Beth Lee, it's my honor to introduce him. He runs endoscopy at MD Anderson Cancer Center. As you heard, he used to be at the Brigham many moons ago. And I consider him to be my brother, even though we're not related. And I'm just truly thrilled and honored to have him here. And he's gonna be talking about, as he mentioned, introduction to ERCP indications and radiation safety. So Jeff, thank you so much. So it's great to be here with the friends. Many thanks for inviting me to this meeting, ASG staff and course directors, Uzma and Linda. So my task today is introduction to ERCP, indications and radiation safety. So part one, indications for ERCP. ERCP began as a diagnostic tool in 1968, and it was for imaging of the biliary ducts and pancreatic ducts. Then it moved on to therapeutic field. In 1974, sphincterotomy was introduced and my mentor and Linda's mentor, Dr. David Carlock, used to say, no, it's actually 1972. That's when I began to do a sphincterotomy. Pre-cut sphincterotomy followed brushing and biopsy, dilation, stent placement, cholangioscopy, intraductal ultrasound, ampulectomy, radiofrequency ablation. So ERCP is no longer a diagnostic tool, but a therapeutic tool. I was having breakfast with Sandra. There's an ERCP weather and she concurred. It starts on a sunny, nice weather. Then at times it gets dark, and Sandy called it stormy weather, and the boat capsized, and you gotta get out of there. So indications for ERCP, benign and malignant. Benign includes choledocholothiasis, primary sclerosing cholangitis, chronic pancreatitis, biliary strictures from surgery, post cholecystectomy, post liver transplantation, post pancreaticotonectomy, aka WIPO. Malignant, cholangiocarcinoma, pancreatic cancer, ampullary neoplasm, metastatic cancers. And let's focus on the benign etiologies. Choledocholothiasis, most common etiology for biliary obstruction. Patients can present with a biliary colic, obstructive jaundice, cholangitis, pancreatitis. So ERCP is an excellent tool detecting stones, sensitivity and specificity are greater than 95%. And then you can perform a porooral cholangioscopy with an intraductal lithotripsy that are effective in 83 to 100%. Endoscopic sphincterotomy and stone extraction are successful in greater than 90% of the time with adverse event around 5% and mortality less than 1%. If ERCP failed, you can try pre-cut or EUS guided biliary access. Biliary strictures due to primary sclerosing cholangitis, MRCP is recommended rather than ERCP for diagnostic workup. To prevent bacterial cholangitis, pancreatitis, bile duct perforation, bleeding and post ERCP hospitalization in this population is as high as 10%. But if you see a dominant stricture in PSC population, brushing and biopsy are necessary to rule out malignancy and antibiotics are must. And if you see a stricture in PSC, you do balloon dilation plus minus stenting if you stent only for a short period of time. Biliary stricture due to chronic pancreatitis, long term success rate is 10 to 30%. Complications include a stent occlusion and stent migration and usually requires a multiple plastic stents and the success rate is as high as a 65.2%. But restenosis rate is pretty high once you pull the stents out. So if ERCP fails in chronic pancreatitis, you can actually do EUS guided pancreatic duct drainage with a technical success rate of 92.5% and clinical success rate of 87.5%. Post-operative eye leak, usually from direct trauma and you do sphincterotomy plus a stent placement. Some people just do sphincterotomy but I prefer sphincterotomy plus a stent placement. Post-operative biliary strictures in case of a cholecystectomy induced stricture, multiple plastic stents are required or a covered metal stent would be helpful. Success rate is a 74 to 90% but recurrence rate is as high as a 30% within two years after removing the stents. Post-liver transplantation strictures, there are two types. One is anastomotic stricture which is dominant, about 80% and the other is a non-anastomotic strictures and that's about 10 to 25% coming from ischemia. So early anastomotic strictures post-liver transplantation is easy to manage with a good resolution but late anastomotic strictures are tougher and it requires a serial dilation and stent placement and the response rate can go up as high as 70 to 100% in one to two years. Covered metal stents are preferred over multiple plastic stents in this setting. Post-whipple stricture, it usually occurs at the colidocal jejunostomy and it's tough to get there. So we use the rigidizing overtube and here you can toggle between flex and rigid and if you set it at rigid, it stays there and as you can see, without rigidizing tube, I'm creating figure eight but with a rigidizing tube, I'm pretty much straight and we can pull the stones out, sludge out and put a stent in. This is a post-whipple patient and I got in and then I'm using basket, I like basket for removing stones and I began to rotate and it started coming. So I kept rotating, it kept coming more. So, most of the stones were removed, there are more to come. Let's move on to malignant indications, cholangiocarcinoma, pancreatic cancer and ampullary neoplasms. I'm gonna just focus on those three. Cholangiocarcinomas are three types, intrahepatic, hylar and distal and we usually see hylar cholangiocarcinoma which is about 50% and this is from a good friend of Linda and mine Linda and mine, and Linda and me and this patient presented with a papillary tumor in the bile duct and it's a courtesy of Dr. Tepesian who I was saying that a good friend of myself and Linda and you can see the imaging is really good because it's not a throwaway, one-time use cholangioscope. It's the prototype and you can really see well and you can put a lot of stents, you can put plastic stents, metal stents, one, two or in combination. Moving down to distal bile duct stricture, pancreatic adenocarcinoma, in 2023, we are expecting 64,000 new cases of pancreatic adenocarcinoma with 50,000 deaths from it. It's the third leading cause of cancer-related death in the US. Unfortunately, 80 to 85% of the patients present with an advanced stage. Overall five-year survival is now six to 11%. The goal of ERCP in this setting is getting into the bile duct and putting a metal stent in. So there are two things for ERCP to be successful. Right indication and right skills. If there is an indication and advanced ERCP skills are needed but no advanced ERCP skills are available in your team, no ERCP will be successful. But if you have advanced ERCP skills in those setting, the ERCP will be successful. So you always have to weigh the benefits and risks. The benefits include relief of obstructive jaundice and risks include pancreatitis, treatment for cholangitis and or sepsis, but it can cause cholangitis, relief of pruritus, it can cause bleeding or perforation. So learning therapeutic endoscopy is a continuous journey. The journey requires a constant practice. And here I'm doing ERCP on Ms. Piggy. Ms. Piggy. And we have a CT scan in the lab and then we're trying RFA before trying it on our patients. So Dr. Lee pick one on the left upper corner. And then we perform the Whipple and looked at the damage we caused. And there was a lot of transmural damage. So we're adjusting the dose of RFA. The journey also requires a team practice. Where's Wen? These are our techs and nurses. I got their permission for this. It's about time that you said, you know what, I'm going. Yeah, you know what? Let's call Dr. Weston. Ryan? Danny. I'm recording you. Okay, let's start all over again. So, we have one of the team members here, our star. Ian, where are you? There you are. So, when you have a great team, things work out really well. And then if conventional ERCP does not work, you can do pre-cut. Then if that doesn't work out, you can do EOS guided biliary drainage. There are two types. One is two techniques. One is through the liver, and the other is going through the common bile duct or common hepatic duct. If that doesn't work, in case of duodenal obstruction, you can do transmural stenting. Here you cannot see the ampulla because there's a tumor infiltration into the duodenum. And then we're performing EOS guided. Then here's another case. Similar, very distorted the distal bile duct, but we are able to put the stent in. And here, the duct is quite dilated, which is very easy to do. And this patient presented with multiple diverticuli in the duodenum, and I couldn't find the opening. So, I asked my ex-fellow, Dr. Weston. Dr. Weston, where do you think the ampulla is? And he said, here, go there. And we went there for 10 minutes, and it didn't work out. So, I found the left hepatic duct. It is at 537 in the afternoon. And then 540 p.m., three minutes later, and then we were able to put the stent in. And then, you see that dark spot on the top? That's where he said I should go. And I agreed. Blind leading blind. And see where the wire is coming out? So, we were able to, and I put the clip on where I worked that. Patient did fine. EOS guided the caudal local duodenostomy. If there is a duodenal obstruction, you cannot do rendezvous. So, you just go directly through the duodenal wall and the caudal hepatic duct and put a stent in. And in this case, the patient had a really dilated bile duct. So, we put a lumen of poison metal stent with excellent drainage. Ampullary mass is another indication for ERCP. Here, about 4 centimeter mass, and I'm cutting it. And we pull the mass out. And here you can see there is a little pulsation. It's the pulsation from me, not from the patient. So, you cut it. And then you really got to think about recurrence. The recurrence rate is about 10 to 15% in one year. So, you're caudalizing the base where you cut. Here, you want to make sure that you have a clear lateral margin. So, you cut a little bit more than the ampullary mass. And here you can see the entire ampullary mass with a negative margin. And PD stent is placed. And the bile duct stent is placed as well. Biliary stent placement in this setting is not mandatory, but I like to, because there is a lot of mucosal edema following ampullectomy. And the follow-up is really important. ESG recommends at 3 months, 6 months, and 12 months, and then yearly thereafter. And here I am caudalizing the base again. A year later, this is what you see. Beautiful new ampulla. Another patient, similar story. And we were able to cut that through and take it out. And then you can see the ampullary mass. Just keep in mind what it looks like. And this is what we cut, identical to what we saw. And this patient had ampullary cancer. So you don't do endoscopic ampullectomy on cancer patients. And we did a rendezvous, and we were able to put the stent in. There have been a lot of advances in technology. And this is a robot operating for us. And what about artificial intelligence? To decide whether we should perform ERCP or not. Here I am at DDW doing virtual EOS. I am actually doing EOS. You cannot see. So we published our paper on artificial intelligence in pancreatic ocular endoscopy and Rochester, New York, I mean Rochester, Minnesota. Mayo group did also. So it's a machine learning. So you pour a bunch of data, and machine can figure out what is normal, what is not normal. So this is a basic example of how AI works. How fast can you find one pumpkin that is different from the others? This is the one that we're looking for. And here, and here. Pretty easy to do. How about here? We need to find a robot that's frowning. And you can't find it here. You can't find it here. What about the second square? Can you find one? You can't. If you do, that's false positive. It's not there. This was published in endoscopy this year. Machine learning model actually did better than ASGE guideline or ESG guideline in indicating for ERCP. The accuracy was 71.5% for the machine, and then 62.4% for ASGE guideline, and then 62.8% for ESG guidelines. So moving on to part two, radiation safety. So here you can see similar units, but one emits radiation from the bottom and the other from the top. So the top scatters the radiation bouncing off the patient. So the distance is really important. You want to be as far away as possible, but you can't really perform ERCP like this. So having a robot would be really nice, but we are not there yet. And the third kind is a C-arm. So mobile fluoroscopy unit, the X-ray tube and image receptor are mounted on the C-arm. Here, the first one is a late image hold acquisition, meaning it's taking the picture and hold it there. The second one is what we usually use, spot image acquisition. And the third one is a spot image with magnification. The radiation dose goes up as you go to the right. So radiation, as you know, will accumulate. The dose that you received today and a dose you received five years later will both remain in your system. The radiation can cause radiation-induced cell death, tissue necrosis, cancer, or genetic mutations. The effective dose is calculated by multiplying the equivalent dose by the tissue weight. The mean effective dose for diagnostic ERCP is 3 to 6 millisievert. For therapeutic ERCP, 12 to 20 millisievert. The dose limit for radiation-related workers is at 50 millisievert annually. And five-year cumulative dose limit is 100 millisievert. And if you look at the lens of the eye, 150 millisievert. So in this study, ERCP was performed in pregnant population. 68 ERCPs in 65 pregnant ladies. No perforation, sedation, adverse events, post-sphincterotomy, bleeding, cholangitis, or procedure-related maternal or fetal death were recorded. But first trimester patients had the lowest percent of term pregnancies, 73%. The highest risk of preterm delivery, 20%. The low birth weight, 21%. But it could be from the hepatobiliary disease, not from ERCP. Another study, 907 ERCPs performed in pregnant population. No reference to the trimester or week of gestation. There was no difference in maternal mortality, fetal distress, fetal loss, when compared to age-matched pregnant population. Non-radiation ERCP during pregnancy. So you perform pre-ERCP abdominal ultrasound, MRCP, or EUS, and you know where the target is. Then you spend least amount of time to get it done. So cannulation is confirmed by aspiration of bile, and at times, oral cholangioscopy can be performed. In this study, five patients with a pregnancy at 20 weeks, mean 20 weeks, between 12 to 32 weeks. The patients were in left lateral decubitus position, and lead shielding was provided from the top, bottom, and both sides. And they used a midazolam. And selective decannulation was confirmed by aspiration of bile. Sphincterotomy and stone extraction were performed. No complications. Here, I had to volunteer myself to be a model. Here is a neck collar and the radiation badge. Then I used a belt to support my back and the goggles. So three principles of radiation protection are distance, time, and shielding. Radiation dose decreases as you get better. Number one, use a pulsed fluoroscopy rather than continuous fluoroscopy. Minimize the fluoroscopy time. Limit the use of enlarged images. Use shielding walls. And wear your protective apron, thyroid shield, leaded glasses. So take on point. Learning therapeutic endoscopy is a continuous journey. The journey requires a constant practice, and also requires a team practice. Thank you.

ERCP

indications

biliary obstruction

cholangiocarcinoma

radiation safety

therapeutic endoscopy