going through, I don't think every gastric varix is amenable to EOS ablation. And that on any given case, it may be very complex to understand. This chart is from a review that was in that TIGE journal from the Mayo Group. And the breakpoint is there is it's, you can see the person has gastric variceal hemorrhage, is EVO or endoscopic variceal obturation. This term means basically putting glue in the vessel, obturating the lumen. So is it feasible? Well, most places it's not feasible. Most places are not doing this. So then they go on to IR and they do a TIPS or something. But if it's feasible, that means you have someone who knows how to do it. You have all the equipment and so forth, then you can pursue it. So I think the reality on this slide is that 99% of places it's red and 1% is green. So people have directly injected glue. You can directly inject glue endoscopically, although EOS-directed glue injection probably is better. This Dermabond or 2-octyl cyanoacrylate is typically used with 1 mL aliquots up to about 6 mL. But the studies usually use a mean of 2 mLs. Putting coils first and then glue is clearly better. There is very high embolization of glue, usually asymptomatic, but up to 40% if you just inject glue. And if there's a patent for amenovale, you can get a big problem with that. So I think if this is going to continue, I think it has to be coiled and glue. The coils people are using are these fuzzy coils. So it's a metal coil with these hairs on it that seem to keep the glue in place and prevent embolization. Ken Bindmuller was the first person to describe this, the coils and then the glue, and I'll show you a video of theirs. But a key thing, which is, it's in the papers, but it's almost like between the lines is generally there has to be like a perforator vessel, a main vessel that's feeding it. And then they target that vessel. But as I was with that anatomic diagram of the venous inflow, what if there's multiple? Then it may be hard or impossible to treat. So there's a little bit of cherry picking in some of these studies. Some of the studies only treat people that have a single perforator vessel. This is from an early study by Romero-Castro, so here's a varix, and in this case there is a vessel that they can target. And what they're doing is they're injecting glue, and you can see this sort of whitish material coming out of the needle, obturating the lumen. It seems reasonable to do, so they inject glue and then pretty much the Doppler signal has gone away. This is from Ben Muller's paper. So this large IGV-1, it's fairly large. It's hard to see an exact feeder vessel, but they target the bulk of the lumen of this thing, and first they're injecting coils. They put one coil down there, and they found this other big leg, put a second coil, and you can see it sort of spinning around in the blood flow. And that's followed by a glue injection. So I think straight glue injection probably, looking at the literature, probably should not be done. If it's EOS-guided, you combine both. I mean, certainly, there are some experts who can just inject glue under endoscopic control, but in this study, they had very good results. This is 152 patients. Interesting, 26 percent of them were for primary prophylaxis, where they had gastric varices but hadn't bled, and only 5 percent were with active bleeding. And you can imagine, if you get called in the middle of the night, I mean, to get the EOS equipment out doesn't seem very feasible. But they used one to four coils, a mean of 1.4, and one to six milliliters of CYA, a mean of two. Very high technical success, re-bleeding uncommonly, and adverse effects were remarkably low, with abdominal pain in 3 percent. And there was one patient with clinical signs of pulmonary glue embolization, but it's very likely, I mean, if they have CAT scans, you probably are going to see glue embolization in a lot more. This was another interesting study. Here, they randomized patients to coil versus coil plus glue. And I wanted to quote this one sentence from the article. So that arrow is pointing at the feeder vessel. So they specifically say, that offending vessel or vessels were followed from the cardia to the proximal part of the esophagus, two to three centimeters above the cardia, to detect the feeding vessel, which was considered to be the convergence of all offending vessels. So does that mean if they couldn't find that, they didn't include them in the randomization? It's not exactly clear, but one theme in all these papers is that they're looking for a feeder vessel and injecting it. And clearly, both was better than coil alone. Another point is that it was one very experienced endoscopist who did all the cases. So it really requires an experienced endosynographer. I think in the acute setting, it's really limited availability. Sometimes when I've undertaken, OK, I'm going to go after this one, it's like this huge lake. I don't see any perforated vessel. And so I've backed off from doing it because it seems like almost pointless to inject this large thing. And they still may require a TIPS. So glial embolization is possible. And I know at some centers, when I was talking to the group from Mayo in the past, part of their protocol is someone had to have an echo to make sure there wasn't a PFO. So it's something with promise, but it's hard to imagine it'll be very widespread. One thing, though, that I would like to touch on, which I think is good to keep in mind, is non-variceal EOS-guided hemostasis. And I remember reading this paper from Mike Levy. They had five patients with refractory bleeding, like a gist, a dulafoil, a duodenal ulcer, hemosuchus pancreaticus. And they directly injected this absolute ethanol that Matt Moyer was talking about, or cyanacrylate, into the lesion. And their endpoint was a decrease in Doppler signal. And it was very intriguing. And so I've actually had occasion to do this two or three times. And here's a video of a case I had. This is a man with an adenocarcinoma of the G-junction, scoping him for bleeding. And I mean, it's like this diffuse bleeding, which is just getting worse throughout the procedure. Nothing really to do endoscopically. So I was thinking of Mike Levy's paper. And I put down the scope. And you can see these feeder vessels. And you can see the tumor as well. And so just a 22-gauge needle, I'm targeting these feeder vessels, injecting small amounts of ethanol, and repeating that until the signal goes away. So I think I did it two or three times. Very straightforward. I think, you know, with our skill set as US docs with a needle, this is something that we could do. It actually stopped bleeding. So it was very successful for immediate hemostasis. And then I can send the person on for radiation therapy. So it's a trick you should think about. And then you can just put fiducials right then and there as a target for the radiation people. So I dare say that someone in this audience is going to encounter this. I think that John DeWitt made some good points about safety. Ethanol, you don't want to inject into the gastric wall. I'm injecting into a tumor. I don't think the stakes are quite as high. Back ages ago, we used to inject ethanol for ulcer hemostasis. And it was in tiny aliquots. It was like 0.2 mils, up to like max of 2 mils. And that's for DU. So I think for a tumor bleed, you can go higher. But I still would probably not go more than, say, 5 mils of the absolute ethanol. But it's something to keep in mind. And I dare say someone will find it useful someday. So final slide. So the take home message is EOSLB, with the right technique, is an excellent means of obtaining outstanding liver core biopsies. And I keep harping on this. Care should be taken to avoid inadvertent splenic biopsy. EOS-guided portal pressure gradient measurement may be favored over transjugular in many cases and can be combined with same-session liver biopsy, allowing a one-stop shop for evaluation of liver disease. And finally, EOS-guided hemostasis with coils and glue for gastric varices is definitely an advance in the field, but is dependent on expertise and availability. Thank you. Thank you, David. That was excellent. That was a comprehensive review of endohepatology. And thank you so much for moving this field forward, as you have done for the last few years. Questions, guys? We can take a few questions before we start the last. Thank you for that presentation. Hit your mic, if you have one. Yeah. All right. Can you hear me? Yeah, yeah. OK. So the question is, just for clarification, regarding the non-targeted use of EOSLB and getting the non-targeted liver biopsies, so it seems as though getting it from the left might be technically easier. But what about complications? Would you worry more about bleeding? And then I noted, maybe it's just the video, you didn't have the Doppler on for liver biopsies. Is that not necessary, because there's less of a risk of bleeding in that situation? Yeah. I find it technically easier to get the liver biopsy I have to say, for a few reasons. One, there's a bigger target. Two, I think the needle's more perpendicular to the duodenal wall, whereas sometimes in the stomach it's more tangential and it's harder to pierce. And it seems like the scope's a little bit more stable. So I find the right a tiny bit easier. Yeah, I typically turn the Doppler on. But the reality is, you generally don't see a vessel with Doppler that you're not seeing just with the EUS. Is there any higher risk of damaging the scope with the 19-gauge needle, with the scope in the duodenum, especially if there's a loop? Yeah, I set the sheath length at the correct amount. And you know what I'm talking about. So the way I set the sheath length is maybe different. I know a lot of people say, oh, it's a Boston needle, so it's at three, or whatever. It's a Cook needle, it's at whatever. With the scope straight, I actually put the needle out and sort of wiggle the elevator until I can see just the tip of the needle moving. And then I'll screw it down. And that's how I set that before I do it. So I don't look at the number anymore, because I can't remember from brand to brand of needle. And scope channels are a little different. And it keeps changing also. In the same scope, with the same needle, I think it's because the flexible portion, the tip of the scope changes in, I guess, a little bit. So you're right. I mean, there's no fixed. So if you elevate, and you've seen the needle tip, then I'm not worried about puncturing the channel. So you get it out, and then you go in and do it. And it's not a lot out. It's just a little bit out, so I can see it. That's a good point. That's a very good point. So when you're taking the biopsy from the right lobe, what is the risk of hitting the higher structures, like the portal vein or the bile duct? And the other question is, when you're doing it and turning the scope to the left, if you see the gallbladder, I mean, you should change the, I mean, is there any risk if you go between the gallbladder and your higher structures to get to the liver? I'm generally not seeing higher structures. And with that, certainly, if I see a gallbladder, I'm avoiding it. I'm avoiding vessels. I'm avoiding the portal vein. But more often than not, I get this huge, like it just fills the screen. The right lobe is just like filling the screen, whereas on the left, sometimes it's pretty big, and sometimes it's a little bit more of a sliver of liver. And I missed that initial heparinization of the needle. What is the concentration you're using there? You know, it's a standard heparin flush. Oh, standard heparin flush. I forget. Like the HEPLOC that they use to keep dying. Oh, the HEPLOC. The HEPLOC. It's ubiquitous in the hospital. So we just get those pre-filled and just flush with that. Dr. Dale, for portal pressure management, the needle that you showed, I understand that there was a staggered release of that needle because I think there were some concerns that the initial data has to be just right and so forth. Would you say that this is something that in your hands or, say, in the community, it's a safe device that we can now use? Because I know there were some concerns earlier on. Right. And the rep from Cook was mentioning it's still, I mean, it's been for a couple of years now, this limited release. They're allowing people, like experienced endosynographers, to try it and so forth. I think it's safe. I think it depends on one's level of skill. But if you can target that vessel, go through parenchyma, I think it's, I feel it, with a 25-gauge needle, it's a safe procedure. I didn't mention also, people generally give antibiotics before that procedure. And we've been doing that. And Dr. Tyberg was telling me they do it at their place too. Dr. Wong mentions getting the superior middle suprahepatic. Is that something you look for, or any suprahepatic? Yeah, there's two ways to approach it. So Dr. Chang, he does a very anatomic approach, looks at, I mean, you certainly want to look at the main hepatic veins. The person could have a bug carrier or something. And he'll trace from there to find a vessel, trace the portal vein to find a vessel. But we're in the proximal stomach. You can see the veins and the exact anatomy, I think. I mean, Dr. Chang is a purist about it, which is good. And I recommend looking at his video. I don't know if you've seen it. He has maybe a 20-minute video on the technique where he reviews all anatomy. But I think one can find your target vessels without that detailed anatomic evaluation. One more question about targeting the feeder vessels. I understand the concern that if it's the feeder vessel, we'll be more likely to be successful. But by the same token, the larger the vessel, is there a greater risk for bleeding? You know, bleeding doesn't, causing bleeding is in the series. Again, all these series are done by tremendous experts in the field, which makes them not as general. Which makes them not as generalizable. But precipitating bleeding is not generally an issue. Having delayed bleeding or bleeding from some other part of the varix can happen. But it doesn't seem to be a concern. So David, I had a question. Because like you pointed out, sometimes it's difficult to find the feeder vessel. I mean, this is such a chaotic situation over there with all these big channels. And you see this bulging varix. And maybe let's say it's bled. And you see a nipple sign over there. And I can't find the feeder vessel. Would you not, would you just come out? Or is there any utility, at least gluing and coiling that obviously visible varix that has probably bled, understanding that, yes, the feeder is still there? Yeah, that could be done if expertise is present. Another thing, I mean, certainly like banding gastric varices is known to have a high recurrence rate. But maybe in the heat of the moment, if you were there with the scope, you could put bands on. And then you might have to do a TIPS or something later. But you would not recommend that gluing. If you don't see the feeder vessel, is your practice you just pull out and you won't glue the obvious visible varix that's bulging into the stomach? Right. I can comment. I do a lot of this. And I think the field is kind of losing its way a little bit. The analogy I make is that we call it a varix, but it's an aneurysm at the end of the day, right? It's aneurysmal dilatation of a vein. And similar to an aneurysm in the brain, we don't target the feeding vessels. We treat the aneurysm because that's what's causing the morbidity. So our approach is very different. I don't go for the feeding vessels at all because, like your point, there's going to be multiple, right? And you're not going to treat them all. But the end result is the bleeding from the protruding varix in the stomach. That's where the bleeding is. So our approach is very different. Not different, but it's just you identify that. Now, you have to carefully make sure you're identifying the one that you saw endoscopically is the one you're seeing under EUS, meaning you need to confirm it's in the submucosa of the stomach, not perigastric. But I think that's a common mistake that's made. When you go down, you'll see all sorts of stuff everywhere. But you carefully, so on EGD, we'll sort of note what's the movement I made in order to get that varix in view. And normally, it's going to be pulling up to the cardia and turning left and kind of burying yourself into that little cardiac crevice. And then, even under EUS, sometimes we'll back away a little bit and look endoscopically. Okay, this is the spot that we clearly saw as the bleeder. And then, I'll target that. And that, I've been doing this for about six years now, including primary prophylaxis. And we'll have our data out soon. But it's quite effective to do just that. And really focusing on what bled. Because it's not like when people get varices, they don't get them everywhere in their body, right? Even though the portal pressure is elevated throughout. And so, clearly, that one was the problem. And if you fix it, it doesn't recur, and so on. Water filling also can be helpful. Oh, very helpful, yeah, very helpful. Do you need to put more? I mean, you need to occlude the varix. Yeah, you need to occlude the varix. Yeah, so normally, you do it with coils. And you'll start, you puncture it, and you start using the biggest coil you have first. And then, keep going until it's primarily, the glue is not there to, in my opinion, it's not the therapeutic part. The glue is there to help, actually, fill in a little bit between, and induce the thrombosis that the coils are trying to induce. That's what the hairs are for. They're like thrombotic surface area, basically. So we'll use one to two cc's of glue. The other main thing I use the glue for is on the way out, plugging the hole. And so, as I would draw the needle, the glue, I find it most useful to plug the hole that I just made. And then, you have to wait a bit. You'll see Doppler flow, right when you do the coils. But if you wait five to 10 minutes, a lot of the thrombosis happens over that period, and you'll see it go away. So it's, I agree, it seems like it hasn't really taken off as a good technique, but I find it very useful. Our transplant program actually refers to us now as primary for prophylaxis, or even acute bleeding. One of the, I'll just end on that. I know I'm the next speaker, so I get it. But one of the real advantages, and it's hard in acute bleeding in the middle of the night, but if, you know, we had a scenario last week, the general GI team scope, they saw just blood in the stomach. And then, you know, it was high risk, or high suspicion. So we came back later in the day, and again, there's just blood in the stomach, but under EUS, you could identify the varics. Because you know where it's gonna be. It's gonna go to the cardiac turn left, and there it was. And in fact, you could see the blood clot overlying the varics, which is what was precluding it, and then treated it that way, and she did very well. So I think it's actually quite powerful, but we need better algorithms, and so on. How do you think it'll progress? Take off? Take off. I mean, how many people at your place do it? Just you, or? It's primarily me. I've trained our two most junior people to do it in a pinch. And so I think, you know, part of it is, the history of endoscopic management of gastric varices in the United States has been very limited. And so I was part of the CGH best practices paper, where we at least tried to make the point that endoscopic, whether it's just direct glue, which is done internationally, routinely, you know, can be used in acute bleeding. And then EUS guided, you know, is in your next step, whether you're debating BRTO, and so on. So I think as we get more comfortable with just, you know, initial endoscopic management, perhaps we can start to add this. But it's actually, and it is quite safe with the coil technique, in terms of embolization. It really does reduce that quite a bit. Now I'm gonna show you a funny one in a minute, but it's, anyway, thanks. Thank you so much, David. Thanks. Thank you.

endohepatology

liver biopsy

portal hypertension

portal pressure measurement

vascular therapies

therapeutic endoscopy

EOS-guided liver biopsy

EOS-guided portal pressure measurement

gastric variceal glue