First of all, this is great to have Ken with us here, and his talk is entitled Endoscopic Resection Techniques in the Esophagus, really from one of the world pioneers in this area. So KKW, take it away. Thanks a lot, Gary, Felice. Way to make me feel wanted. I feel badly. I really wanted to get to introduce you, too. I feel badly. Maybe I'll, like, send you off the podium afterwards, okay? You can do that now. It's okay. All right, everybody. It's getting late. Three more talks left. Well, this is one of them. So we'll get this over with quickly. I'm going to talk about endoscopic resection in Barrett's esophagus. These are my disclosures. Wish I had more. Goals. This is going to be a case-based discussion. We're going to talk about the rationale for doing endoscopic resection, the right resection for the appropriate lesion, and comprehend the anticipated complications. All right, everybody. It's getting late. Three more talks left. Well, this is one of them. So we'll get this over with quickly. I'm going to talk about endoscopic resection in Barrett's esophagus. These are my disclosures. Wish I had more. Goals. This is going to be a case-based discussion. endoscopic resection, the right resection for the appropriate lesion, and comprehend the anticipated complications. So like everybody else today, let's see what the guidelines say. And the 2022 guidelines are similar to those in the past that recommend that we do initial endoscopic resection of any visible lesion before the application of ablative therapy in patients undergoing endoscopic eradication therapy. Note the strength of recommendations about the lowest conditional and the quality of evidence is very low. It's actually primary rationale, if you read the guideline, is improved histology, better agreement between pathologists, and secondarily, improved staging if there's a cancer involved. So that's the reason we're doing them. And I'm going to take a case like this. Say there's a 72-year-old gentleman, iron deficiency, anemia, multiple issues dealing with age that undergoes endoscopy for anemia. As we usually run into, they find by happenstance a 5 centimeter Barrett's esophagus and biopsy show high-grade dysplasia. No evidence of any large macroscopic lesions, some Cameron lesions that may explain the anemia and some adenomas. And you see something like this, a C5-M5 by product classification, you have some distorted mucosal glands in certain areas, maybe slightly elevated, glandular atypia mid-segment. This is a person you know that somewhere along the line, they biopsy showed high-grade dysplasia. Now, this is a flat lesion. One could argue that maybe it's 2B or maybe 2A, if you believe that it's just slightly elevated. But, you know, you should be able to describe this adequately, kind of like Felice was talking about earlier today. We need to standardize our terminology so that the next endoscopist can have some idea of what he's dealing with. But, you know, know your Paris classification. The other thing you need to do is be able to describe the MBI magnification descriptor of the lesion as well. Look at the mucosal pattern. Does it seem regular? Is it irregular? The vascular patterns. These are basically, you know, all the systems that exist, it's basically are they regular or are they not? And if they're not, they're suspicious for neoplasia. They actually have a fairly high sensitivity specificity. As you know, the ASGE had PIVI guidelines and MBI actually made those in expert hands to distinguish dysplasia. The next step would actually be to act on that and potentially resect these lesions. Now, what should you do? You have lesions that you suspect. This is the way I look at it. If you think it's benign, you know, you don't have to be as aggressive. If it's likely to be malignant, I almost always go to a big resection, ESD, because, you know, you never know the margins, stuff like that. No matter how hard we look, everything looks a little funny. If it's a cancer, you want it all out. R0 resection is paramount. If you think it's benign, it's a relatively small lesion, yeah, go ahead and EMR it. If it's excavated or depressed, and this is something the Japanese brought out in their most recent guidelines about esophageal cancer, anything that's depressed or excavated, you really have to go to a more aggressive mechanism to resect. Anything that's bigger, I think you should go to ESD. So you see a lot of ESDs in there. That's because I think we're getting more comfortable in the West with doing these procedures. A lot of you are learning third space endoscopy. How many of you actually do EMR? How many of you do ESD? Yeah, and as you become more comfortable, the indications expand like any other procedure we do. And I think it's something that really is useful, especially as we're dealing with malignancies. Maybe it's me. All right, would you resect this lesion? How many of you would? Just based on the endoscopic appearance, how many of you would resect the area of mucosa? Raise your hands. One. Oh, well, can't convince anybody. The pros are you rule out if there's any kind of cancer present. You eliminate the high-grade dysplasia. One of the problems with ablation is you're never sure you got the right area. And it reassures the patient afterwards when you remove it. Patients love to hear that, we took the lesion out. The cons, of course, are the risk of the procedure. The patient doesn't want to be screaming in pain saying, ah, you perforated me. There's many different techniques which we can use. And they're all actually relatively safe. The most common is, of course, the EMR band technique, suck up the lesion, drop the band, and cut it off. You know, it's basically a polypectomy with a little assistance. Relatively simple to do, relatively safe. You notice in these particular illustrations, there's a submucosal injection, which really isn't in the manufacturer's use diagrams that they give you. Nobody says pre-inject. I do believe in doing this, though. I don't really use it for the cushion. I use it just to lift the lesion up so it's easier to suck up. But I don't see much harm in it, and I think it can do some good in terms of decreasing complication rates. The other one is the old CAP technique, which not many people use anymore. How many of you use CAP EMR? So a few do. Yeah, and the band's much easier. The CAP is cheaper and actually will help you in lesions that don't lift as well. You know, if they are not coming into the CAP very easily, this type of CAP EMR can work to get it out. The toughest thing is step two. In other words, positioning that snare right at the lip of the CAP, that's the most difficult step. Takes a little practice, but, you know, most people can do it. This is my personal experience, 681 patients, 2,500 EMRs. No perforations, bleeding in 1%, stricture rate in 1%. So you know, it's not very high, but this was published in 2013. If you look at something that was published recently, this happens to be a UK experience, only of 221 EMRs, but once again, pretty low risk, 2.5% acute complications, only one perforation and bleeding in a couple of patients, late complications or strictures. So that's really what happens in patients. That's what we tell them. The risk is 1% to 2% of serious complications occurring. And I think, you know, it's something you should think about. Now, going on to another case, we're talking about a mass now. These are serious lesions, 68-year-old, junctional mass, had completed RFA ablation. This is something to keep in mind. One of the big predictors of recurrences is actually high-grade dysplasia or cancer in the initial biopsies. So those people that have the highest risk of recurrence are also those at highest risk of bad disease. So this is a guy who finished Barrett's ablation a year ago, has lots of reasons why he shouldn't undergo surgery, STEMI, COPD, basically the metabolic syndrome. Surgical risk is high. And this is where the surgeons kind of step back and say, well, let the endoscopist try and do something. And the patient naturally wants curative therapy. This is a big lesion. And it looks like it's submucosal, may go deep. You evaluate it, CT PET shows a lesion at the GE junction, no evidence of lymph nodes that are involved or distal metastases. EUS shows a T2 lesion, 2.5 centimeters longest diameter, no lymph nodes. Biopsies show adenocarcinoma, high-grade lesion. A friend of yours that works with you tried to inject it, couldn't lift, so said, gee, you know, you're the local expert. You take a shot at this. Mainly because if there's any complication, this guy could die on the table. So most people don't want to touch these. And these are what we end up seeing a lot of. Should we attempt to resect the T2 lesion? How many of you would do that? A clinical T2? No one? Actually, you know, the EUS isn't very good. If you look at the literature, it's not very reliable on staging early cancers in general. Overstaging of T1A lesions occurs in 40% of the patients, overstaging. And then the positive predictive value of a T1B lesion into the submucosa is about 49%. Not that good. A coin flip. You know, you could skip the EUS and just flip the coin. And in a study that was done in Germany, 38% to 41%. They actually did a retrospective part of this study and then a prospective part. With roughly 40% of EUS stage T2 lesions turned out to be T1 on resection. So you know, a lot of these lesions are resectable. And they could be resected for a cure. So given the morbidity of surgical resection, endoscopic therapy obviously would be preferred in these cases. And that just tells you the weakness of endoscopic ultrasound and being able to predict resectability. You really probably should try to inject it. Now this patient had a non-lifting sign. It didn't lift. You know, so now when we got to EUS, you got non-lifting. But something you got to keep in mind, right, non-lifting could be that you're, what you're most afraid of, that this is tumor adhering to the esophageal wall and you got a big problem. But it's also possible that, you know, you're just not getting into the right space. Whenever I call something non-lifting, I want to see that needle lift something that should be lifting, like little more proximally or distally, and then see it not be able to lift that lesion. If you get no lifting at all, you're doing something wrong. The solution might be a little too viscous. Every one of your assistants does it a little differently despite the recipes they have. Somebody doesn't measure well. If it's too viscous, you can't inject it. There's needle malfunctions. And finally, we have aging populations of both gastroenterologists and our assistants, and some of them just don't have the strength to get that lesion up. You know, you get somebody junior that's a weightlifter. The method of resection, you could try a piecemeal EMR, just kind of nibble on the edges to try to get that. That could be very difficult, and with a non-lifting lesion could be impossible. You could try to do a full thickness resection. If there isn't a vesicle unit out there, I don't know if that's actually going to be on the hands-on. I don't think it is. But it's not, you know, it's designed for the colon. You can get it down a male esophagus. You should size it up first. And ESD can be done too. That actually would be my preference. Give it a shot. See if you can get in there and take it out. Usually what's nice is, as those of you who do third space endoscopy know, if it goes deep, you can see it. You can see these little, like, stalactite type of lesions coming from the tumor into the submucosa. It's really kind of cool. Not for the patient, but for us it makes good pictures. Submucosal fibrosis is something you have to worry about. That could be what's sticking this thing down there. And this just shows you something the Japanese brought out in 2010, a classification system. And it's quite simple. F0 is no fibrosis. The Japanese found all these lesions. This is me and I just used Photoshop. It's a whole lot easier than finding patients. With F1 fibrosis, you got strands of fibrosis in the submucosa, kind of like you can put in there. This is what you go to send the insurance company when they deny coverage. Yeah, I had pictures. And then F2 fibrosis are actually bands. Those are really sticking that tissue to the submucosa. And you can Photoshop that in too. So submucosal fibrosis and ESD in the colon, this was the original study. And what was interesting about it is not only their classification system, but what they found that it did. There is less R0 resection possible once you get a lot of resection. And you get increased complications, both perforation and bleeding. What's interesting is with F0, F1, no fibrosis, with experience, you get better. With F2 fibrosis, they didn't see any change in their resectability range. So this is really a difficult lesion. They've shown this in the colon, in the stomach, and in squamous esophageal tissue, even though this is a relatively small series. It's much more applicable in the US situation, because we usually end up with these patients after ablative therapy, like this case. And we recently did an analysis of looking at fibrosis in the submucosa in patients with Barrett's esophagus. And believe it or not, the most predictive factor was actually ablative therapy, thermal ablative therapy, mainly RFA, not EMR. Basically, the RFA increased your risk of having F2 fibrosis. Now ESD of Barrett's lesions, it's pretty standardized. You mark out the lesion, go around, cauterize a bunch of dots. They start cutting, usually start in the proximal portion. And it depends on the knife you use. This is the complicated part. There's like 9 million different knives. It's kind of like tools and a toolkit. They're all used a little differently. There are knives that are directional, meaning they work better in one direction, like a hook knife. In those, you have to start laterally. There are unidirectional knives that you can start right in the middle and work both sides. There's all kinds of technical things. If you're interested, I suggest you take the ASGE course, or one of the courses available for ESD. Once you make your opening into the submucosal space, that's really cool. You see this blue stuff. You just keep marching your way through there with a submucosal dissection. This is just a sampling of the knives available now. There's probably about three others that are not shown here. And the thing of it is, every Japanese master makes his own knife. So they all promote that knife. And there's a bit of a problem there because different centers use different knives. Some places use lots of knives. You'll see in Korea, they only use one knife because they're only reimbursed for one knife. So everywhere, finances come into play. I personally prefer a scissor-type knife. I think it's easier to use, especially in the States where you're doing occasional ESDs. You're not doing them every day, 50 a day, all this other stuff. I use one knife for everything. This can mark, this can cauterize, this can cut and do the submucosal dissection. Seems to work all right. And we've published that this seems to be safer for novice users in particular. These are ESD-naive people. It allowed us to do same-day discharge. Previously, we held people overnight in the hospital. Since this, there has been no increased complications. This one plays too, but it shows essentially the same thing. And the complication rates are really quite low. They were similar to that of our CAP EMR. So that's why we made this a go-to type of procedure. One thing I do with this that the Japanese also do, and that is I cauterize before I cut. And if you do that, as you can see, you almost end up with relatively bloodless procedures. And that really makes the procedure go faster, rather than cut and then cauterize afterwards. Because once you get blood in the field, it's terrible. That's a vessel that got cut there, so I'll cauterize that later. Everybody can afford to lose a little blood. All right. So the recommended use of ESD in esophageal cancer, there's a practice update a few years ago that we did. And really, it's suggested for relatively larger type tumors. But if they get too big, and there was a recent, actually it was an American study from Memorial Sloan Kettering looking at overall outcomes of T1 cancers. And one of the predictive factors was actually size. If it was greater than two centimeters, much higher risk of recurrence. And if it had lymphovascular invasion. And also poor histology. If you're poorly differentiated, you have a higher risk of recurrence. Morphology, narrow band imaging to get those pit patterns out. By the way, if it's totally avascular, it's most likely an invasive cancer. So something to be aware of. Depth of invasion. Basically, there's different guidelines for whether or not there's squamous cell cancers or adenocarcinomas. Adenocarcinomas tend not to metastasize as fast. So we're able to resect these even if they invade down to about 500 microns. Anything above that is curable with endoscopic resection. With squamous cell cancers, about 200 microns is what the Japanese use. So treat those squamous cancers with a lot of respect. And also, they have skip lesions. So they're hard to find. They won't necessarily metastasize to the local nodal group. They could easily go from mid-esophagus down to the junction or celiac area without involving the nodes in between. So you really have to be careful when you do those. This just happens to be a lesion that was first EMRed. You can't even see the base of it. This is ESD. That shows you what was inside that. It was all squamous, re-epithelialized, but you can't even appreciate that from the resection site, right? So we tend not to want anybody to mess up these things before we take a look at them. So in summary, what's the rationale? Well, we want on-block resection to improve the interpretation of histology and get the right depth, the right resection type. EMR is definitely reserved for smaller lesions, I think, that are not likely to be malignant ESD for everything else. And the complications really are bleeding and perforation that you have to worry about, late complications of strictures. Thank you very much.

endoscopic resection techniques

esophagus

histology

EMR

ESD

resectability