We're going to start the afternoon session with Endoscopic Frontiers in Gastroenterology and Hepatology, which we're going to start with a point-counterpoint discussion where we wanted to evaluate or to take a look to see whether ESD or EMR is the right choice for removing polyps in the colon. So I'm going to have both presenters come up and join us. We'll have Antonio Mendoza-Led from University of California, Davis, and Dr. Shivangi Kothari from University of Rochester coming to the podium to present ESD versus EMR. Let's see who wins. No contest. All right. Good afternoon. I want to thank the organizers of the course for inviting me to speak today. I am going to give you a quick overview about pros and cons of ESD in the colon. I'm going to be quick and to the point to make the most of my 15 minutes up here. So these are my disclosures. So today I'm going to talk about why ESD in the colon, decision-making factors, advantages and disadvantages, indications and implications. So why ESD in the colon? Well, really, because we can at this point. 15 years ago, we would have not been having this talk, and it's grown from then to the point that we have it in the postgraduate course discussion. And there's been a recent boom in training in ESD in the U.S., and now it happens in many more centers than it used to be back in the day, so that's why it's important to talk about it. But how do we really draw the line about who we're going to treat with an EMR or an ESD? It all comes down to the optical diagnosis. We're moving away from tissue diagnosis, a.k.a. biopsy, and moving into the optical diagnosis way of approaching these lesions. When we talk about optical diagnosis, we talk about morphology. We have classification systems like the JANET, CUDA, and NICE classifications that help us make decisions, inform decisions about how we're going to treat our patients. We have to approach these lesions with last generation scopes, with high definition, near focus, ideally magnification, NBI, or any other light filters that would give us more information about these lesions. We need to determine if the lesion is benign or malignant, and we need to know if the lesion invades the submucosa or not. And to do that, we evaluate the borders of the lesion, the surface and the growth patterns, the location, and the relation to folds and flexures. This is the PARIS classification. I know most of you are familiar with it. I'm not going to go into detail with it. But I can tell you that most of the lesions that we talk about when we perform ESD in the colon are lesions that fall within the 2A and the 2C categories. It's important to remember that 2A classification of PARIS gets further subclassified into nodular and non-granular and non-granular lesions. The granular lesions are either homogeneous or mixed when the nodules are different sizes. And the non-granular ones are the ones that are basically smooth, flat, and they're divided into pseudodepressed or flat elevated. The highest risk of submucosal infiltration or invasion lies in those that are non-granular and a little bit also in the ones that have granular type but with dominant nodule, meaning a large out-of-proportion size nodule in the lesion. Talking about the classification systems we have, I'm not going to, again, not go into detail with them, but we have the NICE classification, which divides lesions in three categories based on color, vessels, and surface patterns. We have the KUDO classification, which focuses on the pit patterns of the lesion, and that divides lesions into seven categories. And then we have the JANET classification, which is a simplified version of the NICE classification by a group of Japanese experts. That's why it's called JANET. It classifies lesions into four categories, and it only focuses on vessels and surface patterns. Personally, this is the one that I like to use the most, but the question comes, which classification should I use, or which one is the best one? The best one is really the one that you're comfortable with. There is literature showing that they're roughly equivalent, so my advice is just pick one, stick to it, practice it, master it, and you should be okay. They are, again, roughly equivalent. If you can master all three of them, my hat's off to you. I try to keep things simple, so I only do JANET. So what about imaging? We know that EOS is not necessary to assess the depth into the submucosa of a lesion if it's evaluated by an expert endoscopist, and that's why optical diagnosis is so important. It's critical, really. At this point, we all need to know how to do it because this is how we are going to base our therapeutic decisions. CTs and MRIs are only to determine lymph node metastases or distant metastases, and it's important to remember we need to do them only before any resuction attempts because after a resuction has been done, the inflammatory lymph nodes that can happen after those can be misinterpreted as malignant lymph nodes, so it's important to remember that. So at this point, if the lesion is a good candidate for EMR, just go for it, right? There is plenty of evidence showing that EMR is a good way of treating these lesions, so just go ahead. But what if the lesion is not a candidate for EMR? So what then? So that's when we talk about ESD. So if we look at the numbers of ESD at this point, ESD has an end block resection of above 90% and above 80% are zero resection and curative resection. Those numbers are pretty good. We also have surgery, but you also have to remember that surgery comes with a higher incidence of significant adverse events. It's a little more expensive in terms of length of stay and costs of the treatment itself, and the quality of life for these patients is really not the same, especially if the lesions are in the rectum, so it's something to consider. Advantages of ESD at this point is that it allows us to do end block resections of large lesions that are not going to fit in our single size snare. It allows accurate pathological margin determination, which is critical for therapeutic decisions. It has a lower risk of recurrence. If it is not curative, it is at least a staging procedure, and I love this part about ESD. We know that endoscopic ultrasound is not great to give us the T-staging of the lesion, so if we go ahead and remove the lesion with all the appropriateness of the technique, we can know how deep into the semicolsa the lesion invades, and that gives us a very accurate T-staging of the lesion, and that it is effective in lesions that have semicolsal fibrosis. The disadvantages are that it is time-consuming, it is technically challenging, the learning curve is a little flatter than other procedures, there's no billing code. That's a huge problem, right? So we still need to overcome that. It has a higher risk of adverse events, and it's not available in all the centers. Yes, I started my talk saying it's available in many more centers now, but it still is not in every center that we can do this. So what are the societies saying about ESD at this point? The U.S. Multisociety Task Force recommends doing an ESD, or at least says that ESD is an appropriate treatment for lesions with suspected semicolsal invasion, meaning the large depressed one, and the pseudo-depressed non-granular lesions, which I mentioned in the beginning. Lesions with any kind of semicolsal fibrosis, residual early carcinomas that have been previously resected with endoscopy, and non-polypoid lesions in chronic inflammatory states like IBD. The ESG guidelines say that any lesion that cannot be taken in block with an EMR is a good candidate for an ESD, and especially the ones with high-grade dysplasia or tumor in situ, or SM1, which literally fall into the JANET2B classification of lesions in the rectum or anywhere in the colon with evidence of limited semicolsal invasion. The Japanese guidelines are similar, but they go a little bit more in-depth. They say that an ESD is adequate if an N-block resection with a snare is difficult or not possible, and they give specific examples about those lesions. Those are LSDs, or lateral spreading tumors that are non-granular, particularly the pseudo-depressed ones. These are lesions showing the 6-type pattern in ACUDA classification, carcinomas with shallow semicolsal invasion, large depressed-type tumors, and large protruded-type lesions suspected to be carcinoma. Again, notice that we are talking about shallow semicolsal invasion, early carcinoma. How do we know this? It's all about optical diagnosis, right? That's why it's so important to master this skill. Lesions with semicolsal fibrosis, and again, similar recommendations, lesions in IBD or previous lesions that have been resected endoscopically. This is a video of a rectal ESD of a nodular mixed-type lesion. You can see we've marked the borders of the lesion already. We're going to proceed with the semicolsal injection. Different injection solutions, different blue tangent agents. We proceed to do this into a mucosotomy. We're approaching this in a retroflex position, dissecting as close as possible to the muscle to ensure that we have good margins when we send the specimen. There are different knives, there are different injection solutions, different current settings, different generators. There's a wide variety of instruments that we have available nowadays to perform this procedure. You can see here how we very carefully dissect the semicolsal from the muscle. The muscle is the one that actually chars a little bit when we cut. That's one of the things that guide us when we are dissecting lesions like this. And also in lesions that have been previously resected that have a lot of semicolsal fibrosis, they can also char like that and have that appearance. In this case, the patient, the polyp involved beyond the dentate line. So we had to dissect the lesion from above the hemorrhoid plexus. And the principle is basically the same. We just have to be careful with not hitting the hemorrhoids because then it's not so much the perforation that we fear when we do ESD, not to say that we don't fear them, but bleeding is really something that we watch out for because it can interfere with our resections. So here we're just going to fast forward. We're just dissecting it from the anus. Someone will say, well, this patient is going to wake up in a lot of pain. Yes, I can guarantee you the patient woke up in a lot of pain, but I'm sure I can also tell you that he was much happier to have this done than having a surgery that would have compromised his continence. So it's important to be able to recognize that. And this is just the picture of the final specimen. This is another lesion. I think for the sake of time, I'm going to skip this one. So the implications of doing ESD go much more beyond the technicality of the procedure. Yes, the procedure is complicated. It takes a little time to learn. But the implications of it go beyond the endoscopic part. When you do ESD, you have to be comfortable not only with the technique. You have to be comfortable with the specimen handling and the pathological criteria we use to analyze these specimens. It's critical to have a multidisciplinary team to help you guide decisions, both pre and post resections. And this team should have surgeons, pathologists, interventional radiologists, diagnostic radiologists, oncologists, et cetera. So it's just important to have a good support to make your decisions pre and post. And this is the third point is something that I'm very keen about. And we don't really talk that much about this part at meetings like this. But it is important because it doesn't, it defeats the purpose of you spending two, three hours doing an ESD if your pathologist is not familiar with how he or she is going to analyze the specimen. And I can tell you out of experience, I worked in three different institutions at this point. In all three of them, I've had to sit down with the pathologists and go down, go about the criteria and how we need these specimens reported. That doesn't mean pathologists are bad pathologists. It's just that they're not trained with these criteria. Just like it's relatively new for us in the endoscopy world, it's definitely new for them in the U.S. trained pathologists. So it's worth having this conversation with your pathologist before you embark into a resection like this. Because again, you don't want to spend three hours doing a case and then not get the information you need. You have to be ready to be part of the oncology team of these patients. These are not biopsies in which we're just looking to exclude H. pylori. These lesions are a lot of times tumors, and we need to make sure that we follow up appropriately. We're going to be responsible not only for the resection, but also for the surveillance recommendations for any referrals that are going to need to be done or any follow-up workup studies. These all fall under the endoscopist at the moment you start doing procedures like this. Don't start what you can't finish, right? Because there is a lot of controversy about, at least, whether or not biopsies are a problem for subsequent resection attempts, which there is evidence that they do or they don't. But something that definitely is going to give a hard time for the person who's going to try to resect the specimen if it has been attempted to be resected before. So previous resection attempts definitely make future resection attempts much harder. So if you're going to go for an EMR, go for it. Finish it. If you're going to go for an ESD, finish it. Don't start something thinking, okay, if I fail, I can send it to the other guy because the other guy is going to have a very hard time, and you're going to be doing a disservice to the patient. How do we handle the specimens? We have to mount them in a rigid surface, pin them in a way that they resemble inside of appearance. The pins must go around the margin of normal tissue that we resect. We have to fix them as soon as possible, and give your pathologist minimum amounts of information like where is the lesion, what was the previous biopsy result, the morphology, the size, and which area is most likely to have malignancy. Like in the example I gave earlier, the domino nodule in the mixed type lesions is where typically the malignancy or the invasive part is. Here's another picture of how we pin specimens. So many of you know these criteria, but they're worth reviewing. M-block resection is a specimen that comes out in one piece. Piecemeal resection is the opposite. We resect it fragmentedly. R0 resection is a resection that has lateral and vertical margins that are free of tumor. And a curative resection is an N-block R0 resection that meets pathological criteria for a cure. And which ones are those criteria? Well-differentiated tumors, lateral and vertical margins that are free of tumor. And when we talk about colon tumors, we're talking about a vertical margin that is less than 1,000 microns from the lower border of the muscle or is mucosa. These are very important and critical concepts that pathologists are not familiar with. I was not familiar with when I started doing these procedures. So it's important to know the literature so you can have a good understanding with your team about how you need these specimens reported. There has to be absence of lymphovascular and perineural invasion and absence of low-grade tumor budding. However, there's new evidence showing that the depth of semicolon invasion may not be as important in colon as it is in other organs like stomach and esophagus. This is a meta-analysis of over 20,000 patients that showed a semicolon invasion more than 1,000 microns that was not a significant predictor of lymph node metastasis, as opposed to poor differentiation of the tumor, high-grade tumor budding, or lymphovascular invasion. Another study, a retrospective cohort of about 1,400 patients, 150 of those had lymph node metastasis. They found that semicolon invasion was only a dependent risk factor for it, but it lost its significance once they entered the multivariate analysis. Important to mention in this study, they did not quantify the semicolon invasion. They just subdivided into SM1, 2, and 3, which talks about the upper, mid, and lower thirds of the semicolons of the specimen. Recommendations after ESD, these are all recommendations based on expert consensus, no guidelines, but it is recommended to do a surveillance colonoscopy after one year if your resection was curative. And if it was not, or if it was fragmented, or the margins could not be assessed, they recommend doing a follow-up colonoscopy in six months. However, in this second scenario, I would recommend you discussing with a multidisciplinary tumor board and come up with the best way of monitoring these patients. So in conclusion, based on available evidence, most of the lesions in the colon at this point can be removed with an EMR. However, it is our responsibility to learn how to recognize the ones that need ESD. If you decide to pursue ESD, understand the implications of it, and don't start where you can finish, because again, we want to achieve the best outcomes for our patients. Thank you. All right. Thank you, Dr. Lahrd, for an amazing talk. And my job is to talk about EMR. And as you can see, the title of my talk gives the conclusion, EMR is the way to go. So on that note, those are my disclosures. I will be talking about some good practices in the world of EMR, the assessment, the tools we have, the techniques for EMR, and of course, why EMR is more useful than ESD in the colon-large polyps. So let's look at the guidelines. When you look at the ESG or the Multi-Society Task Force guidelines, not just for small polyps, but even for polyps that are over two centimeters, that are considered superficial, EMR or piecemeal EMR is the recommendation. And ESD is reserved truly for the lesions that are suspected to have early submucosal invasion. We are not here discussing the small polyps. The polyps that we're concerned about are the large two, three centimeter polyps, which can harbor malignancy and are usually found in the rectum or the right colon. The success for resecting these in the colon is over 90%. And that is where the paradigm has shifted, from where these patients needed surgery to now we have great endoscopic tools, techniques between EMR and ESD to successfully resect these and avoid big surgeries in these patients. So Dr. Mendoza-Lahr did talk about the visual aspect and the assessment of polyps. Here you can see in this video by Dr. Friedland, a patient referred for endoscopic resection of a polyp and has clear signs of cancer or deeper invasion. And this is not the kind of lesion you want to tackle. You can see there are depressed areas. This was a frank cancer. So this is not the lesion that you want to consider EMR or ESD on. So question is, when you see this on your screening colonoscopy, what do you do? Do you biopsy these, EMR this, ESD this, or just send the patient to surgery? And there are a couple of do's and don'ts when you see them on a screening colon. Carefully look with white light narrowband imaging. Assess it. See if there are deeper invasive features in the polyp. Take good pictures. Even if you're going to refer it, trust me, a picture is worth 1,000 words. A 2-3 centimeter polyp to one gastroenterologist may be much larger to another. Tattoo if the polyp is not obvious. You do not need to tattoo a polyp that's in the cecum or the rectum. Trust me, the interventionist will find it. It's the one that is subtle or in the transverse colon behind the fold. Then you can consider tattooing away from the polyp, at least two to three folds away on the opposite wall. As Dr. Mendoza-Large said, only begin resecting if you feel you can finish it. Do not partial biopsy, partially resect it, because all those things cause fibrosis. It burns the patient's bridge to a successful resection by you at a later stage or by another physician who's going to step in the care. So pearls for a safe hot snare polypectomy, get a good informed consent. Consent for a large polyp endoscopic resection is very different from a screening colonoscopy consent. So you make sure you bring the patient back. Have a good discussion. Make sure you've allotted enough time to tackle the lesion at hand. Use CO to optimize the anticoagulation. I personally prefer using a clear cap. And locate the polyp and analyze the shape, size, surface. See if the polyp is firm to probing. What's the mobility like? Position the polyp well to the 5, 6 o'clock if you can. And straighten out your loop. Make sure you and the scope are in good position to get a good successful resection. And use submucosal injection to evaluate for non-lifting signs. So if the polyp is really scarred, has those depressed areas, then when you inject it, it's not going to lift. So this is a telltale sign that polyp has more invasive features and is probably a cancer. So here you can see an unblocked resection. We are assessing the polyp. Unblocked means you're going to take the polyp out in one piece. You always want to inject the far wall first. Bring the polyp towards you. And then inject the distal side. Get a good submucosal cushion. I use a firm braided snare to resect it. Move it a little to see if the whole wall is moving with it. And then always check the undersurface of the resection piece as well as the base to make sure there's no target sign. This was in the right side of the colon, so we did clip it to minimize any risk of delayed bleeding. The steps for hard snare polyp piecemeal resections, first is inspection. You want to assess the polyp. Inject it well. And then resect it with cautery. And then there is more of the same till all the visible neoplasia is resected. If you're doing a piecemeal resection, ablate the margins to minimize the risk of recurrence. And retrieve every polyp piece that you have resected because you want to assess it for any kind of neoplasia or anything that changes the patient's surveillance or subsequent plans at management. Prophylactic clipping, if needed. Of course, if there are large lesions, right-sided, patients going back on anticoagulation, these are the patients you can consider clipping to minimize the risk of delayed bleeding. And of course, surveillance is key in six months. And if you have recurrence, the majority of this can be managed endoscopically. So here you can see a large laterally-splitting polyp in the cecum on narrowband imaging. There were no invasive features seen. You want to start at one edge of the polyp, raise it well, and then resect it with a firm snare. You do not want to raise the polyp all at once. Again, the preference for injectate, I personally just use a saline mixed with methylene blue. You could use any of the injectates that are pre-mixed available or what's available in your endo unit. Here I'm using a stiff snare as key for the EMRs. And once you've resected it, lift it up a little so you're always protecting the mucosa. Take a look at the base. And as I said, it's more of the same. Resect the whole thing. I always ablate the margins when I'm doing a piecemeal resection. This one step can help decrease the risk of recurrence of the polyp from almost 15 to 20% down to less than 5%. And of course, we clip it close to minimize the risk of delayed bleeding. Cold snare piecemeal EMR is now being more and more done and talked about. It is mainly to minimize the risk of delayed bleeding or perforation. And of course, a lot of the studies have shown its efficacy and safety. However, there is a risk of recurrence with it, especially with the adenomas. The risk is higher when compared to salicylated lesions, and you could consider ablating the margin after you've done a piecemeal cold snare EMR. So now talking about the good techniques for EMR, we've gone over some of these pros and cons. The pros for EMR is that it's faster, it's safer, and all this is literature-based, and I will go over the literature to support my points. Lesser expensive accessories, it's relatively easier to perform, and of course, the con is, which literature has shown that there is a higher risk of recurrence, but I will also show you techniques on how to decrease that con rates. ESD, of course, can get you on block resection, has a significantly lesser recurrence rate. However, if you look at the cons, it is a longer duration procedure, and in the United States, time is money. Higher risk of PERF, expensive knives, higher skillset is needed, and it is limited to expert centers, and of course, reimbursement continues to remain an issue. So if you look at the indications for ESD, it is, there are very few strong indications. The main indications are predicted superficial cancer. So for your benign superficial pre-cancerous lesions, EMR is sufficient. For the patients that have deep submucosal invasion that have a higher risk of lymph node spread, there is surgery. It is for those patients that are in the middle that have early submucosal invasion where you consider ESD, and of course, lesions that have fibrosis, and ESD gives us the ability to dissect deeper tissue. So talking about recurrence rate, yes, EMR has a higher recurrence rate, especially in large polyps that are greater than four centimeters when you're taking out more pieces, and the goal to decrease recurrence with EMR is that you have to resect all the visible neoplasia when you're doing the EMR and ablation of the resection margins, and these help decrease the recurrence rate to less than 4%. Here you can see central scarred portion, the polyp is being removed with a jumbo biopsy forceps. You can also use the hot avulsion technique by using a hot biopsy to remove any residual small pieces, and studies have shown that there is no significant difference between avulsion and the non-avulsion techniques in terms of recurrence rates. Adjuvant ablation of the resection margin, you can either do it with APC or the snare tip soft coagulation at 80 watts. From Australia, Dr. Berkey's group has reported significantly for the snare tip soft coagulation ablation. You don't have to open another accessory, and both these modalities help decrease the recurrence rate in piecemeal EMR down to 5% from 20%. So both EMR and ESD are effective. There are several meta-analyses that have been done on it, but ESD does have a higher complication risk with bleeding, perforation, and also there are no studies that have looked at the long-term cancer or mortality reduction with ESD. So I think in the colon, ESD still needs to evolve and have that long-term outcomes data. EMR is effective in over 98% of the cases, and even in cases of recurrence, you can resect the recurrence successfully endoscopically to have a long-term adenoma-free survival of 98%. You can see on the right, this polyp was sent to me. There is a tattoo right underneath the polyp, and I was able to successfully resect it, and all the blue you are seeing is not my inject aid. That's all the tattoo. Of course, you bring back the patient for surveillance, and you see a small recurrence at the area which I successfully resected and ablated the site to minimize any recurrence. This is a large rectal polyp, just to match up apples and apples to what Dr. Mendoza-Large showed, and this was sent to me. Patient was not a surgical candidate, so mind you, you're dealing with high-risk patients in this question. Surgeon did not want to operate, and I brought this patient for an EMR, had a good discussion in clinic. We're doing narrowband imaging. There were no obvious depressed areas. Resected the entire rectum almost. It was going in about 10 to 12 centimeters, occupying at least 70% of the lumen. This took me half an hour, and this is six months later, and the patient did well. So the goal is you want to bring the patient back for surveillance at six months, then one year, and then three years after the second exam as per the guidelines. ESD does have a learning curve. There is no formal training. Of course, there are a lot of programs. The ASGE has programs where you can get access to experts and get some hands-on training. Mentorship is not easily available. There is a steep learning curve, and of course, when you look at the studies, ESD has been shown to be significantly more time-consuming than EMR. So in conclusion, my last slide and the first slide say the same thing. EMR is the way to go for benign colon polyps in the United States. It's efficacious. It's safe. It has a shorter learning curve. It's time-efficient. We use tools and techniques which we are familiar with with regular polypectomies that we do on a daily basis in colonoscopy. Limited scenarios with ESD and hybrid procedure. It's essentially preferred when there is, as I said, suspected cancer or significant fibrosis. So just because you can doesn't mean you should for ESD. And as Dr. Uzma Siddiqui says, do not use a cannon to kill a mosquito. Thank you.

Endoscopic Submucosal Dissection

Endoscopic Mucosal Resection

colon polyps

optical diagnosis

en bloc resections

benign colon polyps

recurrence