SGE SuTab tip of the week. In asking can we extend cold EMR to large flat adenomas, the question would be why would we want to and the obvious advantage would be a lower complication rate. If we consider the factors associated with delayed hemorrhage after EMR, those are large diameter and proximal location. They apply to lesions removed with electrocautery and they appear not to apply at all to lesions removed without electrocautery where the size of the lesion and location have no effect on a basically negligible risk of delayed hemorrhage. On the other hand, there could be a disadvantage of a higher recurrence rate with cold EMR. And to answer that, we need randomized controlled trials. A couple are currently being conducted that will tell us if there's a higher recurrence rate and how much higher it is, but that's an actual real risk. A real potential complication is a higher recurrence rate and we need the answer to that in the long run. If we're gonna try cold EMR on large adenomas, what are the best candidates? Well, it's not these. These lesions which have a lot of bulk to them are bad candidates for a couple of reasons. First of all, it's hard to cut underneath them without electrocautery and secondly, bulky lesions, whether they're sessile lesions like the one on the left or a granular lesion with a nodule in it like the one on the right are much harder to predict the presence of cancer by their surface features. And in general, we probably don't want to be using cold techniques on lesions with any significant chance of cancer. These are examples of what the best candidates are likely to be for cold EMR of large adenomas. These are basically flat lesions that don't have nodules in them and don't have any evidence of pseudodepression. So on the left is the homogeneous granular LST and on the right, the non-granular flat elevated, no element of a pseudodepression in that non-granular lesion, no nodule in either lesions. These lesions are gonna be easier to cut through and we can pretty accurately in the great majority of cases predict whether or not cancer is present by careful examination of their surface features. In terms of the technique for cold EMR of adenomas, it's gonna be very reminiscent of the resection of SSLs. This is a non-granular lesion in the descending colon and it's got the flat elevated type. I would say that there's no pseudodepression here and certainly no nodule. It's a little bit difficult to actually see the margin of this lesion. The injection process itself is going to be identical to that that we would use if we were removing this with electrocautery. And then we're starting with a diminutive snare and this is a specialty snare. Again, the diameter of the wire is reduced by about a third from a standard 10 millimeter snare. So it's got a bit of a better cutting action. The technique for this and really the entire process of cold EMR has been championed around the world and in the US by Cyrus Paraka at Henry Ford in Detroit. And he has described in his experience that the recurrence rate with this is low and has some useful videos about the technique that I would encourage you to review. Basically, we're taking a slow and painstaking approach to making sure that we go across the lesion systematically. You can see that the border of this lesion is actually a little bit difficult to discern. It's so flat, but again, that helps with the cutting. We don't necessarily wanna grab such large pieces of tissue that either the snare gets stalled, gets stuck so that we have to cut or so that we have to mechanically tear through them. The overall impact of that is not fully known, but it's kind of discouraged and in general, try to be a little bit conservative with the size of the pieces and just be systematic going from one side of the lesion to the other. The actual snaring process, as I've mentioned in the SSL sections, now what we're doing is just going around the border and checking to make sure that there are no pieces of residual adenoma. Now, that's actually probably normal tissue, but the bottom line is if you have any uncertainty, there's really no downside to removing more, to extending the diameter of the margin, but we want the margin to be at least several millimeters. And then once we take off this little piece, we'll just continue going around the margin. But the time that you lose because you're snaring more slowly is made up for by the fact you don't have to do snare tip soft coag to the margin. The process is cold from beginning to end. And also you don't have to clip close it because the risk of bleeding is so low that clip closure and the cost associated with it would not be justified for cold EMR. This is another patient. And like the previous patient, this patient is participating in a clinical trial, a randomized controlled trial, comparing hot EMR to cold EMR. And this lesion is on the medial wall of the cecum. And one of the questions that comes up is, when should you be doing this in clinical practice? And I have to tell you, I was relieved when this patient was randomized to cold EMR because the medial wall of the cecum is a very difficult place to clip in many patients. The angle of approach is bad and you just can't fix it because the area around the valve is quite stiff and it's very hard to tip the defect toward the clips. And so you may leave the defect open or partly open, not really satisfactorily closed from a clip standpoint, which increases your risk of bleeding. So for this lesion on the medial wall underneath the ileocecal valve, between the appendix and the ileocecal valve, I was pretty happy. And I think that this is an area, especially for PARIS 2A adenomas, to consider the use of cold EMR in clinical practice just because clipping is so difficult, but the risk of delayed hemorrhage with hot EMR is real. And you can see this nice defect that we can leave open. This is a patient that came from a neighboring state for management of this lesion. This is a granular lateral spreading tumor that is in the proximal ascending colon. You can see the ileocecal valve on the 12 o'clock wall. It goes up over the ileocecal valve into the cecum. It occupies about 80% of the circumference. I've mentioned previously that granular LSTs are the largest lesions that we see in the colon. They'll grow laterally for a very long time before they develop cancer. And non-granular lesions, lesions with depression or nodules don't get this big. They turn into cancer before they get this big. Seriated lesions just basically never see seriated lesions that are this large. I thought this was a good candidate for cold EMR. We invited this patient to be in our randomized trial, but she declined. And so I decided to remove this by cold EMR for a couple of reasons. The first one, going over the entire surface of it carefully I could not see any evidence of advanced histology. There were no nice three changes. In fact, the pathology came back as tubular adenoma. Secondly, the shape of it over the entire thing is Paris IIa, it's basically a flat lesion. There's no nodules, no pseudo-depression anywhere. Right now we're injecting into the ileal mucosa, pushing the part that's on the ileocecal valve up away from the valve orifice. And so this is a slow meticulous process using a 10 millimeter snare. It took about an hour and 45 minutes to do this. But another reason to do it is the lower complication rate, which I would say is basically negligible with cold EMR, but with hot EMR, a lesion this size in this location would likely be above 20%. With a patient coming from another state, we don't admit these patients. We typically let them go after the procedure and sometimes they stay in town overnight, but pretty soon she's gonna be home. And if there's hemorrhage, of course, when the patient's in a distant location, you can't really participate in the management of it. So for all these reasons, I thought it was a good candidate for cold EMR. Now, certainly there could be an increased risk of recurrence. We're gonna talk in subsequent tips about managing recurrences on scars, but we can't control the risks associated with the removal of recurrences because they're typically much smaller than this. Whereas we can't control the risk of bleeding here with this lesion because it's absolutely too big to clip close. I did dilute the epinephrine here a bit to one in 400,000, just because I think the very large epi load associated with a lesion this size tends to cause abdominal pain after the procedure and you can limit it by some dilution. We're gonna finally go around the edge, but I think this is the kind of lesion in clinical practice can be considered for cold EMR. Thanks and see you next week on the ASGE SuTab tip of the week.

cold EMR

endoscopic mucosal resection

large flat adenomas

delayed hemorrhage

recurrence rate