In this lecture, Stavros Stavropoulos, MD, looks at the prominence of both Endoscopic submucosal dissection (ESD) and Endoscopic Mucosal Resection (EMR) procedures within the Western population.
Okay. Welcome back. 3. 20. Hopefully, we can do this until 4. 20 on. Then we can do the Colonies day from 4. 20 to maybe before 6. 30 if possible. Get out a little early. So, um, implementing years D in the West. So I'm just gonna speak about how you go about implementing is the in the West, particularly with some focus on barrettes and column, because, frankly, these are the legions that we have thio in abundance in the West compared to gas, turkey s DS, or even square, which is of a deal cancer, which are both easier than bars and colon. So implementing is in the West. So you know, the standard, um nostrum is, you know, you fulfill your preconditions. I mean, you have to be in the center. You can somehow get at least two years. There's a month, ideally, one per week on. You have to have some experience in advance endoscope, er, least Yanmar and an endoscopy unit that is supportive to advanced last extreme endoscopy, including the infrastructure, the stuff support from administration. I mean, I trained a lot of volunteers D, and this often becomes an issue. The support of VSD in that institution as somebody goes through their learning, or some institutions are not willing to invest the time or money on that. Now, Um, Phase one would be basic knowledge acquisition that can be found in tax reviews on courses or even is the videos on YouTube And what not legion assessment knives on Electra Surgical generator settings. I mean, we cover some of that in the course today. He is the techniques, indications, limitations, risks and outcomes. This is the basic fund of knowledge. And then you moved to, ah, preclinical training training in ex vivo animal models. You can often get support if you're promising from ER be Olympus and Boston to do some of that. But the course for ex vivo models is pretty modest. We can probably get a tremendous, um, a box or $2000 and then get the big stomach sent for $150 to hook in there on that one stomach could last you 56 hours at least as a beginner at least, um, then observe life Is this by experts minimum of 10 to 20 cases, what is recommended in general, a trip to Japan will probably give you 40 cases over four weeks, but you can easily see the same amount if you visit my center, maybe other centers. Sergei's 30 cats. And although he's very focused on Colin S. O, you know, the good Japan is not as essential anymore. I personally really developed it without going to Japan. But that was not popular back in 2000 in the early two thousands. So, um, some people like Mitt but reported doing a big ex vivo is D and sending his videos to experts friend of his in Japan to for them to comment on what he should improve. That may help training in live animals. I don't think it's essential, Um, but in live animals, where there is some bleeding control and he has done in the stomach off pigs, usually the SG recommends that, um, you got to read the speed of 30 minutes per five centimeter lesion as a sign of competence and to be able to successfully imagines any perforations or bleeding, and then you go to human experience. Phase three is training toe competence, which may be achieved possibly depending on what cases you have in maybe 40 50 cases. I think in the West it takes longer than that. 20 could be the the test cases where you're trying to reach the competence in the last 2030 you have to prove competence. Um, so basically, that would mean at least 20 consecutive V S DS within block receptions over 80% or block not a zero. An adverse event rates of less than 10% is the competence. Then you can move from companies to proficiency more than 80 cases for sure, where mainly you do self training now to achieve proficiency. And the idea is to reach those benchmarks on block. More than 90% are 0 80% that section speeds of at least nine square centimeters per hour. Some people use square millimeters per minute um, and significant adverse event rates of less than 5%. So here some master classes watching expert cases may help, but there's a lot of effort put by the operator themselves. On Finally, mastery is in the hundreds of cases. Curative resection rate more than 80%. He is the center of excellence involved in training others. So this is really the basic approach on then in terms of the clinical experience, there is too rough approaches. The standard one that is recommended eyes the step up approach. It is also proposed by European guidelines and most Asian experts, which is to start with these areas and then go to difficult. The idea. I start distal stomach rectum and then a sofa good before you do any Kalanick lesions. And this has been followed by many US and many U. S. And European adopters now maverick approach but potentially more applicable to the West and useful is the prevalence based approach. It refers to just including Colin lesions early in the experience as they come in, which obviously are much more prevalent than gastric lesions. And then you could be able to get your once per week. Is the volume a to a more appropriate rate to, ah, learn fast and avoid any injuries to patients. So the prevalence of based approach may get you to competence much faster, which is also good for patients now. This was proposed by Obama Obama, as far as I think initially on, and it was pursued by some early US operators, including myself and also some Northern European is D adopters and has gained traction more recently in the U. S. And has to get development of VSD devices focusing specifically in the Colorado in the West, despite the fact that is considered the most difficult organ to do it and our early phase adoption in the U. S. Still, companies are doing a lot of stuff for the colon. Arise DRS. For the Left column, mainly the dillaman for the right column mainly on then we have now the med robotics robotic platform, mainly for director sigmoid etcetera. So this is the step up approach. Distal stomach, rectum, a sofa, gush, approximate stomach and then the call on on that small bowel is a pretty esoteric domain that should be left toe people with ASD mastery, mainly due a dinner but also potentially the journalism's. So uh, yes, the guidance in the West make it difficult. Thio follow a prevalence based approach. These are the SG once, and then the recent The practice update, which follows them pretty closely. Eso that allow is did for basically any gastric new pleasure. They considered the standard of care master for the Japanese data and criteria. But then, for Barrett's esophagus The bottom line is that t s d is not shown. Sapir TMR Therefore, um are should be preferred. And I'm quoting the guidelines that some exceptions Bigger, Too much poorly lifting tumors. Um, uh, too much for it. With risk of some because of invasion etcetera um Colonna. Enormous. Same thing. Most can be effectively removed by standard political Miriam are. But then again, you have the high suspicion for some because of invasion A particularly if larger than 20 and then religions that cannot be optimally and radically removed by snare techniques, obviously trying to learning as being this kind of legions without having access to the easier legions. He's turning things into their head now. What's the problem with this step up approach? There's just no stomach lesions. So there goes that the CIA database. Out of 43,700 cases of gastric cancer, 1828 were early or 203 per year for this period. That's not enough gastric He has did to go around. Um, this was also identified by mere attendees off hands on courses on yesterday that were pulled in this study. That and this souvenirs, de course, uh, otherwise, about what they perceived as a barrier to ease the implementation in a Western practice. And the server 34 is the hands on course participants and really by far the number one reason lack of lesions. So what does the step up approach look like? This is from a expert in Europe. Might be one of the most prolific experts in Europe. These are study publications 2009, 12, 15 and 17. And if you put them in a grid, it shows you the slow progression of a step up approach until reasonable are zero re sections off. More than 80% are reached, uh, in various cases. And you know, these are spend different periods. This is a all the S. D. S that were done up to 2007. And you see, the R zero rate at the end of that period was still 72%. These are all rectally S D s from 14 2020 11. The R zero proficiency rate of over 80%. Even the easier rectal lesions was reached in the 2011 period. If you look at the eso Fadzir cases by Propst 2007 to 14. You can see again. Um, it took until 2010 to 14 to reach 86% are zero. And the esophagus is definitely easier than, um, some parts of the stomach on, then in the stomach. 100 on 91 legions over 11 years, by the way, Don't steal my slides, please. S 0 191. Oh, at least not without attribution. So this is the in this case, you cr zero section in the 1st 96 cases in the stomach, 59% on the last 100 or so cases, 93% in the stomach, which thought the easiest off places. So he started doing his this in 2003, but he didn't achieve proficiency by our zero until 2010 11. So it's really a 8 to 9 year period as also you on our prevalence based approach data. It took us 3 to 4 years, less than half that Teoh to reach and were training in the same period approximately, like I started in 2005, 67 and then really started doing it in bigger numbers in 2009. Now, this is a U. S. Um, multi center agglomeration of cases from the Double 2016 that shows mostly that these groups were dependent on a step up approach. Seven and a scope is the 304 years days mean, or 51%? Er there were six centers and you can see the 84% were stomach, esophagus and rectum, and only 8% were called on. And nevertheless, even after this 51 cases percenter, the en bloc resection was 78 far below 90. And there are zero rate was 56% far below 80% I with media and he is the procedure of two hours on DA At three months off, half of them had to follow up endoscopy, which showed 11.6% residual disease. That's I mean, that doesn't really happen in is the So this is this is residual disease. So the step up approach slow going. Um, So what other issues have been identified? Um, for the step of operas, this is from props from his latest big gastric cities in 2017 may be the biggest one in Europe, where he said that their data on this gastric Siri's we're still lower compared to large Asian studies, with our zero of 88 versus 90 for in Asia. And he attributed that attributes that to the very low number of suitable gastric lesions in Europe, you can extrapolate this to the U. S. And in their study, he says they needed 10 years to get 189 patients, or 18.9 per year. In a similar study, he says, by Pimentel 162 patients over 10 years at 16 per year. Repeat C 42 patients over four years, 10 per year. The very low numbers on that is the issue. There's another problem with the step up approach, the step of approaches that beginners should focus on the stomach and the esophagus. The problem with that compared to the colony, is over 50% off. Legions in the stomach and the esophagus end up being some form of carcinoma off to on a de one b nature, which means you have beginners now doing incomplete resection or imagine positive receptions on people that in a big number will have early cancers on a positive. Imagine may make the difference between a curative resection and referring them. Thio Salvage Gas Tracked on your ass off objective It's a big impact in a column in the column. The vast majority of the no matter advanced of the normal you get a positive margin usually can be. Usually it's a lateral. Imagine that can be dealt with with another SD or ablation. And in any case, nobody's gonna send for collecting me because a lateral margin of another normal was positive on like a lateral or deep margin of the Barretts carcinoma or a gastric carcinoma. So these are the problem with a step up approach. So what about the prevalence based approach? A zit was proposed by a gunman. He put it very nicely, so I'm quoting him verbatim. Early Gastric cancer is too rare in the West to achieve a useful caseload of at least two years this per month. This record I'll be 20 for a year and as I show you from the Prop state and none of the European centers got above, um, 18 per year, so none of them got to gas tickets this per month. Uh, this recommendation off the step up approach would impede establishment of VSD for decades to the disadvantage of G I cancer patients. Alternatively, a prevalence based approach allows for a reasonable caseload but requires learning is D mainly in the colon on early on in difficult locations. So Westerners might be, however, prepared for a longer learning curve than the one reported by Japanese status because of this need to include colonic lesions and particularly manipulated colonic lesions in the panic. Studies have reported low learning thresholds for gastric and colonies D off 30 to 40 and 80 to 100 cases respectively, for gastric and column. These are likely only achievable in a Japanese training model with continuous tutoring by on site experts and an abundance of gastric lesions, which allows allocation of easier gastric lesions to trainee cases. During the early learning phase, The highly manipulated Western colon polyps that represent a large proportion off the legions encountered in the prevalence based approach in the U. S. Would result in substantially longer learning curves, as is the case I show you with our data now there is data that support this, um, contentions, for example, this is a great piece of data because it involves Doctor Nishimura. Uh, very experienced. He is the practitioner from Japan who moved to the New York, um, two years ago or three years ago and then reported his data in the U. S on SD compared to his data in Japan. Um, and he showed that in Japan, for example, less than half of his years these were called erectile in the U. S. Two thirds of his years. This work, colorectal again, prevalence based approach, Um, in the first year in the US But however, what was really disturbing to him was that his on bloc resection rate for colorectal is this. In Japan, 288 was 96.5%. But in the US, on those 44 1st cases he did in the US, it dropped to 72%. This is from his slight. I did the w Did my ear still get skills get worse after moving to New York? He says fancifully. But the answer is obviously not, um, Western patients are bigger with more than than colons, difficult to move mostly under general anesthesia and, most importantly, highly manipulated. Um, you can see here in previously treated lesions, meaning attempted the Mars bad tattoos he had in a 46% on block right and without previous treatment, 83.8% still below the 96.5. But again, no previous treatment in the US me could mean a lot of biopsies when in Japan that don't even biopsy those politics. And obviously it means big patients with Big Collins that make scope operability difficult. So we're arguing that the case mix in the West is what makes this prevalence of broad, harder a way to prove that would be to do the following mental experiment and see what would be the the learning curve off the US advanced fellow versus the Japanese die fellow if both were tutored by an expert giant Japanese expert. But the U. S Fellow had to deal with the Western case mix and the Japanese fellow with Japanese case mix, what would the results do? And obviously we could predict that the U. S. Fellow would have a much harder time reaching competence on the Japanese fellow on. This is supported by this data. So let's look at the Japanese side first. That's easy. There are plenty of studies that published results on the planet's trainees. This is one from 2017 seven trainees training the SD in this period 45 to 50 Gastric. Yes, these cases portrayed me, obviously in big abundance in Japan. And not only that, they're probably the easier lesions that that institution was getting. They can see Phase one their 1st 15 cases face to the next 15 cases and phase three, the last 15 to 20 cases you can see on the resection speed. They met the proficiency benchmark even on the 1st 15 more than nine squares and emitters per hour. If you look at self completion 86% 96% 98% completed completely by the trainee with no assistance from the mentor. Zero assistance by the mentor on bloc resection. Ah, 100% across the board. Complete resection are 0 96 94 94%. So they actually met most of the proficiency benchmarks on the 1st 15 cases, which means expert mentor, along with appropriate easy cases in the, uh, distal Landrum and what not, Maybe not too big, etcetera. So very quick accumulation of knowledge that due to the appropriate cases. Now, this is an example of the US advance fellow Dr G. Uh, the advanced fellow for Dr Aihara on Chris Thompson. And bring them, uh, 23 years back. We're doing great things now, down in Texas, he, um did he is d training with Aihara. Nakahara is somebody training used in Japan, which considered a pretty much a SD master in that he did one year in one year, 72 years. This nine had no training involvement. 19 had training observation assistance. 18 partially performed by the training on 26. Mainly performed by the training but not defined in the Japanese way, meaning the trainee did almost entire case defined that the train, as the train need doing at least 50% of the case and look at the mix. Very unfriendly. Mix for a nears D trainee. 54%. Kahlan. Mostly right colon, 25% rectum. That's not bad. 14% stomach. That's meaning school. Then even three cases in the duodenum doing the sofa ghoulish on. In terms of previous manipulation, 33% were tattooed on. 43% had previously, um are terribly manipulated stuff in Japan. These numbers are close to zero. Um, so the if you look at what happened with no involvement, yes. This was no training involvement, partially performed, mainly performed by the training you can see here on bloc resection 85.7 70 78. So none off the groups reached 90%. And on bloc resection, if you look at complete are zero section 77.8 a 76 88. None of them reached 90% beds mark and recall the Japanese trainee data which were quite different. Um, and then conversion toe hybrid es de meaning. Here. Marius de big versions in this case is and in this case is none. But, you know, the main benchmarks would be, ah, speed on block and complete resection. So difficult long learning curve with this prevalence based approach because of the case mix, this is a prevalence based approach from Europe from 2014. In the 1st 50 cases of this operator, um, 40 out of the 50 were called a rectum. You see very few stomach and esophagus dinner mostly call a rectum and in fact, mostly rectum. Six out of the 40 call directors were in direct, Um, everything else was calling ascending sick. Um, sigmoid transfers. So prevalence based approach with a lot of difficult colors, and you can see here the progression. The case mix initially had some stomach and duodenum, but the later face it was essentially all colon, and you can see how difficult it is to reach down. Block rate, 75% 67 58 83 still hasn't reached 90% at the end of the 50 cases are 0. 75% hasn't reached 80% dissection speed plateau after 50 cases at 3.3 square centimeters, one third off the benchmark speed of nine. Now let me show you our prevalence based approach, which, after much effort Thio beat Down the Naysayers, was published in CD eight in March. Mind you, the naysayers were saying that we are threshold for proficiency was too high. I'll show you why that's not the case. So these are the volumes that we had 2009 to 2018, every increasing the greenest Kahlan. The dark green is rectum. You can see Initially we had a lot of some epithelial tumors. But then, as we started in 2012 to do stare And the FDR This dropped a little and right about 2013. The prevalence broad based approach blew up the colon. As you see here with now, you know, maybe two thirds of the case is being called on on. You know, this is forget up here. This is some epithelial tumors, and this is where how long it took. So I said it took probably about 10, 11 years with a step up approach. But here it's Ah, from 2009, you can see the very large zero on block on our first block off 50 cases, it went rapidly toe. 80% are zero and 90% of blogged about four years or so. 4, 4.5 years until the beds marks were met. Now, if you look at the blocks off 50 here, it's number is another block off 50 cases. You can see the average rate for that block of 50. It goes up. Um, you can see here the unblock resection rising, the R zero resection rising and again don't block was achieved faster. 90% of block was achieved at about 200 cases and then, ah, the the R zero is a little harder. It needed another 50 cases are 2 50 until it could reliably stay over. Now you can see what a very high level now and then. The adverse event is not an issue if you If you're careful, I didn't show you that on the bare study. But they're also the achieving reasonable adverse event rate of less than 5% or so can be achieved fairly fast for careful operators. What about the the speed of nine square centimeters per hour that also, this is moving average of 50 cases it took somewhere between 2. 50 and 300. Because again, this is a 50 case moving average, um, to achieve the eso every so all the signals point to 200 22 50 for for Siri's that are heavy on Colin. I didn't show. You hear that? That's the consistency of our Siri's 52% call on 88% in the right column off these rectum, 8%. All of the forget, including duodenum, eyes about less than a quarter, and about the fifties of epithelial tumors so very heavy and colon. Now Why does column make things difficulty in the learning phase? This is the median speed in the different organs for the Siri's off 540 s. D. So you can see the Denham very slow, three square centimeters per hour. But the next worst thing is the call on with six in the Sixes. Now the rectum is 10.4 and is the easiest. The next easiest is a suffocation stomach that are about 89 square centimeters per hour for the median for the whole. Siri's would look at this to accuse some graph asking How many is this? Does it take until in a particular program probe in a particular organ, you get to a knee S D that is faster than nine square centimeters per hour s so good it took 170 s D is to reach that speed in a sofa. Delusions. It took 160 years this to reach that speed in stomach lesions. Now this is on Leah's. These because obviously you learn from every single one. So it took 100 60 years these anywhere until you could do gastric lesions faster than nine square centimeters per hour because some people have trouble understanding what they think that we did. 160 stomach leaders in 170 years of idea. That's not what this shows. It shows that we had to read 160 years this before we could get that speed reliably in the stomach. Um, call on 280 cases and rectum is the easiest, with 70 cases to basically start having decrease increasing speeds beyond nine square centimeters per hour. Now we were accused of having to high thresholds rather reviewer that don't publish this because the lawyers would use it. And you know it's to higher number. Well, you know, the number is the number, and no lawyer cares about single, um, retrospective studies that care about guidelines, so that's irrelevant. But, you know, it's the it's the truth. It's the truth. And it is the truth. These are other thresholds from Western studies that also prevalence based already. So you bear who after 50 years D s where was 10% of points before below proficiency and at one third of the speed this sour study 182 colorectal es days that 15% technical failures and in the 85% where they succeeded in finishing they is. The zero rate was 63%. After 182. In this study, they had 301 colorectal. Leah's ds, mind you, 57% rectal. So a fairly easy case mix easier compared to in our cities. You know, you saw director ones were like 80% or something, So, um, 9%. So so you can see with all these rectal lesion. Still, there are zero of 80% was in the block between 2. 40 and three or one cases, and the speed greater than nine square centimeters per hour was in the block between 1 18 to 40 cases smack down in the same number that we figured out on in the marker study. They did 381 colorectal S. D. S that were prior to work concurrently with 39. Forget it. So, basically, these people started prevalence approach, getting all colorectal ones on. Once they became famous enough, that may have started getting some Asafa gastric is this are they even after all this training and colorectal and despite a total of 4 20 years days, there are zero rate in the easier. Forget religions was close to 80% however, but still after 381 colorectal. Yes, this there were able to get to 77% on our on the forget legion. So they experience is not completely interchangeable. But it helps. That's close enoughto proficiency, even though they only had 39 forget. But they were dependent on this vast experience in colorectal. So I mean, I think based on that it's important to avoid this. The world is falling and Westerners cannot learning as the mentality and inferiority complex. When studies published in the West, you know, the the operator really should try toe have done enough cases to reach proficiency before publishing data on E. S. D. So the proficiency birthmarks, I believe now well defined Andi using some arbitrary self serving benchmarks which are studies that look at the proficiency in that way, like looking at plateaus and learning thresholds based on on arbitrary plateaus off things do not represent true proficiency. Unlike poem was, we don't have such clear benchmarks. We have very clear benchmarks on E s D that indicate when somebody has reached not just mere carpenters but proficiency. And I think this would be used liberally on. It's a disservice, I think, to the field onto patients publishing studies without reaching proficiency as it propagates the myth that these benchmarks are not achievable in the West are too difficult to achieve or what not. This is, uh, one of this, you know, Western ASAC kind of meth analysis. The equitorial Madonna looked at Asian studies versus the Western world in Colorectal E S G and Fujio, Look at that correctly is the non Asian nation. You can see the problem on bloc resection Asia 87 Western 71 Asia above benchmark or close to benchmarks. It's a Western well below that benchmark, Um and I'm sorry, that's our zero. So they are above benchmark below benchmark and the scope become block 94% above 90%. Meet benchmark. Western world does not meet benchmark, but this is publicly publications that didn't meet the benchmark now fuck CEO. Same thing on Block 81 93 or 0 71 86 etcetera. So that's about the training. Now let's look at the other arguments other than the that the SD outcomes in the West versus East are in fear for some magical reason. Which really, if you put the same effort, the Asians put in learning SD except in colonic standoff stomach so it can take longer. You shouldn't have in figure outcomes. That's one. But then he has. He has more severe adverse events that lead the surgery. He has this not cost effective. I think these are. These are arguments that we should really look at, particularly as it relates to religions that Westerners would use on the prevalence approach, such as call on our barge. So let's look at the column because, as you saw any prevalence based, Siri's may have a majority of such lesions. So this is one of the often quoted meta analysis regarding complications. They looked at eight studies, six retrospective historical cohorts and two case controls on ER Maar versus CSD. And they sell, you know, um, the late bleeding was, in fact, more on the M R group, but risk of perforation 1.4% the March 5.7 in the SD. But this is all correctable endoscopic li See, there's a there is a need for surgery, um, difference of 5.8% to 9.9% on the assumption might be. And I've seen people make this assumption when the quote this meta analysis is that the perforation led to the surgery. No, This is based only on two historical cohort studies, which means that the easy legions who may have been getting here, Maher and the cancer religions have been getting S d. And you can see this year the mean size or near March 21 to 32 the midsize on New Year's Day 27 to 39. And so on. This historical this we're not case control. They were retrospective cohort studies, So probably SD included more difficult legions. If you look surgery due to perforation 0% 0.9%. Most of these difference was due to surgery for some because of invasion of histology 4.4 vs 7.8. So it's not. It's not the side effects. Now, if we look in more detail, this is the Michael Berk mega studies on Myanmar. There are probably among the best colony, um, are people in the world. So if you look at the study on 803 on Mars, 3% had deep muscle injury or perforation that was dealt with conservatively. 2.2% of legions, amazingly had muscles Arosa or fat in the path specimen. Okay, seven muscle to Saros. A 95 again, a lot of them were dealt with nonsurgical because there are only two people went to surgery. Same thing with the SD. So if you look at our Siri's 407 colonies DS um 63 rectal No surgery for perforation or any other type of adverse event. 4% inconsequential micro perforations type three or four with Michael Berk classifications. And one had the suspected delayed perforation. 23 months later, the patient had a micro APS is in the interesting Terry Christian that may have come from our rectally is did, um, two months before and we have We placed a small drain in this two centimeter collection, so you can count probably this one as a 0.2% If you look at the folks your meth analysis on surgery for any is the complication. Western Studies 3.1% Asian studies 0.8% but it's not even about west or east. When they looked at these differences, um, they found that there were due to learning curve effect because centers with more than two years this per month. And that's a very permissive definition for, ah, learn for high volume center 0.7% Surgery for complications If you look at centers with less than two years this per month defined as low volume about 2% surgery for complications. So that's what I have to say about complications. Now let's look at the cost effectiveness and, you know, to look at cost effectiveness. I think it's important to discuss, um, recurrence data, which is often under looked with a focus being another factors that may have to do with cost effective. So this is from Michael Berk again. All of the normal recurrence 28% of to my two years if you go through the enormous more than 35 millimeters 35% at 18 months, but you can see the graph is still you're still rising, and these were all data with only 88 4% haven't had the six months 4 to 6 months of violence and only 33% of the 1700 legions having reached the second surveillance. And it's still rising here at the end of the study. Um, their argument is that 90% of the recurrences can be dealt with in the skull. Piccoli Asper, their prior publications. Um, of course, that still lives. If we say 30% recurrence, um, 10% off, 30% is 3% over the over alcohol that we live 3%. Still, that may face the prospect of surgery after multiple colonoscopy attempts to EMR the legion. Now, if you look at the africanus after colorectal is actually looks a tacky toya meta analysis recants rates Western study 6.4% Asian studies point for far lower than, um are even for us poor Westerners. But again, it has nothing to do with Western east because if you get a non zero resection, whether you are in the east or the West does recurrences zero. I think it was actually 0.4%. Now, this is our data on the first half, the learning half of our cases in the second half of our cases, we've been able to get fall upon the 84% and 71% eligible patients at one year. And the requirements on our zero legions zero and the recurrence of are always 00.9%. But as you can see, our own block rate on the second half was 97% and are zero section was to 92%. So what does this recurrence happen? Well, because this piecemeal re sections, um, leave small residuals. And I know there's a whole debate about that. People buzzing Now the edge doesn't know that we can address this. Maybe in the Q and A so piecemeal lives this rough little something because a little things sticking out like this that could have micro don't know myself. You get this. But all of this can get any imagines on this And the comparison with the SD Stark, This is a Nate centimeter lesion. 70% of the circumference that we did on the turn off the hepatic fleck. Sh er ascending column. Very good markers for not having malignancy. So we did, uh, S d you can see here a reflection, outline the lesion and then complete resection with 90% of the circumference digs of the muscle here, but we're able to actually circumferential future it. Okay, multi r a procedure, but the patient got this. Hey, lifetime guarantee off non recurrence. This goes all of the normal nine by 11 centimeters. Specimen nine by several Asian lifetime guarantee of non recurrent. You don't even really need to follow this any sooner than three months. I'm sorry that three years, um, and with the structuring lifetime guarantee of no delayed bleeding at 0% so does this. Is this taken care of in the cost effectiveness analysis? Not really. I'll show you why. Uh, most e m r experts focus poor unclear reasons to me on the surgery sparing benefit of VSD in S and one cancers. The idea is you have a cancer with more than 1000 micro meter with less than 1000 micro meter invasion in the sub mucosa which the Japanese tell us has very low risk of live, not metastases. If you get in Myanmar, the specimen is all screwed up. You don't know the exact depth automatically. Sub mucosal invasion means surgery. If you get a knee S d. The pathologist measures less than 1000 micrometers curative resection, no surgery and in this meta analysis by foot to other found from Japanese studies that s M one. Cancers occur approximately 8% off polyps more than two centimeters in the column, and therefore you save 8% of patients, potentially from collecting me. But I'm perplexed. Is that the only way is D six people from surgery? No, it's not. What about aborted the Mars and technical failures due to not lifting difficult positioning of the snare, difficult location or suspected, but not known, some mucosal invasion? This is up to 10% in Syria's by experts like Michael Berk. If you look at the 1700 polyps in the police study, when you look at the flow diagram of the study, I don't remember if it's 8% or 9% were taken off the top because the EMR was intended but not even attempted because of reasons such as these. Now mostly SD experts. A tow level of hundreds of cases where I am and Sergei's can remove most of these lesions that are impossible to remove by Emma. That could be another 10% of a Siri's off legions more than two centimeters. Um, what about recurrence? I mean, I already showed you the issue with recurrence, and I told you that about 3% off the entire group may have requirements that cannot be eradicated and end up with surgery. It could be 2%. If it's 20% who can't? It would be 3% of its 30% who can't. But this man usual, let's say it's 2% 2% plus 8% 10% plus 10%. 20%. So, basically in 20% of patients that, um, have s and one cancer or impossible lesions such as this for EMR or have reconnaissance on earmark that would have led to surgery. Es de can largely saved most of these patients from surgery, looking at number needed to treat off 5 to 7. Now, if you think I would have to collect on it costs something like 50 to 100,000. If there's no complication in the U. S. And on the period study that collected me for benign at the Norma's in the colony had 0.7% 30 day mortality and 14% severe adverse events mailed the leaks. You see how the picture is rapidly changing on the cost effectiveness there. Moving on. Now there is discussed effective study from Michael Burke's group. Um, he has diversity of mark for Colin legions off at least two centimeters. And I'll show you what the issues is with this. That against major league bias. Things in favor of were my in my opinion. So the problem is so they looked at 1000 year Mars in the Australian registry and use their data on that. And then they took E S D data from the literature and created this model and look at cost effectiveness in Australia with Australian assumptions that are British, uh, non replicate a bill in the U. S. So? So. The same causal invasive cancer assumptions. They had a very lowest someone rate. I told you the meth analysis by future 78% sm one legions. They had 3.4 legions in their group, clearly favoring, um favoring, um e m r. Because then you cannot save people for resection due to sm one. Now. This may be because, as I said, 9% were removed off the top. Maybe because there was suspicion off some because of cancer. And they said they said that their specificity off detecting sub mucosal invasion from endoscopic prediction was 98.4%. Whenever they said, Look, this is invasive and in surgery there were 98.4% of the time correct, which is unusual now. Recurrence at 18 months. I don't understand at all. Most of the religions were calling 82% in the model and they put us down as a recurrence that 18 months, 12% a month. Their study from that group by mosque at 20% at 16 months on the study by police later with 1700 lesions, had 23.4% at 18 months. So where does the 12% come from? I'm not sure. Or even the 19% in direct. Um, now there is decide. I showed you a recurrence rate off 0.9%. They put 2 to 3% recurrence on E S D. Now, if you look at the King toy a meta analysis The reconnaissance was 1% at 19 months for every camera, all cameras and 0.4%. If you got the NAR zero section with the majority did. In the Fujiya meth analysis, the overall requires after colorectal es 0.9%. So these numbers are 10 times I'm sorry. 2 to 3 times higher than on this meth analysis. So on, obviously, the mark recognize data are extremely Rosie. Now, other assumptions are problematic. Like they assume that you're Marcus $2000 and years d $4000 mainly caused by the fact that they're doing a man under conscious sedation and not assume that they don't admit anyone And the do yes, deal with Mark anesthesia and automatically give it today. Admission. These are these are again bias towards EMR because, interestingly, the same authors from the same group the Australian Group, when they did a cost analysis comparing earmarked to surgery, their assumption for admission at the m R was that 20% of the, um our patients are admitted for up to three days and 1% more than three days. Why? They assume zero admissions when they were comparing you, Marty S. D. I'm not sure. Now if you look at surveillance, colonoscopy at 16 and 18 months at $861 eats All right, which I was in Australia, a za patient or as an ensure 861 for a colonoscopy. So they assumes a very low cost. Which cats done again? You need a lot more reconnaissance checking with Martin S D. But if the colonoscopy so cheap I mean, you're looking good. Even so, though, they assumed the same surveillance after Ma and Ori Esti, even you have a non $0. They wanted you to get a colonoscopy at six months and at 18 months, this doesn't make any sense. As I said, that reconnaissance after an R zero is D with 80 with 80 to 90% of receptions by SDR is 0.4%. Uh, these patients do not need a six month surveillance. It's debatable whether they even need a one year surveillance or whether to go directly to the three years surveillance. Um, so all these assumptions and still, interestingly, the most cost effective off the off these approaches. See, they had this imaginary 1000 patients on the four approaches is Everybody gets here, Mark. Very few selected ones get PTSD in this approach. Only people with Paris to see or cut off five pattern was basically guarantee cancer. So 43 out of the 1000 then using the app on this criteria for colony s d a little more permissive, you could get 40% of the patients to get the SD 400 out of the 1000 then universally, SD every everybody gets used. The obviously that's very expensive at 6.9 million. But look at the difference between these other three approaches. So the most cost effective was, in fact, selective. VSD at 4.2 versus 4.3 million, but very similar. Now look at this, using Japanese criteria with 400 out of 1000 years d s a great volume to learning as Dion, you get 5.2 million versus 4.3 in Myanmar less than 20% difference. Which means that for every surgery and here, the total surgeries for someone were 27 vs 11 for suspected this because all cancer invasion 12 vs zero so they don't even count the cost of the recurrences because they assume everybody gets to surveillance. Colonoscopy is in both groups number one and to treat a recurrence cost, um, $157. I don't understand that. But that's the probably because since both groups get the same surveillance, if you treat every canIs during a surveillance colonoscopy, all your accounting is you know, the snare, and I don't know the clip or whatever. So, um, so very small difference. Which means that for every of these surgeries that you for everything off this extra 20 patients that you saved surgery from it cost $34,000. That's a very nice number. Saving a surgery for 34,000 extra dollars. I think it's a very acceptable number, and this is all calculated with these assumptions that clearly favor in my mind the, um are rather than SD unfairly. Now. I don't know if I should beat this down any further. You know, I was, you know, a bit, you know, irritated by those assumptions. So I did my New York analysis on the back of a napkin, looking at Let's say you get the 100 patients with larger than know much. More than two and 100 are referred for your mom and I had to Michael Berk or somebody with exceptional quality and 100 or dog Rex, for that matter. And then 100% T o s t on gun. What happens? So as you saw, uh, first of all, you have failed the Mars deep, some because of invasion. 8% of someone 2% failure. Education. You get 18% collect. Amis, you get 28% post yamada normally recorded by two years again. Michael. Birth data. Um, Rick Announcements. At least three additional posture mark colonoscopies before going to routine surveillance. You do it. Colonoscopy one on the surveillance colonoscopy you treated, and then you get to more surveillance is to make sure it's gone for non recurrent. You still get to post EMR surveillance colonoscopies on then hospital days for observation. After treatment, colonoscopies are put 10% two days, which is optimistic compared to the data Michael Berk used on his e m R versus surgery. So the total tally 15 collect Amis 192 post earmark colonoscopies before going to routine surveillance on 30 days off Hospital state. Now you look at the 100 patients sent for U. S. D. 5% will get collect on me 3%. You too deep sub because salvation 1% You do perforation 1% due to failure of the SD and then 90% will have our zero section and then our rate is 95 97%. I'm being very charitable here, which means negligible recurrence that can go to routine surveillance. 10% who are known are zero. I will assume similar to EMR 12 and usually this non zero means you just went close. You didn't leave a piece back. So I'm going to assume that this is similar to EMR with 20% recurrence and four postures, The surveillance is 80% no recurrence and get to post es de surveillance colonoscopies and then hospital day. Let's assume which is not true in our currencies and also you I 80% of treatment colonoscopies admitted for two days. Net result is 10 less collect Amis in the ASD group 159 less surveillance repeat colonoscopies to detect recurrence and treat recurrence and but maybe 128 more hospital days for observation compared to your mark. So at $2000 a day, $5000 per colonoscopy and $50,000 per collector me and mind you in New York, the colonoscopies can often be closer to 10,000 and 5000. You can see basically savings of a million dollars or 10,000 per patient, and these are realistic assumptions based on the literature. So that's my back of the napkin analysis. So now finally finished with few slides on barrettes because that's another controversy. Again, you know the guidelines saying you should do you, um are yes, this not indicated, except in exceptional cases. But why? So there's a lot of us off. As I said, more than half the nodules in Barrett's will have at least t I s t 1 81 Big cancer. If you look at these lesions in the CIA database, half of them are bigger than 15 centimeters. So yeah, I was gonna be piecemeal medium size of Rogelio March specimen 13 by 18 in this study, and then in terms of determining the depth of invasion on EMR, 26% of your march specimens have minimal or no sub mucosal present based on this data, so you get very poor specimens. So why the support of the, um are as the standard of care? Well, there is the publications from the West are horrendous. Okay, so we start here back in 2012 with our zero rate of 39%. All right. Wow. 74 60 ft. All these are below proficiency. Hubble reach proficiency on Coleman. The rest is below proficiency. Um, now? No, he has this group did in 2017 this random. My study with 20 Mars versus 20 years days perspective ran on my style. I mean, the S D R zero section was 59%. Uh, that is not a rate that is ready for a random mice trial in this digestive endoscopy. Start the cohort EMR compared Toa cohort ASD our theory section 20% versus 81%. Now we're talking mhm. And then in this recently, I started from Mayo Clinic case control of the Ahmar versus CSD. 72 patients in this group matched you get I don't know what. Well, there's zero rate on near Mars. Irrelevant. It's never gonna be more than 12 to 20% as we know. But the R zero rate on the S. D was 42%. I'm 42% R zero rate for parrots early cancer. This study used a seizure knife to do the SD of the Barretts. As I said, this is a knife that sends electricity everywhere and cooks everything around it. And it would be easy to use, but nothing that will give you Chris margins. So this is these are these are really this is this are these studies? As I said, public six studies like that gives you a bad name. Really? Aan den. Create the wrong guidelines. Which means you can remove cancers by topping up in pieces something that is antithetical toxicological surgical principles. Um, this is our study that was submitted to do the double 2017. Not surprisingly, a poster. 33 cancers. Hi. Great displeasure. Adenocarcinoma, squamous carcinoma. That's into that cases we have completed until 2016. Very early days. So most of it cancer adenocarcinoma zone here. But you CR zero on the 1st 11 cases. 54%. But then we got better the next 11 cases. 91% on the next 11 cases, 91%. Mind you, we didn't get better from here to here because we did 11 years days. Between this and this along with the 11 barracks these days, there were many others this in other organs. Hence the prevalence based approach. And then you can see procedure time dropping rapidly within a year or two. But again, a year or two of doing other is this as well. So it's very doable if you follow the prevalence based approach to get to Asian levels off our zero section, which would clearly so. The superiority of VSD and some studies are beginning to show this, even in the West. So this is a very interesting study from Stanford that retrospectively looked at their Barrett legions for five years that were expected. Endoscopic Lee, initially with the, um, are maybe the 1st 30 off the 50 legions. And then he has did the next 20 and they look at the last cohort from the previous government. They show what s D? All nice, deep margin. Latter, Imagine a good deal, Maher. It looks nice and not cooked, but this transition here could be right in the middle of the deepest part of the cancer. So the deep margin is affected. The latter imagine, is undetermined herbal indeterminate, and then you can have a bad. Dan Meyer. This is Ah, cooked completely through and through cooked specimen where you know, any depth assessment becomes impossible. So does this have an impact on patient care? Yes, it does. And that's the genius of the study. I think so. They looked What happened on this 9. 31 year Mars in the 20 s. D s r. Zero section, 20% in the EMR, 81% in the S D. So these people, you know, got goto proficiency now, um, legions containing carcinoma 17 in this group. 10, about 54% here, 50% as I said. And then they were removing all nodule. Some of them had low grade. Some of them had high grade. Some of them had cancer. So but about half of the no Jews and barrettes contained cancer. Um, t one or T one B now equivocal path, meaning this notorious assessment by the pathologist. But there is at least in Trump because of cancer. Clearly because they cannot assess. The deep margin was evident in 65% of the 17 cancers in day in my group and 0% of the 10 cancers and the S. D group, which, of course, you can expect from this specimen versus this or this specimen. Now, if you look at the and that obviously has consequences, see equivocal vertical dip margin in 77% of cancers here in 0% of cancers that there should be 10 here instead of 20. So big significance. And that results in this problem totally self object. Tommy was done in five people in this group and three people in this group. These were all appropriate, all three, because 100% were t one B or worse. And then on this group, three out of the 5 60% were inappropriate because they were done. Unfortunately, 41 a or less, including some for high grade dysplasia because of the confusion and only two out of the five g two t one b cancer or worse. So in conclusion, the is The learning curve is longer in the West, where non tutored, prevalence based approaches often inevitable and requires tackling challenging regions such as Colin legions and previously manipulated legions. Once proficiency, however, is achieved, the outcomes are identical, with the excellent outcomes achievable in Japan, including in the column and the surgery sparing benefits of colonies Delivery um are include in addition to the maxed out of this and one cancers, um, our failures in difficult locations here Mari currencies that fail in the Skopje treatment. Additional benefits includes paying patients the anxiety a cost off, multiple surveillance and treatment colonoscopies, foreign currencies. Finally, approximately 50% of buying legions may contain cash. NAMA is the offers definitive resection and diagnostic centrality with negative and accessible deepened lateral margins preventing unnecessary. Asaf objective is now. I had some interesting cases since this came up. This is the final resection with this was this was shown at the video forum, which was virtually a month ago and got the award on the ASD section Um, for best video. So this was that case I was talking to you about 74 year old woman, not this erotic from San Diego. The 23 hours. But this 74 year old woman that took about, I think 67 or eight or 70 74 year old woman with alopecia corner These on Plavix not honking since lymphoma status, post chemo and radiation in that her abdomen, who underwent endoscopy to evaluate this failure. So that's the Nautica sofa goose. It had been died of multiple times with bushy before and then a large to a circumferential lesion from one centimeter proximal to the Z line to 45 after the line and two more cancers. So these are probably, um, legions created by the chemo radiation in the upper abdomen. See, now on surveillance has developed yet another carcinoma in the ant room. So these were three. Is this at the same time, including a circumferential e one and then this one in the greater curvature and this one in the cardia on the same session. So this is the This is the circumferential involvement right at the Z line off this gastric carcinoma. Then this is another legion down there that you barely saw him, that this is so on. Retro flexion. You can see a circumferential region, and then the second legion here can be seen and another legion here. So basically what we're doing is we are we started the sofa gush with a bit of a tunnel technique that's the muscular is because it's very thick again. She had a lot of fibrosis consistent with the previous radiation to the to the G Jackson and upper abdomen from the for the lymphoma. So you can see that because I was not ideal. That made you can see. Now we have done the anterior and posterior tunnels on. Then we're continuing the dissection. Ah, now we're trying to get between the two flaps off the anterior and posterior tunnels. You can see that stuck together. We're using water to separate them on. Try to get through there. Thio do the distal dissection. These are dissection is very difficult to discard. Issue as you see there. And obviously the fund, this is the most difficult area. As you can see, you have a perforation that happens in the funders at some point. So that now the dissection of the Suffolk, which is complete. And now we're gonna work, um, on this difficult funding area, which funds has extremely thin muscle. And you have to dissect for particular to that muscle. As you see there. Now we're introducing our again reported and better than 360 loop of the endoscope to facilitate dissection of the fund is I mean, that's what you do when you don't have enough right reflection. Now you can see we're dissecting here again. We're trying to be careful using a protected knife 92. But nevertheless, there's a little perf there, I guess with his knife. Andi, what we do is after finishing the dissection and getting further from the perforation, we closed a perforation. We don't close it yet. We live the factor until we're away from it. Because if we put an eclipse there now, that will interfere with the resection. So after we complete a little more resection, we are used the micro mental paths technique Thio close the micro perforations. So basically, what we do is we grab the fat and fix it there on that closes the perforation securely because otherwise, very difficult to close. You have to put a clip on muscle that doesn't hold clips. Well, so this is the completed final resection. Uh, we bring out and you see, it was a bit of it took us 30 to 40 minutes to finish the painting of the specimen, Actually. Sorry. You know what I'm doing here is I'm using a grasp er external to the scope to create traction when you do this tunnel technique for the interior wall. You always You always need attraction on the interior wall, Donald Link. So anyway, this is it. And I took a while to pin with the help of the pathologies because you use the same because we made it. Hang through the cork on. We paint the disk and that's I'm scoping now. The specimen there, that's the gastric side with paint. And then that's the closure. You can see the big resection on. Then there were two smaller re sections of smaller carcinomas like that one. I'm not. I'm not gonna bore you with that, because that's far easier. So that's the other one. And then there's the third one. So these are the three re sections of the three cancers, probably radiation induced, because there's a field effect clearly on, as he has a natural one. So that would be her 40 s. Dina is division one doing three there, So that took a while, but definitely not 23 hours. So hold on. So these are the legion one took 424 minutes are zero section listen to 5.5 or 3.5. 55 minutes are zero, and the last leader in R zero Andi. Oh, I think I had the fall of video, Tia. And sorry that everything looks good. So that's the one year fall up along. I saw you. Okay, that's the closure there. So that's a year later, um, on you can see a pretty good result with complete resection. Nice. Jackson, The clip The clip with our mental thought is still there. These are the results here. Now I have ah circumferential donut resection for legions that are around the musical valve. We published seven of those, all with our zero section in surgical endoscopy. So I think you can go. The video is there. You can look at it there because I think we better get going with a colony is D if we're gonna finish at 6. 30 Basically the donut is deep. Produces a specimen like that. That's ilium margin. That outer margin is secret margin and it's removed on block and one donut with your cycle valve fat in between. Um, like that. So that that is a Nazi resection saves the patient from a chemical ectomy. That's the sickle side with the whole of the musical valve. This is a case that I did in the Athens Life Symposium, which is a circumferential e FDR. In the end, he s d turning into the FDR for a circumferential recurrent at that. No mind. A staple line after law, interior section of a cancer and bouts of radiation, chemo and multiple Liam ours. So that needed start as a knee s d and then FDR at the staple line to get the staples with the embedded at the no mind to the scar and then back to USD to finish the resection to get a complete specimen. So these are marking one side of the legion, the in outside with the eternal posterior early using traction. Then that's the posterior tunnel completed. This is a full thickness resection off muscle there, that's what the Tisha. So this is the part of the posterior tunnel that had the FDR. You can see the staples there. And then we went, um, to do that theater and you see the attraction doesn't help. So what you can do is modified with a poorly. That's the posterior tunnels during the complete resection. So you put an extract lip change the victory little. So we can work on the anterior tunnel. A zone here, Andi, then, uh, see, we hold the term while pulling, because otherwise you're gonna tear. You can deploy the clip and then form the public. You use the pulley and then deploy the clip. Important tricks. Anyway, so you deploy this on, then we move on. You can see now are very good access to the interior tunnel on. We complete that. Uh, do you have time to make a complete video here? But this is the start of the anterior tunnel. The entire time, I had all kinds of scar and shooters and clips that we had to dissect with the hook knife off the off the erectile fashion and fat until it was complete. And then there's the end outside all marked, there's the or outside, all removed on block with a staple line muscle. This is the, you know, the results. Here you can see a nice circumferential excision, and then that comes also with a lifetime guarantee of non recurrence. But this was full thickness, but we didn't penetrate the Saros. And so this guy didn't need to stand or anything. He didn't need a bunch of dilation. From what they told me, the Greek physicians that we're following him. This is our data on donut resection, which were seven cases all our zero section. Again. It's in surgical endoscopic anyway, for more videos. You have seven years of Long Island lives with people like hockey. He knew, he repeated. Came one year joke at Savoy's every year, myself doing really challenging cases. We don't hold back in stressing out the master operators. You have seven years to watch on. Let's Go to the Colonies Day now with a dilemma. So that's it for the lecture on. I'm gonna answer questions again in the room.
Related Presenters
