All right. So to continue with the theme from yesterday, we kind of break up all of the didactics into similar subspecialties, but yesterday you learned about interesting cases within each subspecialty. Today it's going to be more focusing on troubleshooting. As you know, with advanced endoscopy, troubleshooting adverse events are unavoidable, so it's important to be aware of and know how to get out of it. So for the first lecture, it's my honor to introduce my dear colleague and friend, Neil Sharma. Dr. Sharma is the Director of Interventional Endoscopy and Endoscopic Surgery at UCHealth, and he's going to go over some of the troubleshooting during third space endoscopy cases. All right. So I consider Sai a good friend, but she's given me 15 minutes to talk about troubleshooting in third space, so I start to wonder, you know, how well your friends treat you. But here we go. So here's my disclosures. I'm a consultant at Boston Scientific, Medtronic, Olympus, and I'm the advisory board for endoscopy now. So I'll talk a little bit about SAEs and ESD, and this is probably one of the largest series published by VariExperience, so I wanted to give you maybe the best possible outcomes. So I spent a decent amount of time, and I'm going back to Japan and doing ESD training, and if you look at their outcomes, they typically will admit patients for about a five-day period of time just before and after doing an ESD, so all the ESDs get admitted. So I'm just giving you this is the best possible outcome, and in this series, we took a look at about 385 patients. If you take a look, significant interoperative or immediate bleeding, and they define that as basically a two-gram drop, happened at about a 7% rate. So that's a decent rate for people who are highly experienced. Postoperative bleeding, so it means that the patient, the procedure's done, they're now in the hospital. And then what's the rate of dropping two grams? Again, that's about a 5.3 to 5.6%, and these are gastric ESDs. Recognize that perhaps rectal ESDs, in my personal opinion, while there's a great safety profile because you're below the peritoneal reflection, I think they're more vascular, honestly, than gastric lesions, and I find myself encountering more bleeding. Interoperative perforation is a rate of 1.2 to 8.2% in gastric ESD. Again, this is with the most volume that they have. I think interoperative perforation, it's really important to recognize where you're at, and we'll talk about this. So what I mean by that is where are you at relative to the peritoneum and the diaphragm? So if you have a perforation full thickness inside the mediastinum, you're recognizing a different set of complications and different bacterial flora that you have to encounter and compensate for than when you're in the peritoneum, right? So if you're doing an esophageal ESD, it's going to be different than a gastric ESD. If you're below the peritoneal reflection, it really probably doesn't matter. To be honest, I go full thickness in the rectum on multiple cases. What I care the most about is to be cognizant of reproductive organs in men and women that fall in different areas, depending on where you're doing that full thickness resection. You can even leave it open when you're below that peritoneal reflection. And then there's other things to think about here. These are longer cases. So peritonitis can occur, especially if you have to go full thickness. Air embolism, thromboembolism, delayed bleeding, delayed perforation, and stenosis. These are all possibilities that can happen inside of complications to be aware of for ESD. So I want to talk a little bit about complications and how you can kind of get out of them, right? I think you're all aware of what these complications look like as advanced fellows, because you've already finished a GI fellowship. So these are probably the five key points that I'd consider for ESD when I put this together. Number one, patient selection. I think patient selection is a big part of how you build out a very successful program. So when I look at the program that we built in Indiana, we started focusing on EMR first and made sure we had a robust EMR program. This is going to be my Indiana bias. Othman's not in the room. We've had this debate multiple times in different venues. But I don't typically do right-sided ESD for adenomas, unless there's high-grade dysplasia and there's a concern for invasive cancer. Because I want to take that out on block and prove that I've done a curative resection for that patient. Otherwise, I was doing, honestly, EMR. And again, I have biases. Al-Haddad and I work closely together, and we have Doug Rex there. But I think the point of this is, be thoughtful that when you're embarking on an ESD, you're doing it for good indications and in the right patient. And ideally, for all of my cancer-related ESDs, they all went through a multidisciplinary tumor board. And they still do, to be honest with you. They go through a multidisciplinary tumor board first. We have agreement upon consensus for stage, which is a combination of cross-sectional imaging and EUS. And then we choose, as a team, to perform an ESD based on NCCN and other database guidelines. And that will keep you out of trouble, because you will get complications. But if you have complications when you have poor selection in any advanced endoscopy thing that you do, it's a lot harder to justify that, hopefully to yourself, but certainly to a patient or a court of law. The goal for ESD is to do an en bloc resection. So you need to be thoughtful about, were you able to complete the dissection and remove this dissection completely from the patient's body? So where I give you an example is, we had a STIR project going on. We're trying to remove gists. And I found myself understanding what the threshold limit was that I could remove a gist, not just en bloc, but get it out of the patient's mouth. Otherwise, you have to have a laparoscopic incision. So just be cognizant of, can you remove it? Can you remove it in one piece intact and get it out from the patient's body without an open incision? If you need an open incision, then it's a combined procedure with surgery, and be prepared for that. Be prepared for bleeding. It's going to happen. If you don't like bleeding, this is not the procedure for you. Know that a perforation can and will occur, even in the best of hands, and just be prepared for how to overcome that. So I think Sai and Mo have put together really nice stations around closure. Closure is a necessary skill set to have with ESD. If you can just cut it out but you don't know how to close things, it's going to be a bit of a problem, especially when you encounter a perforation. And then, understand that postoperative care is part of the procedure or continuum. So a new set of research that I'm trying to embark on is trying to understand, what is the standard of care of how we deal with these patients postoperatively? Because in the United States, I send all my ESDs home. When do we refeed those patients? Do they need continuous antibiotics? If you go full thickness, how do you tell those patients when they should be surveyed? Because most of our surveillance data is based upon guidelines really from EMR. So we haven't really figured those things out, but it is part of how you care for these patients. So I typically do like to see my ESD patients before. You don't have to. They are open access centers that have very robust APPs that maybe see them before. I typically like to see mine before, and then I'll usually follow up with them at least once after the procedure before maybe APPs or others will go back to that continuum. But that is a big part of it. And then we'll talk a little bit about managing perforations as we go through. So again, the goal is on block resection. I know this is the first lecture in the morning, so I'm going to play videos on the sideline. You can ask me any questions I want. I'll try to stick to the time. But are you able to complete the dissection is important. So this is like a duodenal neuroendocrine tumor, ESD. I want to take it completely out. As you can see, the vast majority of this tumor is not actually on the luminal side. So I need to know that I'm going to take this out. It's a carcinoid tumor, EOS, biopsy proven, reviewed at Tumor Board, and it needs to come completely out. So the deep margins for just an EMR, they're going to be positive 40% of the time. The point of this is that you need to have appropriate instrument selection. You need traction sometimes. You need to be careful with the specimen, and you need to communicate with the pathology team. So as you can see on this one, we ended up using traction. This is one of the ways that you can get out of trouble with perforations when you're going really deep. So you have like a clip-in-line traction that's pulling this away from the deep muscle layer, which I have to do a partial dissection of in the duodenal bulb, which is obviously a sketchy place in which you do this type of a dissection. And so understanding tools that will keep you out of trouble will allow you to remove lesions like this. Traction devices, I think, are integral to be able to be successful and avoid complications. I don't use traction all the time. Here's an example of a different type of traction. It's a traction wire where you actually attach a wire to the front portion of the lesion. You grab another clip. You drag the wire backwards, and the wire wants to open like a pop can, so this is actually put together. It's manufactured now by Medtronic, and it's called the Prodigy GI wire. Again, why I mention this in terms of ESD tips and tricks and how to get out of trouble, here to really get through this lesion, you'd have to go completely full thickness because it's so scarred down. It's been partially EMR'd before, and it's a high-grade dysplasia. I think something to be aware of is robotics will probably come into this space at some point in time. It's certainly been in animal labs over the past year where we're starting to work on robotics. So just know what all these tools are and how to utilize them. So when we think about traction, suture line, clip and rubber band like you saw previously, traction wire like you're seeing here, device-based, including over-the-scope devices such as Dialumendi where you can actually create some traction there, and then robotics. So be prepared for bleeding. I talked about this. This is a great example of bleeding. This is a large rectal lesion that has had a partial EMR before. They found high-grade dysplasia, no evidence of cancer on cross-sectional imaging or EOS. The lesion is a little bit difficult to lift, and even when you just touch it and start to lift it, it starts to bleed. So you know this is going to be a tough lesion to get through without a lot of bleeding. It needs to be prepared for kind of going into war with this type of lesion. So how can you get around it? You have to understand your electrosurgical settings, and that's really important. So most of the time when I'm talking about ESD, I really focus on understanding the current principles of current, and that's not what we're going to talk about today. But there's a variety of different settings here and tools that need to be utilized depending on how you want the waveform to act to be able to control that bleeding because you won't bleed during the procedure, and it's important to finish the procedure and to do it on block. This is a great example of some of those items that need to be aware of. Perforation management. So I think this is what freaks everybody out, right? So number one, and I think this is a whole lecture by itself, you need to identify the location. So I talked about above or below the diaphragm, above or below the peritoneal reflection. Ensure that the patient's stable. So how do you know if a patient's stable? Obviously vital signs. Patient's getting very tachycardic. If you're seeing changes in the O2 saturation, but also to feel on the abdomen, right? So if the abdomen is very rigid and you feel like it's at that area, you're really not able to decompress it through the hole in the peritoneum, you may need to do a decompression. I chose not to put a decompression video in here because I think you guys have all probably seen them during GI fellowship. But typically we'll prepare the area with betadine. I'll use a 14 or a 16 gauge angiocatheter. I'll attach a little bit of saline to it, so I'll empty it out, a little bit of saline on a back syringe, and I'll put it into the peritoneum, usually in the left lateral position. I'll percuss, and then we'll allow for some decompression. And I'll usually finish the procedure. Oftentimes, though, these are just tiny focal perforations that occur. And sometimes you can continue. So if it's below the peritoneal reflection, you can keep going. If you're inside the peritoneum, you may need to put a clip and then dissect around that clip. So understand that not all perforations are the same, and it depends on the size of it. And at the end, you have to understand how are you closing this, so that way the patient doesn't show up in the ER later on with more complications from the perforation. So I'm switching tracks here because this is all third space. So here's EPOM. This is SAE rates, again, almost a 400 patient study. And what they looked at is what are complication rates relevant to EPOM. So about 2.4% of patients in this series, again, very highly skilled series of patients being encountered, is about 2.4% AEs that are significant or major. And then about a third of them will have minor AEs. So what are major and minor? And you can see here, mucosal injury, bleeding, hemothorax. Minor AEs would be pneumoperitoneum, which may not require any intervention but can require hospitalization for discomfort. Pneumonia, which is actually the most common encountered overall adverse event. And then followed by pain, fever, and pleural effusion. So some tips and tricks to get out of trouble here. Again, it goes back to the fundamental, I think, of any procedure is patient selection. So in these patients that have motility disorders, I always like to work with my motility team and understand, will they truly benefit from this? And so when I do EPOM, like my recent move to Colorado, I got a lot of, quote, redo patients. So they come back to you. They've already had a heller myotomy. Maybe they've had a POEM. I certainly saw some patients that had POEM. And the question is, should we take this person back and maybe do it on the opposite side, the anterior side or the posterior side? And so I typically will have them go get an endoflip. I'll do a manometry. I'll talk to the patient about their symptoms. I'll assess the ECROS score myself. And I'll figure out, oh, maybe actually the IRP is completely fine. And what's happened here is you have just residual type 1 achalasia. They don't know how to, they're not really, haven't been coached on how to eat, right? And so you don't need to reoperate. So patient selection is extremely important. I think it's the same for GPOEM. Even for cricopharyngeal bars and diverticulotomy that I do a lot of, I like to really understand, do they have true symptoms? Will it be beneficial? Or is this just a radiographic finding? And I think that's really important. And then being honest with the outcomes for the patients with GPOEM is important as well. Know how to work in the tunnel. So if you don't work that tunnel the right way, you're going to get into troubles where you can't move the scope because the tunnel's too tight. You haven't made it wide enough. Or you may get into areas where you don't understand the landmarks of the tunnel. And when you don't understand that, you don't know when the tunnel begins and ends. And that can be a real issue when you try to go to the myotomy. Be prepared for bleeding. I think this continues. Perforation and leaks can occur even though you're using a tunneled approach. And you'll see this with STIR and POEM and so on. Once in a while you can get a leak or a perforation. And then postoperative care, again, is part of the continuum. So these patients, how are you going to manage a patient who did a diverticulotomy on? What's your standard of care of refeeding those patients? What should they expect in terms of throat pain, et cetera? So know how to work in the tunnel is a big one. I think this keeps you out of a lot of trouble. So I like to make sure that the tunnel entry is really smooth and reproducible. So I like to make sure that when I first get in, there's not bleeding in the very beginning of the tunnel, that I can make an incision. I think it's somewhat a preference if you want to do a horizontal or vertical incision for a GPOEM or for an EPOEM. I think for a diverticulotomy, there's a more standardized approach now. When I go into the tunnel, I like to keep the tunnel oriented. So you can usually do that on the shaft of the scope. You can put a little mark, make sure you're looking at the right position so you're not getting rotated in the tunnel. I keep the tunnel very wide. So when I get to the really tight GE junction, I have the ability to maneuver the scope. Otherwise, the scope will move away, and you'll get some problems inside of the damaging potential of the mucosa. And I want people to know their landmarks. So I think that's extremely important. So here's a great example of a landmark, right? So we've created a tunnel. In this brief video, we demonstrate the lower esophageal. And we've gotten down to the GE junction. This is what the GE junction looks like inside of a tunnel. I've seen people go way deep into the cardia, and it's not necessary because sometimes you get a blowout myotomy. So if you don't know kind of where your landmarks are, that can be an issue. I think understanding how far into the cardia it goes is important. I think more and more now we have data that's about one to two centimeters in the cardia. You don't need to go further. It definitely could lead to a blowout myotomy. And this is what the cardia looks like. So again, landmarks, understand when the muscle reorganizes past the high pressure point where you're at, and no stop signs. Because when you get deep into the cardia, you're going to get branches that come off the gastric artery. And those can be difficult to control bleeding if you dissect into one of those. Complications here include dissection of the diaphragm, dissection through the cardia, large gastric vessels that can't be controlled well with the bleeding devices that we have, and pericardial dissection have all happened before in the past. Bleeding control here, you can use coag graspers. You can use spray. You can use PreciseSect. You can also use low force. So you need to understand how to utilize these when you're encountering bleeding. Really large branch vessels where you feel like you can't control it with the knife, certainly you can use coag graspers. I almost never use coag graspers in any poem anymore. I typically use spray for very small vessels. I use PreciseSect and slow down the movement of the knife. And when you slow down the movement of the knife, the algorithm reduces the intensity of the waveform. And basically that allows it to have more cautery effect than cut effect. Or I'll use like a low force setting, 0.5, 0.6 on the VIO3, and just cook the vessel and then cut through it. And I think these are all important. And we skeletonize vessels when we go through. So here you're seeing us skeletonize vessels as we get there. You don't want to just cut through vessels rapidly. It's much more of a surgical principle of skeletonizing vessels. Know what the vessel looks like. Understand what's behind you because you're going to be cutting and you may have a second vessel right behind you. And then slowly, carefully get through the area. And that way, by going slowly, you actually go faster through the procedures. I found my operative times have gone up. I don't have to switch my instruments as often. And I'm able to control some pretty devastating bleeding. Bleeding control can be necessary in a myotomy, so you have to recognize your settings. And if you're going to go full thickness, make sure you understand what's behind you with full thickness. And recheck the tunnel at the end to make sure that there's no oozing when you try to close the tunnel and then it bleeds into the tunnel later. I think for time, I'll skip this little video. So finally, I think I have to talk a little bit about early and late complications here with POEM. Bleeding, leaking into the tunnel, ischemia, reflux, persistent dysphagia, and then potentially now there's data. Most of this is coming from the sages, so our surgical counterparts around Barrett's that's reoccurring, maybe perhaps due to reflux. And again, that's a little early, but it's something for us to be aware of. And I won't go through GPOEM because there's not time. But just for ZPOEM, I think this is a nice image that I tried to put together for you around where the landmarks are. So this is unfamiliar territory unless you scrub in with a lot of ENTs, which I would recommend. This is something I did. But you will get some leaks once in a while. If you get leaks, for the most part, they're conservatively managed by taking the patient NPO, getting a CT of the neck or an esophagram, putting them on antibiotics. You can go back and reclose. I've had two pretty significant leaks in the time that I've done a lot of these. And I've tried to reclose them both and been successful, but it can be difficult. And then time really is important for you. And so this is the bottom. So after you've dissected the cricopharyngeal muscle, this is the deepest area you want to go, the buccopharyngeal fascia. After that, you get to alar fascia and then to danger space. OK. I think I hit everything. We went over. Thank you for your talk. This is more for just a poem just because I saw that you did the drip test and you're going to posterior side. I think the data said there's no difference between anterior and posterior. And I was just wondering what you preferred and why. Yeah, this is a great question. So the data has not shown that anterior versus posterior have any difference in two things. Number one, the long term outcome, if the myotomy is adequate, it's adequate. And reflux. So there've been a few different ways where people have tried to do anti-reflux poem procedures. Another one is a spiral technique that I think Stavros tried to put together, but none of it changed it. I like to go posterior. It's personal preference. I think that if I get into some trouble in really, really tight, so I typically get a lot of really tight patients that have really bad disease. If I have to go a little further, I don't have as many vital structures. Going anteriorly is where I've seen, this is at a Sages conference, someone get into the pericardial sac and so on. But Dennis and I do a lot of research and Dennis likes to go anterior. So I think it's a personal preference. So just for you guys, obviously there's no great data suggesting one better than the other. The bottom line that you should know is that if you're going to do poem, you need to know how to do both and be good at both. Because there's going to be cases where post-surgical anatomy, you got to go posterior. So and then there's going to be cases where anterior may be better depending on the anatomy. So just got to learn both if you're going to do it. And then just one last point to add about Dr. Sharma's talk is if you want to do third space, I think one of the most important skills you need to develop is to be good at hemostasis. You really, really need to know how to control active bleeding, manage bleeding, prevent bleeding because that tends to be the major issue when you do third space. So it happens, and you need to be able to do this, but bleeding happens fairly routinely. So you need to be very good at that. Yeah, I completely agree. There are times I have to go the other way, anterior. I know Dennis has to go posterior, so that's important. And I think the bleeding principles rely a lot upon tools which are advancing and understanding the principles of electrocautery or electrosurgery. There's another question. For GPOME, are you selecting certain etiologies to do it? Has your practice been evolving on choosing those patients? Yeah, I think I talked about this in the ACT post-grad course last year, and I tried to go through all this data really, really carefully. What we report is technical success rates around 99% to 100%, so that means you cut the muscle. And then there's clinical success rates that are reported as high as 80%, and that's based on GCSI scores dropping by two factor. But that might not clinically correlate anything to the patient. So what I try to do is see these patients or talk to them, to a limited degree. They're all funneled through a motility program. I think that's similar to what probably others are doing as well, so the motility individuals are seeing them. I'm honest with my patients, so what I tell them is, look, it's a coin flip, to be honest. You've probably burnt out most of the time when they come to us. They've burnt out through all the medicines, have had side effects from the medicines, and so they're willing to try anything. But I try to be honest with them and their outcomes. If you look at the data, diabetic patients actually do the best in all the data. They drop GCSI scores by four or five factor. The patients that do the worst are pain predominant, where it's unknown etiology, and they don't have diabetes in their background.

advanced endoscopy

endoscopic submucosal dissection

complication management

intraoperative bleeding

third-space procedures

patient outcomes