Burden of liver-related comorbidities in cardiovascular disease

In this video Prof. Sven Francque explores that NAFLD is prevalent in patients with cardiovascular disease, but this does not prove bidirectional causality. Vascular mechanisms contribute to NASH pathogenesis and drugs used in CVD management show some benefit on NAFLD.
Play Video

Hello, my name is Sven Francque. I’m a hepatologist at the University Hospital of Antwerp in Belgium, where I’m head of the department of gastroenterology and hepatology. Now, the burden of liver related comorbidities in cardiovascular disease is very important. So the burden of liver-related comorbidities in cardiovascular disease is a very important topic. Why is it important? Well, I think we have to realize that when we talk about different disease entities that are related to obesity and metabolic syndrome, we should not consider them as completely separate identities, but they are part of a very complex, inter-organ crosstalk and that’s also true for the liver and the problem of non-alcoholic fatty liver disease. 

This is, of course, a very important liver problem, but it’s not isolated. It takes place in the context of a complex interaction between the liver, the adipose tissue, the gut, the pancreas and diabetes, and the cardiovascular system, and we have quite some arguments that once you have not only steatosis in the liver but the associated steatohepatitis, that by the release of a lot of factors – inflammatory mediators, angiogenic factors, impact on lipid profile and so on, that the diseased liver impacts on the cardiovascular system. 

NAFLD in Cardiovascular Disease 

There is also the other way around. We know that already early in the development of non-alcoholic fatty liver disease, alterations in liver vasculature play a role in disease, pathophysiology and in disease progression. Now, if we look into patients that have cardiovascular disease, how important is the presence of non-alcoholic fatty liver disease? That’s one way of looking into it because we usually look at it a little bit the other way around. We have patients with non-alcoholic fatty liver disease and we study the relationship with cardiovascular disease. Here, the viewpoint is a little bit different. We start off with patients with cardiovascular disease and then we look into the fact whether they have non-alcoholic fatty liver disease or not. It’s an important approach as you have to look into patients with diabetes, whether they have our non-alcoholic fatty liver disease or not in patients with obesity metabolic syndrome whether they have non-alcoholic fatty liver disease or not. It’s also important in patients that have cardiovascular disease and but surely when they have a cardiovascular disease event to look also not only to the classical cardiovascular risk factors but also to look into the fact of whether they have non-alcoholic fatty liver disease or not. 

Here you see some data because there are not so many studies out there that have looked into this problem from that angle, so not so many studies that have looked into patients with let’s say as an entry point into the study of having a cardiovascular disease or having to undergo a cardiovascular examination, intravascular cardiovascular examination. There are a few studies out there but not so many that looked in those patients how much occurrence there is, how much prevalence areas of non-alcoholic fatty liver disease, but here you see an observational study wherein patients with cardiovascular disease, and cardiovascular risk factors. They also investigated whether this was associated with indicators of chronic liver disease and you see, for example, if you look into patients that had hypertension that there was an increasing prevalence of patients with increased force or indicator of liver fibrosis. Also in patients with dyslipidemia, there were quite some patients that had an intermediate or high fib4 score, and also in patients with diabetes, as a risk factor for cardiovascular disease, there was a substantial number of patients with a fib4  value that was within the 2 cut-offs or even above the cut-off that raises the suspicion for the presence of advanced fibrosis. 

So it’s one example showing you that in patients, that have cardiovascular disease or high cardiovascular disease risk factors. You also have a high prevalence of indicators of non-alcoholic fatty liver disease, and the more severe symptoms. This is another example and this was about patients that had undergone an intravascular assessment for cardiac vasculature and you see again that if then they also investigated patients, whether they had non-alcoholic fatty liver disease or not, about 22% of those patients that underwent such an invasive cardiac procedure appeared to have NAFLD. And if these patients were then followed up for a longer time, then you see that those that had non-alcoholic fatty liver disease, they had lower free event rate over time although the difference did not reach statistical significance, whether this is due to the observation time, the low numbers or whether there is no difference in the end, that is still something that we need to demonstrate. But at least here, you see that there seems to be a trend towards more cardiac events on the long run in those patients that, when they were assessed at baseline had more presence of non-alcoholic fatty liver disease, compared to those that did not have non-alcoholic fatty liver disease. They also had a tendency towards lower event rates of cardiovascular clinical events on the long run. 

So it tells you that, or it’s an argument to say that it’s important also in patients that come in with cardiovascular disease or high cardiovascular disease risk factors. It is important to look into not only the presence of the classical cardiovascular disease risk factors like diabetes, hypertension, dyslipidemia, and so on, but also to investigate whether these patients have non-alcoholic fatty liver, disease or not. 

Vascular alterations in the pathophysiology of NASH 

Now coming to the real topic of this presentation is what is the role of vascular alterations in the pathophysiology of non-alcoholic fatty liver disease and non-alcoholic steatohepatitis. So again we are more used to looking to the opposite direction of the arrow, how NAFLD the impacts on the cardiovascular system, but here we will look at how cardiovascular alterations have an impact on the pathophysiology of non-alcoholic fatty liver disease and evolution towards steatohepatitis and potentially also fibrosis and fibrosis progression. 

Impairment of blood flow in NAFLD 

Now, what we already know for a long time is that in non-alcoholic fatty liver disease, you have reduced sinusoidal blood flow and what we have shown in our lab and it’s also been replicated by a few other groups, is that this goes along with an increased resistance of the intrahepatic vasculature, normally, the liver is vascular system with a very low vascular resistance, but already early in the development of non-alcoholic fatty liver disease that intrahepatic vascular resistance increases, and this is hemodynamically relevant because it leads to an increase in portal pressure and this or these phenomena I mean the increase in the intrahepatic vascular resistance and the fact that this is hemodynamically relevant and translates into an increase of portal pressure is not related to the severity of fibrosis, of course, the more you go to advanced fibrosis, pre-cirrhosis and cirrhosis the higher, the increase in intrahepatic resistance and increase in portal pressure to will be, but it’s already present early in the disease way before you have advanced fibrosis. 

The underlying mechanisms are in part structural but also to an important extent functional and we have demonstrated that in animal models, but we also have data that this is relevant in patients as well. Here you see an example of the structural alterations in an MCD model. Here you see by electron microscopy of vascular corrosion castings – the normal architecture of the sinusoids centralizing to the central vein. And here you see in an MCD model how this pattern is completely disrupted and where you even get, you have not only the flattened and disrupted pattern of the sinusoids, but you also see this kind of blanks suggestive of neo-intragenesis. 

We have shown that and other groups have replicated that as well in other models too. As I said, structural alterations, but also functional alterations and this is an animal model again, it’s the MCD model, but it’s replicated also in other models and you see an important steatosis . But as I said a few slides before – this is not related to fibrosis. It occurs already early in the disease. And here you see the in that in these models you do not have fibrosis, you have extensive steatosis and you even have no or little inflammation in these models at least histologically at absolutely no fibrosis. 

Despite the fact that we are very early in the disease stage, you see that for every flow we apply to deliver in a perfusion model. You see that it steatosis, the pressure is higher and the difference increases with increasing blood flow, and you have very little increase of the pressure in normal animals because it is a low-resistance vascular bed. But in animals with steatosis, you clearly see that portal pressure is increased reflecting an increased intrahepatic vascular resistance. And there is endothelial dysfunction as you can see. Normally, if you do reconstruction, then you give esiticoli. Thanks to the normal function of the endothelium, you will have an evasive relaxation and hence decrease in the pressure. And you see that this normal endothelial-mediated vasodilation is completely blunted in steatosis, so this shows you that there is an endothelial dysfunction in this and again, early in the disease. What is interesting is that you can of course further investigate what the underlying mechanisms are, it is not just endothelial dysfunction. It’s also an alternate reaction to vasoconstrictive substances and you can modulate that pharmacologically and that is what we have been doing. What we observed was that we had the most pronounced effects with endothelin. And if you then, if both are done, for example, you see, first of all, functionally that you restore normal intrahepatic vascular resistance. 

So the increase in portal pressure on increasing flows is returning to normal. This also translates in a clear amelioration of the liver histology.  So it corresponds with the severity of the steatosis and steatohepatitis that develops in an animal model. We have also looked into other mechanisms, not just the animal filling one. As I said, we mainly investigated that pathway because it came out as the most pronounced or most importantly altered pathway, but you can also either do or try to intervene on the other pathways, the angiotensin pathway, or the cyclooxygenase pathway. So, with thromboxane 2 and then you see again, then you can buy pharmacological modulation, improve liver hemodynamics, improve liver perfusion and that this goes along with an improvement in both, those structural alterations that I previously discussed, as well as the liver histology. Again, the most pronounced beneficial effects were demonstrated by impacting on the endetholin pathway; we had less pronounced, but still some effects when we altered the angiotensin pathway or the thromboxane mediated both ways, but as I said, the most pronounced beneficial effects were seen when we modulated the endothelin pathway. But it clearly shows you that first of all, early in the disease, you have both functional and structural hemodynamics alterations that are hemodynamically relevant – it translates, amongst others into an increase in portal pressure and you can modulate that pharmacologically, and if you modulated that pharmacologically, that also correlates with an improvement in different histology. 

This is also relevant for patients, as I said before, yes, it is, we measured not directly, but indirectly, portal pressure by trans-nuclear hepatic vein pressure gradient measurements or a HVPG measurements in patients because of their body constitution were less suited to undergo a classical percutaneous biopsy. We perform the biopsy transfigurously. And then you see here again in patients that were non-cirrhotic even in most cases, not even F3 Fibrosis. So earlier stages of the disease we could observe in about 1/4 of the patients HVPG, which was above the threshold of 5 millimeters of mercury so, which corresponds to a definition of portal hypertension. So, 1 out of 4 up to one of the three of these non-cirrhotic NAFLD patients do have by indirect measurement, so by HVPG measurement, they do have portal hypertension. It’s rarely above threshold of 10 millimeters of mercury, so it’s not already in most of the cases, it’s not clinically significant portal hypertension, but it is portal hypertension, which is indicative of relevant, hemodynamically relevant increase in intrahepatic vascular resistance, so relevant changes in the intrahepatic vascular mechanisms. What has also been observed when others have closely looked at the biopsies of patients with NAFLD, is that in some patients, not in the majority of patients, but in some patients, you can observe lesions that are are resembling those that we see in portal sinusoidal vascular disease. So, you see shunt vessels that are close to the portal tracks and are also indicative of structural vascular alterations early in the disease. Again, these were patients with non-alcoholic steatohepatitis, sometimes fibrosis, F2 F3, but non-cirrhotic patients. So again, these are not the classical shunt vessels that you can observe in a clear cirrhotic stage, this is early in the disease. 

LSEC capilarization 

Another important aspect or another important piece of data supporting the concept that vascular alterations play an important role early in NAFLD pathogenesis is a sub-analysis that has been done from biopsies in the context of the phase 2 trial of Lanifibranor so not only looking into the patients that were randomized into trial, but looking into the whole population that has been screened. So, including the screen failures and looking then with the CD34 in your histo-chemistry, at the capilarization of the L6. You remember that the liver vasculature or the liver sinusoids, which are the capillaries of the liver, have some very specific structural characteristics in contrast to normal vascular, capillaries, they do have fenestrae, which allow for an interaction between the blood in the sub-endothelial space of this time and they do not have a basal membrane. It’s a very typical characteristic of normal liver vasculature. Now, when the liver gets more and more disease and it’s typically seen again in association with the cirrhotic stage of the disease, characteristics of fenestrae and absence of basal membrane will disappear. And so the liver sinusoids will more and more resemble normal capillaries without fenestrae and with the presence of a basal membrane. And as I said, you can illustrate that with the CD34 chemistry and what we observed in the population of patients, and again, non-cirrhotic NAFLD patients. Some of them screen failed for the trial because their histological severity of the disease was not enough to meet the criteria for entry so it’s larger than the population that was randomized into trial and you see here that the CD34 positive area clearly increases when the disease is more severe. So there is severity in terms of fibrosis. So you see a correlation with the severity in terms of fibrosis? But what is important here is that it’s not just F3. You see already in earlier stages that capilarization occurs and the more severe the disease is, the more capilarization you have or the other way around. It’s of course, what we see here is associative but the important message here is and even in the early stages of fibrosis, you see already capilarization happening. And this is also not only related to the severity of the fibrosis, but also to the severity of the lobular inflammation. So again it has to do with steatohepatitis. And not just fibrosis in the advanced stage and not just cirrhosis. That’s a very important point and it adds to the evidence that we have and also the clinical evidence we have beyond the preclinical animal model data that I showed you that we also have clinical evidence that this is relevant in human NASH pathophysiology. 

Platelet therapy, anti-platelet therapy and statin 

Now, this has also some implications for the therapy of these patients. I already illustrated that, in the animal models with drugs that are used for treatment of cardiovascular disease, risk factors and especially arterial hypertension in patients. If we use these drugs in animal models we see some beneficial effects on NASH. Probably mediated by the effects that we see on the liver vasculature. Do we also have these data in patients? There is some data on those and are hypertensive, but there are not correlated with histology. And just by ways of an example, I will discuss about platelets and antiplatelet therapy. You could argue the same about statins and also about some anti-hypertensive treatments, but I will not discuss that, I will limit it here to the story about platelets.

Now, first of all, this is a study that looked into animal models of atherosclerosis, so that this was not specifically to address non-alcoholic fatty liver disease and it comes again to preclinical data. It’s a model of atherosclerosis but they administered several drugs to these animals, some anti-inflammatory drugs, and also aspirin. And what is interesting in this disease, because in this model it’s of course, not just atherosclerosis, they also developed fatty liver disease and what they observed was that in the animals that were treated with aspirin, not any other animals but in the animals treated with aspirin, they had clearly first of all a lower liver weight, less liver steatosis and less inflammation. So the animals treated with aspirin are better in terms of severity of steatosis and steatohepatitis. 

This interesting study looked into the role of platelets in non-alcoholic fatty liver disease and non-alcoholic steatohepatitis both preclinically and then they also added a small pilot trial in patients, first of all, in preclinical models and they tested different models so it’s not model specific, they could reproduce it in different models. First of all, they could demonstrate in those models of NAFLD NASH that there was an increased presence of platelets in the livers mainly in the livers with steatohepatitis. Not so much with just isolated steatosis, so illustrated with different techniques. So there is an increased presence of platelets. There is also a difference in the functionality of these plates, these platelets are more activated and when they treated these animals with antiaggregant therapy, which was a combination of aspirin and clopidogrel, you could clearly see that the liver fat content in these animals decreased, so pharmacological modulation of platelet and platelet endothelial interaction because it’s not just platelets, of course, but the role of platelets is probably in part mediated by the interaction between endothelial. So you could clearly see an improvement in liver fat content in the animals that were treated with the dual antiplatelet therapy. This was also Illustrated in the liver histology, which clearly improved upon that dual therapy. Also, in terms of the non-alcoholic steatohepatitis score or fatty liver disease activity score I should say, you see that there is a clear improvement by the dual antiaggregant therapy and this improvement is not related to changes in body weight of these animals. So it’s not because these animals eat less, gain less in terms of body weight compared to the animals that were treated with the diet and again this was replicated in several models. But in one of the diet-induced models, there was no difference in body weight between the animals that just received the diet or the animals that received the diet with the dual antiplatelet therapy. So the Improvement was not to be explained by the fact that these animals had less overweight, it was not that metabolic factor that could explain the differences pointing towards the fact that what we observe here is truly related to pharmacological action of the combination of drugs and not to a metabolic improvement and that indirectly could explain the liver improvements. 

Now, this was, as I said in different preclinical models, the study also had a small translational part, first of all looking into biopsies of patients with non-alcoholic steatohepatitis and also here you can see that the number of platelets is increased compared to healthy controls or at least controls with a normal liver. And in that patient population, they also did a small pilot study so they looked into patients that have to undergo cardiac catheter examination. And then based on the result of that examination equations, those patients were put on a mono therapy with aspirin or dual antiaggregant therapy. So, it depended, of course, on the result of the cardiac examination. 

As you can see, small numbers in the end. 11 patients, compared to 13 patients. They regroup those that were immunotherapy or dual therapy, but these patients were investigated at baseline for several different parameters, including liver fat content and then re-evaluated after six months. And although this is a small pilot trial, you can see here, the results showing that those that were on treatment when monotherapy or dual, antiplatelet therapy had a clear and significant decrease in liver fat content and also in liver volume. 

So this at least is a proof-of-concept points again towards the fact that by pharmacological modulation of the parameter of platelet and probably platelet endothelial interaction. You can have an impact on liver histology. Another piece of data comes from a study where they started off with patients that had NAFLD but not advanced fibrosis. So, early stage patients that were then followed a long period and reassessed for the potential occurrence of advanced fibrosis non-invasively. So it’s an incidence study in follow-up, starting off with patients that have a low indices of fibrosis at the start and what you can see here is that if you compare patients that are on daily aspirin, compared to those that are not on daily aspirin, the incidence rate of advanced fibrosis based on non-invasive testing this case is clearly lower, significantly lower compared to patients that are not aspirin and after correction, of course, for quite a substantial list of potential confirms again, illustrating, that aspirin use seems to have a protective effect on the severity of NAFLD and the evolution of NAFLD. 

Conclusion 

So based on all these data and again, as I said, I did not discuss the treatment or the potential effects of some of the anti-hypertensive treatments or I didn’t discuss the potential effects of statins, but the data are there also and they are, let’s say little bit comparable in terms of the message they deliver to what I have told you about platelets and drugs that have an impact on platelets and platelet endothelium interaction, what we can conclude from these data is, first of all, of course, that non-alcoholic fatty liver disease is prevalent in patients with cardiovascular disease. Of course this does not prove any causality as I said, but we have to take into account the scheme that I showed you in the beginning of my presentation where you have that complex and multi-directional interaction between all these different organs. I think it’s important to realize that vascular mechanisms themselves contribute to NASH pathogenesis. So it’s not only NASH that contributes to cardiovascular disease. There are the other way around also, cardiovascular, maybe vascular mechanisms that have an impact on the progression of steatosis to steatohepatitis and steatohepatitis to fibrosis, and drugs used in the management of cardiovascular disease, as I showed you with antiplatelet drugs can also have some benefit in non-alcoholic fatty liver disease. 

So I think this is something that we need to take into account for the current treatment of our patients so we can already use current drugs and probably optimize the treatment using the drugs that we already have in our armamentarium, but I think it is also very important to look into drugs that we are developing now specifically to treat NASH in clinical trials, that we also pay attention to what these drugs are doing on the vascular mechanisms inside the liver and to what extent their impact on these vascular mechanisms can contribute to their beneficial effect on non-alcoholic steatohepatitis and on cardiovascular disease at the same moment, of course, because that’s something also that is very important. We need to have a global approach and take into account that we are not just treating liver disease, but we should also try at the same time to improve the whole cardiometabolic context in which NAFLD develops.  

Would you like to read more?

Prof. Sven Francque

Prof. Sven Francque

Prof. Dr. Sven Francque obtained his MD at the University of Antwerp, Belgium, in 1994 and was subsequently trained in internal medicine and in gastroenterology and hepatology at the Antwerp University Hospital, Belgium and at the Department of Hepatology of the Beaujon hospital, Clichy, France.

Video SERIES

SUBSCRIBE & FOLLOW US

Subscribe Our Newsletter To Get the Lastest Updates