I'd like to start our first session by welcoming my three speakers to the stage. This is the session on pancreatic oviliary diseases. My first speaker is Dr. Brenna Casey, who, Dr. Casey completed her training in interventional endoscopy at Brigham and Women's Hospital and the Massachusetts General Hospital. She's currently director of interventional endoscopy at Mass General, and she's an assistant professor of medicine at the Harvard Medical School. Good morning to everybody, and thank you to the course directors for having me today. It's nice to be with you. We're going to talk about the management of early pancreatic fluid collections, and we're going to do it first thing in the morning, so hopefully I can keep you awake. I have no disclosures. So how are pancreatic fluid collections defined? The revised acute pancreatitis classification working group met in 2012 and came up with criteria for us for contrast-enhanced CT. So we're going to start with acute peripancreatic fluid collections, because that's what we see first. This is actually one of my patients, an unfortunate 35-year-old woman who was postpartum by about six months who got in trouble with some gallstones. You can see here that we have diffuse fluid collecting. It is defined by the normal fascial planes. You can see that we still have preservation of actually the ability to see the pancreatic gland here, and that it's very homogeneous throughout. Later on, a homogeneous collection, as it matures, can develop into a pseudocyst. These are well-circumscribed, round, very fluid-containing without solid components, and usually develop about four weeks or more. Acute necrotic collections, same patient I showed you earlier, she progressed into acute necrosis, occur in the setting, again, of acute pancreatitis. It's more heterogeneous, and you can see both liquid and solid components within the collection. There is no defining wall or encapsulation in the collections at this time, but you can see that we no longer see the pancreas anymore, and the involvement is both intrapancreatic now and extrapancreatic. Well-doff necrosis usually occurs around four weeks, sometimes a little earlier, sometimes a little bit later, but now we have a heterogeneous collection, but you can see that it is now defined and walled off, again, our same patient. So knowing the definitions now, who is the monster that's under the bed? What's keeping us up late at night? The monster is infected necrosis. So about 20 to 30% of our patients with acute pancreatitis will go on to develop necrosis, and about 30% of those patients will go on to develop infected necrosis. Infected necrosis can be defined in under two weeks from a positive gram standard culture, which is obtained from fine needle aspiration, which should be done percutaneously, either under CT guidance or ultrasound guidance. You should not be doing these with EOS guidance, because they'll be contaminated specimens, and you are contaminating a potentially sterile field. Presence of gas in the pancreatic bed and the pancreatic necrosis is also an indication of infected necrosis. More than two weeks after the onset of pancreatitis, you can look for persistent organ failure that's lasted more than 48 hours in patients that are in the ICU. If they're doing well enough to be on a regular ward, look for two inflammatory variables during three consecutive days, so elevated temperature, their febrile, they have an elevated CRP level, or an elevated white blood cell count. Now we know who the monster is. How do we get rid of the monster? Be nice to have a can of monster spray, but unfortunately, we didn't have that. So to learn how we manage early pancreatic fluid collections now and manage infected necrosis, which is our one indication to possibly intervene, we have to see what our experience was later on. The first percutaneous drain was placed for infected necrosis in 1998, and the first transgastric endoscopic drainage was done in 1996. So from the early 90s, what did we learn in the last 24 years, last two decades? Well, we have the multiple transluminal gateway technique study that was described for drainage of symptomatic Waldorf necrosis, and we have to start later in the process to learn how to deal with it earlier in the process. And this was 60 patients in a retrospective study in a single-center tertiary care center. This was UAB's group, and there were multiple transmural tracts created, and they were compared to the single classic drainage group, which is just one tract with a stent. Both had transcystic nasal stents in place to irrigate the cavities. And what they found was more holes are better, so more is better than one. Then there was the dual-modality drainage technique of infected and symptomatic Waldorf necrosis, and they were able to extend this into long-term clinical outcomes as well. And this was, again, a single center, but it was a prospective study, prospective database that they then reviewed. They had a median follow-up of 750 days for 103 patients, so a good population. And surgical necrosectomy was avoided in all 103 patients who completed the treatment. No pancreatic cutaneous fistulas developed, and there were no procedure-related deaths. About the same time, the Dutch pancreatitis group was starting the PANTR trial. Pancreatitis, necrosectomy versus a step-up approach. Again, everything in the beginning was surgery, and we're moving to a less invasive approach of things, and how can we keep these patients alive and give them a better quality of life. So this is our first randomized control trial in which patients with infected pancreatic fluid collections were either randomized to an open necrosectomy or a step-up approach. The step-up arm was managed with minimally invasive techniques like percutaneous or endoscopic transgastric drainage with escalation and minimally invasive retroperitoneal necrosectomy if there was no clinical improvement. The primary outcomes were major complications and mortality. And the result demonstrated that the step-up approach resulted in significantly fewer major complications and decreased healthcare utilization, but there was no difference in mortality. Then we had the MISER trial, again, that came out of the group that moved from UAB to Florida. An endoscopic transluminal approach compared with minimally invasive surgery reduces complications and costs for patients with necrotizing pancreatitis. Again, what we're looking at here is moving away from surgery, and this actually compared endoscopic necrosectomy with either a barred approach or laparoscopic minimally invasive surgery. There was no difference in mortality, and there was no difference in official development. But the mean number of major complications was lower in the endoscopic group. And the physical health scores for the quality of life was better in the endoscopic group, as well as lower, the cost was lower for the endoscopic group overall than the surgical group. The Dutch pancreatitis group followed the PANTER trial with the TENSION trial, which looked at a transluminal endoscopic step-up or minimally invasive surgical step-up for infected necrosis. And it was, again, a multicenter randomized trial. They stepped up a percutaneous drain versus an endoscopic drain. There were two French double pigtails placed along with a nasal cystic catheter for the endoscopic group. The study found no significant difference in the mortality, but did show a decreased rate of pancreatic fistula that was significant and a reduced hospital length of stay. Then they extended their follow-up from the initial six months of the TENSION trial to seven years. And what they found was that overall there were fewer pancreatic cutaneous fistulas in the endoscopic step-up group and fewer reinventions after the initial six-month follow-up. So now we need to learn how to pick our battles wisely. We've explored what was studied in the later groups. So most of these studies came in patients that had pancreatic fluid collections greater than four weeks. Those are not early pancreatic fluid collections. Those are considered later or postponed. So now we have the POINTER trial, which helps us pick our battles today. Postponed or immediate drainage of necrotizing pancreatitis. It's important to note that in the immediate drainage group, 93% of those patients underwent drainage within 24 hours of randomization. The postponed drainage group was not intervened upon until they had developed walled-off necrosis. So that would be four weeks approximately or greater. The primary endpoint was the complications index. Interestingly, there's no difference in the mortality between doing it now, which we always felt pressured to do. The patient's got pancreatic necrosis. The patient's got fever. The patient's in the ICU. We have to intervene. But there was no difference, significant difference, in the mortality between the two groups. In fact, it was a little tiny bit higher in the immediate drainage group, the do-something-now group. There were no differences found in the incidence of major complications, nuanced organ failure, bleeding, perforation of a visceral organ, fistula development, or wound infection. The mean length of stay was a little bit longer for the immediate drainage group than it was for the postponed drainage group. And the length of stay in the ICU did not differ. The mean number of surgical, endoscopic, and radiologic interventions, both catheter drainage and necrosectomy, were higher in the immediate drainage group than they were in the postponed drainage group. Feels backwards. In the postponed drainage group, 39% of the patients were treated conservatively with antibiotics alone, without any need for drainage or necrosectomy, and 17 of these patients survived. Necrosectomy occurred in 28 or 51% of the immediate drainage group and 22% of the postponed drainage group. Again, it feels like it's a little bit backwards than what we always thought it should be. At the six-month follow-up, there were no differences between the groups. So if you intervene now, which is what we always felt like we had to do in the past versus later, the patients had no increased development of endocrine or exocrine pancreatic insufficiency, and their total hospital costs were comparable. So they followed their patients up long-term as well. To the primary outcome, they added, in addition to the CCI index, death, because CCI was really developed primarily to look at complications in the short-term interval. What they found was that the median number of interventions was higher in the immediate drainage group overall than the postponed drainage group. And at the end of follow-up, there was no difference in the quality of life for the pancreatic function, which leads us to believe that, really, if you can manage to do it, you ought to be waiting until later to intervene on these patients, not pushing to do it right now. So what do we take away from the studies that we've had in how to manage early pancreatic fluid collections? One, recognize the monster under the bed, and an early pancreatic fluid collection that needs to be dealt with is infected necrosis. Two, recognize how best to deal with the monster, and monster spray could come in the form of antibiotics. And perhaps you don't need to do anything else. Just because we can do it doesn't mean that we should do it. And three, choose your battles wisely. And unless you have clear clinical deterioration, try to avoid early interventions in these patients because overall they do better. So I wanted to leave you, in your slide set you will see when you can access it online, there are some appendices to refer you to the ASGE Guidelines on Treatment of Inflammatory Pancreatic Fluid Collections, as well as the AHA Guidelines. And then there's a very nice paper from the Acute Pancreatitis Task Force on Quality, referring you to quality indicators that you can look back at to make sure that you're doing a good job, which is very important. Thank you very much.

pancreatic fluid collections

early management

pseudocysts

necrotic collections

intervention