We're going to discuss current controversies in the definition of non-sorotic portal hypertension and a diagnostic approach. My name is Theo Heller, and I have no disclosures. I'd like to welcome you to an exciting program. This is the first collaboration between the Pediatrics and the Portal Hypertension 6, and a lot of people have done a lot of work, and I'd like to thank everyone, especially the speakers. The first session is what we're about to embark on, and we're going to discuss an approach, genetics, vascular causes, global epidemiology, and then a case. The second session will entail discussion of monitoring adults and children, an intervention radiology, and a surgery approach, and lastly, there'll be a case elucidating the principles that have been discussed in the presentations. Why care, and what is non-sorotic portal hypertension? Is it because hepatitis C is done, or hepatitis B is not as interesting, or because NASH is really an endocrine disease? No, I hope over the next 17 minutes to show you how non-sorotic portal hypertension is both important and interesting. You will notice over the talks that management and physiology are intimately linked, and while a lot of principles have been derived from the management of cirrhosis and the study of cirrhosis, non-sorotic portal hypertension parallels, but does not overlap cirrhosis. When we approach non-sorotic portal hypertension, we think about it as pre-hepatic, entropatic, or post-hepatic, different causes occurring at different sites. Entropatic is divided into pre-sinusoidal, sinusoidal, and post-sinusoidal, and this is where I'd like to focus today. Of particular importance, this is of particular importance because this is where the greatest diagnostic dilemmas and the greatest controversies lie. A fundamental concept in non-sorotic portal hypertension is that synthetic dysfunction is atypical. In cirrhosis, one finds both synthetic dysfunction and portal hypertension. In non-sorotic portal hypertension, it is all about the pressure. The simplest definition is portal hypertension that occurs without cirrhosis. Portal hypertension is manifest by ascites, varices, things like that, and cirrhosis is excluded by etiologic factors or when needed by biopsy. All of this work has been laid down and developed, and these concepts have been developed by giants, and we stand on their shoulders. These are people who've been working in India, Japan, America, Canada, and Europe for years, and you will hear from some of them today. One of the greatest problems in non-sorotic portal hypertension is the paucity of the literature, in particular prospective literature. There's a near universal disagreement on nomenclature, and a lot of cases are rare manifestations of rare diseases, making it difficult to study. Complicating this is that, in genetic diseases, there are multiple phenotypes that can arise from a single gene type, and there are multiple genotypes that can give rise to single phenotypes. The fact that non-sorotic portal hypertension often occurs in the setting of another disease makes management far more complex. It's also possible to see evolving disease patterns in a single patient. One might see vascular changes, followed by regenerative changes, and eventually fibrotic changes, in the same patient over time. Disease patterns vary by geography and socio-economic status, making practice patterns difficult as well. Diseases can coexist. For example, a patient could have CVID, common variable immunodeficiency, with hepatitis C, and nodular degenerative hyperplasia. Unrelated to the hepatitis C, portal vein thrombosis is a special case, as it in itself can give rise to non-sorotic portal hypertension, but also may result because of intrapartic causes of non-sorotic portal hypertension. Patients are also living longer, and this gives more time to develop liver-related complications. As an example of this, cystic fibrosis has more recently been appreciated to have a second peak of liver disease in adulthood, the pediatric peak having long been appreciated. This has now been described in other settings as well, such as cystinosis, chronic granulomatous disease, common variable immunodeficiency, and others. The most challenging diagnosis in the category of intrapartic non-sorotic portal hypertension is the spectrum of disorders that we call idiopathic non-sorotic portal hypertension, or portosinusoidal vascular disease, or non-sorotic portal fibrosis. These all refer to essentially the same thing, and they represent the final common pathway, much like cirrhosis does, of a number of disorders. This final common pathway might obscure the route taken to get there. This makes it difficult to lump things together, because in cirrhosis you would not say that the natural history of hepatitis C is the same as the natural history of NASH, and it also makes it difficult to split things apart to study them carefully, because they look the same at the end stages. There are multiple terms that have been used for intrapartic non-sorotic portal hypertension. Some of them are different, but many of them overlap, and this has further confused the field. However, in terms of histology, there's increasing conformity in terminology, and I'll give you some examples. This is an example of portal venopathy, where the portal vein is a slit-like structure near the bottom of the image. This is an example of incomplete septal fibrosis, and lastly is an example of nodular regenerative hyperplasia. You can see on gross specimen in panel A that the liver appears nodular, and even on the H&E stain, which is atypical, you can see nodularity. The meson stain in panel C shows an absence of fibrosis, and the reticulant stain, which is mandatory in the setting, in panel D, shows nodularity. If we look a little bit closer at the nomenclature, the commonest term that has probably been used in the West is idiopathic non-sorotic portal hypertension. This is a misleading term. We might not know the exact mechanism that hepatitis C uses to cause cirrhosis, but it's not directly relevant. We would still say that the cirrhosis is induced by hepatitis C, or caused by hepatitis C. In non-sorotic portal hypertension, there are often clear associations, making the term idiopathic incorrect. For example, azathioprine-induced nodular regenerative hyperplasia is not idiopathic. The risk with idiopathic is that it stops us asking why, and that prevents us from mitigating risk factors. A further problem is that not all hepatitis C patients have portal hypertension. In the same way, not all non-sorotic portal hypertension patients have portal hypertension. This sounds strange because it seems built into the term, but non-sorotic portal hypertension is more analogous to cirrhosis with portal hypertension, and the term leaves us wanting, in terms of descriptions, for earlier stages of the disease. A more recent term that has been proposed is portosinusoidal vascular disease. This is a much better term, and it solves many of the problems, but it doesn't address central vein damage. In one series of 88 biopsies in chronic granulomatous disease, 63% of patients had central venopathy. Here is an example where the central vein can be seen almost completely obliterated. Further, in non-sorotic portal fibrosis, abnormally dilated central veins were seen in 64% of patients. A term that has been commonly used as well is non-sorotic portal fibrosis, and in one series published two years ago in Hepatology International, periportal fibrosis was seen in 77% and perisinusoidal in 61%, making the fibrosis non-universal, and the term then slightly inaccurate. So let's be practical. How do we step aside from all of the disagreements with nomenclature and the confusion and take care of our patients? We would take the same approach that we all know so well. We should apply standard principles. We should have an index of suspicion, maintain it in our differential diagnoses, the right patient at the right time, a careful history, physical, laboratory exam, imaging and biopsy, especially when intraepatic causes are suspected. There are some exceptions, for example, in autosomal recessive polycystic kidney disease, congenital hepatic fibrosis is universal, and the diagnosis is made because of the syndrome. Furthermore, there is no staging for the biopsy findings, making the biopsy even less useful clinically. Staging is an important concept, and here traditional markers don't work. Albumin and bilirubin are typically normal, PT at most is slightly elevated, and fibrosis by definition is not a factor, making something like an ISHAC score or a METAVERE score not possible to stage patients. Through the talks today, you will hear about transient telastography, hepatic venous pressure gradients, MELD and CPT scores not being useful, and you will hear about the problems of the biopsy findings. And you will hear about the promise of platelet counts and spleen size as markers to follow patients for the development and progression of non-sorotic portal hypertension. And as an example of some of the problems we face, here are some patients from my practice. The top two panels and the lower right panel reflect patients with chronic granulomatous disease and non-sorotic portal hypertension. The lower left panel is a patient with cirrhosis. If I asked you which patient had varices, it would be difficult to say. If I showed you the hepatic venous pressure gradients, it's not much simpler. It was, in fact, the patient with a pressure gradient of 8. If diagnosis and staging are a problem, what is it that we need to manage? What is the clinical problem beyond that? There are a number of factors here. The first is what we're really trying to do is manage non-sorotic portal hypertension complications. The ascites, the varices, splenomegaly, things like that. But we don't really know what the true incidence is or what the denominator is. Is it that most patients we're presenting are just the tip of the iceberg and we're missing far more? Or is it that those who are presenting are really the bulk of the patients with the disorder? The literature is clearly biased towards the latter, but it's not clear. We also don't really understand the rate of progression. Traditional liver problems are seen, but with a slightly different bent. Varices are common in both and are typically the presenting feature in non-sorotic portal hypertension. Ascites is less common in non-sorotic portal hypertension. Hypersplenism and cytopenias are more common and typically more severe, and here's an example of a child with congenital hepatic fibrosis and massive splenomegaly. Infection is more common. Encephalopathy, if present, may not be as severe and it's often subclinical. Further manifestations are acute on chronic liver failure, but here it has a different flavor. What we would typically see is the development of ascites in patients with infection. And as the infection resolves, hepatorenal syndromes can occur. Hepatopalmy syndromes and portopalmy hypertension should be screened for. And a long-term sequelae such as progressive liver dysfunction seen in extrapodal venous obstruction or portal biliopathy should be anticipated. These are some of the varices seen in my practice and this is an example of a patient with nodular regenerative hyperplasia and a massive duodenal ectasia from this ectasia. Mortality is often hidden. It's seldom as obvious as the patient I just outlined. Infection in primary immunodeficiencies is a good example. They're often ascribed to the underlying disorder instead of the portal hypertension and here it's possibly the double whammy of two diseases that makes the infection so much more serious than it would have been without the portal hypertension. Mortality also depends on association and it's pretty clear that a patient with CVID is not the same as a patient with azathioprine sorry, a patient with CVID and nodular regenerative hyperplasia is not the same as a patient with azathioprine induced nodular regenerative hyperplasia and the azathioprine can be stopped. Mortality has been looked at in a number of ways and I'll just highlight some studies. Skoten et al. looked at 62 patients and showed that survival was worse than the general population. Siramol Paiwat looked at 69 patients and showed that survival was less than that which would be anticipated but probably better than anticipated for other forms of liver disease. Feld et al. took a different approach. They took 194 patients and they divided them into patients who had features of liver disease specifically elevated alkaline phosphatase elevated ALT and declining platelet slopes suggestive of progression in portal hypertension and they separated these patients from those patients that had no features of liver disease and you can see that the survival curves are very different. Instead of taking patients with non-sorotic portal hypertension it took patients at risk for non-sorotic portal hypertension and divided them into those who appeared to have it and those who did not. What do we want? In the absence of a winning lottery ticket we'd like to help our patients. We'd like clarity. We'd like to make an early diagnosis. When patients have advanced disease it's not that difficult Non-invasive tests are not possible yet The natural history is very difficult Prediction is difficult especially when it comes to the future is one of my favorite Danish proverbs Especially difficult because we're lumping different diseases together as we've already noted as a thyroprene-induced non-sorotic portal hypertension is not the same as common variable immunodeficiency and lastly we'd like more effective therapies So ideally we'd like an international consensus working group to get together to come up with a series of definitions that would make it easier to study and work with these patients However, in the absence of that which doesn't help us right now we still have the possibility of deriving a standard approach We'd want to diagnose patients by an anatomical site where appropriate pre-hepatic, intra-hepatic, post-hepatic If it's intra-hepatic we'd like to define the dominant histology as specifically as possible understanding that multiple abnormalities can be seen in a single biopsy specimen We should note the etiology and association because that will define the prognosis We should also notice the stage or complications defined clinically For example, we might say that a patient has nodular regenerative hyperplasia in common variable immunodeficiency complicated by varices We should notice whether the patients are adult or pediatric because the management will be different as you will hear later in the day and we should also notice whether the problems are acute or chronic because that will affect management as well We should not neglect the team approach Often the underlying disease needs a multidisciplinary approach and patients would benefit from this The essence is that we should have an index of suspicion We should not ignore signs of liver disease even if atypical We can make a diagnosis We can define the anatomy, histology and underlying disease and if we use different terms it's okay as long as we understand what we mean when we use those terms and what it means for the individual patients Where needed we should seek out expert help We should especially seek out expert help for pathology As we study these patients we should notice whether they have portal hypertension or not We should understand the variation of traditional liver disease as I've outlined them and we should also understand that in the setting of other diseases these disorders may not be so benign I want to thank you for your attention and hope you enjoy the rest of the sessions Good afternoon The title of my talk is Precision Medicine Diagnostic Approach Unique Therapeutic Opportunities My name is Silvia Villarino and I'm an assistant professor of medicine and pathology at the Yale School of Medicine I'm a physician scientist at the Yale Liver Center and our research focus is to use human genetics and genomic approaches combined with mouse and cell-based models to uncover the molecular mechanism underlying a variety of liver diseases that we still don't know the cause I have no disclosures Precision medicine is an approach to disease prevention and treatment based on patient's individual differences in genes, environments and lifestyle With the completion of the human genome in 2003 followed by major advances in next-generation sequencing technologies precision medicine has been prevailing in many fields of clinical medicine and pathology is no exception The work from our group at Yale as well as work from the Mayo Clinic has shown evidence that there is a clinical utility in genomic analysis both in the diagnosis and management of with liver disease that we don't know the cause In a recent review we proposed that patients with undiagnosed liver disease with onset of clinical features before the age of 40 with a typical presentation with congenital or syndromic features with multisystemic disease as well as patients who have a positive family history of liver disease of a known cause should be considered for further evaluation with genetic testing namely whole exome sequencing and that genome rounds or a multidisciplinary approach where genotype and the detailed phenotype are combined has been fruitful in providing diagnosis to these patients and the diagnosis many times will translate into therapeutic opportunities or options as well as more informed prognosis and adequate family counsel Idiopathic non-sterotic portal hypertension is a very good illustration of a liver disease that we don't know the cause Over the decades several contributing factors have been reported such as exposure to medications chronic infections immunological disorders which include both immunodeficiencies and autoimmune syndromes but for the most part the pathogenesis remains elusive For this reason our group and others hypothesized that genetic factors may underline the onset and development of portal hypertension in a subset of these patients and to address this question we used one application of next generation sequencing named whole exome sequencing which is an effective way to capture and sequence the coding parts of the human genome So here you have a cell with a nucleus with a chromosome and when you zoom in you find the genomic DNA which is composed by these grey lines that depict the introns known coding parts of the human genome and the smaller rectangles that are the coding parts of the human genome and they are called the exons and the exons only represent 1% approximately of the human genome but it's estimated that will harbor up to 85% of all the mutations with large effects in human disease and that is the reason why we decided to use this genetic test to better understand the molecular pathogenesis of non-serotic portal hypertension So the sequencing data is then analyzed through a computational pipeline that is able to give us annotated information of rare genetic variants in these patients. Applying this technology, our colleagues from the University of Amsterdam were the first ones to report that the de novo mutation in a gene called KCNN3 was found in a family with idiopathic non-cirrhotic portal hypertension. This gene encodes a small conductance calcium activated potassium channel three, and is involved in the regulation of both arterial and venous vascular tone. The authors identified a rare heterozygous mutation in this gene in four affected individuals in the same family, namely the father and three children. And they propose that these individuals that harbor these mutation have an increased predisposition to develop endothelial damage and possible microtrombi. But further studies addressing the molecular mechanism as well as identification of additional families are in need at this point. By the same time in 2016, our group also through the application of whole exome sequencing to patients with idiopathic non-cirrhotic portal hypertension implicated a recurrent recessive mutation in DGO kappa, an amino acid alteration of N to S at position 46, which impairs ATP binding and reduces the catalytic activity of the deoxyguanosine kinase, which is required for mitochondrial DNA replication. So we found that patients that have decreased activity in this enzyme develop non-cirrhotic portal hypertension. And here are some histological pictures showing features of hepatoportal sclerosis. Interestingly, the danosine, which is a nucleoside analog, used in the HIV treatment regimen in the past, has a black box FDA alert for possibly causing non-cirrhotic portal hypertension. And interestingly, when cell lines in vitro are treated with this drug, DGO kappa levels are decreased. And this really brings the hypothesis that maybe this molecular pathway is involved not only in inherited forms of non-cirrhotic portal hypertension, but also in acquired forms. But further studies are required to test this hypothesis. In late 2017, a new gene has been implicated in idiopathic non-cirrhotic portal hypertension. And this time, the word comes from colleagues from France, where they identified heterozygous mutations in a novel gene that they named FOVP, which stands for familial obliterative venopathy. And they found six patients from two families, as well as an additional unrelated patient with mutations in this gene and features of non-cirrhotic portal hypertension. The mechanism remains unknown, but the authors provide data that shows that at the RNA level, expression of this gene is found in the intrahepatic portal vein, supporting a vascular etiology for this disease. I'm glad to share with you the results of a collaborative effort where we identified that recessive mutations in a gene named GMAP5, which is a GTPase with a still unknown function, cause idiopathic non-cirrhotic portal hypertension. We found four families with one or more individuals affected with non-cirrhotic portal hypertension who had homozygous mutations that were rare in the same gene. And as you can tell, all the mutations are different in different families, but they are overall in conserved areas of the genome. Here I just represented images of an HNE from one of the patients that showed some vascular abnormalities, as well as a minimal lymphocytic infiltrate, and the reticular stain from one of the patients that show classic features of nodular regenerative hyperplasia. I should mention here that four of the patients, actually one from each family, have been submitted to a liver biopsy, and they were features constituent with nodular regenerative hyperplasia, but the spectrum of disease is variable, which one can argue that you are seeing different stages of the natural history of the disease, or it could be related to sampling issues. But they all have the same genotype. Moreover, when we stained for CD34, we saw an aberrant liver sinusoidal CD34, expression in the four liver biopsies from patients, as compared to the unaffected control. And we were fortunate to have access to mice that are homozygous for a loss of function mutation in the same gene, which recapitulates many of the liver features seen in GMAP5 deficient patients. And again, also CD34, as shown in brown stains, is seen in the sinusoids as early as two weeks of age in these mice, whereas in the control mice, CD34 is only detected in the large vessels. Moreover, we were able to show that GMAP5 is also expressed on liver endothelial cells, in addition to immune cells, as it was previously known. And hepatocytes have no expression of GMAP5, as well as other parenchymal cells of the liver also have no expression of GMAP5. So to further understand the role of GMAP5 in liver endothelial cells, we perform single cell RNA-seq of these cells, both from control mice, as well as GMAP5 deficient mice. And what was striking is that the liver sinusoidal endothelial cells, for short, I will use LSEC, which represents 75 to 80% of liver endothelial cells in the control livers, and are represented here with this mustard color, are completely almost gone in mice that are GMAP5 deficient. And there is an emergence of capillarized cells, CD34-expressing cells, in the mutant mice that are barely seen in physiological conditions. In this chart, it's very clear the appreciation that in red are the GMAP5 deficient animals, and you can see that there is an over-representation and emergence of capillarized cells, and almost an absence of sinusoidal endothelial cells. More interesting, the transcription factor GATA4 that is essential for the differentiation of liver sinusoidal endothelial cells is also significantly downregulated, supporting that GMAP5 might be a regulator upstream of GATA4 to control the identity of liver endothelial cell, sinusoidal endothelial cells. So in summary, absence of GMAP5 in liver endothelial cells leads to a near absence of LSACs, an emergence of an abundant capillarized endothelial cell population, and an expansion of lymphatic endothelial cells. And we believe that this is the cellular and molecular mechanism that supports why patients that are deficient in GMAP5 develop portal hypertension. Are there other genetic disorders associated with idiopathic non-cerotic portal hypertension? And the answer is yes. There have been reports of non-cerotic portal hypertension in patients with Adams-Olivier syndrome, as well as Turner syndrome. Also, I should mention that despite not being the focus of this talk, when you apply genomic analysis, you are also well-equipped to identify genetic causes of thrombophilias, which as we all know, are associated with non-cerotic portal hypertension. And moreover, also genetic analysis will facilitate the diagnosis of primary immunodeficiencies. And for many years, nodular regenerative hyperplasia, plus or minus portal hypertension, has been described in patients with common variable immunodeficiency. And by applying exome sequencing, as well as other next-generation sequencing technologies, it has been identified in 2013 that the subset of these patients actually harbor mutations in genes in the pathway of the PI3 kinase delta. By making this discovery, these patients were then eligible for clinical trials for personalized treatment with selective PI3 kinase delta inhibitors that actually have been previously developed before other malignancies. So this is an elegant example how going from molecular diagnosis that was facilitated by genetic analysis can really open new therapeutic opportunities that were personalized to their molecular disease. And I should also make a point that in the case of the GMAP5 deficiency, because we found that there is an intrinsic liver endothelial cell dysfunction, in these patients, the modulation of the immune system by means of more neurotransplantation or other methods will most likely not correct the liver disease. So in summary, the increasing application of genomic analysis is the delineating non-sterotic portal hypertension at the molecular level, and therefore enabling its stratification into distinct groups informed by genotype, refines the taxonomy of idiopathic non-sterotic portal hypertension at the same time that uncovers novel genes and its functions in portal hypertension. So I should emphasize that as physicians, we should understand the data that supports the cause of our patients' non-sterotic portal hypertension. And when we are not satisfied, when we don't know the cause, then we should consider a referral to a medical center where genomic analysis can be offered. Timely diagnosis can prevent complications and ensure adequate preventive and therapeutic management and family counsel to our patients. Thanks so much for your attention and please feel free to get any comments or questions through the chat. Thank you. Thank you for the invitation to speak to the new virtual version of the liver meeting about this interesting issue of unique diagnostic challenges in vascular causes of non-sterotic portal hypertension. These are my disclosures, which I think are not relevant to this current presentation. I would also disclose that I am a pediatric hepatologist. I will therefore approach this from a pediatric perspective, although try to speak more generally about relevant issues. Our objectives are by the end of this session to have an understanding of the diagnostic test abnormalities commonly seen in vascular causes of non-sterotic portal hypertension and be able to choose an appropriate diagnostic approach for patients presenting with non-sterotic causes of portal hypertension. Now, this is a broad topic. There are numerous causes of non-sterotic portal hypertension and I have only 17 minutes to address these issues. So I will be focusing particularly on these main conditions, portal vein thrombosis, idiopathic, non-sterotic portal hypertension, which of course has gone by many names over the years, including nodular regenerative hyperplasia, obliterative portal venopathy, non-sterotic portal fibrosis, portal sclerosis, et cetera. Also sinusoidal obstruction syndrome and hepatic venous outflow tract obstruction and Bugchiari syndrome. Now I recognize of course that there are many other important causes, some of which I've filled in on this table, but unfortunately we won't have time to cover those in sufficient detail today. So let's begin with thinking about typical clinical presentations. There are really two buckets here. The first is those patients who are identified through incidental finding of laboratory abnormalities or splenomegaly such as mild elevation of liver enzymes, thrombocytopenia. And those who present more dramatically with symptomatic presentations that may include GI bleeding or cites or dyspnea. When we think about the asymptomatic groups, data from this study in France from a few years ago is illustrative of the kind of enzyme abnormalities that may be seen with ALT and alkaline phosphatase elevated in about half the group, but mostly in the one to two times the upper limit of normal range, so just mild elevation. GGT elevated in a few more and perhaps a little higher up to four times the upper limit of normal in most of those. And platelet count reduced in this particular study in about 56%, although this course depends on inclusion criteria and other studies have found much higher rates of thrombocytopenia. Among those presenting symptomatically, if we look at this study from Barcelona, 69 adults with idiopathic portal hypertension from which things like portal vein thrombosis have been excluded, symptomatic presentation occurred in 42%. Most of those were GI bleeding. A few had ascites and a small number had dyspnea that was secondary to hepatopulmonary syndrome or portopulmonary hypertension. And then of course, because this is a broad topic that we're covering today, there are patients with Budghiari syndrome who will have acute severe presentations with abdominal pain, ascites, tend to hepatomegaly, etc. Clues to the presence of non-sorotic portal hypertension are usually found in the clinical history, key as always to understanding our patients. We should be thinking of country of origin because idiopathic non-sorotic portal hypertension is much more common in countries of Southern Asia, India, Pakistan, Bangladesh. And then there are many other associations that can lead us to the right diagnosis. So for example, for portal vein thrombosis, we think of neonatal umbilical catheterization or umbilical sepsis as predisposing factors. Abdominal trauma, surgery, intra-abdominal inflammation such as pancreatitis, inflammatory bowel disease may predispose to vascular trauma or thrombosis. And then for both the large vessel and the small vessel intrapathic disease, we have to think about prothrombotic coagulation disorders, lymphoproliferative and myeloproliferative diseases, immune mediated disorders, including autoimmune disorders, rheumatological disorders, connective tissue disease, et cetera, and immunodeficiencies that may be primary such as common variable immune deficiency or acquired such as HIV infection. There are numerous other genetic disorders that may be associated with some of these conditions. And then of course, the important group of patients exposed to different types of medications and amongst those important groups include antineoplastic agents and thiopurine derivatives. So where do we begin? Once we get these referrals, usually we're first interested to look at the initial ultrasound and Doppler imaging. And this can give us a clear indication immediately whether we're dealing with a portal vein thrombosis with these sorts of findings shown on the slide or a hepatic vein or super hepatic inferior vena cava thrombosis causing a Badkiari syndrome type presentation where we may see absent, reversed or turbulent hepatic venous flow, intrapadic collaterals, sometimes subcapsular in location, a web at the hepatic vein ostium, hyperechoic cord replacing a hepatic vein and this finding of a prominent chordate lobe. In idiopathic non-sterotic portal hypertension, we may find subtle changes that could lead us in the right direction to the diagnosis right from the get-go with this first ultrasound. And the findings of note are this hyperechoic area adjacent to the larger branches of the portal vein within the liver. And it's demonstrated in these three cases. The one on the left showing the left portal vein in longitudinal section with this hyperechoic boundary to it. Two cases on the right show similar findings in cross-section. And then in addition, the second image from the French group on the left labeled B shows a reduction in the lumen of a branch of the portal vein with this hyperechoic area probably thought to represent fibrosis around the portal vein. And then hypoechoic area beyond that, which again is a recurring feature in idiopathic non-sterotic portal hypertension. And these are the subtle findings to look out for in this condition. For sinusoidal obstruction syndrome, diagnostic criteria for this, you will probably be familiar with the Seattle or the Baltimore or the European criteria which focus on clinical observation after a bone marrow transplantation, looking for things such as jaundice, tender hepatomegaly, ascites or weight gain due to fluid overload. There are also ultrasound criteria initially dating back to this early study by Lassow in 97 and subsequently more recently followed up in a small study looking at multiple ultrasound criteria aiming to help the sensitivity and specificity of the traditional Seattle, Baltimore or European diagnostic criteria. So usually the combination of clinical criteria plus ultrasound imaging is sufficient to make the diagnosis of sinusoidal obstruction syndrome. So then when do we need further imaging with CT or MRI or invasive phonography? And this differs by the sort of diagnosis that we're chasing down. So first of all, for portal vein thrombosis and bud chiari, these diagnoses are usually reached by ultrasound imaging in the right clinical context. And further imaging is mostly reserved for those patients in whom there is diagnostic uncertainty after the ultrasound, or for those patients where there are additional comorbidities to be more accurately defined such as a tumor or in patients where planning for therapeutic interventions needs additional imaging. So for example, if we have a patient with portal vein thrombosis in whom we're considering a mesorex bypass, we may need to undertake cross-sectional imaging such as this CT scan shown here demonstrating patency of the left portal vein adjacent to the portal cavernoma. Or in addition, we may need invasive venography such as the wedged hepatic venography shown on the right demonstrating patency of the left portal vein highlighted by the white arrowhead and in continuity with the right portal vein highlighted by the black arrowhead. And we may hear more about that later in the session. In bug chiari, cross-sectional imaging may help to demonstrate this peripheral hyperenhancement on a contrast-enhanced CT scan with compression of the IVC shown in the white arrowhead there. And invasive venography may help to define the status of the location of the thrombus and the nature of the vascular pathology such as shown on the right with intrapathic collaterals showing and lack of outflow from that hepatic vein into the IVC following percutaneous venography into the liver. Now for idiopathic non-sorotic portal hypertension, cross-sectional imaging again can provide some clues for us such as shown in this study in which the liver is somewhat shrunken but has smooth contour. There is clear evidence of portosystemic collaterals and significant splenomegaly. And in this condition, the splenomegaly is often more prominent than seen in cirrhosis. We may also see this interesting change in portal vein morphology from which that term obliterative portal venopathy perhaps comes in which the main branches of the portal vein are seen to be quite dilated. And then there is a clear step down in caliber of the branches of the portal vein as shown in this scan and also in this further case, illustrating the same features on the right there. So if we come out of our ultrasound and cross-sectional imaging, we may have this situation where we have demonstrated a patient with mild increased enzymes with parenchymal imaging abnormalities and evidence of portal hypertension. The parenchymal imaging abnormalities may include the shrunken liver that we saw in one of those images. There may be perfusion differences through the parenchyma giving a heterogeneous appearance to the liver. There may be nodularity arising due to nodular regenerative hyperplasia or more focal nodular lesions that leave us in doubt whether we are dealing with a case of cirrhosis or of non-cirrhotic portal hypertension. And here there is a need to go further with one of two approaches or probably both approaches at this point, I think. The first is elastography. So is there a role for elastography to help to determine whether we're dealing with non-cirrhotic portal hypertension or cirrhosis? There are only a small number of small studies of this question in the literature. The first that I'll show you is this study from 2013 in which cases were divided into those with idiopathic portal hypertension or IPH, liver cirrhosis, LC, chronic hepatitis, CH, and normal controls on the right. NC. And you can see that for the first graph on the left, the liver stiffness was highest in the cirrhotics, lowest in the normal controls and intermediate in those with idiopathic portal hypertension and chronic hepatitis. This study used ARFI as the elastography technique. Spleen stiffness, on the other hand, was highest in the idiopathic portal hypertension group. And you can see them progressively lower in the cirrhosis, the hepatitis, and the normal controls, such that if you looked at the ratio of spleen to liver stiffness, this was highest in the idiopathic portal hypertension group. And the area under the receiver-operated curve for the spleen to liver stiffness ratio was 0.92 in this study to differentiate idiopathic portal hypertension from the other diagnoses. So quite an impressive diagnostic accuracy there. In a more recent small study reported last year using a more up-to-date technique of MR elastography in the hopes of avoiding the observer variation that complicates some of the ultrasound elastography techniques, this small study looked at 11 patients with non-cirrhotic portal hypertension defined as nodular regenerative hyperplasia and 11 patients with cirrhosis. And you can see here, again, significantly higher stiffness in the cirrhotics compared to the non-cirrhotics for liver stiffness. Spleen stiffness on this occasion was very similar between the two. But again, if you look at the ratio of spleen to liver stiffness, much higher in the non-cirrhotic group compared to the cirrhotic group. So then what about liver biopsy and its role in helping with this? I guess it gives us a more definitive answer. We should, of course, see changes of cirrhosis when we do a biopsy, although we have to remember that sometimes sampling error can still leave us confused. There are specific features that can enable a diagnosis of non-cirrhotic portal hypertension, and those include the changes of nodular regenerative hyperplasia, but also some venous vascular changes that I'll highlight here. First of all, portal vein narrowing shown in the two left sections, incomplete above and complete loss of portal vein lumen below. Herniation of the portal vein, where the portal vein has extensions out into direct contact with the hepatic parenchyma, as you can see in those two examples. And then hypervascularization of the portal tract, in which there are additional vascular lumens within the portal tract, shown in the two sections on the left. And periportal abnormal vessels where those additional vascular lumens occur adjacent to, but outside the portal tracts in the liver parenchyma. So in summary, the take-home message is really around our diagnostic approach. We begin with a clinical history, physical exam, and a list of associated disorders and drug therapies that gives us a sense of where we're heading. Liver labs, ultrasound imaging, and the need for a cirrhosis workup to rule out underlying cirrhotic-type liver disease. Our ultrasound is usually enough to diagnose hepatic venous outflow tract obstruction or Bucchiari syndrome, sinusoidal obstruction syndrome, portal vein thrombosis, although cross-sectional imaging may help to define the comorbidities or therapeutic options for those conditions. Commonly, we need to pursue further cross-sectional imaging or liver biopsy if we are not finding those first three diagnoses. And these should help us to determine, finally, whether we are dealing with a cirrhotic liver disease or a non-cirrhotic portal hypertensive condition, such as INCPH or one of these other conditions that I haven't had time to talk about today. With that, I will finish, and thank you very much for your attention. I'm from the Institute of Liver and Biliary Sciences. I cover global epidemiology of NCPH and impact of infectious diseases. I have nothing to disclose. I'll try and cover terminologies, epidemiology, pathogenesis, and the way forward. Several disorders are clubbed together under non-cirrhotic portal hypertension. In different parts of the world, they include NRH, obliterative portal venopathy, INCPH, NCIPH, NCPF, IPH, hepatopulmonary, or PSVD. NCPH is a heterogeneous basket of diseases characterized by rising portal pressure of more than 10 millimeters due to intra- or prehepatic vascular lesions in the absence of cirrhosis and Batkiari syndrome. Importantly, portal pressure is higher, and HEPG is near normal. When we assess such patients, whether there is a lesion in this planic vein, portal vein, or presinusoidal area, you will see that the hepatic vein pressure will be normal. HEPG will be normal, while the portal pressure would be high. So both prehepatic and presinusoidal diseases will raise the portal pressure. However, HEPG will be normal. There can be several causes. Dominantly in prehepatic will be EHPVO with portal vein thrombus, hepatic will be EHPVO with portal vein thrombus, while in hepatic causes, presinusoidal distinct diseases would be like non-sedotic portal fibrosis or IPH. Other diseases have their complete natural history by themselves. You can also have diseases which have sinusoidal component, or post sinusoidal like HVOTO or venal occlusive disease, or post-hepatic component. But most of these are liver diseases by themselves. And one of the presentation is non-serotic portal hypertension. I will mainly discuss two diseases, IPH and EHPVO. IPH or non-serotic portal fibrosis as called in the Indian and Asian continents is a disease where the main portal vein gets thrombosed, replaced by cavernoma. So NCPA for IPH is a hepatic venopathy involving the small and medium branches of the portal vein resulting in development of portal hypertension. Spenomegaly moderate to massive presence of viruses and collaterals with a patent spinoportal axis, normal liver functions, and in HVPG, as I said earlier, normal or near normal with high portal pressure. Liver histology and nose cirrhosis of parenchymal injury and obliterative portal venopathy is the most important part, and this is evidence-based. These patients have well-tolerated episodes of very sealed bleed, no viruses, no chronic liver disease, and the portal vein is thickened with periportal hyper-echoic areas on ultrasound or other imaging techniques. When we look at the East and the West, the proportion of portal hypertension due to non-serotic is much higher in several parts, several Eastern countries. These patients are generally younger, and in Japan there are female dominance, otherwise all over it is male dominance. Infections remain a major cause, and infections can induce immunological changes, growth thrombotic states, and of course HIV remains an important cause in the West. Now this is to show you in the autopsy series how portal vein is thickened and is sclerosed. This is hepatoportal sclerosis, thickened narrow portal vein branches. If you look at the wedge biopsy, you see classical obliterative portal venopathy. The portal vein, the whole of the portal vein is actually occluded with formation of neo-aberrant vessels and periportal fibrosis is seen. And in the space of this, there is deposition of collagen. See these sclerotic thickened portal vein branches. So in a normal person, the portal vein, the blood flows in the prehepatic, the sinusoidal area mixes with the arterial blood and goes into the inferior vena cava. In non-serotic portal fibrosis, because of the injury in the presinusoidal area, there is thickening, and in the sinusoidal bed also there is thickening. So both presinusoidal and perisinusoidal injury is there. There is a gradient between the spleen, ISP, and the liver, and between the liver and the wedged area. So there is one gradient here and one gradient here. If you put your balloon here for measuring HVPG, then you will realize this is normal, but on this side, there is portal hypertension. These serotic patients, if you measure the wedge pressure, let's say it is 20, the intravaricial pressure by direct puncture of the varix will be the same. But in non-serotic, you may have an intravaricial pressure of 25 or 30, but the wedge pressure may be lower and near normal. This indicates that the site of resistance is presinusoidal and perisinusoidal. Here, the gradient is between the spleen and the liver, intrahepatic pressure gradient, and here it is between the spleen and the wedge area. And these two gradients put together decide how much is the difference between the spleen and the near normal wedge pressure. This is in a normal person, you will find when you put a balloon here, the free hepatic vein pressure is between 2 and 7, and the portal vein pressure is 4 to 10 millimeters. While in non-serotic or IPH, you will see the portal pressure here is much higher, while the wedge pressure remains 2 to 7. So HVPG is not a measure for portal pressure in non-serotic. In fact, intravaricial pressure, portal pressure directly, or intrasplenic pressure represent this kind of patients. In these, as I said earlier, there is thickening of the portal vein, there is a narrowing and aberrant vessels, portal vessels will open up. Here you can see that CD34 positive endothelial lining, this whole portal vein is now replaced by aberrant vessels. These are normal, but in patients with IPH, you can see that there is multiple areas of new vessels, and there is thickening of the portal vein, the portal vein diameter is reduced compared to what you see normal. Here one more time, there is a muscular hypertrophy in the large size portal vein. So what happens in the beginning, these are the normal, here is the portal vein, the portal vein gets obliterated due to injury and the infection, over a period of time, the portal tracts come closer, fibrous tissue gets laid down, and this is obliterated portovenopathy that you see. This can finally lead to linkage and incomplete septal cirrhosis in these patients. So whether the infections occur in your natal period, umbilical vein, catheterization, or umbilical sepsis, or in the young children or in adolescent period, there is either blockage of the portal vein, that is EHPVO, or the second, third order branch of the portal veins. And there is endothelial cells, they transform to di-fibroblast, and you can see that this leads to narrowing and obliterated portovenopathy. You require animal models to identify the disease process, because by the time patient comes to you, he already has his spleen varices. Several models were developed, first by Kunio Koda in dog, of E. coli and T. coli. Over a period of time, we developed several models of schistosomiasis, and others, but I would like to stretch, stress upon the endotoxin-induced portal hypertension. In a rabbit, you cannulate the gastrosplenic vein, and this cannula comes out from the nape of the neck, and can stay for a few months patent. Through this cannula, you keep injecting libopolysaccharide over time, and wait for six months period, and you see that sphenomegaly develops, these animals develop hemodianandrogens, and you can see that the sphenomegaly develops, portal hypertension like non-serotic. You see rise in portal pressure, mean arterial pressure comes down, heart rate remains normal, and the liver histology is safe. You maintain this for six months, compared to the sham, you have endotoxin-induced portal hypertension, portal pressure remains high. So this is a true reflection of endotoxin-induced portal hypertension, and development of non-serotic portal fibrosis. So one more time, the endothelium of the portal vein, which has the inner lining, there is a change of TGF beta-1, there is conversion into myofibroblast cells, and endothelium to mesenchymal transition occurs, and there is obliterated portal venopathy. Nodular regenerative hyperplasia, a term which is commonly used in many parts of Europe, US, and in Japan. The incidence of this is between 27% or so, and obliterated portal venopathy, which is seen in about 2% in all autopsy series, and about 0.5% of all liver biopsies. People, the patients can have an overlap between NRH and obliterated portal venopathy, and especially seen in patients with HIV who are on heart therapy. This study, published in Hepatology of 43 patients, had 22 as diagnosed as NRH, 13 as portal vein thrombosis, and miscellaneous verrate. What is slightly disturbing is that advanced compensated chronic liver disease may have almost similar LSM range as of NRH or PVT. Therefore, liver stiffness is not a reliable indicator in them. Liver biopsy is helpful. When you look at NRH, there could be multi-acinar lesions of liver in the absence of any fibrous septum. So these are multi-acinar lesions, and there is mainly ischemia and obstruction of the small hepatic vessels. We can see here, there is thickening of the portal vein and periportal fibrosis. Liver functions in these patients are normal. Spleen is massive, varices do develop. HVPG is normal. And this can be because of drugs like azathioprine, myelodysplastic diseases, autoimmune diseases, or common variable immune deficiency syndromes. Such patients very often require surgical interventions. This data has also been shown by Dr. Brenner and several other series from Japan. Here is a recent case report of chronic arsenicosis. You can see the skin lesions, the raindrop kind of picture seen of arsenicosis. Large spleen, development of collaterals. Again, the obliterated portal venopathy, periportal fibrosis. These are seen rarely, but often still seen in some parts of India. There are three Japanese surveys I'm reporting, 1999, 2005, and 2015. And three diseases, IPH, EHPVO, and butt caries syndrome. You can see that the incidence per million of IPH or NCPF has come down. EHO is also down because of the improved hygienic condition, but Chiari syndrome on the other hand remains same. So in Japan is still you have 1000 patients of IPH seen over a period of time. Let me briefly touch with you EHPVO where the main portal vein is replaced by cavernoma formation. And this occurs due to infection in young children where the main portal vein has blocked and the HEPG would therefore be normal, spleen is moderate to massive, no parenchymal dysfunction unless it is long standing and albumin and iron are normal. Here you can see portal cavernoma, in the CT scan there is no portal vein, it is replaced by cavernoma. There is ascites, liver may look a little shrunken irregular, but it is due to parenchymal extinction. Another patient same. Now local factors, both in pediatric and adult series have been studied. You can see that in pediatric umbilical vein catheterization or sepsis has been commonly reported. However, in adults, this is not being reported. Majority of patients may have unexplained portal vein thrombus. Umbilical vein injury and EHPVO, there are several series have been shown. However, in some no portal vein thrombus, so I think it is the endothelial injury rather than the infection per se. Prothrombotic states, both in pediatric and adult series have been studied and the deficiency of protein CNS has been commonly reported. However, several gene mutations, MTHFR and others have been commonly reported. So there is an underlying prothrombotic state in these patients. I give you a unifying hypothesis. Infections are the major player of the portal vein, portal pyemia. If it occurs in newborn and is severe, it can manifest as portal vein thrombus in the childhood itself. And this can lead to very severe bleed and many other like growth retardation and other complications. If the infections are mild, they are chronic, they will manifest in adulthood. Chronic antigenemia remains the major, it can have other exposures of arsenic or copper. This presinusoidal fibrosis leads to stimulation of RE system, non-serotic portal fibrosis develops. So this presinusoidal syndrome of increased resistance has to have a prothrombotic state in the vessels. This leads to portal hypertension and EHPVO and NCPA are two major manifestations. So let me summarize, NCPH is a vascular disease. It is present in 2 to 20% of all cases of portal hypertension and CPF, one of the variants of NCPH has well tolerated bleeds, anemia, massive spleen, liver functions normal with the patent portal vein and the major central lesion is obliterated portal venopathy. If in Japan, females are dominant, otherwise males are dominant. Chronic endotoxemia is the major player and HIV can add to some of these in adults. I have not covered drugs and chemicals which are commonly seen with NRH and PSVD. Pre and perisinusoidal gradients are the two major gradients, HVPG remains normal. In EHPVO, there is thrombosis of the main portal vein due to portal pymia and endotoxemia occurs in young children and they develop portal kevalin. To take home message remains, NCPH should not be ignored, should be suspected, managed and these patients can have near normal life. Thank you for your attention. Good afternoon, everybody. Now, we come to the last topic of the first half of the session on controversies in the diagnosis and management of non-serotic portal hypertension in children and adults and I would be presenting a case with a diagnostic dilemma. I am Anshu Srivastava from HDPGI Lucknow, India and I have no disclosures. Going on to the case, this is a 12-year-old female who has been symptomatic for last three years, presented to us with three episodes of hematomasis and melina and a progressively increasing lump with mild driving discomfort in the left upper quadrant. The episodes of hematomasis were large volume, painless and associated with postural symptoms, but all the three bleeds were well tolerated with no jaundice, ascites, edema or altered sensorium. No fever or NSAID intake before the GI bleeds and she was managed outside with intravenous fluids and blood transfusion, but no endoscopy was done. This child did not have any history of bleeding from other sites or any neonatal events or any episodes of intra-abdominal sepsis or recurrent skin, GI or respiratory infections or poor scholastic performance or any autoimmune features and she was relatively healthy with a good appetite and no weight loss or effort intolerance in between the episodes. On a personal front, she was a student of class 7, developmentally normal, attained menarche a month ago with no chronic illness or intake of alternative medicine and the family history was non-contributory in terms of liver disease, splenomegaly or hematomasis. On examination, she was a conscious, alert, well-oriented child with stable vitals. Both the weight and height were above the 50th centile. She had pallor, but she did not have any itchiness or any other stigma of chronic liver disease. She was maintaining saturation in room air and the JVP was normal. Abdominal examination showed a large spleen about 12 centimeters or just 1 centimeter palpable liver with a span of 11 centimeters. No prominent veins over the anterior abdominal wall or the back. There was no ascites. Kidneys were not palpable and the cardiovascular, respiratory and central nervous system examination was non-contributory. So, to summarize, we have a patient who has presented to us with a large spleen with hematomasis, which takes us towards the presence of portal hypertension and there was no growth failure or stigmata of cirrhosis of chronic liver disease. So, Simon, at this stage, I would like to ask you as to what would be your differential diagnosis and how would you like to investigate this patient? Thanks, Anshu. So, interesting presentation, a 12-year-old girl with upper GI bleeding. And as we go through the history, it's really notable the lack of positives. So, no real indicators what's going on, even to the point of including the possibility of a GI cause for the bleeding rather than a hepatic cause. Then we get to the physical exam and we have splenomegaly, very pronounced splenomegaly and relatively much less pronounced hepatomegaly. So, in thinking through how we classify this kind of presentation, well, we have the possibility of GI causes, but much more likely now is a hepatic cause, which could be cirrhotic. We could be thinking of an autoimmune or a PSC or a Wilson disease, less commonly a viral hepatitis in children, other causes of cirrhosis. Or a non-cirrhotic cause, and the most likely one there would be a portal vein thrombosis with the possibility of other non-cirrhotic causes of portal hypertension. Although, again, in the history, there's nothing to give us clues, so no associated conditions, no prior malignancy, no interesting drug therapies, and no associated rheumatological conditions, for example. So, how would I investigate? Well, I would like, first of all, an ultrasound scan to look at vasculature and to look at liver parenchymal texture, nodularity, et cetera. And I'd like a blood workup to include the basics of liver enzymes, albumin, INR, CVC, to see where the platelet count is at. And I would usually, at this point, put in the initial workup for causes, the more common causes of cirrhosis, autoimmune markers, viral serologies, cerebral plasma, et cetera. Okay. All right. So, on investigations, this child had anemia, leukopenia, and thrombocytopenia, suggesting the presence of hypersplenism. The liver function tests were relatively preserved. Synthetic functions were normal, and there was only a mild elevation of SGOT and NPT. Renal functions, electrolytes, and sugars were normal. The endoscopy showed the presence of large esophageal varices, which were mandibular, and this child also had gastric varices, as you can see in the picture. The ultrasound Doppler showed that the liver was 14 centimeters, mildly enlarged for this age. There were no intrahepatic collaterals, portal vein was patent, there was no cavernoma, all the hepatic veins and IVC were patent, and there was no ascites, and the kidneys were normal. So, at this point of time, after having excluded more or less EHPVO and Buttkerry syndrome, we are basically boiling down to an intrahepatic portal hypertension, more likely to be non-sorotic than sorotic, and for confirmation of diagnosis, we have the options of etiological workup of cirrhosis, transdupilar biopsy, advanced abdominal imaging, transient elastography, and HPTG measurement. All would have to be done in one or the other order. And so, I would like to ask Simon as to, considering the age of this patient, and especially if the patient would have been a little younger, how easy or difficult it would be to do an HPTG measurement under transdupilar biopsy? Yeah, thanks, Anshu. So, transjugular liver biopsy is certainly something that is well reported, several case series showing that we can get good tissue in children through the transjugular route, and that it is a safe procedure. Age ranges down in most of those series to mid-infancy, and I think the risks of adverse outcomes increase in the smaller newborns and preterm newborns. So, a transjugular liver biopsy, definitely something we would consider in the setting of ascites or significant coagulopathy that complicates an attempt to do this through the percutaneous route. Hepatic venous pressure gradient measurements in children have been achieved. There are a small number of small case series in the literature looking at this measurement in children, showing, I think, primarily that it is feasible and that it is safe. These are retrospective studies, I think, capitalizing on cases that are having transjugular biopsies done. What's much less clear is what we should do with the results of hepatic venous pressure gradient measurement in children and what those results mean, particularly in the non-sorotic portal hypertension setting. I think it's such a rare condition in children, and we just don't know what the HPPG measurements mean. I have concerns about the venous-to-venous shunting that has been shown in the previous studies of HPPG measurements in children, particularly common in those with biliary atresia, complicating further our ability to interpret the results. Okay. So, Theo, would your approach be different in the case of an adult? What would be your... I think you'll have to unmute yourself. Sorry, not very different. Once we have the signs of portal hypertension that you so elegantly laid out, we will proceed with the workup, the most important part of the workup being a liver biopsy. Prior to the liver biopsy, as Dr. Ling alluded to, we would already obtain serologies to look for other causes of liver disease, and we would image the patient looking for, particularly at the vasculature and for clues. The larger spleen would make us think more of non-sorotic portal hypertension than you would see in cirrhosis. But two caveats. The first is that the liver biopsy should be read by an expert pathologist. The second is that sometimes once the diagnosis is made, the patients are best managed in centers with expertise in managing non-sorotic portal hypertension. Dr. Sreen, you have a lot of experience in doing HPPG, so can you comment upon that a little further? First, my compliments, Anshu, for a very good workup, three or four clinical tips for people in our part of the world. It's an easy case because you have a 12-year-old girl who has achieved menarche, and she has normal menstruation and normal growth. So the differential diagnosis of this patient would be non-sorotic. In non-sorotic, whether it is portal vein thrombosis, EHPVO, or it is non-sorotic with patent portal vein. And also you show that in ultrasound, the portal vein is patent. So it remains only one possible diagnosis that is non-sorotic portal fibrosis, NCPF. Now the role of HPPG in transducer, even in children and adults, this is required that you measure the pressures because NCPF and EHPVO, both are vascular diseases. They are not parenchymal diseases. So normal HPPG confirms, it excludes cirrhosis. And having done that, you have a transjugular biopsy, which will tell you. So I think both are required for a final confirmation of the diagnosis. So we went on to investigate the patient, and as you can see here that the workup for Wilson's disease was negative. The autoantibodies were negative, and the B, C, and HIV were also negative. Simultaneously, a transjugular liver biopsy was done, which showed that there was unremarkable hepatocytes with no evidence of cirrhosis or congenital hepatic fibrosis, and no features on histology to suggest the presence of NRH. As you can see from these slides, that the portal tract was showed expansion with some delicate fibrotic scepter, and there were multiple thin-walled aberrant channels which were located both inside the portal and at the portal hepatocyte interface. The liver architecture was largely maintained on the reticulum staining. On going on to the further higher magnifications, we can see sinusoidal dilatation, and we can see these multiple aberrant microvasculatures. The portal vein showed some fibrosis around it, and some of the branches of the portal vein were reduced in caliber. These features suggest the presence of an obliterated portal venopathy, although may not be the very classical ones. And simultaneously, a triple phase CT also was done, which confirmed the presence of a patent portal vein, and the splenic vein was dilated, and the liver was mildly enlarged, and there was no ascites. So, as you can see in these pictures, the spleen was enlarged, there was presence of splenic infarcts. This is a dilated splenic vein, the portal vein was dilated and patent, and this is a typical find where the left branch of the portal vein showed narrowing of the caliber, and the hepatic vein and IVC was patent. The transient elastography study showed that the liver stiffness was slightly raised at 16, splenic stiffness was not available, and HVPG unfortunately was not measured in this patient. On the immunological front, there was no suggestion of an immunodeficiency on history. The gamma globulins, HIV, and autoantibodies were negative, and thrombophilia workup was not done. So, now I would request Dr. Sareen to comment upon how useful is the ratio of splenus to liver stiffness or the actual value of liver stiffness in differentiating between cirrhosis and idiopathic non-cirrhotic portal hypertension? Well, this is a difficult question because the pediatric probe of the elastography is not very well standardized, and as you see in this particular patient, the LSM is 16 kilopascal. And it would put anybody to say, yes, it is cirrhosis of liver, but the biopsy doesn't show it. So, if I have a child with that biopsy, I would think it is NRH rather than cirrhosis. Number two, splenic stiffness certainly is very helpful to diagnose portal hypertension, but not really NCPH. So, both splenic and liver stiffness are large range. I would not recommend them as of now into our routine diagnostic armamentarium. I certainly think HUPG should have been done. If not, it should become the routine because it differentiates the two gradients. One is spleen to the liver, that is splenohepatic, and one is presinusoidal and perisinusoidal to the postsinusoidal, that is the second gradient. And this explains why in these patients, you may have slightly higher LSM. As far as workup is concerned, if you were asked immunological workup, well, in a 12-year-old child, it may not be needed, but yes, in several parts of the world, there are reports that even children may have some immunological deficiency, so you may do it. But by and large, I think thrombophilia workup is required, but that is very important to find out the cause. Dr. Silvia, would you like to comment upon this workup of immunological causes and thrombophilia? Thank you. Yes. So, I think what is really remarkable in this case is the fact that we still don't know the cause despite all the comprehensive workup that you did very well for this child. So, in these cases where we still don't have the cause, I would favor, in accord with Dr. Serene's comment, to go ahead and get the thrombophilia workup to see if we can have more information about why this child developed this severe phenotype. And I would argue as well that in some parts of the world, like I would include the U.S. where I practice, that it should be part of the algorithm to further investigate these patients that still don't have an etiology for their non-sorotic portal hypertension to be evaluated for immunological diseases, as more emergent literature is finding a leak between immunological effects and non-sorotic portal hypertension, and the spectrum of disease is still not clear. Okay. So, basically, our diagnosis was a non-sorotic portal hypertension with hypersplenism and variceal bleed. As we have evidence of portal hypertension, we have excluded EHPVO and bud theory. We have no evidence of cirrhosis on all fronts of evaluation. The liver histology is suggestive of a non-sorotic portal hypertension, and there is no CHF. The etiology in this case appears to be most likely idiopathy, although we have not done a thrombophilia workup. So, what is the role of genetic testing in this scenario, Dr. Silvia? Would you like to comment upon that? Thanks for your question. So, as I discussed in the beginning of this session of lectures, emergent data is evolving, showing that some genetic defects can actually be the cause of cases that are now labeled as idiopathic. So, in cases where we don't have a cause, despite the comprehensive workup, we should definitely consider genetic testing. I acknowledge that this is not something that can be done in all hospitals around the world, but if a physician finds him or herself in this situation, she should definitely refer to a center where proper genetic testing could be done, and I would argue that whole exome sequencing would be the most cost effective in these cases, and I'll be happy to answer any questions or discuss it further. Okay. So, I would like to summarize by saying that idiopathic non-sorotic portal hypertension is a disease spectrum which ranges from the patients with no portal hypertension to portal hypertension. Under good liver biopsy, which has been reported by expertise, people with expertise is absolutely essential. A cross-sectional imaging measurement of HVPG is very vital because it helps in making the diagnosis. An immunological and thrombophilia screen should be done in all patients, and a genetic workup in idiopathic and familial cases, and maybe with more knowledge we might be needing it to the indications would become more clear to us. And with this, I would like to thank all my panelists and the audience also for patient listening. Thank you very much. Thank you.

non-cerebral portal hypertension

diagnostic approach

cirrhosis differences

causes

manifestations

genetic mutations

KCNN3

DGO kappa

imaging findings

portal vein thrombosis

case study

12-year-old female