So, these topics are always, you know, it's interesting to show your complications, but at the same time, it's humbling, too. And these are all educational sort of opportunities for us in the field and for anybody coming into ERCP, because you learn a lot from how things are not supposed to happen, but also it's good to step back and have a background of what to be aware of as you enter the field. So, briefly, I'll go over some of the risks and adverse events. We don't use the term complications. We use the term adverse events because of the simple fact that no one's trying to complicate anything. These things happen, and this is the nature of the game here. Specifically associated with ERCP and the principles of management. So, one of the initial six, you know, Peter Cotton, who's like one of the forefathers of ERCP, said ERCP is most dangerous for people who need it least. And that still stands true. And it's not just about the adverse events. You have to step back, and I think that's a recurring theme today. Is the ERCP really indicated, or is there a better or a safer way to figure out what we're trying to figure out? Did we get the appropriate consent? Did the patient, you know, were they told about what the procedure actually involves, what the risks are, the benefits are? They literally show up, and you ask, why are you here? Because my GI doctor said I need to have this procedure done. So you have to start from zero and say, okay, well, this is what they're referring you for. This is the procedure. This is the potential risk. These are the risks. And you can't downplay that. You have to be honest and upfront. The limitations, whether it's endoscopy, equipment, anesthesia, fluoroscopy, we've touched on those as well as anatomical limitations. And these have clinical, not just clinical implications, but real-life implications in terms of the financial burden. And we talk about healthcare systems and hospitals, but to the patient and their family members. A patient who gets bad pancreatitis is in the hospital, but elective procedure to being in the hospital for three weeks, the burden of that on the patient and their family. Incomplete or suboptimal procedures, needing for referrals, need for referral to other specialists or other hospitals, and then the delay in subsequent career. And the big one is mortality. ERCP is associated with mortality, and you just cannot downplay that. And every consent that we do, we teach our fellows, you have to bring in mortality. You cannot walk away from that patient telling them there isn't a risk of death, because we owe it to the patients. So that's something you need, obviously, people who are in the field need to know about. So I'll go over multiple complications, but these are the big three. Pancreatitis, bleeding, perforation. Just for my understanding, how many of the people in this room actually have had any sort of exposure, like I think you mentioned, with some sort of ERCP? Like you guys see ERCP, or you've in the past done it? Okay. So you've all seen, and how many people would say that they have a good understanding of what pancreatitis is? This shout out, when you say, when you hear, we keep throwing this term at you, right? What is your impression of pancreatitis? What does it mean? It's missed? And what does that mean, like in real terms? So what- Inflammation. Yeah. It's food or hormones to control blood sugar. Okay, good. So yeah, that's more on the chronic side, right? Longstanding pancreatitis, staying at home. What about acute pancreatitis? They're the ones in the hospital, right? Point that it just doesn't drain properly? Yeah, so essentially, I'm going to come to some of those things, but it is an acute change in the pancreas' function that leads to, or acute inflammation in the pancreas that leads to a systemic response. So it's no longer limited to the pancreas, but the entire body. And that's why patients get so sick. And I think just for, I'm going to give you some background on pancreatitis to make you understand why we worry about pancreatitis so much. So if you look at the definition of post-ERCP pancreatitis, firstly the burden, 35,000 cases per year, over $200 million cost. The definition is, in the setting of an ERCP, new onset or worsening pain with an enzyme level greater than three times the upper limit of normal. And the carton definition says more than two days of hospitalization. But if you step back and we look at overall pancreatitis, what is the definition? Or what does pancreatitis imply? So as everybody was saying, there's a pointer here. This is your pancreas. The pancreas sits behind the stomach. And the bile duct from the liver runs right through it at the end. That's why, by its nature, the pancreas is going to be in the field, right? It's in the firing zone. And everything we do can have an implication on the pancreas. There are two types of pancreatitis, interstitial edematous, where the pancreas preserves its circulation. There is no harm done to the circulation of the pancreas. The organ survives. The one we worry about mostly, in any setting we worry about it, you know, mild too, what we worry about is necrotizing pancreatitis. These are the patients where part of the pancreas dies. So here you have a patient, a young woman, right upper quadrant pain, gallbladder's gone. I keep having pain. Doctor says, let's do an ERCP. She shows up, elective procedure, has an ERCP, gets severe pancreatitis, organ death in the hospital for a month. Worst case scenario, dies. Young patients die. That's what we're talking about. That's what we worry about the most. Of course, we worry about the mild too. In order to understand why is everybody always harping on about pancreatitis in this field, that's what you worry about. And those are the patients that are perhaps most suboptimally treated. The mortality, the mild interstitial is not associated with much mortality. What we're concerned about is the necrotizing pancreatitis, which when it becomes infected, as you were saying, big collections, those are the patients that have higher infected pancreatitis mortality. Then this whole evolution concept. You start off with necrosis within day one to five, whatever, and then the patient's not getting better. Immune systems like rev depth, multi-organs are involved. You can have multi-organ failure. And in the process, you have this necrotic collection, which becomes infected. And these patients are the ones who stay in the hospital for a long time. So I just wanted to step out of ERCP and just give you a background on pancreatitis before we move forward with why we're so paranoid and worried about post-ERCP pancreatitis. Secondly, the difference in definitions. If you look at what I was just showing you, that's the Atlantic classification. That's what we rely on for all pancreatitis. What you see is that the cotton criteria is slightly different. The definition here is pain. The enzyme level is greater than three times the upper limit of normal and imaging characteristics. Here there's no mention of imaging characteristics. We're talking about hospitalization. This is more of a soft definition, because it goes back to that original definition that Cotton described. This definition has been revised, but we haven't really quite revised the post-ERCP pancreatitis consensus cotton criteria definition. And that has limitations, because we know post-procedure, a lot of patients will have some degree of enzyme elevation. So we expect that. They may have pain, but the pain may not necessarily be from ERCP. It may be from other factors. It's a spasm stent. I have a physician who I was just texting. I did an ERCP. He says post-ERCP for pancreatic duct stones, I've had this pain in my right side, goes up to my right shoulder. He's a gynecologist. And I said, I can't remember if I put a biliary stent in you or not. Let me check. And we're going by text, and he's like, yeah, you did put a stent in. It's the stent that's causing the pain. It's got nothing to do with his pancreatic duct stone. He said, this is a new pain. So every little thing we do can cause problems. And then you have admissions can be for other reasons. It doesn't have to be for pancreatitis. So you lose your sensitivity and specificity, which makes it hard to really truly understand and study the burden of the complications or adverse events caused by ERCP. So what we did was a study where we looked at symptoms. We said, OK, what are your baseline symptoms coming in? And then what are your symptoms post-procedure? So we're moving away from just this cotton consensus definition. And we're going to see how you are clinically. And what we found was on a standardized questionnaire where there is a scale of seven questions, zero to 10, what we found was that the most accurate sort of cutoff was a score of greater than seven from baseline. And when you see that, when you do that, when we did that for 679 patients, by the cotton criteria, only 4.7% of patients met the criteria for pancreatitis. But actually, if you looked at symptomatic burden post-procedure, that's 20%. That's one in five. So one in five patients has some sort of significant increase in symptoms, but our current definition only captures five. On top, there isn't good overlap. So it goes to show our definition isn't perfect, but we'd work with what we have. So all the data you're going to see is actually based on a somewhat limited definition. Not flawed, but limited. But what it goes to show is that there's more to it that meets the eye with respect to patients. Regardless, let's just focus on the definition. Has the incidence of post-ARCP pancreatitis changed over the course of time? This is an interesting way of looking at it. They looked at a meta-analysis of—everybody familiar with a meta-analysis, right? You take a whole bunch of studies with a similar theme, and then you analyze the data together. 145 studies, and what they found was that over the course of time in randomized trials, the rate of post-ARCP pancreatitis does not seem to be changing at all. And in terms of high-risk patients, it's still at that 14%. So no matter what you do, if your patient is the highest patient, you're going to be at a higher risk of giving them pancreatitis from the procedure. And mortality persists, too. So this, I think, is where the crux of the whole pancreatitis debate gets to. Dr. Freeman, who's a big advocate of figuring out how to fix pancreatitis, especially procedurally, has come up with the five Ps. And we ran out of the last one, so he said, you know, let's use Ph. So patient-related, procedure-related, pancreatic stents, pharmacology, and fluids. These are the five that we need to remember. And if you look at various guidelines and various studies, they will mention the different patient-related and procedure-related factors. I think the big one I want to bring up to everybody's attention for procedure-related is pancreatic intervention, whether it's contrast, wire, wire passage. On the one hand, guidelines say use wires for cannulation. On the other, we say multiple guide wire passages cause pancreatitis. So inherently, our procedure is going to put the patient at risk of pancreatitis, right? So we just need to know about these. I won't harp on too much about these. And then the ASC guideline that Renee was part of, sort of leading, you know, has these summarized. And we're going to hear more about the guidelines. But moving on to the actual prophylactic measures, what can we do for these patients, right? So this was this landmark study by Joel Munzer looking at high-risk patients undergoing ERCP and randomizing them to no rectal endomethasin, which is a non-steroidal anti-inflammatory medication, versus the medication itself. And there was a clear benefit in giving rectal endomethasin to patients undergoing high-risk ERCP. So this sort of set the scene for all patients undergoing ERCP getting rectal endomethasin or another type of anti-inflammatory. And if you look at meta-analyses of various studies, there is definitely a benefit as outlined in these sort of tables here. And I'm going to come back to them. So I don't think we're done with rectal endomethasin. Not quite yet. PD-STEN, I think we mentioned this. It's been shown earlier. This is a cannulation I did. By the way, that trauma is not from ERCP. We did an EUS first, so that's just the scope against the papilla. Got the wire to the pancreatic duct, second atome wire into the bile duct, and then we're putting a soft STEN into the pancreatic duct. And following this, we can get on with our biliary intervention. So that's the principle of a pancreatic sort of STEN placement. There's a whole wide variety of STENs. I would encourage Medtronic to also get in the game. If you're going to get into it, just come in all in rather than partially. That said, I think Praveen had mentioned this. Multiple orientation STENs, generally speaking, I would say most of us prefer to use soft STENs for prophylaxis. These bigger, harder STENs are usually for therapeutic conditions, such as strictures and leaks in the pancreatic duct. This is the Hobb stent that Freeman was part of developing, and these are quite popular now because they're very soft and they're slippery. Praveen mentioned the issue of follow-up. These tend to fall out quite easily. So these are good stents. And then how do you make the decision which one to use? So here's a simple sort of schema. If you have the wire out to the tail, then you can use pretty much any stent. If you have a long one, you can go long. If you want to go short, you can go short. If you go short with flanges, stiff stents will not fall out. Soft flange stents do fall out, so it depends on the type of stent, too. And soft, long ones, with or without a flange, tend to fall out. But we have these very difficult anatomies, right? You don't necessarily need to go all the way to the tail. If you're stuck here in what's called an ansa loop, you just need to leave the wire short and just put in a short stent, and you can get out. So you don't have to push it to the point of the stent causing pancreatitis or your attempts of getting the stent and causing pancreatitis. So this is where the data gets interesting. So a Munzer study was published in 2012, and what you see is prior to the era of, say, 12, 13, 14, post-ERCP rectal endomethasin 6.4%, PD stent usage 37%. Here comes the wider acceptance of endomethasin. Already, endomethasin use has gone up. Stents drastically dropped to 8.5%. And if you look at the rates of post-ERCP pancreatitis, they haven't necessarily changed. They haven't gotten better, maybe slightly gone and gotten worse. So there was this whole concept, well, maybe we can replace stents with endomethasin. But actually, the data started to suggest that really we're not really changing our rates of post-ERCP pancreatitis. And so how do we sort of reconcile that? I'll come to that in a second. But in the meantime, we started this. I'm just giving you a historical perspective. We started also figuring out that fluids actually are helpful in patients with post-ERCP pancreatitis. And if you hydrate them well, you can reduce their risk. And usually, it's lactated ringers because it neutralizes the acid environment. But then there was this huge effort, and with colleagues in the room who were part of this, of doing, again, Almanzar being a forefront leader in the field, did the study of comparing endomethasin to stents. He said, let's just try and figure this out. Which is actually superior, endomethasin alone or endomethasin with stents? Because we're trying to get rid of stents, but is that really the right strategy or not? And what they wanted to do was a non-inferiority study, meaning that they just wanted to show equivalency. What they found was that their results crossed over the line of the non-inferiority margin. And because it crossed over, they came up with the conclusion that actually stents plus endomethasin is better than no stents or endomethasin alone. So it wasn't like a humongous difference, but it was enough to say that actually a dual modality approach is better than a single modality. So now there is a movement back towards putting stents in, in patients especially who are considered high risk. And so how do we put it all together? We basically say, well, patients, you look at the risk factors, you look at the procedural risk factors, you give endomethasin to everybody if you can. PD stent, if you get access, high risk. You put in the stent and you give them fluids. So you do all these. So post-traumatic pancreatitis is not a single modality sort of approach. You have to look at all of these. That's PEP in a nutshell. OK, moving on to bleeding. Of course, there are patient-related factors that can inherently increase the risk, such as cirrhosis, anticoagulation, low platelets. But there are procedure-related factors. So low procedure volume. Do some research on who you're getting your ERCP from, if possible. Procedural bleeding, that and unsuccessful cannulation leading to pre-cancellation. I'll show you some examples. This is a case of a 12-year-old kid. I had done a sphincterotomy, came in with biliary obstruction, had actually an anomalous junction with a choledocal cyst. This is a congenital malformation of the bile duct. And we had done a sphincterotomy. Everything went well. Cleaned out her duct. But two days later, she was bleeding. And if you see, there was a plot on the site of the sphincterotomy. That's where patients bleed from. When we cut up towards the bile duct, we cut across a small arterial vessel. And so there was a bleeding from there. And here we're deploying a stent. The issue is deploying a clip, rather, through a duodenoscope. It's somewhat challenging because it's not coming on force. But it can be done. And when you sort of put a clip on successfully, you can basically deal with the vessel. Even though the clot is there, the vessel is dealt with. And of course, you're going to have the vessel on the other side, too. Sometimes you'll see it on both sides. This side, I decided to use quartering instead of the clip. The pancreatic orifice is going to be down here at the base. So we're safe from the pancreas. But this kid got treated with two different sort of approaches. But bleeding can be significantly worse. This is a patient with a liver transplant with an anastomotic stricture, referred for ERCP. A colleague, these are also the ones where these kind of cases, when you show them, you want to make sure if it's not yours, you point out that this wasn't me. But a colleague did a needle knife, very reasonable. And one strategy is let the tissue necrosis, let the edema resolve. And then you go back in a couple of days later. So the plan was to repeat it 48 hours. But the next day, the patient started bleeding to the point of hemodynamic instability. And we got called, can you take a look? And so this is us going down into the duodenum. And what you see is this blood clot in a frank, bright red arterial blood. So this is somewhat of a blind cannulation with a sense for where the papilla is and the blood is coming from. Unfortunately, I don't have the fluoro to go along with it. But we got the wire into the duct using a smaller wire because I wasn't sure where we were going to end up. And you can see there's a stricture here as well as the dilated duct. So I wanted to dilate the stricture before attempting stenting and then also dilated the orifice and then placed a fully covered metal stent across the stricture as well as the papilla where the bleeding was. The idea being that covered stents will tampon out the bleeding and also treat the stricture. Now, this was somewhat of an overkill. But when you have minimal oozing at the time of synchrotomy, we can simply just inject some epinephrine into the tissue and get some sort of vasoconstriction there. With the neurotopical agents, we can also spray those. But you don't want to clog off the PD. So we'll sometimes use those agents too. But a lot of times, we'll just use some epinephrine. So if it's mild interprocedure, just epinephrine. Any concern that's sort of standard nowadays is putting a covered metal stent and not lose sleep over it. Because the burden of the patient coming back with bleeding, hemodynamic instability, blood transfusion, et cetera, is much worse than a $2,000 stent. So low threshold if our level of suspicion is high. So basics, you know, access, maintain access as much as possible. Epinephrine injection spray, hemostatic agents, and clips, and then the covered stents. Perforation, this is not pretty at all when it happens. Altered anatomy, again, sphincterotomy, pre-cut. That's when you'll get perforations. And also, if there are strictures, say in the duodenum, or altered anatomy, then you can get bowel perforations. This was a patient at another hospital who was referred to us later, but basically underwent an ERCP for a suspected stone. She gets an ERCP. The pancreas is a retroperitoneal organ, as is the papilla. So what you don't see is ear in the peritoneal space. You see retroperitoneal ear. So this can be hard to see. But she woke up with excruciating pain. Clearly had a perforation. They watched her for 10 days or so. She developed cholangitis. That's when she got transferred to us. By that time, things had improved a bit. We went in and changed out the stent, brought her back. And this was actually pancreatic cancer. So this is where we're talking about delay of care. Pancreatic cancer patient gets an ERCP for a stone, which wasn't there, and gets a perforation. So this has significant implications. This was one of mine. Recurrent acute pancreatitis, stenotic papilla. Everything seemed to go well. Looks a bit beat up. But difficult cannulation, double wire, PD stent, biliary stent. Unfortunately, the patient gets pancreatitis, goes home the next day or a few days later, presents to the outside hospital three days later. And that sphinctrotomy site just opened up even more over time and has this huge collection in the retroperitoneum. She came back to us. We had IR put a drain in. We went back in and put in a metal stent across the orifice to seal up the leak. And then she had the drain for at least, I think, two weeks before she got better. And this is what we, I think, Praveena brought up. You know, transpancreatic sphinctrotomy. So basically what you're doing is you're actually wedged into the pancreatic duct, and you're cutting up, creating space for your bile duct access. This was another one of my endeavors. Now for full disclosure, I got so confused under the circumstance. This duct was so dilated that actually I thought I was in the bile duct. So this is also not last week. This was many years ago when I was a bit more junior. So didn't recognize the pancreatic duct from the bile duct. Kept cutting and saying, why can't I see bile? And then I see free air. And of course, what do you do? You call a senior colleague. So this is, I called Dr. Freeman from clinic. And we secured the pancreatic duct, got a PD stent in, and we're probing, and we're just in the retroperitoneal space. And all you can see is contrast in here in the retroperitoneum. And this is where you thank God for EUS. We had the EUS scope, you know, did a rendezvous, found the duct, accessed it, and put in a covered metal stent. In terms of perforations, again, a lifesaver in the last sort of decade or so have been covered metal stents. If you recognize it intraprocedurally, we just put a stent in, the patients wake up, they don't even realize they had a perforation. If you don't recognize it early, that's a problem. Then they come in with all these complications. But you put in stents, you give them antibiotics, you observe, you communicate clearly. If the hospitalist gets a CT scan on that patient, there's going to be panic all over. The surgeon is going to be called. Patients have been taken to the OR for no good reason because the problem has already been fixed, but there's air on the CT scan. They're setting expectations, and then for the patient and family. Infection, briefly, if we don't drain, any manipulation can lead to infection. If we don't drain them well enough, such as hyaluronabstruction, they can get infection. And chalangioscopy we heard about. This was just recently in our system. This is actually interesting because the gallbladder is already distended, right? And the patient has a very dilated bile duct, pancreatic tumor, gets a fully covered metal stent, and a month later presents with pain, new ascites, and air around the gallbladder with a concern for perforation of the gallbladder wall. So it doesn't have to be simply that this could, you could argue, well, this is a progression of his cancer, but we could have caged off that cystic even more than it was already caged. Or this has progressed to the point of perforation. The patients will come back with complications such as cholecystitis that we need to be aware of. This patient ended up going to the OR for surgery. So to summarize, avoidance of unindicated ERCPs, that's number one. Clear communication with the patient and family about the risks and benefits of the procedure. Understanding of the complications that occur and their associated risk factors. Knowledge of optimal peri- and procedural interventions. So fluids, indomethacin, stent. You know, those three are a must apart from the patient-led factors. And then early recognition, staying calm. You know, we cannot, as physicians, lose our cool because then that sets the pace inside the room. And then if you're in the room, then you get in the firing line, which is not really nice at all. Because I've seen that happen. It was the rep's fault. You know, he distracted me. Really? Because you're the one who just kept talking anyway. So, you know, so tailor it to the situation. Preserve normal flow. We try and preserve pancreatic flow, biliary flow as much as possible. And then involve others. Call a colleague. I actually ask anybody in the room. I'll ask the tech, what's your thought? Ask the nurse. And also sort of call other colleagues. And then communicate openly and clearly, you know, both with the patient, their family, our colleagues in the field, and, of course, partners in the field as well. Thank you for your attention.

ERCP complications

informed consent

adverse events

post-ERCP pancreatitis

preventive measures

patient communication