Welcome, and good evening. The American Society for Gastrointestinal Endoscopy appreciates your participation in tonight's webinar. My name is Ed Dellert, and I will be facilitating this presentation. Our program tonight is entitled The Use of NBI in Colonoscopy. Please note that this presentation is being recorded and will be posted on GILeap, ASGE's online learning management platform. You will have ongoing access to the recording in GILeap as part of your registration. Before we get started, let's take a brief tour of this new platform that ASGE has launched this past weekend so that you are aware of the many resources available to you during and after tonight's program. On screen, you should see the main lobby of our new platform. You'll see many features that are in here. You'll see our corporate partners that is on the signage on the left-hand side. You will see our banner here. You'll see meeting information, which has our agenda for tonight and some resources related to our topic. You'll also see a satellite symposia from this past weekend's program that's available for your viewing pleasure. You have already managed to get into the auditorium, which is the main, and you can navigate by using any of the buttons on the footer of your screen here. I would call your attention to the exhibit hall. We have a number of exhibitors that are available, including ASGE. Feel free to visit during and after and logging back into this program to visit those sources. We have a resource room where you can, if you're into history of endoscopy, you can find resources there for that or meet the masters from our video GIE journal is available to you for your viewing pleasure. Access to our guidelines are available here. If you like gaming, we have some US and ERCP games to play around with and then access to GIE. In our networking lounge, you will notice that there is our survey for tonight's webinar. We would encourage you to come here at the end of the webinar and fill this out. It'll only take you a couple of minutes. And then obviously there's some other networking social pieces that you can do with the people participating tonight. You can chat with them and then you can add things to your briefcase. So I'm going to stop sharing here and come back to our main part of our presentation. Now it is my pleasure to introduce tonight's presenter, Dr. Doug Rex. Dr. Rex is a distinguished professor emeritus of medicine at Indiana University School of Medicine, Chancellor's Professor at Indiana University Purdue University in Indianapolis and Director of Endoscopy at Indiana University Hospital in Indianapolis. He graduated from Harvard College in 1976 and from Indiana University School of Medicine in 1980. He served as chief medical resident at Indiana University Hospital and joined the faculty at the university in 1985. He has received many teaching awards throughout his career. He has authored more than 250 original research papers, 60 book chapters, 120 invited papers, 60 editorials, 40 guideline papers. He has also been the associate editor for journal Watch Gastroenterology for 19 years and serves on the editorial boards of 11 gastroenterology journals. He is also the editor-in-chief of ASGE's Journal Scan. I would encourage you to view that if you have not already. He has also served in the American College of Gastroenterology as chairman of the board of governors and as past president of ASGE. He is a master of ASGE in the American College of Physicians and he is currently president-elect of ASGE and a recipient of ASGE's Master Endoscopist Award and the Rudolph Schindler Award. It is our great pleasure to turn tonight's presentation over to Dr. Rex. Dr. Rex? Thanks, Ed. I'm going to give a very clinical presentation about narrowband imaging in colonoscopy. Let me just start by giving sort of an overview of what I'm going to talk about. So NBI, narrowband imaging, which people often talk about the science behind this, which is shining a couple of wavelengths, but predominantly blue light on the colon. I have a nasty habit of just saying, you know, turn on the blue light. The advantage of it is that it tends to highlight blood vessels. The blood vessel pattern in the background, normal mucosa, and also vessels that are present in the surface of lesions, which is going to be the main focus for us tonight. So we're going to talk about the use of NBI. We'll focus on the NICE classification. NICE stands for NBI International Colorectal Endoscopic Classification. This is a validated classification that divides lesions up into three groups, the type 1 lesions, which are the serrated lesions, the type 2 lesions, which are adenomas, and then type 3 lesions, which are lesions that have deep submucosal invasion. Okay, so type 1 lesions are the serrated class lesions, type 2 are the adenomas, and type 3 are lesions that have deep submucosal invasion. So the NICE classification within the type 1 lesions does not attempt to make the distinction between hyperplastic polyps and sessile serrated lesions. There's a separate classification for that, the WASP system, which I don't think has a huge amount of clinical application at the present time. And within the type 2 lesions, it also makes no attempt to distinguish lesions according to whether or not they've got high grade dysplasia. There's a variation or extension of the NICE classification called J-NET, which is commonly used in Japan, and less so in the United States. J-NET 2A lesions tend to have low grade dysplasia, J-NET 2B lesions have a higher risk of high grade dysplasia or superficial submucosal invasion, but we tend to not use that. I'm not going to talk about it tonight. And then type 3 lesions, which are lesions that have deep submucosal invasion of cancer. So as I hope everyone knows, in the colon, when we talk about colorectal cancer, our clinical definition of it is submucosal invasion. Without submucosal invasion, there really isn't cancer. So I find the terms carcinoma in situ, intramucosal cancer that are used in the Vienna classification to be confusing for patients and physicians, and so I don't use them at all. I think they should just be called high grade dysplasia. Once we have cancer in the submucosal, we divide that into superficial, which is basically less than a millimeter of invasion versus deep submucosal invasion, more than a millimeter or a thousand microns of invasion. So the deep submucosal invasion by itself, in the absence of things like lymphovascular invasion, high grade tumor budding, etc., that it's not a really important predictor of lymph node metastasis. But I think the general trend has been to think that it is. So when this is present, it's generally an indication for surgical treatment as opposed to another form of treatment, endoscopic treatment. Okay, so we're going to focus initially on our first use of NBI, which is the identification of NIS3 features. So this is an obvious cancer, but I think it will show us what NIS3 looks like, and it's got a big area of ulceration. And one of the first things that you'll find to be almost universally true is that when NIS3 is present, there's some morphologic change on the surface of the lesion that goes along with it. The NIS3 features are present in some form of ulceration. So the right side of this lesion, what we're looking at right now, this is NIS2. The blood vessel pattern is regular. On the far left edge, you can see some of that. And then as we go down into this ulcer, we'll start to see that the blood vessel pattern is basically absent. Terms that are often used to describe it are amorphous, disrupted vascular pattern. It's very irregular. This is basically what NIS3 is. So when we see this kind of disrupted pattern in what is obviously a cancer, this indicates that we've got deep submucosal invasion. Again, traditionally, this would be a reason to send the patient to surgery. If we think about this in lesions that look to be more like potential candidates, the last lesion was just an obvious cancer. Of course, it's got to go to surgery. This is a granular lateral spreading tumor in the cecum that was sent to me for endoscopic resection. So the first use that we put to MBI is to try to check the entire surface of the lesion to see whether the vascular pattern is intact. So over here on the lower right, we have NIS2, which is an intact vascular pattern. And then down here, again, you can see this morphologic area where it looks superficially ulcerated. The surface features are sort of effaced. And as we get down into it, you'll see that the blood vessel pattern that we saw up on the other parts of the lesion is basically absent, disrupted. It's very amorphous. So this is NIS3. This is the presence of deep submucosal invasion. And when you biopsy this lesion, it's essential that you take the biopsy from the portion of the lesion that shows the abnormality. So what happens in, for example, in my referral practice, is commonly that the patient is sent with a biopsy that shows adenoma, tubular adenoma, tubulovilous adenoma, something. And then when we see the lesion and we inspect the entire surface and we find some part of it that has NIS3, and then we biopsy that specific part, we get cancer. And here you can see cancer. And you can actually measure this at more than two millimeters of invasion in the central part of this slide. So the key is to check the entire surface and then biopsy the affected area. So the biopsy from the referring physician, of course, came from some other part. So here is a lesion in the transverse colon. Now this lesion has a very smooth appearance. We would characterize this, I would characterize this as a non-granular lateral spreading tumor, has a smooth surface rather than a lumpy bumpy surface. Non-granular lesions tend to have more submucosal fibrosis and a greater risk of cancer. And if you look at the periphery of the lesion, you can see NIS2 features. The central part of it has a whiter color. That's one of the features. If you look back at the table of NIS3 features, the color of them often has sort of a white, whitish area, at least patchy white areas. And this typical type 2 pattern is completely absent from the central part of it, particularly from this little sort of dimple of ulceration in the middle of the lesion. And again, this lesion had been biopsied probably from the edge and it had come back as adenoma. But in the inspection, we see that the vascular pattern is disrupted, biopsy it from the appropriate part, and this again shows cancer. So NIS3, in my view, is an indication for cancer. I think most people would currently think that that's the case. Now, one question is how sensitive and specific are these findings? And the sensitivity for deep submucosal or for submucosal invasion is fairly low. It's only around 50%. So as we all know, we sometimes send a lesion to the pathologist that we didn't endoscopically think was cancer and we get a report back as cancer. And that happens because we don't have any endoscopic features that are highly sensitive. On the other hand, they're very specific. In some studies, 95% plus of lesions with NIS3 features have cancer. So not very sensitive, but extremely specific. So when you see it, then probably the patient should go directly to surgery. Let's look at one more of these lesions. This is another granular lesion. You can see this relatively whiter color or lighter color in this ulcerated area. This is again a granular lesion. Up on the edges, if we look to the right and left, we'll see the intact vascular pattern. That's NIS2. But down in the base, that's NIS3. This again, another lesion where some other part of the lesion has been biopsied, reported by the pathologist as adenoma. Then the lesion is sent. But when we inspect the entire thing, we see evidence of cancer. Now, sometimes it's very difficult to inspect the entire lesion. I have been more than once been caught having have removed half or three quarters or more of a lesion. One time I remember probably more than 90% of the lesion and got down into a little nook on the other side of a fold and found NIS3. That's, I feel really unfortunate because it exposes the patient to all of the risks of endoscopic resection without any benefits because the patient is going to go to surgery anyway. Now, this is a pedunculated lesion. Pedunculated lesions that show NIS3 features are still okay to resect endoscopically. There's some chance at least that you could resect such a lesion and it might not have unfavorable histologic features. The unfavorable features, unfavorable features means that when you see them in the pathology report, you should recommend surgery, all other factors being equal. If the patient is a really poor surgical candidate, you still might not choose to do it. But in this case, we've got intact vessels along the edge and then disrupted vasculature over the middle, this ulcerated area on the top of it. So it's okay to resect this. In my experience, most of these lesions will still have unfavorable criteria. Most of the pedunculated polyps in my experience that have favorable histologic criteria, the cancer comes as a surprise. All right, so that's one really important use and I brought that up first because our first job when we see any lesion is to try to inspect the entire surface. But let's talk about the clinical utility of type 1 versus type 2. So first of all, with regard to large lesions, we are in the midst of what is often called the cold revolution. I like that term. We're moving more and more toward cold resection. And this is, I think, clearly established for the SSLs, less so for the adenomas. But there's a group of adenomas, particularly flat Paris 2A lesions, where it's quite reasonable to perform cold resection. Bulky ones and ones that have pseudodepression adenomas really can't do. But if there's very low risk of cancer, cold resection may be appropriate, but it's appropriate for basically all the SSLs and we can use NBI to distinguish those. Within the diminutive lesions, leaving distal colon type 1 lesions in place, these are basically mostly hyperplastic polyps, and then resect and discard. Now, resect and discard, I won't say a huge amount about tonight because although it's been endorsed by the ASGE, the tech committee of the ASGE said it's basically ready for prime time, the ESGE has said the same thing. In the United States, we don't do it very much. It's not because the science is not there to do resect and discard. It's because the politics are not there. And the big problems, I think, are that a lot of people perceive it would have medical legal risk and there's virtually no financial incentive for it. In Japan now, there is a code, basically a payment for the use of enhanced imaging, including NBI, but we don't have anything like that and we're unlikely to get it. So let's talk a little bit about these uses and see if we can learn how to do this. Okay, so here we're focused on type 1 versus type 2. The type 1 features lighter color, either no or lacy vessels, and then usually dark spots or that are surrounded by white. You can have white spots. This may depend on the angle that the light enters the crypt. And then type 2 lesions tend to be browner. The vessels are thicker. And there's, in the highest confidence lesions, I think, structures that are oval, tubular, or just variable in shape. Ideally, the pits are much more variable than they are in type 1 lesions, and these are the adenomas. So as I mentioned, for narrowband imaging, as well as for some other optical imaging techniques, the evidence overall suggests that the data is quite good. We have in the ASG the PIVI criteria, which say that if the technology can be used and you can reach the correct surveillance interval, you know, more than 90 percent of the time, then it would be reasonable to do resect and discard. Again, we don't do that. Same thing for distal hyperplastic polyps. If the technology can be used and with negative predictive value for adenomas more than 90 percent, then you can do it. So the data is there that it works. So this is an adenoma. And just to review the main features, first of all, it has some brown color compared to the surrounding mucosa. And you want to try to feel confident that the brown color is the result of blood vessels, as opposed to just inflammation, because redness in a polyp, like an inflammatory polyp, will also appear brown when you look at it in NBI. The other thing is that the pits are somewhat variable in shape. You can see pits that are tubular, oval in shape. And so this is a typical adenoma. Here's another one. The vessels in an adenoma are also characteristically thicker. So the brown lines here tend to be thicker than the vessels that we'll see in hyperplastic polyps. You can see that the pits are variable in shape. That's a key component of high-confidence adenomas. Overall, I think that about 80 percent of colorectal polyps are high-confidence either adenomas or hyperplastic polyps. And this is another one, at least on the top part of it, quite brown vessels, variable-shaped white structures. Those are presumably the pits. Here's just an extremely high-confidence adenoma. Now we see pits that are quite long and tubular. Some of them are branching, sort of like CUDO4. I will say, in general, that NICE2 correlates well with CUDO3 and 4. NICE1 correlates with CUDO1 and 2. The CUDO classification was developed for dichromoendoscopy in combination with high-magnification scopes. And the NICE classification was developed for just high-definition scopes. So they don't necessarily correlate perfectly, but in clinical practice, a lot of people sort of use them interchangeably and apply the CUDO classification when you're just using high-definition scopes. So this would probably be CUDO3L and 4. It's NICE2, extremely high-confidence adenoma. And here is another just classic one, lots of brown. The brown is all from vessels, these thick vessels. The pits are quite variable in shape. This one is just overwhelmingly classic, CUDO3L and 4, NICE2, lots of brown. All the brown is from vessels. You look down in the lower right, you can see how thick the vessels can be in adenomas. Here's a very small lesion. So this lesion is also an adenoma. It's got brown. You can see that the pits are tubular on it, and the vessels are surrounding these pits. Now, this lesion, I would say, is probably about three millimeters in size, somewhere between two and three millimeters. If you can imagine the tip of the snare sheath approaching this lesion, the tip of the snare sheath is 2.4 millimeters in diameter. So it's a good way to judge if you push it right up against one of these polyps. Now, we find, and I'll come back to this, that about 15% of the adenomas that we see endoscopically and verify by the NICE classification are reported by our pathologists as normal. And that is primarily because we send the polyp as a much larger piece of tissue. When we cold snare, often we have a piece of tissue that is five to 10 times or more bigger than the actual size of the polyp. The pathologists don't see the polyp grossly when they cut it. And so we actually, in clinical practice, if we have good photography of polyps, we will still call them adenomas, treat them like adenomas, even if they're not verified as such. So one use that I put this to right now in practice is, like say I have six lesions that all look to be high confidence adenomas, and only four of them are verified by the pathologist. I would still treat the patient from a surveillance standpoint, like they had six adenomas. They would come back in three years as opposed to five years. We can now, according to our guidelines, bring people back with three or four adenomas in three to five years. And if they're all diminutive, I usually choose five years. But if I know there are six, I probably will still choose three years. So my point is that I don't necessarily rely on what the pathologists say when we're dealing with definite diminutive adenomas. I think another factor that contributes to this is fragmentation. The specimens can fragment on the way back into the trap. So that's another use. Here's another tiny adenoma. It has a little bit of a brown spot in it that I would call the valley sign. I'll come back to that in a minute. So again, we're looking primarily at type 2 lesions. These are adenomas. I usually refer to them as conventional adenomas. Here is another one, brown color, tubular white structures. These are both adenomas. And the one on the right I like because you can see how thick those brown lines are. And this is something that you want to get used to looking at is asking yourself how thick the vessels are. The same thing is true really over on the left. You've got some on the left. You've got some vessels there that are just really thick. You won't see vessels that thick in general on type 1 lesions. It's just something that you don't see. If you look at the left lesion, you can also see this brown furrow that runs across it. There's two of them, one sort of on the right side and one running across the left side. That is the valley sign. And the valley sign is very specific for adenomas. If you look at that close up, you'll see that there are a lot of very small vessels in there that give it its brown color. And that's not really a depression. I would not even call that a pseudo depression. And so it's been named the valley sign. And it's not that sensitive for adenomas. It's seen in about 40 to 50% of adenomas, but it's highly specific for adenomas. It's something that's kind of like NIS3, not much sensitivity, but very high specificity. Now, this is a lesion that I show you that if you initially, if you ask yourself, what is this? You probably would say this is an adenoma, but it's an example of the criteria don't work perfectly. And this is a hyperplastic polyp. And a couple of hints are the vessels are not very thick. Sometimes you see the vessels, there's actually a tiny vessel and a hyperplastic polyp basically surrounding every pit. The other thing I think is in the central portion of it, a lot of the pits are relatively uniform in size, but I just point this out to say that there isn't any system that's foolproof. And this is an example of where there might be a hint, but you could easily make a mistake. I will say that in general, the variability in pit size is something to pay attention to. If you have what you think are prominent vessels and brown color, but you don't see hardly any variation in pit size, I tend to make that a low confidence adenoma most of the time. I mean, yeah, low confidence lesion. Okay. So looking at some type one lesions then, what do they look like? So here's the typical hyperplastic polyp, and you'll notice the little black dots. Those are the pits. The pits of most hyperplastic polyps are dark as opposed to light, but they can be either. And nothing on the surface of this lesion is a blood vessel. There are no visible vessels, at least in this magnification. Here we can see a couple of blood vessels in the lower left, but for the most part, what we have here is a pattern of black dots. And very importantly, the black dots are nearly uniform in size. So it's the uniformity of the pits. And the fact that these blood vessels are very thin, you can sort of see them in the lower left part, just around a few pits, but this is a very high confidence type one lesion, probably hyperplastic. This is also hyperplastic. You see more dark spots, but those dark spots are actually pits. We'll come back to these large open pits here. This is a fairly small lesion to be seeing a lot of open pits, few lacy blood vessels in the lower left. This is also a serrated class lesion, type one. Now this is actually, I think, a confusing lesion. And one that could reasonably be called low confidence. But especially if you look in the lower right, you'll see that the pits are almost entirely uniform in size. Part of the trouble is that there are parts of this where some of the pits appear to be quite long. But over the most part, the pits are quite uniform in size. They also, in many cases, have a little central dot in the middle of them. And that little tiny central dot is typically not seen within the pits of adenomas. Now here is a hyperplastic lesion in which the pits actually, for the most part, look white. They have a little white centers in them. A little bit different, but still a high confidence type one lesion, hyperplastic. Here's another one. Some a little bit darker pits in the middle. Very typical hyperplastic. And here's a slightly bigger lesion. You can see off to the left, a few large open pits. These are the large open pits that we'll come back to that suggest sessile serrated lesion. And here's another such typical hyperplastic. Really no vessels. This one has lacy vessels on it, and also has this bumpy appearance that's often referred to as the cloud-like appearance, which is characteristic of sessile serrated lesions. But these are lacy vessels with a little bit more magnification. You can see that those are just coursing around pits. They actually are sort of an interconnected lace work, but in some areas they're slightly more prominent, and you can see them from a little distance. Here's a couple of hyperplastic polyps. Again, no vessels. And so we've talked about the main uses of this, potentially type two for resect and discard. Currently, you might be able to either ask for recuts if you're confident that a lesion was an adenoma and it comes back as normal. I don't typically do that anymore. When they're recut, they still may miss it. Sometimes they'll get it, sometimes they won't, but we just count them as adenomas. And then the hyperplastic lesions, which in the distal colon, the rectosigmoid, when I see hyperplastic lesions, we all see these and leave them in, I think. I think from a medical legal standpoint, it's better to not make any mention of them if you decide to leave them in place. If you see 10 of them and you decide you're gonna take maybe a couple of them out or biopsy a couple of them, just say that you saw a couple of them. From a medical legal standpoint, it's better to not suggest that you left anything in the colon because some people may not understand that they are perfectly benign lesions. Okay, let's go through a few little tests here. So ask yourself, what are the main features of this lesion and what is histology? First of all, is it brown? Yes, it's brown. What is the brown color from? Is it from vessels? You can see that it is. And then in terms of the pits, are these pits variable in shape, some of them tubular, oval? Yes, I think also there's a bit of a valley sign toward the top here. So this, you should be thinking very high confidence adenoma. So we saw this lesion before, almost no vessels, black dots of uniform size. So you should be thinking type one. Here is a lesion. First ask, what's the color? Is it brown? Yes, it is browner than the normal surrounding mucosa. And it is from vessels. There are some actually quite a few punctate vessels. And I think there's also a fair amount of variability. It's overall the variability of the surface that suggests that this is an adenoma. I would still call this a high confidence adenoma. Here we saw this one before, tubular structures, lots of vessels surrounding those, variable shape. This is an adenoma. We saw this lesion before, again, focus on the bottom right, uniform pits, hyperplastic. Actually, it could be an SSP. Here's a new one, probably three millimeters, overall brown color, lots of tubular white structures surrounded by vessels that are quite thick, extremely high confidence adenoma. Now this one has, again, the brown color, some tubular structures, and it has a valley sign that's occupying pretty much most of the surface. A lot of punctate vessels often constitute the brown color in the valley sign. Again, high confidence hyperplastic. And we saw this lesion before. Just to point out again, these are not vessels, except in the lower left, you can see a trace of a vessel. Those are the large open black pits of a serrated lesion. Here we've got something mostly brown. You can see lots of tubular structures, the brown thick vessels surrounding the tubular structures, probably a valley sign in the middle, extremely high confidence adenoma. Somebody tells you that's not an adenoma or the pathologist says so, they're wrong. Here is a larger serrated lesion. You can see lots of open pits on the surface of the lesion. This is a type one, probably an SSP. And this one should be easy now. Brown color, variability of the pit structure, brown vessels surrounding the pits, very high confidence adenoma. Again, this lesion, probably only three millimeters in size, has a central brown valley in it. So again, highly specific for adenoma. On the edges, long tubular white structure surrounded by brown vessels that are quite thick, very high confidence adenoma type two. So we were talking about clinical decision-making. So here is a lesion that is clearly an SSL. And so this is, I think, an excellent candidate for cold resection. Doesn't necessarily have to be done with EMR, but these can be resected cold with a very low recurrence rate. I prefer to remove them by EMR because you can see the edges so well. Some adenomas, again, candidates for cold EMR, cold resection, some larger ones. Obviously everything that's under 10 millimeters, we're gonna resect cold anymore anyway. I even resect pedunculated adenomas that are under 10 millimeters without electrocautery for the most part. So this is one use that we make is deciding on cold resection, certainly for the SSLs. I wanna briefly touch the WASP criteria. The WASP criteria, I don't think, have a lot of current use in clinical practice, but they're fun and interesting. So let's touch on them. So these are the four criteria that indicate SSL as opposed to hyperplastic. Irregular surface, large open pits, a cloud-like surface, and indiscreet or indistinct edges. So here you can see these lesions, both of these are quite irregular on the surface. That's one of the WASP criteria. And then the arrows on the left picture are on these large open pits, sometimes called type O pits. And that also is a feature of SSL as opposed to HP. This is a smaller lesion, but you can see the large open pits. Those are type O pits, again, suggests SSL. Recent studies suggest that overall, the WASP criteria don't work fantastically well for diminutive lesions. And I don't think we have a great way to distinguish SSLs from HPs in diminutive lesions, which is a lot of the reason why many people think that if we ever start doing resect and discard, it's going to apply predominantly to adenomas. Diminutive adenomas, we could call them, take a photograph of them, take them out and throw them away. But since we don't have great criteria to distinguish HP from SSL, we might take the type I lesions, especially those from the proximal colon and still submit them to pathology. Here is a large lesion. The rectangle is around some large open pits. The arrows are on the edges, it's supposed to be sort of just point out the indiscreet edges here. The indiscreet edges is part of the reason that it's nice to inject them before you remove them if they're large, also on a regular surface. Here you can see the arrows are on these large open pits, little bit of bumpiness to the surface and these kind of white lines, that's all part of the cloud-like appearance. Here you can see a nice example of a cloud-like appearance with the bumpy surface of the lesion. There's some lacy blood vessels on there, but I think that's a nice cloud-like lesion. So here, these are my favorite lesions in the colon. So I'm gonna show a few of them because this is the SSL that has endoscopically recognizable dysplasia in it. So the lower left part of this lesion is an SSL. You can see the large open pits, a virtual absence of blood vessels. And then in the upper right, we see the part that looks like a conventional adenoma, nice type two in the upper right, nice type one in the bottom two thirds. This is the endoscopic appearance of an SSL with cytological dysplasia. The upper right part, the adenomatous appearing part is the dysplastic part of this. Michael Burke often says that he worries that people see the adenomatous part and take it out and miss the rest of the lesion. Here is the lesion under the microscope. And histologically, the dark area here at the top here is the dysplastic portion. There are actually a couple of dysplastic parts of this lesion, the lighter parts of the SSL part. You can see sort of in the middle center, a couple of boot-shaped crypts of the SSL. And here's the dysplastic part better. You see on the end in the lighter crypts, the boot-shaped or lateral growth of crypts, that's the SSL. And then up at the top, the more hyperchromatic cells, the typical appearance of an adenoma. And then here's a focus, even higher power view of the adenomatous or dysplastic portion. So here's another just classic lesion. I just love to see these things. They're not extremely common to see endoscopically. I certainly don't see them every day, but the entire tip of this lesion has the NICE type II pattern. And the rest of the entire thing is NICE I and SSL with cytological dysplasia. Some of my slides say SSP, but the terminology now, the favored terminology is now SSL. For those of you who are fellows, it's very critical that you understand that an SSA and an SSP and an SSL are all exactly the same thing. So here is the dysplastic part, but you have to recognize that this is on top of a very broad serrated lesion. Here it is after injection. You can see nicely the dysplastic part is to the left and the serrated part is to the right. Down at the bottom, you can see where someone took a biopsy of the lesion. This is the SSL with cytological dysplasia. This is a more advanced lesion than an SSL. The dysplastic part tends to have microsatellite instability in it more often. It's believed that this lesion is closer to turning into cancer. And again, you can understand why it's important to recognize all this SSL part. You know, this particular lesion, 95% of it is SSL, and only a small part is that dysplastic part. But it also nicely demonstrates here in MBI, the nice two pattern versus nice one. Okay, follow up of resection. So another important use of MBI is to evaluate EMR scars at follow up. And right now, we are commonly clip closing lesions that are 20 millimeters or larger, proximal to the splenic flexure removed by electrocautery. And when I'm teaching fellows, one of the things that they almost all sort of have a tendency to jump to the first time they see it are to think that these bumps that are present on these scars are recurrent polyp. And in fact, those are clip artifact. They are artifacts of having the, of clips placed at the time that the EMR was done. Here we're just biopsying to verify that everything is normal, but there's no recurrent polyp here. So what happens is that the clip gets it, gets an inflammatory mound of tissue around it. And then later the clip falls off. And when that mound heals, it leaves a little bump of tissue. And in the actual scar, the crips, the pits are often just a little bit bigger than in the normal mucosa, but still they are very uniform in size. There's no variation in them. This is very typical clip artifact. And you have to recognize this if you're doing clipping because you don't want to apply more thermal injury to one of these scars taking off tissue that is basically normal. In my experience, if you do like to cold biopsy scars, and we still do that, and you biopsy this off, when you come back the next time, you'll see less of it or all of it will be gone. But it's very common to encounter this in clip scars. About 30 to 40% of them will have some. And so here is another scar. And I just want to point this out to you. There you can see the pit pattern of the normal mucosa, the little white dots. And then when we look at these little bumps, we will again see that they've got, but that right there is actual adenoma. So there's the difference between adenoma sort of in the upper center and these other little bumps, which are a clip artifact. And you can see how easily that's distinguished. Here's another scar at the very top. There's some clip artifact. And here there's actually a couple of areas of recurrence. And that's the ileocecal valve at the bottom, but you can see a couple of areas that need to be resected, easily distinguished by their adenomatous NIS2 pit pattern, as opposed to the clip artifact, which just has the pit pattern of normal mucosa. So I've tried to cover some of the ways that I use MBI in clinical practice. Resect and discard, as I mentioned, I hope we'll get to that someday. I think the evidence to support it is already there, but it's just hard. Leaving distal hyperplastic polyps in place, that's ready right now. You can do that. But just like I said, don't mention the ones that you don't remove. My own practice is I will actually remove everything that is NIS1 in the rectosigmoid that's over five millimeters in size. And I will remove anything that has large open pits in it that might have a greater chance of being an SSP. In our experience, if there's just typical NIS1, then only about 2% of rectosigmoid type I lesions are SSPs. So we don't have to remove them all. So we also want to be able to use the features of deep submucosal invasion to identify those patients that should be triaged to surgery. We can use MBI to identify lesions for cold EMR because we can recognize very easily that they are serrated class. We take photographs of most of the lesions that we remove, and you can use that as a record of adenomas. I actually do that in clinical practice. Sometimes it might change the surveillance intervals, other times not. But I think actually for things that are three millimeters and smaller, we're already more accurate than the pathologists are just because of the technical problems for them of finding the polyp. And then the final important use I've described is identifying clip artifact and recurrent polyp. And so now thank you for listening, and I hope you'll have some questions and comments. Thank you, Dr. Rex. Great presentation. We have a few questions here, and I will just remind everybody that I am monitoring the chat box and the Q&A box. So if you do have additional questions, please submit there. Our first question for you tonight, is there a difference in stage, depth, or prognosis between NICE-2 that come back as CRC as compared to obvious NICE-3 lesions? I would say in general. So the first question when that happens is always, did you really fully inspect the surface of the lesion? And so that you'd know that NICE-3 wasn't present. But in general, you would say if it's NICE-2, there should be more of a tendency if cancer is present for it to only be superficial. And that does not to say that you're not gonna sometimes get deep submucosal invasion, because I said NICE-3 is not perfectly sensitive. But in answer to the question, yes, on average more superficial cancers compared to NICE-3. Very good. Thank you. Our next question is, we know that tonight's talk was on one specific type of image enhancement MBI, but for those of us using scopes from other companies, do you believe that their image enhancement modalities are equally accurate and can be used in the same manner as MBI? I definitely would say for the Fuji equipment, yes. I think that there's a lot more data in general with MBI than with any other sort of imaging modality for predicting histology. But Fuji has a couple of systems that are non-white light. One of them is called blue light imaging. And it's very similar to MBI. And they have validated a classification system called the BASIC system, which actually has a lot of similarities to the NICE system. It has, and I think if you know the NICE system or you know the BASIC system, it's very intuitive to switch from one to the other. And I think that the imaging is very high quality. So I feel very confident that the Fuji scopes will provide the same clinical sort of performance. I'm not quite sure that we have as much data about the Pentax scopes and the electronic imaging there. You know, we've got some data for detection, but I don't think quite as much. So I can't say with certainty about the Pentax and I hate to be too specific about it. I think all three of the major manufacturers make very good definition scopes, but I feel a little bit more comfortable with the Fuji data. Very good. Thank you very much. Our next question is just overall in your own practice, how do you use MBI? Should we turn it on during the entire withdrawal phase of an exam or just in the right colon or only after you found a polyp with regular white light and want to investigate it more closely? Yeah, so we didn't talk about detection and I mostly talked about differentiation and that's not fair. Cause I was, you know, the title of the talk was MBI in colonoscopy. So I will say that one of the changes that happened when we went from the 180 series scopes to the 190 series scopes is that the illumination became brighter. The 180 series scopes, it was quite hard to do detection with MBI because the lumen was dark and there was not really any improvement that was substantial in detection with the 180 series. But if you look at the 190 series studies and there's a very nice individual patient meta analysis in gastro from James East group from I'd say a year or two ago, in which they showed that the 190 series scopes with MBI are associated with higher levels of detection than white light. And they are, and that's particularly true if the bowel prep is excellent. One of the problems with looking for polyps in MBI is that if the prep is not good, it's very unpleasant because the fecal material and the retained mucus look pink as opposed to the green color. And I think the pink seems more opaque than the green mucus does in white light. But if the prep is excellent, really high quality, then you can actually get an improvement in ADR from using MBI if you're using the 190 series. So I think you can do it. Having said that, I don't do it on a routine basis. I've done a couple of trials. I think I just am so used to using white light and there's just a tremendous comfort level about examining the colon in white light. That's not really a very good excuse. I do sometimes, if I'm examining the colon twice or if I'm looking for serrated lesions, I think there's actually some data that suggests that MBI improves detection of serrated lesions. I will go back over the colon. I routinely now in everybody, I examine the right colon twice. And sometimes I'll do the second examination in MBI. So that's not really a fair answer. If you want a data-driven answer, you could say, if the prep is excellent, it's perfectly reasonable with the 190 series to do the entire exam in MBI. I just think there are not many people that are doing that. Very good, Dr. Rex. Thank you. Our final kind of overarching question really is, is there any good references that you would recommend for further read or other resources that you would recommend for further information for the audience? Yeah, if you want to look at pictures, there's a nice atlas of MBI that John Cohen has put together that, you know, if you just want to get some familiarity looking at a lot of pictures, I think that's one of the best sources that I can come up with. And I think there may be some material on our website that will help with this. And I think this talk will be up there if you want some pictures to look at. I will say that we are going, in the past, we've had a STAR course for MBI, and that STAR course has been created. We sort of held up putting it up on the website until the FDA approved, you know, MBI for differentiation of, or predicting histology of diminutive polyps. We're probably going to come back to that at some point. So that, at some point in the future, that will be available on the ASU website. That'll have a lot of pictures and videos for getting experience in, you know, and using the differentiation schemes, the NICE scheme. Yeah, thank you for mentioning that last part. Dr. Rex, just for the audience's information, we are working with Dr. Rex on that MBI STAR and its updates, and our plan is to launch that NGI LEAP later in 2021. So keep an eye open for that information accordingly. Dr. Rex, thanks for an awesome presentation tonight. You're always a fantastic resource for the GI and endoscopy community worldwide. In closing, I want to thank everyone for their participation in tonight's presentation. But before you do log off, as I mentioned earlier tonight, would really appreciate your feedback on tonight's presentation by going to that network lounge in our platform and completing that evaluation. It'll only take you a couple of minutes, and your feedback really is looked at, and it helps us in our future webinars for the remainder of this year. So thank you for doing that. This does conclude our presentation. We hope the information that was provided is useful to you in your practice. And as a reminder, you will be able to access this recording of this webinar by logging on to GI LEAP next week and gaining, and you can gain access by learn.asge.org. That's where you will find it in the menu bar under webinars. You do not have to be an ASGE member to access this content, as our goal is to provide this information and education from our Thursday night webinars topics as an open source resource to assist gastroenterologists and endoscopists globally in improving their practices. Our next webinar will be next Thursday, March 18th at 7 p.m. Central, where Dr. John Martin from the Mayo Clinic will be presenting some ERCP cases. We thank you again, and we wish you a good night. And everybody.

narrowband imaging

colonoscopy

NBI

NICE classification

serrated lesions

adenomas

deep submucosal invasion

EMR scars

clip artifacts