Let's talk about pancreatitis. Is there anyone that does not take care of pancreatitis in their ICU? Depending upon where you live, it either goes to the MICU or it goes to the SICU if the patient is sick enough. And it doesn't matter where they go. Doesn't matter who takes care of them, because the therapy is basically the same. Same disclosures. We'll talk about common presentations, diagnosis, how you decide how sick they are. We'll touch on nutrition support, antibiotic use, and then a lot about management. Well, you can't hide from this stuff regardless of which servers you're on, because people will have it for a wide variety of causes. When you look at mortality from acute pancreatitis, this has taken a sharp nosedive. So when you look back in 2008, mortality was almost 29%. Now it's much less. We're doing something right. And what we're doing right is principally not operating. I know, hard to hear that from a surgeon. Diagnosis by history, habits, medication. Were they recently involved in a motor vehicle crash with a chance fracture? What is their past history? All kinds of physical examination clues. And if you have two out of the three, and the third could be amylase or lipase, there's more than three times normal or consistent imaging. You have acute pancreatitis. So you don't have to have all of them all at once. Which the following suggests severe acute pancreatitis? Amylase of 5,200, Apache 2 of 7 with a CRP of 100, CT scan imaged pancreatic necrosis of 25%, a tympanic membrane temperature of 40, or blood gas that has a PO2 of 58 on 50% FiO2. Who wants to go for the amylase? No, no one likes the amylase? How about the Apache? CAT scan. Couple of takers for CAT scan. Everyone, yep, you're going to go with that one? Ear temp of 40, hypoxemia. Well, two of them. You thought it was just going to be one, didn't you? I have to throw you a curveball. And so there are many different things that will indicate it, but you don't have to have anything in particular other than what has been highlighted in bold. CRP will have some thresholds. We'll talk about that shortly. But let's talk about why you get it. Many causes, from alcohol through hyperlipidemia, toxins, gallstones, far and away out of these common triggers which should be part of the history that you select, it is gallstones and alcohol, 65% to 75% of the time. The rest occur much less frequently, including autoimmune toxins, pancreas devisum. I've seen two in my professional career, and rarely ischemia. Criteria and severity assessments. Everyone should be familiar with the Ransom criteria. Broken down admission versus the initial 48 hours. And the more criteria you had, the sicker you were. And if you had above a certain threshold, which was determined by your particular ICU, because everyone adapted the Ransom criteria to their practice, that would get you into the ICU. Important pieces. If you're over the age of 55, you have a higher risk. And therefore, you tolerate it less well. I'm 62, so I'm at risk. You can see very different criteria. Changing your chromatic rate, because you require a lot of fluid resuscitation. The decrease in calcium based upon pancreatic saponification of surrounding tissues. And old criteria, like a base deficit more than 4, which we now know you can make a base deficit that you'd like, depending upon how much chloride you give them and how much of a robust hyperchloric metabolic acidosis you would like to engender. But one of the durable things, you need more than six liters of fluid to resuscitate you. You're generally pretty sick with that. In 2012, for those of you that have been to Atlanta, there's the revised Atlanta classification, mild, moderate, and severe. Mild is just what you would think. No organ failure, no complications. Moderate, less than 48 hours of organ failure, so it rapidly resolves. Whether there are or not complications, and severe is everything else. This has stood the test of time over the last decade of re-evaluation. CRP, CRP that you had is of 100 in the last question. There are some data. When you look at those who have mild, moderate, and severe, these are three different groups in terms of their C-reactive protein level. And so that you can take a look at either the absolute level or the change in CRP from admission over the next 72 hours. And 48 hours seems to be a really good point. If your delta in CRP is 90, so if it goes up by 90, or the absolute at 48 hours is 150, or at any point in time you're over 190, that's very consistent with the inflammation that accompanies acute pancreatitis. And if you're above 190, that's generally severe pancreatitis. So that's a useful additional laboratory marker for you. In particular, if you want to say that person should come, or I'm sorry, the CRP is only 110, we're not accepting that patient, they can go to the floor. Imaging is helpful for complication recognition. And so if they have mild pancreatitis, you could get an ultrasound just as a baseline to see what their pancreas looks like. For moderate and severe, a contrast CT is indicated if you're not certain that that's really the diagnosis, or they're not getting better by 72 hours. And repeat CT, of course, is failure after they improve, and you're not quite sure why, or they've had some intervention. Question two, 56-year-old presents with new onset pancreatitis. Some selected labs, the admiral ones, are hypoxemia on room error, elevated white count, hyperglycemia, and a calcium of 7. What should we do with them? Admit them, get a contrast CT, give them some oxygen, control their glucose, and start them on imipenem. Or we can do those things, but not start them on imipenem. Or we can do ICU admission, oxygen plasma volume expansion and glucose control, or all those with some additional antibiotics. One, who wants one? Two, two going once, one taker, three. OK, lots of people for three. Anybody want four? No, three is, of course, the right answer, because this person needs care. And in fact, they need critical care. Generally, we're not going to put the hypoxemic patient with hypocalcemia and hyperglycemia on the floor. 14 different randomized controlled trials, almost 1,000 patients, increased multidrug-resistant organisms with a substantial increase in C. difficile colitis. Some of those required emergency colectomy. And no reduction in any metric of infection or pancreatic necrosis. And it didn't help with mortality or the need for subsequent intervention. There are terms for you to know. And you'll see them listed here. I will call your particular attention to necrotizing pancreatitis. Some of your pancreas is dead. There's an acute peripancreatic fluid collection. There's fluid plus the pancreas looks edematous. No well-defined wall. It's less than four weeks since it started. Pseudocyst is a collection that's there for more than four weeks. And it has a well-defined wall. Acute necrosis is necrotizing pancreatitis plus fluid and necrotic solid material. Not there for more than four weeks with no well-defined wall. And walled-off necrosis. It has that necrosis more than four weeks since the onset. And there's a wall. Sounds confusing. Every consultant that you have will know these. Because it drives what decisions they make and what kinds of care they deliver. This is a pancreatic pseudocyst. Note radiologist arrow sign. Importantly, it has no gas in it. That's not an infected collection. But you can see there's a wall there. It's a nice rim. This is dead pancreas. Normal pancreatic architecture is gone. It looks like a big fluid collection and maybe a shell of a pancreas is present. Necrotizing pancreatitis. Now you can see that there's necrosis adjacent to some normal pancreas. And there's a very well-defined wall. And that's infected necrosis. It has gas in it. Note that it is immediately inferior to the posterior wall of the stomach. This will become important. Because it will help drive management. Overarching issues. Figure out who's sick. Do they have concomitant organ injury besides the pancreas? They need monitoring for hemodynamics because they often have a huge fluid sequestration. And they are very high risk, in particular, of intraabdominal hypertension. Not only from secondary ascites, but also from organ edema. Therefore, this place is at high risk for acute respiratory failure, as well as acute kidney injury. It's helpful to support nutrition for those that have severe pancreatitis. And this will be augmented by management of pain control in the context of plasma volume expansion and whatever therapeutic maneuvers you undertake. Let's talk briefly about intraabdominal hypertension. We see this a lot less than we used to. Because we use a lot less crystalloid. If you look at the influence of acute pancreatitis and intraabdominal hypertension in black, the upper bars, these are patients that have intraabdominal hypertension. And note that the time frame is very short. It's within the first 12 hours that that develops. So that when you look at what happens with those patients, mortality is high. It's 80% at 72 hours if you don't relieve it. Seven of those patients in this group had the abdominal compartment syndrome. Who's had a patient in their medical ICU opened at the bedside for secondary compartment syndrome? So much fun for me. Very messy for them, right? And did you think that it happened early enough? Or do you think this was a late finding and it was a Hail Mary event? Hail Mary event, right? So this means you should monitor specifically those that have severe acute pancreatitis for the development of intraabdominal hypertension. So that if you find it early, you can undertake steps to mitigate it. Infection and organ failure also have influences where infection drives mortality. This should surprise no one in this group. When you look at pancreatic collections that are sterile versus infected, you can see that mortality more than doubles. Your odds ratio for death is 2.47, and odds ratio for organ failure is almost threefold higher. So concomitant infection is a problem. Necrotizing pancreatitis and organ failure, what you're looking at in blue is overall, in orange, the SCCM color is sterile, and in gray is infected. And so when you look at the extent of necrosis, it almost doesn't matter. When you look at the gray bar, because whether it's 30%, 30% to 50%, or more than 50% of necrotic pancreas, the percent organ failure is almost the same when there is infection. If you don't have infection, the extent of necrosis stepwise increases with the more pancreas that is involved. But if it's infected, it doesn't matter. Nutrition, in a patient with severe acute pancreatitis and moderate protein calorie malnutrition. Somebody got in nutrition labs whether they should or should not. Which of the following statements is most accurate regarding support in the ICU? Avoid gastric feeding to prevent triggering pancreatic exocrine work. TPN should only be provided using formulas without intralipid. Voluminal nutrition support formulation does not impact outcomes. And elemental formulations are optimal during the acute episode. Who wants the first one? Second one? Third one? Sanjay likes the third one. Fourth? Yeah. Why is that the right answer? Think about what you do. You're doing it right now. You are feeding your GI luminal cells, all the cells that migrate from the villus crypt to the villus tip. And what they do is provide mucosal barrier protection, but also your intestinal glycocalyx, your luminal brush border enzyme system. Most of these patients are acutely ill, severe acute pancreatitis. They are NPO. They get resuscitated. They have abnormalities of flow. And the one thing they do is in every 72 hours, and you do this also, you turn over the entire lining of your GI tract. Where do you work? Medical Institute. Which institution? I'm in Indiana, in a small community hospital. OK. At your small community hospital, you're going to run an experiment. And you're going to collect stool samples from all of the nurses that work in your ICU. And then we're going to desiccate all of the stool samples. Your hospital lab is going to love this. And then it's going to parse the dry content of their stool into its component parts. What will be the greatest component of their dry weight of stool? I guess the GI tract? Yeah, the shed enterocytes. Exactly so. And so when you shed all of those enterocytes, you can't process complex luminal nutritional support formulae. But you can directly absorb an elemental diet with simple sugars, medium-chain triglycerides, and single amino acids. As a bonus question, in what kind of pet food will you find medium-chain triglycerides? Cat food, actually. They do it routinely. Not for dogs. Go figure. So nutrition, you have to throw these things in. Mild pancreatitis, rapid resolution, doesn't matter what you do. Just feed them the regular diet. Severe or prolonged problems with your pancreas that can take four to six weeks to resolve. The decision is really only one. If they can tolerate enteral feeds, you should use their gut. All for these kinds of reasons. It was amazing that they got people to volunteer their stool samples for this. In every way, shape, and form, enteral outperformance per enteral, mortality, infection, organ failure, the need for intervention, the relative risk is less if they can tolerate enteral nutritional support. Doesn't matter which study you look at, in which country, in which patient population, for moderate to severe pancreatitis, enteral outperforms parenteral. What do we do when we intervene? So we always want to know, should I get an ERCP for this person because of the high incidence of biliary lethiasis? If they have gallstone pancreatitis and they have obstruction or cholangitis, they deserve an ERCP. 24 to 48 hours, that's the benchmark. It resolves cholangitis, it decreases the amount of inflammation, and it absolutely improves survival. When do you take out their gallbladder? Sometime before discharge when things look better. They do not need to be normal. Or if they're not suitable right then, within two to four weeks after discharge because of the very high incidence of recurrent gallstone pancreatitis. Now there are some that are absolutely medically unfit. You know these patients. They have heart failure with reduced ejection fraction. They have multi-morbidity. Maybe the only thing they get is an ERCP and a sphincterotomy so that they can then pass their stones. But they don't get a cholecystectomy. Should you get an ERCP if they don't have cholangitis but a high probability of stones? It is debated. It will depend upon your place. An MRCP can rule in or out ductal stones. And it is certainly less invasive than an ERCP. And that is the predominant pathway in most institutions. So gallstone pancreatitis, the high-risk patient, the person who needs an ERCP and sphincterotomy leads to a 0% to 8% 30-day recurrence. If you don't do this, recurrence can be up to 40%. So there is clearly a benefit from ERCP and sphincterotomy in a variety of patient populations. Let's talk about the pancreas and its ducts. Intervention indications. It's infected. It has its own complications. I'm not quite sure what I'm dealing with. Or I thought I managed the biliary ductal system, but I didn't really do that very well. You can either obtain a sample with a fine needle aspirate. You can endoscopically biopsy or drain things. You can have IR debrided. And the last thing you should do is take the patient to the OR. You will see why. Infected necrosis. You saw an image of that before. They get antibiotics. They get CT-guided drainage or endoscopic transenteric drainage because the posterior wall of the stomach puts you right in that space. Or what is known as the step-up procedure. And we'll talk about that. The last thing is a laparotomy or a laparoscopy to do this. When you look at a surgical step-up approach versus an endoscopic step-up approach where there are serial debridements using different access pathways, there are pros and cons for each. And most commonly, most surgeons will say, that's not me. You should do this endoscopically because the outcome after surgery is less than ideal and extends everything that you don't want it to extend. In particular, the surgical step-up has a very high risk of pancreatic fistula. Most collections can be accessed endoscopically. Surgical step-up gets you percutaneous drainage to a minimally invasive necrosectomy to an open necrosectomy, which you should avoid. Open necrosectomy looks like this. This is a case I did. Stomach is up. Transverse colon is down. And that's the dead pancreas. It looks like there's just dead rubberized tissue. You do this every day until you don't have any more dead pancreas. Or you take out something that's not pancreas, and it's really a major blood vessel. That gets very exciting. So all the tissue planes are unclear. You cannot easily identify things. We try not to do this. When you look at fistulas, drainage to necrosectomy time frames, days in hospital, endoscopically outperforms the surgical approach in every single way. I would go with the 5% risk of fistula rather than the 1 in 3 risk. Laparoscopic transgastric necrosectomy, so that you don't have to dig the stomach off of the pancreas, has been described. So this is a minimally invasive approach that goes through back wall of the stomach into that collection. And you can see that identified here. And you can do that to Bredemont without having to open up that space. This is much less common than doing it endoscopically. And what you're looking at now is the inside of the stomach. These are your gastric folds. And this is a stent that's placed through the wall of the stomach into the collection. And what happens through that are these little probes that are advanced endoscopically. And you can see that they have all of this wire mesh here in a ball. And those are twirled around in there and rubbed back and forth. And the dead tissue gets caught in all these interstices. And so this is a lovely endoscopic step-up approach. IR can do this also, coming through the retroperitoneum. But this is far away favored. And so in conclusion, there are imperatives from severity assessment to when people should have ERCP, when should they have surgery. Never. Enteral is always better than parenteral. No antibiotic prophylaxis. And endoscopic is much better than, did I say it, don't operate on these patients. Questions? Yes, Sammy. Can you explain special therapy again? It goes from, I don't do a whole lot. Maybe I just drain you, to maybe I do a little more. Maybe I debride you using some non-open surgical technique to I finally get to an open surgical technique. That's the surgical step-up. Endoscopic, I've endoscopically placed a drain. Then I've endoscopically done a little bit of debridement. And then I've endoscopically done a lot of debridement. And so the extent to which you do these in terms of dealing with infection that you get rid of, dead tissue that you get rid of to decrease the metabolic response, and then finally clearing that space and allowing it to collapse in on itself is a sequential process. Days to weeks, as opposed to the patient you saw that I'd operate on, that was a three-month adventure. Yeah, I mean, and a lot was done at the bedside because I just couldn't go back and forth to the OR. It's hard to find OR time. Other questions? You're just stretching? Question. Question, OK. Is there a role for post-pyloric end-to-end feeding in? Oh, my NPs love that because they have an iris device and they can put it in particular places. Doesn't make a damn bit of difference. They love it and like it. Yes, they absolutely love it. It's a toy. Many of them thought they were going to be guests for a neurologist. And they could be the person at the other end of the scope. Now we've given them the scope. Absolutely no outcome difference, no difference in aspiration rates. You'll always find these small studies that say, well, we feed our patients post-pylorically. And here's our aspiration rate compared to historic controls. We're so much better. But not a head-to-head comparison. And the patients are never really the same. So if their stomach works, great. They have diabetic gastroparesis, please feed them post-pylorically because they won't empty. Your stomach is compressed by the gigantic pseudocyst and you can't fill them with enough food? Sure, put it post-pylorically. But for the average patient, it doesn't result in an outcome benefit other than the ability to bill for your iris catheter placement. Other questions, comments, concerns? Great. Now you're done with me.

pancreatitis

ICU management

revised Atlanta classification

nutritional support

endoscopic interventions

ERCP