Without further ado, Dr. Mishra, Professor of Medicine at Wake Forest, and also Chief of the Division of Digestive Disease and Hepatology and Executive Director of the Digestive Disease Service Line will enlighten us on probably the most sought after topic in all of the U.S. Pancreatic cysts, everything you need to know. Welcome Girish, and such a great pleasure and honor to have you, one of the true masters in endosynagraphy, and everybody I know is looking forward to this talk. Thank you, Vivek, that is an amazing introduction, thank you to the ASG and to Ed for asking me to speak. It's really humbling to be part of this amazing group of speakers and a group of friends and folks that I can always learn from and always do. So we'll get started because I think I've got a lot of slides to cover here, all right. These are my disclosures, and Vivek asked us to do a top 10, and I think these aren't earth shattering, but this is my perspective having been doing this for about 23, 24 years. You know, it always goes back to our, you know, what we're taught to be great clinicians is, you know, get a detailed history when a patient is presented to you with a pancreatic cyst, you know, the age matters, the presentation matters. So I'll show you, and you know this, a 25-year-old woman that presents with an enlarged cyst is probably unlikely going to have a mucinous lesion, and if she has never drank, it's probably not a pseudocyst. So those things are important. You know, I always look at every single slide that I make by myself because we do have rows, and I've gotten pretty good at looking at slides, same thing with imaging. I think all of us are used to looking at our own CT scans, and the reason that's important is you'll pick up things that the radiologist misses. The other sort of corollary is that it's great to have a report on the cross-sectional imaging, but always use your EUS to help guide you, right? So the CT scan may say this appears to be a cirrhosis diagnoma or a mucinous cyst or a side branch, but once you do the EUS, that is very powerful, and so don't bias your thought and results just based on what the CT scan shows. That's a piece of the puzzle, but at the end of the day, the EUS is going to be the most powerful tool that you have in addition to all those other things. Before I jump to the risk of FNA, Dr. Call has already shown to the group that you should biopsy the solid portion, and I think now I've gone to doing an FNB in the solid portion. I've found that to be very, very useful and very diagnostic when you have a cystic lesion. So if you have a lot of cystic fluid and no solid components, I go in with an FNA needle, but if I have some solid components, I'll go with the FNB in the solid part. The next point is that, especially in terms of surveillance, if I have a cyst, and it's a large cyst, but a low CEA, if I follow that, and we'll go over some surveillance, I may not do an FNA every time, or I may get a serial FNA with CEA level, and if that's not increasing, that in itself is quite powerful. I think the risk of pancreatitis is quite low, but all of us who do EUSFNA of pancreatic cysts can remember probably one case a year, and they can be bad, and it's not related to the needle size, it's not related to necessarily the cyst characteristics, although there's some sense that maybe pseudocysts don't give you the pancreatitis that amuses a cyst lesion, but just because you have a cyst and you're following it up, it's okay to just observe and look at the cyst characteristics and the size, and not necessarily FNA. So that's the next point, is if you have a low CEA fluid in the cyst fluid, you may not have to FNA. For right now, we do recommend antibiotics prior to FNA of a cystic lesion, but perhaps this guideline may change as their data is somewhat conflicting on whether you need to, or the value of antibiotics for cysts. The only time that I'll FNA a cyst that's less than a centimeter is if I think it's a cystic islet cell lesion, otherwise you're probably not going to get enough fluid to get a meaningful analysis. And we've always been taught when you're aspirating a cyst, try to suck it dry for many reasons, and if you don't have enough fluid and you have to choose one test, I'll usually send off for a CEA because, or a glucose now that we're learning based on more recent data. So those are sort of my top 10 approaches. You know, this is what we are confronted with all the time. So you have a 46-year-old patient who was seen by their PCP for hematuria, and then they get a CT scan, and lo and behold, you're told that there's a low-density lesion in the head of the pancreas. This may represent either a pseudocyst or a cystic tumor and a follow-up MRI or MRCP. So this is the epidemiology. We know this. We have an aging population, better cross-sectional imaging, and so we discover more cysts, okay? So the question is, how many of these cysts are pseudocysts versus what percent are cystic neoplasms? So if you look at the studies, the true prevalence is not really clear. You can go to some series that show it's really kind of low, 2.4, whereas other series where they say it's as high as almost 50%. However, most asymptomatic cysts are still pretty small. Very small fraction of cysts are going to be larger than two centimeters. So we're seeing increased number of cysts. So sort of how do you, in your mental thought process, how do you compartmentalize an asymptomatic cyst? So this schema, which all of you have seen, really just kind of highlights trying to differentiate mucinous from non-mucinous. I think that's an important distinction because especially early on in your career, you get so bogged down with, is this a malignant cyst or not? Really, none of the tests, outside of doing a biopsy of a solid portion of the pancreatic cyst, are really looking at malignancy. They're more looking at whether a cyst is mucinous versus non-mucinous. And so you can see in this schema that as we go from left to right, pseudocysts are very, very benign. You'll also see these other cysts that we don't talk much about, but you'll see it in your practice, the lymphoepithelial cysts. Those tend to be quite small and, in my experience, much more anterior portion in the pancreas where you'll get these retention cysts. And then you can see pseudocyst adenomas further on down to frank ductal adenocarcinoma or neuroendocrine tumors in terms of the malignant risk. So this is a patient that was sent to us, a 71-year-old woman with a 3.3-centimeter cyst, and they said an enhancing nodule on MRCP. So when we see the term nodule, we're already thinking that this is going to be a pretty aggressive type of lesion. The size is large. The patient's a little bit older. So mentally, you're thinking that this patient's going to have some sort of aggressive type of cyst. However, here's a corresponding image. So I think, unfortunately, we can't go back to, you know, raising and getting a poll from the audience in terms of what you think, but I can tell you that almost all of you who've done EUS can walk in the room and say that emphatically, this is a serious cyst adenoma. So that's the power of EUS without FNA. So rarely can you use EUS without an FNA to just give you an answer, and I've just shown you a case. So the other sort of histologic or more the didactic components of these cystic lesions, you all know, you can read, you can look up in the textbook. So I'm just going to kind of gloss through this, sort of give you more of a bigger perspective. But these are serious cyst adenomas, these classic honeycombs, and you can read these. These have very, very little malignant potential, 25 reported cases in the literature. This, however, is a different type of cyst. So this is a patient with a large cystic lesion. Here's a corresponding EUS. This patient was in her 40s. This was in the tail of the pancreas. What I'm showing here is dopplering just to make sure, you know, it's always good to doppler the needle tract. I mean, I think I'm guilty of this after having done this for so many years. You can kind of feel like you may not have to doppler, but it's important to doppler for safety reasons. You also need to document it in your report for medical, legal, and billing purposes as well. So what we did here is this device is not available to us anymore. It was a brush, but there are other devices which I'll go over, and Vivek has already spoken to us all about this. So this was a mucinous cyst adenoma. These are quite unilocular, ovarian stroma, more common in women, greater in the body and tail than the head. And there is a risk of malignancy in these patients over time. I think these are lesions that, you know, you can get the classic sort of fish mouth appearance. You can almost see the mucin extruding from the pancreatic duct. This is a case of a cystic IPMN of the main duct variety, which carries a pretty high risk of malignancy. In these patients, you pretty much need to get your surgeon involved. We'll go over the branch duct, which has a lower prevalence. We used to think it was zero, but it's not zero. It's definitely not zero, but it's not as high as the main duct. Then you can have the mixed duct, and you want to treat those as a main duct IPMN in the mixed duct. So you have several different cysts or several different types of IPMN. And again, you can look at this in the textbook and the radiology, but just for completeness, this here is a main duct where you look at the duct, the main duct. Let me just use the pointer here and the laser pointer. And you can see here, this is the main duct and this large cystic lesion. So this is a main duct IPMN. This here is a branch, sorry, this is a multifocal IPMN. This is a branch duct IPMN, and this is a combination of branch duct and main duct. So these are fairly classic that you'll see in your practice. Here's a video of a patient that was referred for a cystic lesion. We're measuring the pancreatic duct here. And this is in the head of the pancreas. And you go in and out, and then you're seeing a couple of important features. If you look, there's no solid component in this lesion here. So you look at the wall of the cyst to look for solid component. And then we measured this cyst. And this is a branch duct IPMN that's about almost three centimeters. Doesn't have any high-risk features, but we did perform an FNA on this lesion. So the other types of cysts, I mentioned that the age is going to be extremely important. The old term for this, or what I remember during fellowship was a solid cystic pseudopapillary, but they took the cystic term out. But here's a young woman that has a very large sort of almost solid, but you can see the cystic part right here. And so we FNA this lesion, and this was a solid cystic pseudopapillary tumor. So I'm trying to get rid of my laser pointer there. And you can read here that these are more common in young women. And you can see these nice papillary projections. Sometimes it's hard to differentiate these from a neuroendocrine tumor. And uniformly, these patients should undergo surgery. I can't imagine when they would not be sent for a surgery. So if you look at the summary of the different cystic neoplasms, I've kind of gone through this in fair rapidity here. I'll come back to the pseudocyst story a little bit. But we've talked about the cirrhosis adenomas, the mucinous, the IPMN, and the solid papillary neoplasms. And you can look at this in terms of which patient population is this more prevalent in. I don't want to take our time with that. So the next sort of aspect of a cyst. So now you have a nice sort of morphologic view of what a cyst looks like, is really the FNA. How are we going to use our FNA to help us? And Vivek has already gone over very nicely the FNA and how we try to FNA this. The only other thing here, this is a moray or microforceps biopsy that is on the market that's used with some degree of success. And we can talk about sort of the use of it in our panel discussion. But really, it's the linear EUS with FNA that's going to help us in terms of trying to get a diagnostic yield. Now, just a word about features. Now, we talk about how EUS features by themselves are not that helpful. You need to perform an FNA. However, these are slides from Dr. John DeWitt. And I think early on, I didn't give credit to Dr. Sri Kamanduri. He and I did a Meet the Professor Luncheon together. Some of the slides that I've taken are from Dr. Kamanduri at Northwestern. So credit to him. These slides are from Dr. John DeWitt at IU. And this is a really neat characteristic that just by looking at the EUS, you can make a distinct distinction between whether a lesion has a mucinous component or an intramural nodule. And here, this very nicely marked apacean at Mayo showed that if you have a true lesion in the wall or a solid component, these tend to be hyper-echoic when you compare this to the adjacent soft tissue, very irregular edge. However, if you see this, I mean, this is like a beautiful, well-rounded lesion. This is hypo-echoic compared to the adjacent soft tissue, smooth edge with a hyper-echoic rim. So this would be a intracystic mucin, very different than a solid component in a cystic lesion. So just some little tidbits here, just based on the appearance itself. So really, you need to perform an FNA. And you can see here, we're performing an FNA of this component in the sidewall. And this here is a nice depiction of that mucin component within the cyst itself. So what do we do? We get a cyst fluid analysis, send it off for the usual markers. And I'll talk about this. I think this is quite safe. The risk of pancreatitis is low. Often, I'll quote, one to less than 1%, but it's not zero. And then the risk of infection. And right now, we are using antibiotics in these lesions. So the classic landmark study was performed by Dr. Bill Brugge at Mass General. And this is an ROC curve analysis. And basically, what they looked at is, it's a sensitivity specificity analysis. So at what cutoff do you gain the sensitivity? Again, it's looking for mucinous and not malignant. So the 192 is a cutoff. Now, nothing in science is black and white. And certainly, nothing, well, maybe in the US, things are black and white. But what we need is that you've got to use it not in a vacuum. You know, a cyst fluid of 194 doesn't mean that it's something that's worrisome. But I look at big extremes. Is it 5 or is it 5,000? That's where I think the CEA is helpful. And there's emerging data that glucose in the cyst fluid is useful. So how do we put all this together? You can look at cyst fluid CEA levels. Cytology, as we know, is not very sensitive for diagnosing. And then there's DNA. Asif Khaled from University of Pittsburgh did the sort of very elegant Panda studies to try to see if the sort of DNA molecular markers can help us distinguish a mucinous versus non-mucinous. And so that's still being used. You can look at cyst fluid color to help you. I mean, I think the classic, once you puncture a cyst and it comes out sort of this yellowish brown color, you know it's a pseudocyst. Or if the viscosity is really high, we puncture the cyst and it's so thick that it won't even suck up. That's probably a mucinous lesion. So those things are quite helpful. So all these different characteristics, this is adapted from Asif's study, as well as Bill Groogie, who really was a pioneer in helping us distinguish and try to study pancreatic cysts better. But these are the different features. We look at the cyst fluid CEA levels. We look at, you know, really you look at everything in a package, in the complete package. The age of the patient, the look of the cyst, the FNA results, the cytology, all those are quite useful. So now let's just shift here to guidelines. Are there guidelines that can help us distinguish? And the first set of guidelines came in 2006. Then we had a consensus guideline in 2012, 2013. I mean, these are the different guidelines up to the European guidelines. So what is sort of different with these guidelines? And here's where the difference is. You really want to hone in on the surveillance imaging. So the 2017 guidelines differ on the surveillance intervals, when to stop and when to refer patients to surgery. So anything that's over three centimeters in terms of a cyst, that's going to require very close surveillance. You're talking about an MRI or an EUS in every three to six months. And then these ACG guidelines were, again, it's this follow-up of these larger cysts. Now, this is very different than the AGA guideline, which is a little bit more conservative. And it's, you know, gotten a lot of sort of amongst the very avid endosynographers, we feel that treating cysts in a minimalistic approach can also have dire consequences. Because in this guideline, they represent a very sort of more conservative approach. So again, you can read these guidelines, but the difference in these guidelines is in terms of the surveillance based on the size. Some will say you need to continue to survey. The AGA guideline says that if there are no worrisome features, then perhaps you don't need to be as aggressive or do an EUS on these patients at all. And we can talk about that later. How about when an EUS is not needed? So if a cyst is less than a centimeter outside of it being a solid sort of cystic islet cell neoplasm, cysts arising in the setting of acute pancreatitis, poor surgical candidates, and when you have classic findings on imaging. I will say that in our practice, we tell our referring doctors, but the patients are so keen on getting that endoscopic ultrasound. So it's hard for me to convince patients not to get an EUS. But if I have an 80-year-old woman with a 6-millimeter pancreatic cyst, I just tell them we don't need to do any further imaging or EUS for that matter. So Vivek asked me to talk a little bit about pseudocyst strain, just because we weren't going to cover this in any other aspect of our lectures. And I think nonetheless, Dr. Basad Wadi can go through this in great detail. And her slides are just so elegant and beautiful. But you get the picture. You have a big, large cyst. You have an EUS. And then you puncture it. And then you have a lumen apposing stent. So these are the indications, symptomatic, obstruction, complication. Traditionally, these were drained by surgery or IR. But I can tell you, and I think in almost all centers, I would say probably 85% of all pancreatic pseudocysts that are symptomatic are drained by our interventional EUS folks. And then you can debate whether you need plastic stent or not. And as soon as you puncture, you get the classic look of this. And these are the stents that are placed. And Dr. Basad Wadi explained all this in her elegant lecture. So Shaman Group out of Orlando showed that endoscopic versus surgical cyst gastrostomy, this was a randomized controlled trial. And the key thing is that the endoscopic group had a much lower length of stay compared to the more interventional surgical group. And there's no difference with technical success or complications or reported sort of complications or reinterventions. There was a slightly improvement in physical and mental scores in patients who underwent the endoscopic approach. So here's a necrosectomy that I think Dr. Nathalie and others can go through this in great detail. But this has become pretty much the standard approach for doing a lot of these types of cases. So just to wrap up here in the last minute or two, these are sort of some novel approaches of what is on the horizon. So there's EUS guided cystoscopy, where you can go directly into the pancreatic cyst. The other is a technology called needle focal laser endomicroscopy. Dr. Krishna Someshaker at Ohio State has an NIH grant on this. This is where with the 19 gauge needle, you can go into the cyst and their classic sort of findings of what different cysts look like based on NCLE. This is more of a cystoscopy type look of what these look like. And here's a corresponding needle based confocal laser endomicroscopy. Are we there yet? I don't know he's doing this study. So if you look at combining all these different FNAs, cystoscopy, needle based, then you start increasing the sensitivity. Now, it's not widely available. So right now we rely pretty much on FNA and other markers. And here's a picture of dysplasia. So I know I've gone kind of fast, but I just want to make sure to stay on time. A couple of take home points. Cysts are quite common. They're increasingly diagnosed on cross-sectional imaging tests. We know that. The key here is to remember we're distinguishing mucinous versus non-mucinous. Not necessarily malignant versus non-malignant. Clinical and imaging characteristics are often entirely unreliable. So that's where the FNA and cyst fluid analysis comes. There are some evolving guidelines from our societies. And I think on the horizon, we have some novel EUS guided imaging and sampling modalities. That may help us to help diagnose and distinguish these cysts. So thank you for listening.

pancreatic cysts

patient history

age and presentation

endoscopic ultrasound

fine-needle aspiration

fine-needle biopsy