Good morning, everybody. Thank you so much, Dr. Verna and Dr. Tannen, for having invited me to participate in this session. My goal in this session, since I'm a portal hypertension person, is to show you that if there's no portal hypertension, there's no ascites. So, just to put you back and set the stage of where we are, oh, man, we have to think of two stages in cirrhosis when we think about cirrhosis. Make it personal. No. Oh, yeah. You have the compensated patient and the decompensated patient. These are two entirely different disease entities. Decompensation is defined by ascites, variceal hemorrhage, and encephalopathy, overt complications. And they're entirely different because while the patient remains compensated, the median survival exceeds 12 years, while the moment a patient develops any of these complications, that survival goes down to a median of about two years, entirely different. Now, which of the three complications that come from, that define decompensation, the most common by far, and there's only one of several studies that show this, is ascites. And the main mechanism for you to have ascites is an increase in sinusoidal pressure that drives fluid into the peritoneal cavity. And so how do we measure sinusoidal pressure? We measure it by measuring the hepatic venous pressure gradient, or HVPG. It's a measure of hepatic sinusoidal pressure. In normal, HVPG is three to five. So this is a very small study, but a very significant study by Dr. Blend, this group, that looked at patients with cirrhosis. The yellow dots had ascites, and you can see that every single patient with ascites had an HVPG of at least 12. Remember, normal is up to five. And there was an inverse correlation between the degree of portal hypertension and urinary sodium excretion, so that urinary sodium excretion was lowest in patients with the highest pressure. The other evidence comes from TIPS. TIPS, you put a bridge between the hypertensive portal venous system and the systemic circulation. So you're decompressing the portal vein, for sure, so it's good for varices, but you're also decompressing the sinusoids. And there's evidence that, and this is from Dr. Bosch's group, again, these are patients that had a TIPS, but then were followed after TIPS placement. And you can see that patients that rebled despite TIPS all had an HVPG, it's no longer the HVPG, it's the portal systemic gradient greater than 12. And in patients with ascites, it was the same thing. So patients who developed either de novo ascites, which are the dots, versus recurrent ascites after TIPS placement, you can see here that all of them had a portal systemic gradient of greater than 12 millimeters of mercury. So pressure drives ascites. So what is the main driver, in fact, of decompensation in something that we call clinically significant portal hypertension? It was based on the study by Ripol et al from the TIMOL study. And this is defined as an HVPG greater or equal to 10. When we divided patients in the TIMOL study between those that had between five and 10, or more than 10, these patients tended to develop more decompensation. And guess what was the most common decompensating event? It was ascites. So let's build what happens here. Just so you understand, and I know that Dr. Wong has already done this very nicely, I'm sure I'm gonna go back to many of the issues, but I wanna show you vis-a-vis different treatments. So with cirrhosis, you have an increase in resistance, you have fibrous tissue, you have nodules. All this increases the resistance. It leads to an increase in pressure in the portal vein. But at the beginning, this is mild. So the HVPG is six to 10. But at the same time, these splactic capillaries are sensing an increase in pressure. So they synthesize nitric oxide, among others. This leads to splactic vasodilation. This increases the flow into the liver, increases the pressure even more. So now they reach a state of clinically significant portal hypertension. At the same time, and then this leads to the formation of collateral. But at the same time, this vasodilation leads, like Dr. Wong said, to splackening. So not only splackening, but systemic vasodilation that leads to effective hypovolemia, activation of neurohumeral systems, running angiotensin, aldosterone, et cetera. Sodium and water retention, hypervolemia, that increases cardiac output, increases flow even more. Not only that, these neurohumeral systems act inside the liver by causing vasoconstriction and increasing the pressure even further. Now you can see now then that if there's an increase in flow through the varices, you'll get variceal hemorrhage. If you have now more shunting, you'll have encephalopathy. And if you have more sodium and water retention and the increase in pressure, you'll have ascites. So our thing is about ascites. So how can we prevent ascites from happening? Well, we act on portal pressure. And we act on patients who have clinically significant portal hypertension. How can we do this? We can do this by either acting on flow or acting on resistance. How do we act on flow? And for many years, more than almost 40 years now, the current mainstay of treatment of portal hypertension has been based on non-selective beta blockers which act by decreasing flow. How do they do this? These are the traditional ones, propranolol, they act by beta-2-adrenergic blockade, which is the main factor that causes splenic vasoconstriction. But also by decreasing cardiac output, which is the beta-1 blocking effect. Additionally, carvetolol is a special type of non-selective beta blocker that also has vasodilating effects, has a greater effect on portal pressure because it also causes an effect in vasodilation inside the liver, we think. So this is sort of a proof of concept study. These are all patients that were included in several randomized controlled trials. These were patients that had cirrhosis, had varices, but did not have ascites. And the objective was to prevent first variceal hemorrhage. But we looked at this, we asked the authors for the specific data, and we looked at the clinical events that were beyond variceal hemorrhage, mainly ascites, first variceal hemorrhage, or encephalopathy, and we found that in patients in whom the portal pressure decreased, the ATPG decreased, and in this case it was to a decrease of less than 12 or more than 20% from baseline. The aggregate of these studies showed a decrease in all of these events in those who decreased portal pressure. This is a reduction in the rate of decompensating events of what, 70, 82%. So clearly you reduce portal pressure, you reduce ascites. And this also happened in patients who had bled. So these are not entirely compensated, but these are patients who had bled without ascites. And you see here the same thing. So these are all the studies that were for secondary prophylaxis, that is patients who have bled, but we selected, we asked them to send us the data on patients without ascites, and you can see here that the outcomes that we're measuring is re-bleeding, because these patients had already bled, but also ascites or encephalopathy, and patients who had an HPG reduction had a significantly lower rate of these events. This is a reduction of 59%. But the ultimate proof is a randomized controlled trial, and I'm showing you the results of a prospective multicenter randomized controlled trial that selected patients with compensated cirrhosis, but who had a higher risk of decompensation, i.e. those with clinically significant portal hypertension. They included patients with no or small varices, because patients with large varices would have needed beta blockers. So these are patients that traditionally we would not treat with beta blockers. They were randomized to placebo versus beta blockers, and two-thirds were propranolol, one-third was carvetolol, and you can see here that the outcome was the development of any decompensating event or death. And in the overall assessment, and this is using non-liver related deaths as a competing event, there was a significant reduction in any decompensated event with the use of non-selective beta blockers. When you look at the specifics, so this is the overall results of the trial, said 27% decompensation in the placebo group, 16% in the beta blocker group. This was significant. When you look at the specific decompensating event, look at it here that the only event that was significantly lower in the patients was the development of ascites, 20% in the placebo group versus 9% in the beta blocker group. The first study that shows primary prevention of ascites. And interestingly, in the study, they measured HPG throughout the study, and the risk of ascites, not surprisingly, perhaps was lower in patients who had had a decrease more than 10% or to less than 10 millimeters of mercury at one year. 6% in those that reduced the pressure versus 24% that was statistically significant. Now, what about working on increased resistance? And here we don't have perhaps a strong data, but increased resistance is structural, 70% of it, but it's also functional. So there's an element that is reversible where there's a decrease in nitric oxide inside the liver that causes an increase in intrapartic resistance. Statins ameliorate the endothelial dysfunction and increase nitric oxide synthesis inside the liver, and thereby will decrease portal pressure. And in this proof-of-concept study, you can see here these were four weeks of treatment with simvastatin, placebo versus simvastatin, 20 to 40 milligrams. HVPG did not go down in patients on placebo. It went down modestly, but significantly in patients on simvastatin. The hepatic blood flow did not change, so if the pressure goes down and the blood flow doesn't change, it means that it was a decrease in resistance. Interestingly, and as opposed to other vasodilators, what was seen is that vasodilators deliver so that the clearance of endocyanin green was improved in patients on simvastatin, which potentially means an improvement in liver function. And also, the mean arterial pressure, which with other vasodilators is a problem because the systemic pressure goes down. With simvastatin, there were no changes in mean arterial pressure. So there's proof-of-concept that statins may work. This is a retrospective study, propensity score match study that we did with a database of VA compensated cirrhosis from hepatitis C. And we looked at, we propensity matched them by their likelihood of receiving a statin. So that includes, they were matched by severity of cardiovascular, hyperlipidemia, and also severity of liver disease. And you can see that the statin users had a lower probability of developing decompensation and also lower probability of developing mortality. When we looked specifically at the compensating events, and we're talking about ascites here, you can see here that the event rate in the non-user was significantly higher than in the statin users. So statin users had a lower rate of development of ascites. This is a decrease in the risk of, what, 41%. So there's evidence, retrospective evidence of this where we could be awaiting prospective trials. So now, I'm gonna go back. So this is not obviously the only factor, but I've shown you that it's an important factor. I will not go into this very much because Dr. Wong already went about this. You have inflammation, vasodilatation, decrease in affected arterial blood volume, you're more activated on your humeral system, you have sodium retention. When this gets worse, you not only get, you're gonna get avid sodium retention, so now the patient has refractory ascites. If instead of just the vasoconstrictive substances, you also have ADH, you'll get watery tension, you'll get dilutional hyponatremia, and if there's huge activation of vasoconstriction, you get renal vasoconstriction and hepatorenal symptoms. So it's part of a continuum in a way. So then this, ascites is perhaps a complication that you can better exemplify this because the decompensated patient has ascites alone. The state that I call a further decompensation has the patient has advanced to refractory ascites, hyponatremia, and renal failure, and the driver here I showed you is clinically significant portal hypertension, and the driver here, like Dr. Wong said, is probably more related to bacteria and inflammation that leads to further vasodilatation. Nevertheless, I'm gonna show you some data that shows that decreasing portal pressure, even at these advanced stages, may result in better outcomes. And this, again, from our individual meta-analysis. Now, I'm talking about primary prophylaxis in patients with ascites. So this patient had ascites, had various, but had never bled from them, and we looked at further decompensating events, meaning variceal hemorrhage, encephalopathy, but also refractory ascites, SVP, or HRS, and we found that patients in whom the portal pressure reduced with the use of beta blockers had a significantly lower rate of these events. This is a reduction in 62%, 72%. Anyhow, the other evidence that reducing the portal pressure will actually improve outcomes is through tips, again. So tips in this study was associated not only with a better survival, but since this talk is about ascites, let me show you what happened to ascites after successful tips in patients with recurrent ascites. The number of paracentesis decreased from 10 to one. The volume extracted decreased from 64 to six. And actually, albumin infusion, which was huge in the albumin group, did not make a difference because these patients still did better. So, as Dr. Wong told you, bacterial translocation leads to inflammation, and so the issue is, can we also act on this and prevent ascites? But this is already further, when the patient probably already has ascites. Primary prevention, I showed you, the only studies are non-selective beta blockers. How about preventing getting worse, all right? So one idea is to prevent this by using non-acetyl antibiotics, and Dr. Fernandez showed this in a while back in patients. These were patients with advanced cirrhosis. This patient had ascites, plus they had all these other parameters. And you can see here that compared to placebo, norfloxacin could prevent the development of HRS at least transiently. However, in a more recent study, including child C patients, most of whom had ascites, you can see here that the primary endpoint was not met, and norfloxacin was not associated with an improved survival. Now, what about enoxaparin? Enoxaparin is purported to act on bacterial translocation, although it may act also on the production of these microthormbi that Florence was talking about. But in this study, so they included child B, C patients. So again, these are decompensated patients, and they looked at cirrhosis progression. And cirrhosis progression was defined as development of ascites in those patients who did not have it or worsening of ascites, encephalopathy, SVP, or variceal hemorrhage. And they found that the patients on enoxaparin had a significantly lower rate of these decompensating events. And not only that, survival, the probability of survival was greater in patients on enoxaparin. In this study, Dr. Avila looked at markers of bacterial translocation and inflammation in blood of these patients and showed that they were lower in the patient on enoxaparin. Now, what about inflammation? And Florence talked about this as well. How about albumin? Albumin could decrease inflammation, could actually expand the volume, so act here in effective hypovolemia. And at least there's some evidence that suggests that albumin may actually cause vasoconstriction, although this is controversial. So what is the data with albumin? This is the ANSWER trial. You can see here that the overall survival was better on patients treated with weakly albumin infusions compared to placebo. However, this was not a double-blind study. And if you look here, these were all patients that had what, and I'm quoting from the paper, persistent, uncomplicated ascites despite diuretic treatment. So they were already decompensated again, but not refractory ascites. And you can see here that the rate of the need for the first paracentesis was significantly lower in patients who received the albumin. Again, food for thought. So to conclude, I have shown you that increasing hepatic sinusoidal pressure is fundamental mechanism in the formation of ascites. If not, it will not accumulate in the belly. I've shown you the proof-of-concept study that demonstrate that lowering portal pressure result in decreased ascites formation. I showed you clinical trials in analysis of, in the retrospective analysis of these clinical trials that demonstrate that non-selected beta blockers prevent first ascites in compensated patients. And in decompensated patients, lowering portal pressure has been shown to decrease, again, bad outcomes for decompensation. And that TIPS can also prevent recurrent ascites in the decompensated patient. Therapies targeting inflammation are hopeful and may have an effect in preventing recurrent ascites in the decompensated patient. Thank you so much for your attention. Thank you.

portal hypertension

ascites

cirrhosis

sinusoidal pressure

treatment options