Everybody in here believes we should be screening everybody with chronic reflux disease for Barrett's esophagus? Show your hands. Cynics. Everybody in here believe once we find Barrett's, we should survey everybody out there? Show of hands. So, the majority of us, it's what we've been doing for a long time, and what I'd like to do is maybe spend the next 15 minutes talking about a change in that paradigm, what we probably should be thinking about. So, I want us to question the practice of screening for Barrett's esophagus, not just dismiss it, but question the practice, and I want us to consider where it's appropriate to enroll all patients that we discover with Barrett's into a surveillance program, and then I want us to also think about if we are doing surveillance, how the heck should we do it, and how should we optimize the detection of the neoplasia, because that's actually what it's all about. We really don't care about Barrett's. We care about neoplastic Barrett's and eliminating that. That's really what the goal should be. So, the burning question, should we screen? Should we survey? And if we do screen and or survey, how the heck should we do it? Let's talk about the first one first. This is an interesting study that I think is really was one of the starts of us questioning screening practices. We were really under the, I think, the misunderstanding that 5 to 10 percent of people with Barrett's developed cancer during our observation period. This is a registry from Northern Ireland looking at 8,500 patients for almost nine years, and you can see that the actual incidence of neoplasia in Barrett's is much, much less than most of us were trained to think about. Along those lines, too, Denmark's an interesting country because everybody is under one healthcare system and all the records are centralized, so all 11,000-plus patients with Barrett's are followed in this Danish registry, and over a five-plus-year period, you can see that the incidence of neoplasia is extraordinarily low in a Barrett's population. So these two population-based studies, as well as many other population-based studies, are being thrown back in our face, questioning whether we could ever justify screening and surveillance of Barrett's esophagus from a cost-effective standpoint. You know, when we turn to the three professional societies, at least in North America and worldwide to some degree, and we look at the guidelines related to screening patients for Barrett's esophagus, they're relatively soft. And if you really look at the evidence, they are fair in treating the evidence, stating that can be considered, but you probably should be telling patients we really don't know if it is any good. The guidelines also suggest perhaps we should be pedigreeing our patients and applying screening and surveillance in higher-risk patients, but we really don't have firm evidence that we should be doing this, nor do we have guideline support that we should be doing this universally. So the answer to the first question is, in general, no, but it's an option, and it may be appropriate in some patients. And I want to go back and look at this, because I think it's time we actually really did look at reflux patients and decide who we should screen and who we should survey. Look at the impact sex has. Women have almost no risk of developing cancer in Barrett's esophagus. Have we ever seen cancer in a woman with Barrett's esophagus? Of course we have. But look at the difference of risk between men and women. It doesn't matter if you look at the Northern Ireland Registry or the Danish Registry. Women have a very, very low risk. Now look at that second red arrow as well, and I want you to concentrate on that. Many of the guidelines suggest that five to ten years of reflux, we should be screening patients for Barrett's esophagus, but in patients under age 50, which is the majority of patients I see getting endoscopy for chronic GERD symptoms, it is almost zero risk of having neoplasia. They may have Barrett's, but they have almost no risk of neoplasia. So what good can we possibly be doing that population? And as a matter of fact, if you look at this from a data perspective, young patients, and especially young women patients, their risk is almost that of the general population, which is pretty darn close to zero. So I would actually propose that in general, no, we shouldn't be screening everybody, but there are people at higher risk, and that includes white males over the age of 50, especially those that are obese. Those are the highest risk patients for neoplasia, and I think it's about time we probably concentrated our resources in that specific population. Now that I really annoyed you with that comment, let me talk about surveillance, because this gets even a little dicier. Should we be surveying? If you look at this kind of crazy diagram that I came up with, screening's tough to justify, but it makes more sense, because that initial screening exam of that 50-year-old fat white guy is going to pick up all the prevalent neoplasia cases up to that point in time. But look at your three-year exam. You're actually only moving into a very small incidence group, that 0.1 percent per year incidence group, to pick those up. So if I was going to argue screening or surveillance, I'd take screening any time over surveillance. Surveillance is much harder to justify. And if you go back to that Northern Ireland cohort, you can see that the subsequent surveillance exams are going to pick up hardly anybody with incident neoplasia that you didn't already discover in that screening exam. This is a beautiful study. Once again, Pratik has done so much work in this area, in his group, in his protégés, in his mentees, and everything else. But this looks at the risk of adenocarcinoma or high-grade dysplasia over time. What's fascinating about this is not only the low risk in the U.S. population, a group of five centers that looked at these patients prospectively, but in this population, if you had three negative surveillance exams, the risk of developing neoplasia was zero. And so even if you really do believe in surveillance, there is a time in the patient with continued no dysplasia that we probably ought to cut the cord, because their risk has become zero. And I don't think we've had a frank discussion about that as endoscopists and as professional society members for a long time, and I think it's time that we actually had that discussion. Going back now to the Danish registry, look at those red arrows. So green is screening. You pick up all the prevalent neoplasia at that point in time. The red arrows are your subsequent surveillance exams that you're doing over the lifetime of this patient, and you can see you're picking up no further neoplasia. You're not doing those patients any good. And if you look at the BEST study, you know, this large U.S. consortium, after those three exams, those patients in the United States had zero risk. It leads you to question what we should be doing with surveillance. How many neoplastic patients we need to survey, but is there any other high-risk populations in that group that we should be continuing to survey? If you go back and you look at the professional societies, you really don't get much help here. Again, they're very, very soft recommendations based on very poor evidence for surveillance. But if you also go back now to this consortium that has collected data from five U.S. sites and looked at where the risk factors were in those that we were surveying, probably the greatest risk factor was found in the length of the Barrett's esophagus. In those patients that had lengths of Barrett's under 3 centimeters, the risk was extraordinarily small. But in those who had lengths greater than 3 centimeters, you can see there was a linear increase in risk over time. Based on that, I would try to answer this question this way. Should we survey all patients continuously for the rest of their life with Barrett's esophagus? No, especially those with short segments and those without dysplasia. But indeed, when we find dysplasia, that's a no-brainer because those patients are not only surveyed, they need to be ablated. But also, I would argue still that surveillance probably should be continued in those with longer segments. I think that their risk of neoplasia is substantive enough to continue that practice. Now, to kind of tie this up with Betsy and Pratik and Luis asked me to talk about, we are still going to survey. I'm not naive to that. And there are patients that should be surveyed. But I'm not sure we've always been doing it right. So in the last few minutes, I'd like to talk about if you're going to do it, how do we do it? Whatever you do, do it right and do it well. So is there any data how we should do this? You know, it wasn't that long ago that we didn't even think about how long it took coming back from the cecum. But then Doug Rex and many others showed us that if you take your time coming back, you find more adenomas and you prevent more cancer in that patient population. So it's become a quality metric that we all recognize, that six minute or longer, which are all time from the cecum. Again, another Sharma study looked at a post hoc analysis of a study that was looking at optical illumination of Barrett's tissue. And what they found was somewhat surprising, but shouldn't have been surprising. If you spend a little more time looking at Barrett's, the yield of neoplasia goes up remarkably, even in this enriched population, that was true. As a matter of fact, if you took about a minute or longer for each centimeter of Barrett's, look at the difference in neoplasia detection. So this linear relationship between how long you look at the tissue and how much you find of important stuff is real in Barrett's as it is in the colon and everywhere else we look with an endoscope. And I would propose it's time that we start looking at this as potentially a quality measure. If you're going to do endoscopy, if you're going to survey Barrett's, you've got to do it the right way, and that means taking your time looking at it and probably documenting that in the near future. The other area that I found fascinating is how often I see patients refer back in for surveillance of their Barrett's that are well outside the guidelines. This is a beautiful study looking at this, again, in a consortium of U.S. patients treated at three centers. Sixty-five percent of the patients received endoscopy more often than the guidelines suggested, and that added up to 2.3 endoscopies per patient, more than they needed. Not only should we not be surveying most of these patients, we're using it as an annuity, and that practice has to discontinue. You know, we need to be intellectually honest with ourselves that we're not doing much good, but we're also scoping people that don't need to be scoped. And then, again, when we do scope them, we need to talk the same language, and I see an enormous number of Barrett's referrals from outside my institution, and sometimes I can't tell anything about their Barrett's from the endoscopy reports. The PROD classification has been recommended by all the professional societies, the experts in Barrett's. It's been validated. We need to start using at least the right tool when we communicate with each other about Barrett's patients. Well, what else is there out there? And I want to finish up, because I've been doing endoscopy and practicing medicine for almost three decades, and I learned on a fiber optic endoscope. We used to see Barrett's, as you're seeing on the left of the screen, but we needed to see what was on the right of the screen, and that is the histology. We approached that by doing endoscopic esophagectomies, you know, four-quarter biopsies every one to two centimeters until you couldn't see anything else because there's so much blood in the field. But what we really wanted to do was see the tissue ourselves at that point in time. And this brings up the issue that it's really time to abandon white light endoscopy as the only way we look at Barrett's and wait for the pathologists to tell us what we're seeing, because we have enhanced imaging techniques that actually change this detection of neoplasia that is our central interest point. Now, let me talk to you a little bit about that. Everybody in this audience uses one of these scopes, and I don't care if it's an MBI, BLI, intelligent chroma, or eye scan, all of our instruments have one of these buttons on them. And that changes what we're looking at when we turn that button on. As a matter of fact, a lot of the newer generation scopes have magnification with it as well, which also changes our ability to image Barrett's tissue. This is an example of MBI highlighting that area on the left that looks a little irregular, but once you turn the MBI on, it makes it very easy to detect. I want to show you these data, though. I think it is now time to make the statement that it's a negligent practice not to use advanced imaging. This is a meta-analysis that shows that if you use advanced imaging of any sort in a patient with Barrett's esophagus, you increase the yield of detection of neoplasia 34%. The goal of the endoscopy is finding neoplasia. This increases it 34%. How can you not use it? And how can it not be standard of care? And again, if you're going to do it, do it right. Well, the era's on us, though, where we may not even need to do that. We may never need a pathologist because we can also use optical biopsy techniques that are available today and being developed and will become available tomorrow to look at the tissue in real time. Spectroscopy, I think, is still the greatest promising area. I've been fascinated by this for over a decade. Light behaves funny when it hits tissue, and depending on that tissue, spectroscopy can give us a lot of information, be it autofluorescence in Barrett's or different light-scattering spectroscopic approaches. And then we have this whole era of confocal microscopy, and if you've never used it, it's an incredibly fascinating technique. I don't know which of these techniques is going to end up being the one that we employ that is the best technique, but it is time to start using at least enhanced imaging, if not one of these techniques, in the near future. So I'm going to try to answer the questions that I proposed to you at the beginning. Should we be screening everybody for Barrett's? And the answer is no, except those at higher risk. And for the most part, that means older patients, at least age 50, probably means males, and of course, the overweight male over age 50 is even at higher risk. Should we be surveying everybody that has Barrett's? And the answer is no. Clearly, those with existing neoplasia, longer lengths, or at least after a couple negative surveillance endoscopies, we should be cutting these patients loose. And if we are going to do it, do it right. Do it slowly. Take your time. Don't do it too often. Use a validated, descriptive methodology, and use enhanced imaging. It's 2014. We should practice 2014 medicine. Thank you very much.

screening

surveillance

Barrett's esophagus

neoplasia

advanced imaging techniques