The next presenter is Dr. Amir Rumin, who is a gifted interventional endoscopist who trained with us and is now an assistant professor of medicine in the section of digestive disease and nutrition at the University of Oklahoma, and we have the pleasure of him speaking to us on the best ASGE lower GI abstracts. Thank you, Amir. Thank you so much, Dr. Shami, and thank you to the ASGE for this wonderful opportunity to present today. So like my colleagues mentioned, it's kind of a tough task to pick a lot of good abstracts this year, but I did manage to pick three themes, and we have two abstracts per theme, so we're going to jump right in. I have no major or no financial disclosures relating to this talk. So the first theme relates to improving your adenoma detection rate, which is something that is important for all colonoscopists. The first abstract looks at artificial intelligence-aided detection in colonoscopy, and in this particular study they looked at non-expert endoscopists. So we have had a lot of abstracts this year looking at artificial intelligence, and we've already heard some of those applications in the foregut. The colonic applications are probably the most established, and in the U.S. we have FDA approval for a computer-aided detection system specifically designed for adenoma detection. So this particular group had already published last year about using artificial intelligence in colonoscopies amongst expert endoscopists, and so this year they come back and they report using this technology in non-expert endoscopists, which they defined as endoscopists that do less than two or have done less than 2,000 colonoscopies. So they included 10 endoscopists in a prospective RCT design that is powered for non-inferiority, and the colonoscopies were done for screening, surveillance, or diagnostic purposes in patients aged 80 years old as the maximum and 40 years old as the minimum. So essentially what they demonstrated is that the artificial intelligence system increased adenoma detection rates of the non-expert endoscopists by 22 percent. You can see the baseline adenoma detection rate in this cohort was 44.5 percent. They reported no significant increase in withdrawal time and no increase in the amount of polyps, non-neoplastic polyps resected. When they did their comparative analysis between the two cohorts, they found that the benefit of the artificial intelligence system was equivalent for both expert and non-expert endoscopists. So I think this study really shows that a computer-aided detection system improves ADR for both expert and non-expert endoscopists without increasing withdrawal times or resection of non-neoplastic lesions. I think AI is here to stay, and certainly a lot of the abstracts presented this year raise a lot of questions about cost analysis. And I think in this particular setting, it would be very interesting to see how this technology can help some of the low ADR colonoscopists and potentially some of the non-gastroenterologist colonoscopists that may have lower ADR rates. So moving on from the extreme of very high-tech to the extreme of very low-tech. So this is a very interesting study from Greece, looked at acetic acid chromoendoscopy to help the detection of right-sided CES-ulcerated lesions. So we all know that CES-ulcerated lesions can be very difficult to identify. So this is just a brief video clip that shows you a very subtle lesion here with some loss of vascular pattern. And essentially, when you apply acetic acid, that causes a conformational change in the tertiary structure of the surface proteins, and you get a whitish staining of the surface pattern, and that can help you identify the edges of these lesions. And so in this particular RCT, they looked at screening or surveillance colonoscopies with adequate bowel prep, and essentially their technique involved doing a double exam of the right colon. So during the first exam, they did a standard high-definition white light endoscopy examination in the left lateral decubitus position. They examined from the hepatic flexure down to the cecum, and they did polypectomy of all visible lesions. Then they performed a second exam, they turned the patient on their right side, they irrigated the colon with either normal saline or a solution of acetic acid, then they put them on their back for a minute to allow for the solution to cover all walls of the colon, and then they turned them again into the left lateral decubitus position and performed a high-definition white light examination. So they show a significant increase in the polyp detection rate in the acetic acid chromoendoscopy group compared to the control, and that difference was driven by increased detection of hyperplastic polyps, sessile serrated lesions, and traditional sessile adenomas. There was no statistically significant difference in the detection of adenomas in this study. As expected, they report a very favorable safety profile with no significant difference in adverse events between the two groups and no significant electrolyte abnormalities relating to acetic acid use. So in summary, irrigation of acetic acid solution improves the detection of right-sided SSLs, HPs, and TSAs, but not adenomas. Most of us probably do our colonoscopies under deep sedation, and probably that technique of application has to be adapted for that setting. It's very difficult to turn patients under deep sedation. I think we need to sort of think about the added withdrawal time and consider the role of other adjunctive technologies such as optical chromoendoscopy, CAP-assisted colonoscopy, and CUP-assisted colonoscopy, which was excluded in this particular study. So my second theme centers around colonic endoscopic resection, and the first study looks at the safety of cold snare polypectomy in patients that are on active anti-thrombotic therapy. So there's no doubt that there is a cold revolution when it comes to polypectomy, and there are actually multiple very nice abstracts in this year's DDW looking at this technique, and the rationale behind the safety of cold snare polypectomy is that you eliminate a lot of the cautery-related delayed effects. So when it comes to anticoagulation, our society guidelines suggest that we stop anticoagulation before polypectomy, but that is something that can be problematic or impossible in certain high-risk patients. And so this particular study was a single-center, single-endoscopist prospective observational study that looked at the safety of doing polypectomy on full anti-thrombotic therapy. So they looked at both cold snare polypectomy and cold snare EMR, which was decided at the discretion of the endoscopist, and they closed all their polypectomy defects with through the scope clips. They followed up their patients at seven days with a phone interview and saw them in clinic within 30 days. So they included 47 polyps in 16 patients. 12 of the patients were on some form of antiplatelet agent, and five of those were on dual antiplatelet therapy, and they had four patients on some form of DOAC. All the polyps were Paris 2A, and they were all relatively small polyps with a mean size of 6.3. Most of the resections were cold snare polypectomies, although they did perform five cold snare EMRs, which involved lifting with methylene blue-tinted saline and epinephrine solution. And clip closure was performed for all defects with a median number of clips deployed of one and a range of one to five. So this particular study showed no significant delayed bleeding events, and specifically no readmission, re-intervention, or transfusion, no thrombotic events, and no 30-day mortality. So I think, you know, the study shows us that it's potentially safe to remove these substance meter lesions using cold technique in patients on full anti-thrombotic agents. I think this is an area where shared decision-making between the endoscopist, the patient, and the other physicians involved is very, very important. And obviously, we need larger studies to really characterize the safety profile of this technique in these high-risk patients. Obviously, anytime we think about clips, there are cost considerations which must be factored into the analysis. The next study I thought was a very interesting study, and as clinicians, we don't necessarily think about these type of studies very often. This is a cost-consequence model where they compared piecemeal EMR to endoscopic submucosal dissection. And so the ESD technology is now increasingly adopted in the West, and there are, again, multiple studies this year that looked at outcomes. But one of the biggest barriers to wider adoption of ESD is the cost-effectiveness, and if you are trying to start an ESD program at your institution, you might be faced with a graph like this from the hospital administration that basically shows that the average cost of the ESD procedure is up to double what the cost is for piecemeal EMR. And so this group built a hypothetical cohort of patients with polyps measuring over 20 millimeters and randomized them to either receiving ESD or piecemeal EMR. And their methodology is very interesting. I encourage you to look at it in the full program. But essentially, they pooled a lot of probabilities of various outcomes, bleeding, perforation, recurrence, requiring surgery, so on, from previously published studies and plugged them into this model to sort of try to simulate what the final procedural cost would be for these two technologies. So not surprisingly, the initial cost of ESD, the upfront cost of ESD, is two times more expensive than EMR. But what this study showed is that the overall treatment cost can be lower in EMR, so you can realize the cost savings of up to 12% with ESD due to the higher R0 resection rate compared to EMR. And these cost savings are realized by eliminating surgery for SM1 cancers and requiring less colonoscopies for surveillance and retreatment of recurrence. So here, they show the number of laparoscopic colectomies and the number of additional follow-up colonoscopies that are generated by 100 ESDs as opposed to 100 piecemeal EMRs. And I think this really shows a nice bird's eye view of what this technology can offer. And sometimes in our healthcare system, things are very siloed. And so this data, I think, is very welcome, particularly when you're trying to convince your institution to adopt ESD technology. So obviously, we need real-world data to quantify cost savings, but I think this nice modeling study is a good start. So finally, I'm going to give some updates on some devices. So the first abstract looks at endosponge and managing anastomotic colorectal leaks. So endosponge has been used for a few years now in the foregut. And there's increasingly recognized indications in the colon. So basically, what the system involves is a suction tube with a black sponge that is cut to size and sutured and delivered endoscopically into the cavity. And once you have it secured in the cavity, the suction is applied and the sponge is generally exchanged every few days until the cavity granulates over and heals. And so this is a systematic review and meta-analysis that was done to evaluate the safety and efficacy of endosponge in managing colorectal leaks. The primary outcomes that they looked at were technical success and clinical success. So technical success, basically getting the sponge in the right place, and clinical success is the resolution of the collection. And in a secondary outcome analysis, they looked at adverse events. So they included 17 studies with just under 400 patients. And you can see that the mean number of procedures was eight, with a range of 12 to 16. So this is a pretty labor-intensive procedure. So as far as technical success, it's essentially 100%. Clinical success on pooled analysis was also very high, 85%, and the pooled rate of adverse events was 8%, mostly driven by bleeding complications. So endosponge therapy is definitely labor-intensive, but seems to be an effective and safe procedure for the management of colorectal leaks. And obviously, we need some prospective data with comparison to other treatment modalities. And the final abstract is the rigidizing overtube. So we already saw in the previous talk, using this rigidizing overtube in the foregut. And very briefly, because it's mentioned before, basically this tube is a tube that you apply suction to, and when you apply suction, it becomes 15 times more rigid. And when you use it in the colon, it can prevent loop formation or loop enlargement and stabilize the position of the colon. So in this study, which was a single-center prospective study, they looked at difficult colonoscopies, which they classified into two groups, either colonoscopies that were very difficult due to extreme looping, where they could not reach the cecum despite position changes and abdominal pressure, or there were colonoscopies with planned therapeutic polypectomies that were challenging due to poor position. They looked at success or failure of the procedure as their desired outcome, and they used this very interesting NASA task code index to assess the endoscopist's perspective of the demand of the procedure, and so that's mental, physical, effort, temporal performance, and frustration-related demands. So they were able to successfully complete all of their procedures, and their endoscopists reported significantly lower demands across all measured domains with using this device. So I think this is a very interesting technology that has the potential to help salvage difficult colonoscopies and endoscopy cases in general, as well as decrease the workload of the endoscopist during these difficult procedures. I think we already saw some potential uses for this in the foregut relating to deep endoscopy, and I think there has been a couple of other abstracts looking at its potential uses in advanced resection methods and even in ERCP and altered anatomy. So I think in future studies, it would be very interesting to look at implications of reducing mechanical strain and MSK injury with using this over tube, both for the endoscopist and for the staff that have to apply all this manual pressure, and I think it would be interesting to see how it affects the procedure and anesthesia time. So that's all I have, and I'll be happy to take questions at the end. Thank you again.

American Society for Gastrointestinal Endoscopy

adenoma detection rate

artificial intelligence

acetic acid chromoendoscopy

colonic endoscopic resection