All righty, guys, so now that you're energized from your snacks, we're going to move on and talk about, this is a little kind of pancreas mini-session here. I'm going to start off by talking about managing pancreatic fluid collections, and then Dr. DeWitt will talk about celiac plexus block and neurolysis. So this is a case of a patient, 54-year-old male with severe gallstone pancreatitis, five weeks ago, transiently intubated, now on the floor with abdominal pain, and can't eat. No other organ failure happened, and this is, of course, why he can't eat. You see this big, walled-off necrotic cavity that's squishing his stomach. And feel free to honestly interrupt me as I go along. If you have questions, I don't mind if you guys do that. Now, I think most all of you know this, but just a quick reminder of terminology. Of course, we have interstitial and necrotizing pancreatitis. In both situations, you can get acute fluid collections in interstitial pancreatitis. We call that acute peripancreatic fluid collection. And in necrotizing pancreatitis, acute necrotic collection. Both of those situations, if you fast forward usually about four to six weeks, those collections can mature and become a pseudocyst in interstitial pancreatitis and, of course, walled-off necrotic cavity in necrotizing pancreatitis. And either of these can become infected at any point in time. So here, looking at these two CT scans, I hope you can appreciate that the one on the right is showing you kind of straightforward interstitial pancreatitis with all the fat stranding around it. Pancreas appears viable, though. It's nice and bright as it should be. And on this image, you start seeing some amorphous fluid around the pancreas, around the tail here, and you see a thin sliver kind of around the neck, going down to the head as well. As you scroll down more on the CAT scan, you see more and more fluid. So this is the acute peripancreatic fluid collection in interstitial pancreatitis. And again, you fast forward, and suddenly it's coalesced to form the pseudocyst with that nicely defined wall there. Contrast that with acute necrotizing pancreatitis. You see here bright pancreatic head, neck, tail, but then the body, you see it's dark there. So that's the necrotic pancreas. You can almost imagine this kind of dead pancreas tissue being vomited out into the space there. And then again, fast forward about four to six weeks, and this can become a walled-off necrotic collection. So back to my patient who had necrotizing gallston pancreatitis and can't eat now. So how we handle these fluid collections really revolves around a couple of decision points. One is the patient symptomatic from this or not? Because if they're not, you don't need to intervene upon it. However, of course, if they do have symptoms, that's when you need to do something. And then the second point is whether or not the collection is mature. One thing about sampling these fluid collections, this is from the ESGE guideline that said they recommend against routine percutaneous FNA of pancreatic collections. FNA should only be performed if there is suspicion of infection and clinical imaging signs are unclear. A weak recommendation, low quality of evidence. You don't have to obviously routinely do this, but if there is a suspicion and you're wondering, certainly can have your IR colleagues do this. One word about conservative management of infected necrosis, because it can work. And this is defined as using antibiotics with or without percutaneous strain. So not endoscopic or surgical approach, of course. And studies suggest that it can work in up to even two-thirds of patients. You see there about almost 30% ended up needing endoscopic necrosectomy or surgery. So if they do need some kind of intervention, as with many things, we have endoscopy, radiology, or surgery. And when I was training, it was surgery was the kind of first thing that we did, because there was no such thing as endoscopic necrosectomy. But now, things have completely flipped around the other way. And this is just one of the randomized studies that have been done comparing endoscopic to surgical necrosectomy. And they all pretty much say the same thing, which is that major complications is significantly higher with the surgical group compared to the endoscopic group. So therefore, nowadays, surgery is really reserved as last resort. Now, of course, there's IR guided approach as well with their drains. And there's not a lot of studies comparing the endoscopic to the percutaneous approach in managing walled-off necrotic cavities. The table I show you on the left is from our group at the Brigham. Chris Thompson published this. And the one on the right is a more recent experience. But they both pretty much say the same thing. And these are retrospective. These are not randomized studies, certainly. And what our group, what Chris found, was that in the percutaneous arm, which is on the right here, more procedures were needed per patient. The length of ICU stay was greater. And the clinical resolution rate was significantly lower, which is echoed by this more recent retrospective study comparing the EOS guided to the percutaneous approach. And in this particular recent study, the mortality rate was significantly lower in the EOS guided approach as well. So the endoscopic approach seems to be favored. However, it doesn't always work. And these are some of the predictors of when the endoscopic approach may fail. And this is, again, from our group. Chris published this just this year. And what they found, looking at our series at the Brigham, was that these were the three predictors of failure with the endoscopic approach. If the Apache's score was higher and the percentage of necrosis was higher, they used 50% or greater. As their cutoff subsidy used 40% or greater. And if there was evidence of pericolic gutter extension. And this is just a CT scan showing you an example of a patient with this huge cavity that's extending down into the pericolic gutter. Other studies have suggested the cavity's really large, over 18 centimeters. That's another predictor for a failure with the endoscopic only approach as well. So some things to bear in mind when you have a patient in front of you and trying to figure out, OK, should we do this purely endoscopically or should we be doing something else as well? Now in terms of if the patient is symptomatic and you need to do something, one of the first questions, of course, is when should we go in and try to do drainage? And usually we like to wait for the collection to mature, which basically means that this amorphous collection has developed this nice wall around it that you can see on imaging. Usually takes about four weeks and some patients it can happen a little bit sooner. Some patients it can take a little bit longer. And this is one study that tried to look at, well, gee, can we do this even earlier, even before the wall has formed? And I just wanted to point out the mortality rate was higher in that group that were done soon within that four-week period. Their median length of stay was also higher. And in the paper, they were saying, well, perhaps it's because the patients in that group were sicker. They had more pulmonary issues and whatnot. But I think it's still not standard of care to go in before a wall has fully formed at this point. So I would say we still need to wait until that wall has formed. Now if you've decided, OK, the patient needs drainage, we're going to go in, EUS or no EUS? Well, there are two randomized studies that have shown that EUS leads to greater success. It's of course helpful because you can identify vessels that might be in the way. These patients, of course, can get splenic vein thromboses and whatnot, and develop varices, et cetera. And it's also helpful in trying to help identify the optimal entry point where you're going to deploy your stent. In terms of the type of stent, you certainly can use double pigtail stents, fully covered self-expandable metal stents, and then, of course, more recently, the lumen-opposing metal stent, of which we only have one available in the US right now from Boston Scientific. But that really has revolutionized the procedure and really opened the door, of course, for all of therapeutic EUS. Now studies have tried to look at this and found no significant difference in treatment success between plastic stent or the use of LAMs, the lumen-opposing metal stents, whether you're draining pseudocyst or walled-off necrotic cavity. Where we do see differences, of course, is that the older method of using plastic stents requires multiple steps. That's one of the great things about the LAMs is that it's all just in one, whereas before you needed to gain access, dilate, et cetera, et cetera. So it is a little bit longer if you do it that route. However, again, clinical success seems to be very high with both approaches. In terms of complications, maybe a little bit lower with a plastic stent. There are differing studies that have differing conclusions on this. Of course, the one thing that we fear with the lumen-opposing metal stent is leaving it in too long, right? Because studies have shown that if the collection has collapsed and then the LAMs phalange is rubbing up against the wall of that necrotic cavity or the pseudocyst cavity, it can lead to potentially devastating bleeding complications. So you don't want to leave these in for longer than about three to four weeks, whereas the plastic stents, you can leave them in forever. If somebody has disconnected pancreatic duct syndrome, they're just leaking, leaking, leaking, it's OK. Just leave the double pigtail stents in forever, and that's OK. Of course, the cost is hugely different between using these few hundred dollars, not even a few hundred, $100 plastic stent versus the lumen-opposing metal stent. However, when studies looked at the overall cost of the patient, it seemed to be similar. Now this is a video from my colleague showing the older technique. So you see the collection there, and there you see accessing with a 19-gauge needle, aspirated first to set enough for culture, injecting with contrast to define the cavity and actually expand it out, coiling a wire in there, and then dilating with a balloon going into the cavity and then doing some debridement. And as you can see, when we first started doing this, we used to do a lot of debridement in the beginning, and I'll talk about that more too. And then finally, double pigtail stents being deployed here. And you can watch that with the newer approach, where you see the lambs has already gone into the cavity, and you see the first phalange being deployed, pull it back, and then deploying the other phalange where you see the pus coming out now, a lot quicker. So now that you have drained the cavity somehow with plastic stents, et cetera, whatever, the question is, do you debride or do you not debride? And there's a few questions to answer here. Do you want to do the debridement at the time of establishing drainage with your stent, or do you wait? If you wait, do you bring these patients back on kind of this protocolized schedule? OK, I'm going to bring them back in, what, three days, seven days, whatnot? Or do you do it as needed, depending on how the patient is doing? And then finally, if you do do debridement, what kind of technique is best? A lot of these questions we don't have really good answers for right now. And in terms of the question of, do you debride now versus just put the stent in, establish drainage, and bring them back later, as I said, when we first started doing this, we used to do a lot of debridement at the first session. Things have really kind of changed a little bit. And when you look at the studies, what they have found is that complications tend to be a little bit higher if you're doing the debridement from the get-go, which you might imagine, right? Because you're going out into the cavity, trying to remove, break down the necrotic pancreas tissue. You see there the various complications that can happen, perforation, fistula infection, air embolism. You don't want to use air, of course. You always want to be using CO2 when you're doing any kind of therapeutic procedure. And mortality as well. Now, on the other hand, studies have suggested that if you do do initial debridement at the time of establishing drainage, potentially there's fewer necrosectomy sessions that are required down the road in these patients. So I think you can consider doing initial debridement if somebody has a really large cavity or maybe there's over 40%, 50% with solid debris that we already discussed is one of the risk factors for failure of endoscopic approach. So perhaps in those patients, you might want to consider doing this from the get-go. But I'll tell you, we've really changed our practice where typically we just place the stent and then see how the patients do. And our practice at the Brigham, which I'm sure is very different from what a lot of people do across the country and the world, is we don't bring them back for scheduled necrosectomy. We just watch and see how the patients do clinically. And obviously, if they're not doing well, febrile, or ongoing pain, then we'll image them and bring them back to do the necrosectomy. Now, if you're going to do necrosectomy, as I said, CO2 is a must. Therapeutic gastroscopy is usually what's used. We usually give them prophylactic antibiotics as well. And in the beginning, we had snares, baskets. Now more recently, there's this powered endoscopic debridement. And I'll show a video of this tomorrow afternoon. But this is showing you just this is a not a great picture, but this is the endorotor device where basically it literally sucks in the dead tissue in there. And then those are some of the devices that we have, not a lot, mind you. And then we used to do antibiotic lavage, warmed bacitracin, which of course we can't get anymore. There's a lot of more recent literature about the use of hydrogen peroxide, which seems from all the literature that's been published so far seems to be safe and potentially efficacious. And if the patient's been on a PPI, we stop the PPI because we believe that the acid may help to encourage kind of chewing up the dead pancreas tissue there. So I know in our country, no one likes to have anything put down their nose. I mean, people hate NG tubes and whatnot. But elsewhere, they do seem to do this. And a lot of the studies, actually all the studies from the Dutch pancreatitis folks, in their protocol, they actually use nasocystic irrigation in their studies, which is why I'm putting this in here. And it's unclear if putting this down, down the patient's nose into the cyst cavity and doing lavage is superior to other necrosectomy techniques. And they infuse about 500 to 1,000 cc's of normal saline. So now, how long should you leave the stents in? And I already mentioned how with the lumen-imposing metal stents, you don't want to leave them in too long because of the fear of the bleeding complication. And so typically, you want to get repeat imaging in about three or four weeks to see what the status of the collection is. And then if there's still somewhat of a collection left, then bring them back to take the lumen-imposing metal stent out and replace it with double pigtail stents. As I said, for plastic stents, you can leave them in until the collection is resolved. There's no time limit on plastic stents. And you can leave them in forever in somebody who has disconnected pancreatic duct syndrome. The before stent removal, you can consider imaging the pancreatic duct. The ESGE guideline does suggest using secretin-stimulated MRCP because, again, they're looking to see if a patient might have disconnected duct syndrome. I have to admit, we don't routinely do this, but something that you can think about. So back to our patient who had just drainage. And then about four weeks later, this is what his CAT scan looks like with it gone now. So what if the initial drainage fails? Let's say the repeat CAT scan did not look so great, and the cavity's still there. It's maybe half as large, but still there. Then there's a few different options. So if you did not do necrosectomy the first time around, you can certainly bring them back and do necrosectomy to help encourage the cavity to continue to resolve. And another thing that you can do is if the cavity is really large, over 12 centimeters in size, and if you had just placed one lumen-imposing middle stent or whatever stent that you've placed there, you can do the so-called multiple gateway technique that Cheyenne Varadarajulu published on a while ago, where literally you're basically just putting multiple stents along this huge cavity to encourage drainage, as opposed to just the one. And then we already talked about how if there's pericolic gutter extension, that's a predictor for failure of the endoscopic technique. So in that kind of situation, you might want to consider getting your radiology colleagues involved to get a percutaneous drain down in there. So now in terms of this is from the Dutch pancreatitis group, where they compared so-called endoscopic step-up approach to the surgical step-up approach. And again, in terms of the endoscopic approach, they did all use nasocystic drain, and they used double pigtail stents. They did not use lumen-imposing middle stents for this study. But what you see here is that they found that the rate of major complications or death were comparable for both the endoscopic and surgical group. What was favorable for the endoscopic group was that there were fewer cardiovascular complications as well as pancreatic fistula. The number of drainage procedures was lower, although one more necrosectomy was needed. The days between first drainage and first necrosectomy was lower. And then the days in the hospital within the six months of randomization was significantly lower in the endoscopic group. Now the Florida group did a similar randomized study. The thing that was different is that they didn't use nasocystic drainage, and for the first couple of years of the study, they used double pigtail stents. But then when the lumen-imposing middle stents came out, then they switched to using lumen-imposing middle stents in this study. And then they used that multiple gait procedure kind of at the endoscopist's discretion if they felt the collection was really large and they wanted to establish multiple drainage points. And what they found in their study was that the endoscopic group had significantly fewer complications than the surgical step-up group. Similar to the Dutch folks, less fistula forming, and also significantly—actually, their length of stay was not significantly different, although there was a trend towards decreased length of stay in the endoscopic group. And you see the mean total cost was significantly lower with the endoscopic group as well. So if the patients have a symptomatic and mature collection, endoscopic drainage with double pigtail stents or LAMs with or without necrosectomy really is the method of choice at this point. And if you're doing the EOS-guided approach, you want to use CO2, prophylactic antibiotics, and avoid PPIs. Consider initial necrosectomy if the collection is really large or filled with a lot of solid debris. And consider other hybrid approaches if there's pericolic gutter pelvic extension. And then you do want to repeat imaging, especially with the LAMs, at about the three- to four-week period to check on the status of the collection and then take the LAMs out at that point. If need be, you can change out to double pigtail stents. So quick word on disconnected pancreatic duct syndrome, which happens—there's not a lot of studies on this, and it can happen. You see a huge range of incidents there in acute necrotizing pancreatitis. And basically, you don't have a duct there. And so the upstream parenchyma, which is functioning, is just pumping out pancreatic juice and can, of course, lead to persistent fluid collections, fistula, ascites, et cetera. You can diagnose this with MRCP or secretin MRCP, as I mentioned earlier, CTEUS or ERCP. And typically, you see that a lot of times it happens in the neck, followed by the body and tail of the pancreas. Management can be tricky, right? So in these patients, antibiotics, nutrition is important. Octreotide doesn't help, although a lot of us wonder whether it will or it doesn't. You want to avoid percutaneous drain. Surgery is often needed, although endoscopically, there are some things that we could do. If there is, of course, a pancreatic fluid collection associated with this, we can certainly do EOS guided drainage and then just leave double pigtail stent in long term. There was a study looking at this and found that there was lower recurrence of the fluid collection if you did this approach, rather than just removing the lambs, which of course makes sense. Transpapillary stenting, really not a huge role for that unless there is no fluid collection or a small fluid collection. And then there are more complex techniques that you can try to use as well, but oftentimes these can be difficult. Now for a partial pancreatic disruption, I just did want to mention that the treatment of choice is to stent across the disruption. This is from a long time ago at the Brigham, David Carlock did this study. And then if you can't bridge a disruption, you can just leave a short stent across the ampulla if you want to do something, because you see there's a success rate for that compared to getting a stent right up to, but not bridging this disruption is pretty much similar. And this is just an example of a patient of mine where there was all this fluid draining out from the disruption, and then we got a stent across the disruption there. So just to conclude, if the patient, again, is symptomatic from their pancreatic fluid collection, you look at if it's a mature collection or not. If it's early and they need some kind of intervention, really you go to your radiology colleagues for a percutaneous drain at this point. If however, the collection is mature, then we can certainly do EOS guided drainage. You may want to consider doing a combination of EOS and percutaneous drainage if it's a large collection that's extending down into the pelvis. And you may want to consider the multiple gate approach, again, if it's a very large collection, rather than just one stent into the cavity. Now, if the collection is a pseudocyst or there's not a lot of solid debris in the walled off necrotic cavity, in terms of what stent to use, certainly reasonable to use double pigtail plastic stents, much cheaper, although procedure takes a little bit longer. But the success rate is comparable, at least based on the literature that we have so far, or you can use aluminum posing metal stent. I don't think you would need to do debridement, certainly not in the pseudocyst, but not even if there's just a little bit of solid debris. And then again, you want to repeat imaging in a few weeks to check on the status. If the collection is gone and you're not worried that they may have a disconnected pancreatic duct syndrome, then you can just bring the patient back and take your stent out. However, if the question is resolved, and for some reason you think the patient may have disconnected pancreatic duct syndrome, this is where you just want to leave the, if you initially placed double pigtail stents, you can just leave that in. However, if you place the lambs, you want to bring them back, take that out and put in double pigtail stents. Of course, if the collection is not resolved, this is when you would need to bring the patient back to do debridement with whichever technique you wish to use at this point. And if it's not resolved and you believe they have disconnected duct syndrome, again, do necrosectomy, but then leave the plastic stents in there indefinitely. So thank you very much. And I'm happy to take questions now, or we can just move on to Dr. DeWitt's talk, and then we can just take questions all together as well.

pancreatic fluid collections

endoscopic approach

pancreatitis

drainage

symptomatic

follow-up care