So now, we're going to move on to the next speaker. It's my great pleasure to introduce my dear colleague and friend as well, Dr. Shivangi Kothari, who's the Associate Director of Endoscopy and Co-Director of Developmental Endoscopy Lab at University of Rochester. And Dr. Kothari is going to be going over some of the adverse events and troubleshooting during ERCP. So, very important. All right. Thank you, Sai, and Mo, and ASG for inviting me. It's been a great two days, and we had fun at the hands-on station. It's nice because I had the ERCP station, and I'm talking about all the complications today. So, we'll go over some of the common complications, how to troubleshoot, and a procedure that all of you are now learning, well-versed with, one of the first procedures you'll be learning. These are my disclosures. And we will be going case-based. I have some challenging cases that I will review, and we will talk through them. And if you have any questions, feel free to ask. Disclaimer, not all the complications shown here are mine. Just saying. So, of course, ERCP, we know the nature of the beast. And it's a truly advanced endoscopy procedure, primarily therapeutic. It is resource-intensive, and I said this at my station yesterday, too. You should know your wires inside out. You should know your tools inside out. Being a good therapeutic endoscopist is being a good technician. You are the quarterback in the room. You have to know your devices, and that is what will help you troubleshoot. It is invasive, and it is life-altering, not just for you, but also for the patient. Every complication will leave a mark on your soul, and as well as have patient implications. So make sure that you are carefully selecting your patients. Make sure you have a strong indication to do the ERCP, because it has a high litigation potential when things go south. It does require a commitment, and that is not just by training or getting into the bileduct. I was told very early on in my career when I interviewed for my job, and I asked my then chief who hired me, Dr. Call, and I said, oh, you know, I'm a new graduate. If I don't get in, will one of the partners come and help? And his golden words were, Srivangi, your troubles as an advanced endoscopist are not cannulation. They will start after the cannulation. And those words are so true, and they still hold true. And of course, ERCP is extremely gratifying when you're successful. Pus comes out, you get the stent through, but it is extremely painful when you're working for three hours, you've not got in, and things are going south, and everybody's heart rate in the room is rising. I had the unique privilege of writing the privileging document for the ASGE, and Ashley was the lead author. And the ASGE does recommend at least 200 ERCPs, 80 in sphinctrotomies and 60 stents, and these numbers are just not arbitrary. These are because over time, they have realized that you do need a good number to be proficient, and again, it's hard to assess competency, but to be proficient and to be able to troubleshoot. So I always like to put this, Dr. Cotton's famous words, things will always go south when that procedure is needed the least. And it is so true, and I will show you a few cases of that. So your standard 10-minute ERCP, you only feel as good about your ERCP as your last case. You got in in 30 seconds, sphinctrotomy's done, stent is in, okay, you feel like I'm on the top of the world, I'm in control, but it's when things go south, when the bleeding starts happening, when there is a perforation or your stent migrates, that's the test of time and that's the test of your patience and how you creatively find the solutions for that. So I'll present our first case, a 57-year-old guy with PSC. The billy was high. Anybody think there is a dominant structure here on this cholangiogram? Yes, no, maybe. What is the dominant structure in PSC? So to keep things moving, B is where the dominant structure is, and by dominant structure you mean that there is an upstream blockage and dilation of the system because there are the other structures also in PSC, but the dominant is when it's truly causing upstream dilation and blockage to the bile flow. So now one could argue, well, in PSC you could just get away with dilation of the structure, and we did dilate, and we chose to put two long stents. And now if you look at the literature, they say stenting is actually not the best thing to do in PSC and just dilation actually has good outcomes and lesser complications. They probably looked at our case. So case goes well, patient comes three weeks later, patient is having pain, fever, we're like, what happened? Let's go in. The stent is probably clogged, less likely, three weeks later, stent getting clogged. One of the stent migrated and perforated the duodenal wall. So we're like, okay, let's try to pull it. We try to use a rat tooth forcep to pull it. The stent is jammed into the wall of the duodenum. Now you've not consented the patient for this. You told him he's coming in and you're just changing the stent out. So we punted, we get a CAT scan, the stent is actually perforating through the lateral duodenal wall and the arrow is pointing at the duodenal wall and a good chunk of the stent is outside. So at this point, we're like, we're not able to pull this stent. The stent is going through the wall literally, we can't pull it up into the intrahepatic because there is a stricture above it. It's below the stricture and it's out the wall. So what we thought about doing is cut the stent. So with the use of the endo scissors, we cut the stent in the middle of the shaft. And these are creative solutions. Patient is on CO2, you've talked to the patient about the risks at hand, your patient was given antibiotics because once you pulled that impacted piece, you will see the amount of force it took with the rat tooth forceps to pull the perforating part of the stent out. And that's why it's benign as seven friend stents could even be, not as benign as you think they are. And then the clips were ready to go and in a zipper like fashion, we closed the perforation. This was a double channel scope. So just as we pulled the perforating part, we closed the perforation and this is what it looked like. Patient came two months later, did fine. And of course we removed the stents. This is a second patient with a duodenum perforation, 57 year old, cholangiocarcinoma, bilateral seven friend stents place. Again, the longer the stent, the higher the risk of distal migration. So choose the length of your stents very carefully. And here you can see one stent is in good position, but the other stent has migrated down and that was a stent that was perforating the duodenal wall. This stent was a little more forgiving. So we loaded the over the scope clip. We're able to pull the stent using a rat tooth forceps through the channel of the scope and using the over the scope clip, you can close the perforation. Here you can see the frank perforation in the center of your lumen. As I said, you're using CO2, antibiotics, you're watching these patients. You can shoot a little dye after this as well to make sure that you've gotten good closure of the perforation. But the goal is to keep calm and the clip was loaded. So it's not like I pulled the stent, then I'm loading the clip, then I'm going down. And that's why the x-ray image shows the over the scope clip is loaded. You're pulling the stent, you're deploying the clip and all in once you're done and patient did well. This is a 78 year old lady that came with right upper quadrant pain, severe COPD, EUS showed filling defects in the CBD. And we thought, okay, this looks like a bunch of soft tissue. This is probably sludge. I do a balloon sweep to pull the sludge out and lo and behold, the balloon sweep does not pull out any sludge. It pulls out a four centimeter polypoid lesion out of the bile duct. The biopsy from this showed tubulovilous adenoma. So and that's the entire polyp that was pulled out with the balloon sweep. So patient, again, multidisciplinary discussion, not a candidate for surgery. The surgeons like you manage it endoscopically and we're like, okay, you know, this looks like a adenoma, we can do like an ampulectomy, put a little snare around it. So we've published this case. That's why our thought process was like this animation. We'll pull it out of the bile duct, we'll put a little snare around it, cut the polyp off and life is going to be good. Already had a huge biliary sphincterotomy and but of course life is not that simple in the world of ERCP. So here as we see, we put a snare around it, we cut it, cut goes well and we have two holes. One hole is the biliary orifice. The second hole is a free hole in the duodenal wall. Again, the goal is to identify it immediately. Now what do you do? Do I put the clips? Do I put the stent in first? The main goal here was the bile duct is nice and open and this was a case, again, I did with my mentor. This was in my like second month of my job and I was like, where are these cases coming in Rochester from? So my senior partner, Dr. Cole, was in the room. We did this case together and of course as the perforation happened, we were prepared. Our first thought process is let's secure the PD. You don't want to blindly close the perforation and not secure the PD. So we first secure the pancreatic orifice, put the PD stent, then used endoclips, closed the perforation and then you put the bile duct stent. Again, you're thinking on your feet to troubleshoot in case things goes out. What is more important right now? I have to close the PERF, but I cannot close the PD orifice. So you secure your PD, you put your clips in and the patient did fine. The pathology actually showed intramucosal cancer with tubulovilous adenoma. We brought the patient back a couple of weeks later. This is how it looked. She did great. We watched her overnight and there was no residual dysplasia or adenoma on biopsies. This is an ERCP scope, July effect, that the fellow drove across the GE junction into the retroperitoneum, happens, but the goal is you identify it. Suddenly I'm looking at FAT and I was like, where are you? Pull the scope back, a 15 millimeter PERF right at the GE junction with the side-viewing scope. So we identify it, we put the over-the-scope clip, but your problems in ERCP don't end that quickly. As we inject contrast, the contrast is not making it across the pseudopolyp. So we put a fully covered stent because how will the patient tolerate the secretions? We call the surgeon, we're like, this has happened, we've put the ovasco, there is no leak, but the contrast is not going across the GE junction. He said, put a fully covered stent because otherwise the patient is going to aspirate. How will the secretions go down? So we put a fully covered stent. This is a 91 year old patient. The ERCP is on the side now, the ERCP is still hovering on our heads. Luckily patient wasn't cholangitic, there was stones, non-occluding, so when we brought the patient back, I pulled the esophageal stent out and we were able to complete the ERCP. And this is how it looked at the end. I was talking about guidewire PERF at my station yesterday. This is a pancreatic mass. Now pancreatic mass is like, you know, they chew up everything, the anatomy is completely changed, completely different. Anybody think anything wrong in this cholangiogram? You see the stricture, the sphincter tomes there, you're seeing some lighting up of the bile duct. So again, it's hard to do it in a still image, but you keep monitoring your images and your cholangiogram, the wire was in a different plane than where the contrast was going. And I readjusted the wire because I was like, the wire is going, but the contrast is faintly going in another space. So we readjusted the wire and there was a guidewire PERF. So initially the wire was in a different plane and the contrast was in a different plane, and this can happen in cancer patients. We've had three cases now sent to UFR where after two hours, wire goes in, everybody's happy, you put the stent in, and this was from the community, sent to us, the stent was put in the portal vein. Because the mass had chewed up everything, so the wire goes right in the direction of where the bile duct would be, and you put the stent in, you quickly get out, and when the bile doesn't improve, you get a CT, the stent was in the portal vein. So things can happen, you have to be vigilant, always watching your cholangiograms. We put a fully covered stent across this and the patient did fine, but the goal was to identify that the wire is not in the plane of where your contrast is going. Coming to bleeding, we've done enough holes now. So this was a patient with a large ampullary adenoma. This is Dr. Kohl's case, so I thanked him for sharing it with me. And again, if you talk to your different mentors, everybody will do the ampullectomy a little differently. So we put a PD stent first and then cut it, then some will say just cut it and then worry about the PD. Here we are using a big braided 20 millimeter snare to do the ampullectomy, and you're almost putting it like a garland around it. You want to make sure you've grabbed every last bit of the bottom of the tissue. And we do the, okay, so we do the ampullectomy, you want to get it on block, and the first thing actually when you do an ampullectomy is grab the tissue. So you want to grab your specimen before you do anything, PD stenting, CBD stenting. Specimen retrieval is as important as resection. So even if you're doing EMRs, ESDs, grab your specimen, then a PD stent is placed in successfully, then a biliary sphincterotomy is performed, case goes well, everybody's happy, you put the CBD stent in and looks good, case is done, not so much, and you start getting pulsatile bleeding. At least at this point, you've secured your PD, it happened right at the end, you watch your area, your tools that you use for hemostasis are the ones that you're going to use routinely, little bit of epi is injected to slow down the bleeding, it gives you time to think what you want to do, and after the epi was injected, clips, endoclips were used, again, I have no shares in Duraclip, but the Duraclip works well on the elevator. So if you have to do any post-sphincterotomy or bleeding, the Duraclips work well, make sure you're not closing off the ducts. So as easy as you think clipping may sound, you saw two cases now where we wanted to make sure that the pancreatic orifice was not being shut down. And after putting the clips, the bleeding was successfully managed. You want to watch these cases, you want to lavage because you are there, that is your moment to manage that complication, that adverse event. You don't want to be in a hurry to pull out, and then your partner is going in the middle of the night to salvage because the patient kept bleeding. So when you are in there, identify the problem early and take your time to fix the problem. Watch that area for two minutes. Only if you're satisfied, then pull out. There are some cases of using hemostatic powders and gels for post-sphincterotomy bleedings. I think you can use it as the last resort, but I always like to make sure that the PD is protected. I have my concerns if the powder gel would clog up the PD and you get, now you've got another problem and you're doing, you're getting post ERCP pancreatitis in addition to the bleeding. You can't create another problem while fixing one. Here's where multimodality therapy comes into play, papillary large balloon dilation is being performed for a CBD stone. Just as the balloon is deflating, significant bleeding. So you immediately want to secure your access, so you have a wire, right, put a stent in. So a plastic stent was placed, epi was injected, balloon tamponade performed, the bleeding didn't stop. The plastic stent had to be removed and finally a fully covered metal stent was placed, which finally tamponaded the bleeding. The goal is do not lose access. If you get torrential bleeding, say you're doing a sphincterotomy or you're doing a large balloon dilation, that wire is your savior. So do not pull the wire. You can put a stent in, you can put a fully covered stent in, but that wire will help save your day. So always make sure you keep that, do not lose visual of that wire. And then lastly, I'll end on this note. As much as everybody tells you through your GI fellowship and advanced fellowship, all bleeding eventually stops. This is a minor papilla sphincterotomy. Patient with pancreas divisum, minor papilla sphincterotomy done. You can see how small the cut was. The bleeding started, of course, immediately we think, okay, let's put a stent, that will tamponade it, didn't stop. What co-act forceps use, didn't stop. So the stent was removed to do a balloon tamponade, didn't stop. And the bleeding persisted two stents later, three clips later, the bleeding continued and patient went to IR. There was an unidentified pseudoaneurysm. So I said that yesterday, IR and surgery are your friends. If you have to punt and send the patient to IR, do what's right for the patient. It's not a personal loss. It is a resource intensive procedure. Make sure you have appropriate indication, you have a good informed consent, you have all your tools ready to go and early identification is key. You identify that your wire is in the wrong place or there is a poor for you've driven the scope into the retroperitoneum and you fix that problem at hand and when you have your scope down, that is your window to do it. Lo and behold, if you're not able to do it, have that favorite surgeon and that IR colleague have the adequate backup and work clearly with your team. Do not yell at your team, do not yell at your nurses and techs as much as on the inside you're getting frustrated. You set the tone in the room for especially when things goes out. If you are flustering, you're getting angry, the team will start to fluster. So keep your calm as much as it's easier said and done. You are the quarterback in that room and own your complication and then manage it adequately. Thank you. Questions, concerns? Yes. So the base was right at the ampullae, it was just so big that it was kind of flopping in and out of the duct and there was no extra duodenal portion. So it is actually not that common. We actually have a series of three cases of intraductal adenomas and the anchor point is at the major ampulla, which is why we thought, oh, we can just put a lasso around it and cut it off and it should be fine and there was a beautiful biliary sphincterotomy. So little did we know that there was some portion stuck to the duodenal wall, which is why it perfed. Generally speaking, if there is any concern of an intraductal extension, we should not resist that. I mean, if you extract the polyp from the duct and then you got it outside, I mean, it's protruding outside now and you cut it, there is some intraductal extension, you can have a perforation from the side of the polyp. Right. We have also reported use of like the RFA catheter to treat the intraductal portion. If you are able to get it unblocked, you could use a fully covered stent and bridge that leak. It will all depend on where the anchor point is for that polyp. Are there any tips and tricks to avoid the distal stent migration and the perforation as you have shown? So carefully choose the length of your stent. Don't put two very long stents in short distal strictures. I use my scope diameter to gauge the length of the stent I should be using. And then, you know, sometimes it's just a matter of luck. I was talking about it yesterday. I do prefer the stents that have the natural bend, the duodenal bend in them, because they kind of follow the path of the duodenum, less likely to perforate through the lateral wall. The other tip I would say for that is consider using double pigtail stents. Those may be a little gentler. I mean, that's still not 100% guaranteed, because like the tip of the double pigtail could get buried, but that's something else to consider. And which is why you saw once we removed it, we switched to double pigtails after we closed that perforation.

ERCP procedures

adverse events

endoscopy

patient safety

case-based examples

collaboration