especially the first group, have been here for a while, so I'm gonna try to be quick and swift and hopefully cover things in endohepatology. It's tricky to do troubleshooting. One, endohepatology procedures aren't as common, as robust as the other stuff we're doing. Two, a lot of them are sort of more diagnostic, which we probably see less complications of, but still, I wanna run you through the day-to-day problems that I encounter, be it clinical or non-clinical, or even setup-wise, doing endohepatology procedures. I have no disclosures, and we're gonna go over the sort of maybe 90% of what endohepatology is, which is EUS-guided procedures, so liver biopsy, portal pressure gradient measurement, and coiling of gastric varices. We won't be going over elastography, which I think it's too early to see any troubleshooting in, and we won't be going over variceal bleeding that's refractory to common management. I think that's probably more within the realm of GI bleeding, but certainly still an overlap with endohepatology. So EUS-guided liver biopsy, you guys have probably gone over this with Dr. Ken Chang. It's a straightforward procedure. The patients are usually sedated with MAC or GA. I always like to review imaging that definitely plays a role when you're doing complicated cases like liver transplants or patients who have cirrhosis. You could do a left or right lobe. Some people are doing a bilateral or a bilobar approach. We have data that using a flushed catheter or the wet technique is superior, and 19-gauge needle is what we should be using. Some people go up to three actuations. Different protocols exist. And always look at your core, and this is kind of what you want to add. Does this work? I think so. There? Yeah. So this is sort of the core that you're trying to hope for when you're doing these liver biopsies, and that's kind of the goal when you're doing this. Okay. So your patient has cirrhosis with portal hypertension spenomegaly. This is very, very important, and you will rarely do this because you really don't need to do a liver biopsy in a cirrhotic patient, but if you ever do because your hepatologist wants to, you have to kind of come in with a different mentality. You have to do a very detailed examination. These patients have sort of a different anatomy in terms of their livers are quite shrunken and their spleens are quite big. And you have to be cautious because even early on in your career, it's easy to mistake the two, and it's happened before. So things that I always do is I will always document the spleen in every single EOS guide liver biopsy that I do. I'll take a picture of it. I'll label it as a spleen. I'll go back to the liver, and then I'll go back to the spleen, and I kind of keep it in my brain as this is how things look like because when I'm ready to do my needle, I'm focusing on the needle, but I have to make sure that it's the right structure that I'm sticking. So identifying that your vessels are very important, using your Doppler flow, using your pulse wave to kind of see how things look like is extremely important. You guys can see here the liver. Clearly, you could see the caudate lobe with some bile canaliculi and blood vessels, while on the right side, you have the splenic hilum, the spleen, which is sort of more of a paucity of blood vessels except in the hilum. And in patients who have cirrhosis portal hypertension, the spleen could be so big that it kind of wraps over the entire anterior abdominal wall and that's most of what you see. So keep that in mind when you're doing this liver biopsy, not because of this that happened recently to the surgeon, but it has been documented within the US guidance that we can inadvertently go into the spleen and you could have some adverse events and it's something that you need to be aware can happen and be ready to deal with if it ever does. So you'd perform an US-guided liver biopsy of the caudate lobe, but this is the yield that you get. This is sort of almost an everyday when I was starting off my career. Not the best sample. The lower one was one I did recently and I just didn't think it was something that I was happy with sending to the pathologist. So one, it's not unlikely, but there are things that you can do. One, repeat it in the contralateral lobe if you haven't done a bilobar approach. If you've already done bilobar approach and it's safe, it's fine to retry on both lobes or in a single lobe. If you did only one pass for one actuation, you can increase that. Try to go another stick and do two to three actuations. And always look at the specimen with your nurse. Make sure that they're handling it correctly. I tell them to be careful when they take it to the lab. The last thing you want is someone shaking the specimen and it's arriving all fragmented. And the one thing that you really need to make a habit of is always read the path report. Not to get the clinical diagnosis, but see if the pathologist is commenting on the quality of your liver biopsy, meaning how many portal tracks it's lengthened. Sometimes they'll just say suboptimal. And if they're not doing it, call them and see if you can go over it because you want to know that your samples are good. We do about 150 to 200 liver biopsies a year. And I'm always curious to know what the liver quality of our biopsies are. So my wife's a transplant hepatologist and she sits on all of these path meetings. And so I end up just asking her every week, you know, what was our yield? And she's like, oh, a few were good, a few were bad. And so there's always room for improvement if you have to. But go over your specimens and find someone in the inside to see how your samples are. So you do a U.S. liver biopsy. And when you pull back your needle, you see some blood tracking through your needle track. So this is common. And in fact, it's somewhat almost normal if it's transient and short-lived. Observe, it'll usually kind of go away within two to three minutes. If it doesn't and you're worried, meaning the patient is clinically acting like maybe there's a bleed, you could see an expanse-style sort of bleed happening within the parenchyma where you stuck. Then at this point, there's a few things that you can do. One of the most easiest things is just bring your sheet out and kind of apply that pressure and bring your scope angle down, almost like tampon-adding that area in hopes that you can stop that bleed. If that doesn't work, then, you know, there's good data or minimal data, but good and quite nice that I'm sure Ken Chang had showed you about putting back 20 to 25% of what you took, almost like a blood patch, into that track in hopes that maybe that can coagulate it. And if that doesn't work, then, you know, start to think of different things you can do. You can, you know, use Puristat or absorbable gelatin sponge or gel foam if your needle is still in there to kind of pack that area. Since it's all intraparenchymal, it should be able to tampon out that bleed. Again, quite rare, but it's something to kind of keep in mind if it does happen. You're planning to do an EUS guide to liver biopsy and you see this on exam. So ascites, right? Be odd to do a liver biopsy in a patient with ascites, but it's happened sometimes where I go in, I'm like, okay, there's some ascites. Or you see something like this where it's a pretty much significant ascites. And so if it's mild, it's usually fine to proceed. Usually the ascites is pooling on the outside and you shouldn't see it in the caudate lobe. If it's moderate and they really want you to do it, then, you know, do a paracentesis at that point. IR can do both. And if it's large, then, you know, probably not the right patient to do an EUS guide to liver biopsy due to the risk of infection. But not only that, there's some studies that if you have some ascites, the yield of your path is actually less. And that's been studied recently in a trial where you guys can see there. So keep that in mind when you're looking at a patient's livers, whether there's any sort of fluid surrounding that area. Moving on to EUS guided injection therapy. So my favorite topic of all time. So EUS guided coil therapy is increasing in the US. We're finally starting to see more of it. It's being adopted to approach varices that are not amenable to IR, or if IR don't have the robust services to do it. Here, this is a recent EUS coil that I did. We're doing a quick retroflexion. You could see sort of networks of varices and then one that actually has an ulcer in it. When they're small, I like to tag them. So I just put a clip nearby, so I can see that clip floating within the lumen of my stomach. I'll instill two to 300 CCs of water. So based on Dr. Marvin Yu, who taught me at the Brigham when I was doing this. And then I'm looking at these network of varices and assessing the largest one. Sometimes it's hard to know, where do I go? The key thing here is, I'm sorry guys. The key thing here to keep in mind is, you don't want to stick the varices that are beyond the stomach wall. Those are extramural perisplenic varices. If you get to those varices and something happens, you really can't salvage yourself. You're targeting the submucosal vessel. So everything that's sort of above that line of the stomach is where you're supposed to go to target these varices. So here we often measure, we'll bring your sheath out. Your coil, and we'll go over this in a second, is already pre-deployed or pre-loaded on. You're jabbing into your network of varices and then your assistant is passing that stylet to get that coil into. And you guys can see it there, nicely filling this variceal nest. The next key is, if you want to put more, don't take your needle out. I've seen this happen before where they take a needle out and they load and they go back in and jab. Keep your needle in. If you have a good assistant who's comfortable in loading, they should come next to you and reload it immediately while you're in the varix. And that's the second coil. Filling the space, the Doppler flow significantly improves. And then you decide from there and then whether you want to chase it with any injectate, which we can go over. You guys can see here a very nice image of near complete cessation within that variceal nest. This is the most thing that I get is, we're all well-versed in EUS-guided FNI and FNA. It's really what EUS-guided gastric varices is. But there's something about injecting a blood vessel with a needle that I'm sure we're all worried about. And none of us really do it as much in training until we're out there in our faculty jobs and someone says, hey, do you want to consider doing this? So one thing that we recently developed and we're hoping to publish this soon in video GIE is you can simulate this in an ex vivo model. And we recently did this in a few courses. I started this in a course abroad and then most recently at our WashU course. And we actually have it today for the second group who did it. But take a glove, fill it with some water, make it nice and tight and seal, and then use the digits to kind of flip over the digits and make sort of almost like a cluster of grapes or sausage-like digits, which works very nicely. Then take a rubber band once you're happy with how your nests or your varices look like and secure your digits or your grapes or your varices all together. And then in your ex vivo model, place it sort of almost in the fundal area, so just below the GIE junction. So that's how it looks there. You're going to take this, Michelle, our fellow who's supposed to be here today, did this great video. So we're just going to place it right under our GIE junction to simulate true varices. And everybody thought I was crazy when I did this, but it works. And so then take your EUS probe and you can just kind of practice seeing how varices look like under EUS guidance. So this is really how they look like under, you know, they don't have Doppler flow if you were to turn Doppler, but they look like they're varices. And the most nice thing is when you inject them, they don't pop. So here we're measuring them. And then the next thing is we're doing our coil. So the next step that's very important is loading. This is actually the hardest part of doing an EUS-guided variceal coiling is getting your coil that isn't made for us, for our EUS scopes, it's actually made for IR, to transfer it onto our sheath. So you want to open the coil, it's fixed in a catheter. You're going to take the protective sheath and the covering out there. You're using a 19 or a 22 gauge FNA needle. And you're going to take out the stylet that comes in your needle, and you're going to attach this sheath that houses the coil. And you're going to just sort of advance the connector there and seal it. And the goal is to transfer your coil from this sheath into your needle. So they give you a stylet that comes in the kit. You're going to advance that stylet. And now you've hopefully transferred the coil. I always like to do an internal check. And so what I'll do here is I'll either advance the stylet of your EUS through and make sure it's definitely in. And then the other thing I'll do is I will take out that sheath, which you guys can see here. And as another internal check, I'll pass the stylet through that sheath to make sure that the coil is not there. The last thing you want to do is you're in this vessel, and you're stressed, and you don't have your coil inside your needle. And you're like, where's the coil, where's the coil? So now the coil is high up in my needle, and I'm advancing it as close as can be down towards maybe 10 centimeters away from the tip. Make sure it doesn't come out of the tip. You cannot backload it. You're going to have to put a full new one. And to be honest, if you're doing this for the first time out on your own, try to just do one outside of the body. See how it looks like. Make sure you're comfortable and your tech is comfortable. So this is Dr. Yu here with me doing the case. So these are still the gloves that we used. And he's just about to inject here. And you guys can see very nicely once he accesses it here. So there's the needle there coming out. And so I'm advancing my sheath, and you guys are going to see that coil pop right through. This is quite a small coil, but you could see it pop through, and then you could see him here in a second, kind of rotating and showing you there's the coil right there at the end. And you can actually then simulate injecting it. You know, you can inject saline or gel foam or PureStat to see how that feels. But it's a great way to get comfortable doing this, and one that you guys can do today for those who already did it with us or for the next group. And the nice thing is these don't usually pop. You can actually open it up and see what your coil looks like inside. So something to keep in mind. I'm not sure what coil to use. How can I know? So we have two main types of coils, either Tornado or Nestor coils. They come in different flavors. And it doesn't really matter. They both look pretty nicely. Whatever you have available should work. The key things is they look different. Tornado coils look like a tornado. Nestor coils are sort of more cylindrical. But the thing that I want you to keep in mind is the diameter of the coil. If you were to take the coil and you were to stretch it out, it's either 0.18 or an 0.35. An 0.35 will only go through a 19 gauge. An 0.18 will go through both. The second thing is if you were to just put the coil on the table and it were to kind of coil on its own, it's here five millimeters, but you can have six, eight, or 10 millimeters. And the last thing is the length. So if you want to stretch it out, you can have it up to 20 centimeters. And the reason this is important is the last thing you want to do is put a super small coil in a very big varix. The standard that we use in binomials is this study, but it's sort of at least 30 to 50% longer than the actual diameter of your varix. Do I inject the perforator vessel or the varices or both? This one's tricky. You see both, what do I do? And really, we don't have much data. Just one study in Egypt looked at doing the perforator rather than the varices, and they thought that maybe it worked a bit better in terms of the number of sessions and similar adverse events. But there's something about injecting a main vessel that's feeding into your varices. My practice till now is I'll go for the varices itself, unless I feel like it's not working, then I'll go for the bigger perforator vessel. But sort of something to keep in mind. While injecting a coil into a varix, I see blood passing into the fundus. So especially if you have some water in there and you inject it, sometimes piercing the vessel itself, you'll see some blood start to pass through. I only saw this once during my fellowship with Dr. Yu that he was doing the case. And the key here is obviously stay calm, but continue doing what you're doing. Don't take your needle out and try to go down with an upper. Try to put that coil in and maybe put another coil in to induce thrombosis. These coils, if you guys ever see them and play with them, they're very thrombogenic. They have sort of these filaments that are projecting almost like hair-like filaments. And those, once they come in contact with blood, they really wanna thrombose pretty quickly. So here, if you have to chase it with some glue, and then at the last stage, just switch to an upper and make sure that you're looking at the area quite nicely and you don't have to use any adjunctive endoscopic interventions. What do I do? Do I do coils? Do I do both? Do I just do gel foam? The reality is we don't have very strong data comparing combination therapy. We looked at a meta-analysis at the Brigham that shows that maybe combination therapy was superior, but the reality is do what you're most comfortable with. Remember that embolization or systemic embolization, the highest risk happens when you're using glue, but there's some data that coils can embolize. The group at UNC used a Puristat recently that works. At the Brigham, we were using gel foam or absorbable gelatin sponge. So that's another option. This one's very important. You're doing a gastric variceal case and suddenly your patient is tachycardic and tachyarrhythmic specifically, and you don't think it's a bleed. At this point, consider embolization of coils. And we haven't seen it much, but it's starting to show up in the literature, at least two or three cases this past year where they've actually embolized. One of them actually was a cardiology group at Columbia that said that they had to go in with interventional cardiology means to retract the coil that had embolized and caused a V-tach or a V-fib of some sort. So it's something that can happen, which is why fluoroscopy is very, very important. Because you can actually see whether your coils are in the varix or whether they've migrated. US portal pressure measurements. So one or two more minutes here and we should be done. So you guys have seen the technique yesterday. I don't want to go over it here today just in the interest of time, but you're accessing the hepatic vein and the portal vein. Not a lot of things can go wrong here, but a few things that you can notice or try to troubleshoot when you're doing this. The one thing that I've struggled when I started doing this is, I can't see the hepatic vein very clearly. It's thin, it's hard to get to. I'm sure, Sai, you've had this as well with all your patients. And know that you have the inferior vena cava nearby. It is an off-label use. This is not something that we suggest to do, but it is going to give you a measurement of your hepatic pressures. The key thing here is make sure you're actually seeing the IVC and accessing the IVC. The IVC can look very similar to the aorta, but make sure it's not because it needs to be passing through the liver. You have to see parenchyma on either side. In fact, you should never be doing a stick to a vessel that's non-parenchymal. So the fact that you can go through liver is always a good thing because if a bleed happens, that parenchyma should be able to tampon out. And always look at your pulse flow to see that you have a monophasic flow rather than that classical aortic appearance. But this happens, and the IVC is a good surrogate to get your measurements when you're struggling with the hepatic vein. Some of my measurements don't make sense, are not consistent. This happens. You're like, what are these numbers even mean, right? So not uncommon. Ensure your patient's very sedated, and usually GA is what we use. Some people are starting to use MAC as long as they're very comfortable. Make sure your transducer is leveled at the patient. The company's now providing sort of a stabilizer that you can put to make sure it's at the level of the heart. Make sure that you're flushing it, that there's no bubbles in the system. Ensure your needle is not in the parenchyma or in the wall. If that's the case, then you're not really getting a true pressure. And then if you have any outliers, those usually don't make sense. So don't include them in your mean or your average. So if you're getting a 15, 15, 16, and then your first one or your last one was a 25 or a 26, that's probably a wrong read. Discard that and get your mean for the first three that you got. And that should be it for my side. So hopefully not too long. And thank you guys for staying. Thank you.

endohepatology

EUS-guided techniques

liver biopsy

portal pressure measurement

gastric varices coiling