So, this section is called Management of Liver Disease Complications. We're going to do this together. All right, neither of us have relationships that we need to disclose. So, the objectives of this presentation are to define liver disease complications, to describe different types of liver disease complications, to discuss the implications of specific liver disease complications, to learn the liver disease or liver disorders, I should say, associated with specific liver disease complications, and to determine how to work up patients with liver test complications. Let's start with a question here. Upper GI varices can be caused by, is it A, cirrhosis specifically due to alcohol use, or B, hepatic artery thrombosis, or C, tapeworm infestation, or is it D, any disease that results in elevated pressures of the portal vein? Awesome. Particularly awesome because it's early on a Saturday morning. I know that it's barely crossed the 7 a.m. line out on the West Coast, and you're kicking it already. So, indeed, the answer is D, any disease that results in elevated pressures of the portal vein. Portal hypertension results in portal hypertensive esophageal varices, gastric varices, splenic varices, all kinds of varices, and these are varicose veins that result from that venous obstruction or the obstruction of the portal venous flow. Remember that the portal vein is a very special vein. It's called the portal vein because it transports blood that is rich in nutrients from the intestine up to the liver. So, while it's bringing blood into the liver, it's called a vein because it's taking blood out of the intestine and porting it up to the liver, kind of like a conveyor belt, so that the liver can process those nutrients and then release all of that good stuff into the systemic vein, which is the hepatic vein. Remember, the liver has two veins, the portal vein and the hepatic vein. The hepatic vein leads blood out of the liver, and the liver has one artery but two inflows of blood, the hepatic artery and the portal vein. In fact, the portal vein provides more blood than the hepatic artery does, and the portal vein, even though it is a vein and is carrying somewhat oxygen-depleted blood, the total oxygen that the large portal vein carries to the liver is probably greater than the amount of oxygen carried by the hepatic artery. Isn't that right, Andrea? That's right. Thank you. So, there you go. Interesting to remember your hepatic vascular anatomy. Okay, let's do another question because you're up and you're sharp. Antibiotic therapy is indicated for, A, patients undergoing elective esophageal variceal band ligation, B, patients with cirrhosis presenting with presumed portal hypertensive bleeding, C, patients with acute alcohol intoxication presenting with upper GI bleeding. Or is it D, a cirrhotic patient who is found to have a recurrent bleed from known duodenal ulcer? Man, I'm trying so hard to get you to have to work really hard and you're still nailing these things. So, the answer is cirrhosis presenting with presumed portal hypertensive bleeding. Now, if you have a scope down there and the patient's bleeding from a duodenal ulcer and they happen to be cirrhotic, which is kind of what I'm implying with D, that patient doesn't need antibiotics. This is portal hypertensive bleeding in the setting of cirrhosis. And the need for antibiotic therapy there, prophylaxis, it has nothing to do with the patient's need for endoscopy. You're not prophylaxing against endoscopy-related infection. These are patients who have cirrhosis and have portal hypertensive bleeding, and those patients are at risk for being infected with gram-negative organisms just from that presentation. And therefore, they need antibiotic coverage for gut bugs, whether you're going to scope them or not. You're going to scope them. There's no question about that. But the scope isn't the reason. It's the bleeding in the setting of cirrhosis. And that's the point that I'm hoping to make with that polling question. So, we know the answer to that. So, then let's do a case together. A 40-year-old woman, now think about this. A 40-year-old woman is brought to the emergency department via ambulance with acute multiple episodes of bright red hematemesis, which began just 15 minutes ago. Her husband of two years, who witnessed the event, says that she's being treated for, quote, liver disease from drinking a lot of beer for a few years. And he says that she had three beers yesterday. Her husband endorses that she's had one prior upper GI bleeding episode years ago from what he thinks was an ulcer. Her medication list reveals spironolactone, furosemide, potassium, lactulose, PRN naproxen. Always got to throw something in there to make you wonder, right? And identifies only one allergy, which is to disulfiram, which causes nausea and headache. Okay. That med list I'm just going to clue you in should tell you a lot about this patient. And I'm going to let you look at that again. I'm going to be quiet for a second. That is a big clue to you. Okay. She's followed at a clinic that is not a part of your EHR network. So you can't see the clinic notes from the clinic where she's followed. Okay. So on exam, she's lethargic. She's oriented to name only. She has blood and clots around her mouth, her nares, her neck, and the front of her shirt. Her vital signs are temps normal. Her pulse is tachycardic and regular. Her respirations are a bit elevated. Her blood pressure's kind of soft, right? Her systolic's 79, her diastolic's 55. And her oxygen sat's 95%. Her head and neck exam is normal except for that blood residue we just talked about. Her neck is supple. Her chest demonstrates a few telangiectasias on the upper chest. We said that she's tachycardic on her heart exam, but she's regular. Her lungs are clear, thankfully. Her abdomen exam is significant. It demonstrates bulging flanks, a fluid wave, although she's not tense. And her liver span is six centimeters at the midclavicular line, and she has a palpable spleen tip. And her extremities reveal petal edema, which is bilaterally symmetric. What are you going to do next, and in what order? Consider the following. On her labs, her hemoglobin is seven. Her MCV is up. So her hemoglobin's low, right? That's a crit of 21, a hemoglobin of seven. So that's pretty low. Not normal by any stretch of the imagination. Her MCV is high. Her sodium is 125, her potassium's 3.7. Her BUN is 25, her creatinine is one. Her ammonia is 64. And her acidic fluid, when aspirated with a needle through the anterior abdominal wall, demonstrates no neutrophils. What would you order? Andrea, any comments so far before I go on? Yeah, so obviously you look at the medication list, and this patient has well-established liver disease with fluid issues and a history of HE or hepatic encephalopathy. So that's the clue. She's obviously with the addition of potassium. They're managing her electrolytes due to the diuretics. So, you know, this looks like a bleed, coupled with, of course, that hemoglobin that's low. It's hard to not imagine she doesn't need intensive care, but certainly needs that upper GI endoscopy by my endoscopic colleague, Dr. Martin, to help her situation. Well, I appreciate that. And the vignette that you just put together there, Andrea, is exactly what we're after here. This lady needs to be in the hospital, and we are really, really going to need to jump on her and take care of her. Because she clearly bled, right, and clearly had an upper GI bleed. And while her husband says that she apparently had an ulcer years ago, she's on a lot of medications that suggest that she does have chronic liver disease. And she's got blood covering her mouth and her nares, and she is encephalopathic. So this lady absolutely needs to be in the hospital. She might need to be in the ICU, depending on the institution. And, you know, at a lot of institutions, she would be sent to an intensive care unit where she would be tanked up with crystalloid and blood products. And obviously, we would look to stabilize her hemodynamically and to get her set up before we perform the endoscopy. We always want to make sure that the patient is in a vital sign setting where it's safe to sedate her and safe to perform endoscopy. It is team dependent and somewhat institutionally dependent when it comes to the decision of whether you would endotracheally intubate a patient like this to protect their airway, to perform the endoscopy in case they have active upper GI variceal bleeding and are at high risk for aspiration or may even have food content in their stomach, or just frankly not be able to control their airway because of their impaired cognitive function. So the first thing would be to get this patient admitted probably to an ICU and to get her euvolemic and to get her hemoglobin up to a level where you at least have some wiggle room in case she's still bleeding or if bleeding recurs. Any thoughts, Andrea, about what medication you might think about giving from the get-go? Obviously, we kind of discussed that really in a recent polling question, but it never hurts to hammer this in over and again. Yeah, you know, I think that this is a situation where obviously we have to be, you know, get her stabilized and to get the fluid and get her back to a safe place. I'm a little less well versed on sort of the acute management of these patients. Obviously, they're managed by a hospitalist team. So I might defer to you on that question, John. Well, you know, all of what you mentioned is correct and is the right thing to do. And the only thing I'm really going to add to that is to make sure that you give this lady either a fluoroquinolone or an appropriate cephalosporin to cover for gut organisms, because we're assuming that she is a cirrhotic with the medication list that she's on and her physical exam presentation. And we're also assuming that she's having an upper acute, upper GI bleed. And we know that one plus the other equals antimicrobial therapy. All right. So then we admit her, we stabilize her, and we're going to perform, as Andrea just ordered me to do, an upper GI endoscopy. And it demonstrates the following. And I think you can see a couple of things here. You see the positive arrow sign twice. The yellow three-tipped arrow, what you're seeing there is varicose veins in the esophagus, which we call esophageal varices, and they are almost always due to portal hypertension. And when portal hypertension occurs and the blood cannot percolate through the liver from the portal vein to the other side, which would be the hepatic vein, and is backing up, it may shunt through little vessels that connect the portal system to the systemic outflow around the liver, and these will become way fuller, let's put it that way. They will carry much more blood than they are normally made to carry, and as a result will become thin-walled, under high pressure, and at that point are at risk for bursting and bleeding. And that is what an esophageal variceal bleed is. What you are seeing, again, at the three-tipped arrow, yellow arrow there, is the esophageal varices. However, what you are seeing at the top right with the black-tipped arrow pointing at it is a little white or cream-colored thing that is sticking out, and that is a fibrin plug, which is an indication to you that there is actually point source bleeding from that spot, which has self-resolved for now because the clotting cascade and the thrombotic function of the platelets has led to a fibrin plug forming there as a finger in the dike. That is a spot that can bleed again, so that is almost a gift showing you, hey, look over here, I just bled. I know you can't see it coming out right now, but I was pouring out just a few minutes ago. Okay, so that's a real big clue to you that that is actually not just esophageal varices, but esophageal varices that were actually the cause of the acute upper GI bleeding event. So what are your endotherapy options and how do you choose? I'm going to go to the next slide here and show you. The most commonly used device on the endoscope to eradicate esophageal varices is a rubber band. It's a low-tech device, so this is called esophageal variceal band ligation. Back in the bad old days, you could only load one rubber band on at a time, and you would have to take the scope out, reload another one, and go back down. But now, there are multiple of them that are loaded onto a little cap or hood that is wedged onto the front tip of the scope made out of clear plastic, which is what you see here. If you can see my cursor moving around, that clear thing ends right here. And that's just a little clear tube that extends about a centimeter or more in front of the tip of the scope. And there are a couple of threads that weave between each rubber band. And as you turn a spool or fishing reel that is attached to the strings, as you turn that thing, the strings tug on the next rubber band that is loaded onto that hood or cap and cause that hood or cap to roll off onto the varix when you suck said varix up into the hood. And that is how you rubber band ligate an esophageal varix. So there is the fibrin plug, and there is the varix, and a band has just been placed on it. And it's sort of a clear white color and is buried under the mucosa, the esophagus, because it's tight, and that's why you can't see it. But that's what makes this flat-ish varix with the fibrin plug now look like a polyp. It's because you've just choked that thing off with a rubber band. And, of course, what happens is that staunches the flow of any further blood through that varix, not just here, but also upstream and downstream from it. And, of course, it is the clot that forms in that stagnant blood that is the ultimate long-term hemostasis that results from band ligation and other esophageal variceal occlusive therapies. So I just showed you that. This is what the rig looks like, the apparatus on the outside. There's the strings that are weaving in between each band, and you can fire one off at a time until they're all gone. And if you haven't done the job, you can load yet another refill right on there and keep going. And this is the fishing reel I was talking about that turns, wrapping these strings that go through the scope up to make this tighter so that you can, in succession, fire one after the other. You can also inject alcohol-type agents either into or alongside the varix, which causes inflammation immediately and clotting of blood in that varix to make it stop bleeding and to prevent future bleeding. You can also inject glue, a medical form of super glue, into the varices. That product is not FDA-approved for variceal injection in the United States. So in the U.S. of A, we do not actually have access to the variceal glue formulation. But you can off-label, use cyanoacrylate or medical super glue that is marketed in the U.S. for skin-closure purposes and inject that into varices. That is off-label and not performed at all institutions and has risk factors, including the risk of embolization of that glue causing a stroke, which is rare but reported in the literature. A TIPS is the use of a metal stent, similar to a metal bile duct stent, that is driven inside the liver, an IR vascular procedure, and they drive that metal stent through the liver between a tributary of the portal vein and the hepatic vein to create a portal to systemic shunt within the liver using a metal stent. And that will also decompress and stop variceal bleeding at any location, whether it's esophageal or gastric, by taking some of that backed-up portal blood and shunting it through the stent to the systemic side, to the portal vein side, thereby reducing the portal vein, backed-up portal vein blood pressure and reducing it. That is not an endoscopic modality. Obviously, that is done in interventional radiology, which we abbreviate IR. What you see in the picture on the left, for your interest, is an actively bleeding varix. When that stops bleeding, if it does so spontaneously, which it will sometimes do, you will see that little whitish fibrin plug, if you're lucky enough to find it with your endoscope. Here is a picture to the right of a varix ligated, and you can see that that happens to be a black rubber band, so it's a little bit easier to see. Here are some of the devices, in case you wanted to see. This is the glue. It's usually tinted with a dye so that you can see it in your needle catheter that you use to inject it. If it were clear, you wouldn't be able to see it in the clear needle catheter very well. It's important that you see it well, because if you put too much of it out while you're pushing this, about to push this down the scope, you could glue your scope with superglue, which would be a multi-thousand dollar damage to the scope, and you'd have to go fetch another scope while your patient's bleeding, so you wouldn't want that. The glue is tinted for that reason. This is what a needle injector looks like. It's basically nothing fancy. It's just a hypodermic needle that's connected to a lung catheter. We talked about being able to do sclerotherapy before esophageal variceal band ligation appeared in the early 90s. We used to inject oil alcohols like ethanolamine, which you see here, which comes in glass ampoules, and various other oil alcohols to cause immediate inflammation, intra- or perivascular, to cause clotting and hemostasis. We also talked about forming a TIPS, which is a metal stent-created shunt between the portal vein tributary and the hepatic vein outflow, and this can be done by the interventional radiologist accessing the hepatic vein and driving a needle and guide wire between these two, balloon dilating the tract over that wire, and then exchanging the deflated balloon off the wire for the stent, which is positioned across both vessels after balloon dilation over that wire, and then expanded here, which allows some amount of that blood to bypass the liver to the systemic side. That has its own risk of complications, short-term and long-term, associated with it. Happy to discuss those during our Q&A if you're interested in things of that technical detail. All right, so just because you took care of the acute bleed doesn't mean that you're done with the job. This is a patient, and when you have a patient, you have a relationship, and so what you do after you deal with acute management of esophageal variceal bleeding can be more than unifold. There are pharmacotherapeutic options to consider and additional band ligation to consider. When it comes to additional band ligation, you not only want to get rid of the bleeding that's happening today, but you, of course, also want to get rid of the risk of future bleeding, and so you will usually want to bring the patient back after a couple of weeks, allowing the ulcers that happen as a result of the initial banding to heal, to go back and look again to see if there are any residual varices, and if you are, you're going to want to perform additional band ligation to the point that all of the varices are obliterated. If the patient has had portal hypertensive bleeding, you'll usually want to place them on a non-selective beta blocker to also help prevent future portal hypertensive bleeding of any form, and if the patient's never bled and you put them on beta blockers, that is primary prophylactic pharmacotherapy. If they've already bled and you're going to place them on it, that is a secondary prophylactic pharmacotherapeutic use. Now, what if the bleeding was actually coming from a gastric varix? How would the management differ? And to cut to the chase here, I mainly want to point out that band ligation is by far the most common and most useful treatment for esophageal varices that require endoscopic treatment. Sclerotherapy is usually only for salvage any longer. However, gastric varices aren't particularly amenable for a number of anatomical reasons to successful management with band ligation. In the United States, the most common treatment for gastric varices that bleed is to place a TIPS, that shunting metal stent that we just talked about. In rare instances, in certain kinds of gastric varices, banding might be a good thing to reach for immediate control, but it will not be likely to be successful for long-term management. In other countries, particularly where TIPS isn't all that available, glue can be a very effective treatment and is far more frequently used than banding because it's a better treatment for gastric varices than banding is. We have much better access to TIPS placement in the United States than we do to glue injection. Andrea, I'm going to hand off to you because while I have expertise in endoscopy and biliary endoscopy, I am not a hepatologist and you are. With that disclaimer, all yours, my partner. Thank you, Dr. Martin. For clarification, I'm not a hepatologist, but I am a hepatology NP. I certainly manage these patients commonly in our outpatient practice. Six-week follow-up on this particular patient. Regrettably, they have gained some weight with more abdominal fullness. We perform an abdominal ultrasound, which is identifying fluid. They're also struggling with pitting edema. Thankfully, sodium, potassium, and creatinine are normal. Now, a couple of comments. Dr. Martin recently showed the primary prophylaxis option. In this particular case, as that patient presented with a bleed, we're looking at secondary prophylaxis, but many of these patients are indeed placed on a non-selective beta blocker, so either Natalol or Propranolol. We recommend that this be provided at bedtime to minimize any sense of lightheadedness that they may experience following the dose. The other thing that we know is these patients are already oftentimes at fatigue. This certainly doesn't help. One of the issues with non-selective beta blockade is it may also contribute to some degree of fluid retention. Now, with this particular case, as we're working through managing the active bleed issue, we're certainly not going to hold the non-selective beta blocker, even if some fluid develops, because that's not just due to the non-selective beta blocker. It's likely due mostly to the advanced stage liver disease. In this case, with the moderate amount of fluid that was seen sonographically, we would certainly go ahead with paracentesis. Our standard approach is to do diagnostic and therapeutic on the first paracentesis just to make sure this patient doesn't have evidence of SBP. Seeing the patients clinically, we do discuss the low-sodium diet of not to exceed 2,000 milligrams per day. Now, some patients are particularly careful in reading labels and are very astute at managing their sodium intake, but I found that for the majority of patients, it is helpful to have them meet with a dietician to talk about those sort of hidden sources of sodium that may be contributing to challenges with managing their fluid. Diuretic therapy for patients with advanced stage liver disease who are retaining fluid, we would typically start with both furosemide and spironolactone, oftentimes at doses of 20 milligrams of furosemide and 50 milligrams of spironolactone, monitoring their labs, and then increasing the therapy as needed over time. Some patients won't need much once they're much more strict about their diet, and we can get away with relatively low doses, but others, it needs to be pretty active management with periodic paracentesis and robust diuretic therapy. Follow-up recommendations, so I do recommend that patients monitor their weight. If they're so inclined, they can certainly measure their waist circumference, but most don't do that. I've also, you know, just monitoring for the lower extremity fluid retention just to keep track. If there is a fair bit of variability, I encourage them to think about how we might fluctuate that dose of furosemide slightly, meaning 20 to 40 milligrams on a particular day, or if they're starting on 10 milligrams, 10 to 20. So I do like to empower them to do some of the management, but to keep us posted with what they're needing to take on a regular basis so we can adjust appropriately. Of course, for some of these patients, once they're adequately ablated to obliteration, we could, in theory, think about discontinuing the diuretic therapy, but oftentimes we'll keep them on low dose over time. Okay, John, I'll turn it back to you for case two. Great, thank you, Andrea. I didn't know that you engaged patients in self-titration. That's actually fantastic. All right, so here's a 48-year-old man who's seen in his primary care clinic where he describes intermittent passage of dark bowel movements over the past month, most recently a couple of days ago, and then when queried specifically, the stools are intermittently jet black, and they don't really have any red or maroon blood in them. He denies chest pain or shortness of breath and has had no hematemesis, nausea, vomiting, or abdominal pain, even after he just had lunch consisting of a turkey sandwich and fries an hour ago. His medical history includes hepatitis C, which isn't presently treated, and it was only intermittently treated in the past because he was intolerant to the medications, and so he wasn't compliant. Medications now include hydrochlorothiazide and potassium. He takes ibuprofen a few times a week for knee and back pain, started aspirin once a day because someone told him it'd be good since his dad died from a heart attack. So a six-month follow-up visit. The wife describes changes in his sleep pattern and forgetfulness, and he notes difficulty concentrating. Alert, oriented, fully comfortable, no acute distress. Vital signs are, as you see there, nothing really particularly ominous. And you look at the exam, there isn't a whole lot there. Liver spans 10 centimeters at the midclavicular line. Extremities are normal. So what do you do next? Right, this isn't an acute GI bleed. The last lady had blood around her mouth, around her nares. She was encephalopathic. This patient is really neither one of these things. It's not acute, and he's just fine as you look at him, but he's having black tarry stools, and this is clearly melanin. So you can think about the labs that you could get. You want to make sure that he has an adequate hemoglobin. You want to know if he's bleeding chronically, which you can tell from the indices. You can think about some imaging studies to get a better idea of what's going on, but I'm going to profess to you that endoscopy is probably going to tell you what you need to do. This man's stable, so he probably doesn't need any sort of an admission, and he is hemodynamically stable, so he probably doesn't need any fluids either. So let's get some labs, and you see that his hemoglobin is the better part of 28. It's not too bad. It's not normal. His MCV is low at 79, so he may be iron deficient, so he's probably chronically losing blood. So is endoscopy indicated? I think it is, but it's probably going to need an elective sort of scheduled endoscopy rather than urgent right away sort of thing, and he just ate anyway, so it's not like you can scope him today. You know that your affiliated endoscopy center a block away often has openings on an as-needed basis, but you probably don't want to do an endoscopy today if you just had a turkey sandwich and fries an hour ago, right? Because we know from yesterday's talk that after a solid meal, you need to wait six hours, and by then, the endoscopy center is going to be closed, so then you're going to be talking about a hospital admission or the ED as a spot to do the endoscopy. If you're ED, he'll let you even do that, so the earliest you're probably going to be doing it is tomorrow, and you're going to ask him to be NPO before midnight. So let's say that this is what your endoscopy shows, and this is the stomach. It's the body of the stomach, and you see sort of a reticulated or a network type of pattern with some little punctate hemorrhages in those areas between the white lacy appearance. This is portal hypertensive gastropathy. If it's mild, you won't see the actual red bloody look. If it's moderate, you'll see that. If it's severe, then you'll actually see oozing blood, actively oozing blood from it, so this is probably moderate portal hypertensive gastropathy. Your pharmacotherapeutic options for this are generally beta blockers, which Andrea has described to you already. There isn't really good endoscopic treatment for diffuse presentations of portal hypertensive gastropathy like this. However, if the patient's losing substantial blood from severe portal hypertensive gastropathy, then you can ask your radiology IR colleagues to consider a TIPS. Now this is a variant of that that is called, well, this is, first let me say this is the antrum of the stomach, okay, and this is a form of portal hypertensive gastropathy, but it is nodular, and it's nodular gastroenteral vascular actasia, and if this, which can be associated with portal hypertension, is present, this is exquisitely amenable to endoscopic therapy in the form of band ligation, as I'm demonstrating to you here. Now this will obliterate what you see, but over time, since the underlying condition persists, this will recur and will require repeated endoscopic management. So some practice pearls for you before we move to Q&A. Portal hypertension can lead to multiple sources of upper GI bleeding. Acute bleeding requires proper workup and stabilization before considering intervention, whether pharmacologic, endoscopic, or radiologic. Additional interventions and surveillance are often required longitudinally after acute intervention and may be coupled with pharmacotherapy. Not all portal hypertensive upper GI bleeding sources are amenable to or require endoscopic intervention, and ongoing management of decompensation include therapy for ascites and hepatic encephalopathy. Andrea and I thank you very much. Andrea, any additional comments before we switch over to Q&A? No, it looks like we have several questions, so I look forward to us working through those. Me too.

portal hypertension

varices

endoscopy

TIPS

pharmacotherapy

management

gastrointestinal bleeding

evaluation