Introduction

Portal vein thrombosis (PVT) is common in cirrhotic patients. The prevalence ranges from 1%, at earlier stages to 30% in candidates for liver transplantation. The ADAM (a disintegrin-like and metalloproteinase) proteins, with their subfamily of ADAMTS -13 (A Disintegrin and Metalloprotease with Thrombospondin-13) are produced exclusively by hepatic stellate cells and play an important regulatory role in coagulation. The ADAMTS-13 protein is an enzyme, Willebrand factor cleaving protease. In vascular injury, von Willebrand factor (VWF) together with platelets initiates formation of thrombus. Functional deficiency of ADAMTS-13 results in presence of unusually large VWF multimers in plasma, which promotes platelet aggregation and thrombus formation.

Aims & Methods

The aim was to investigate the relation of serum level of ADAMTS -13 and portal vein thrombosis in patients with HCV related liver cirrhosis.The study was carried out on 56 Egyptian patients with HCV related liver cirrhosis (LC) (LC group: 28 cirrhotic patient without PVT and LC+PVT group: 28 cirrhotic patient with cirrhosis and PVT) and 28 healthy volunteers as controls from Menoufia University Hospitals (Egypt). The study was conducted in period from March 2020 to April 2021. They underwent history taking, physical examination and laboratory investigations (CBC, liver profile {ALT, AST, serum albumin, bilirubin, INR}, urea, creatinine, HBsAg, HCV Ab and serum ADAMTS-13 assay). Abdominal ultrasound with Doppler of portal vein (PV) was done for all patients and controls. Triphasic CT of the liver was done for patients with PVT to exclude hepatocellular carcinoma (HCC). Child-Pugh score, MELD score, AST/Platelet ratio index (APRI score) and FIB-4 were calculated for all patients. Patients on anticoagulant therapy, age less than 18 years, inflammatory or bleeding events in previous 2 months, history of personal unprovoked thromboembolism (familial in first‑degree relatives), etiology of liver cirrhosis other than HCV and HCC are excluded from the study.

Results

Serum ADAMTS-13 (ng/ml) was significantly lower in LC+PVT group (0.66±0.44) than in LC (0.92±0.99) and control group (0.96±0.92) (P=0.0001 for both) but there was no significant difference between LC group and controls (P=0.2). ADAMTS13 at cutoff level 1.62 ng/ml have 71.4% sensitivity and 62.5% specificity for prediction of PVT. In LC+PVT group, level of serum creatinine (mg/dl) (1.5±0.5) and INR (1.7±0.6) were significantly higher than LC group (creatinine 1.28±0.37) (INR 1.29±0.34) (P=0.03 and 0.001 respectively). There was no significant difference between LC+PVT group and LC group as regard, history of encephalopathy, history of gastrointestinal bleeding, presence of ascites, CBC, liver profile (ALT, AST, serum albumin, bilirubin), urea, APRI score, FIB-4, and MELD score (P˃0.05). Regression analysis showed that ADAMTS 13 level and advanced degree of Child-Pugh score predict occurrence of PVT by 1.26 and 4.66 fold respectively. In LC+PVT group, there was a positive correlation between ADAMTS 13 level and INR (r=0.41, P=0.03) but there was no correlation with other parameters (age, platelet count, Hb, albumin, AST, ALT, bilirubin, urea, creatinine and INR, APRI score, FIB-4, MELD score and Child-Pugh score (P˃0.05). In LC group, there was a positive correlation between ADAMTS 13 and Child-Pugh score (r=0.41, P=0.03) and MELD score (r=0.46, P=0.02).

Conclusion

Low levels of ADAMTS-13 are associated with pathogenesis of PVT in liver cirrhosis. ADAMTS-13 concentration is potential biochemical marker in diagnostic strategy of PVT in patients with cirrhosis.

References

1-Hugenholtz GC, Adelmeijer J, Meijers JC, Porte RJ, Stravitz RT, Lisman T. An unbalance between von Willebrand factor and ADAMTS13 in acute liver failure: implications for hemostasis and clinical outcome. Hepatology. 2013 ;58(2):752-761.
2-Takaya H, Uemura M, Fujimura Y, Matsumoto M, Matsuyama T, Kato S, Morioka C, Ishizashi H, Hori Y, Fujimoto M, Tsujimoto T. ADAMTS13 activity may predict the cumulative survival of patients with liver cirrhosis in comparison with the Child‐Turcotte‐ Pugh score and the Model for End‐Stage Liver Disease score. Hepatology Research. 2012 ;42(5):459-472.
3-Lancellotti S, Basso M, Veca V, Sacco M, Riccardi L, Pompili M, and De Cristofaro R. Presence of portal vein thrombosis in liver cirrhosis is strongly associated with low levels of ADAMTS-13: a pilot study. Internal and emergency medicine. 2016; 11(7):959-967.
4-Mikuła T, Kozłowska J, Stańczak W, Sapuła M, Różyk A, Wiercińska- Drapało A. Serum ADAMTS-13 levels as an indicator of portal vein thrombosis. Gastroenterology research and practice. 2018; 1-12.

Introduction

The advent of direct-acting antivirals (DAAs) is a major breakthrough in hepatology representing the therapeutical standard of care in patients with chronic hepatitis C virus infection. Despite high rates of sustained virological response (SVR), DAAs therapy doesn’t eliminate the risk of thrombotic events.

Aims & Methods

We aimed to assess the prevalence of nontumoral portal vein thrombosis (PVT) after SVR.We prospectively analyzed a cohort of patients with HCV-related liver cirrhosis treated with paritaprevir/ritonavir, ombitasvir and dasabuvir (PrOD) ± ribavirin and ledipasvir/sofosbuvir (LED/SOF) ± ribavirin for 12/24 weeks, in a gastroenterology center from Romania, between January 1^st 2016 and July 1^st 2021. All patients with presumption of thrombosis were evaluated by vascular Doppler, abdominal ultrasound and confirmed by CT scan.

Results

The study included 730 patients treated with DAAs, of which 35 were diagnosed with non-malignant PVT after-SVR (15 men and 20 women, mean age 57.86 ± 7,068 years), corresponding to a prevalence of 4.8%. The mean time from SVR to complication was 290.00 ± 116.639 days. Most patients with nontumoral PVT received LED/SOF (71.4%), while the rest received PrOD (28.6%). During the study, an improvement in the Child-Pugh and MELD score was observed at the SVR. The evolution changes slightly at the 48-week assessment, with a slight increase in the proportion of patients in the Child B class and MELD ≥ 15. The pro- and anticoagulant factors evaluated reflect the classic hemostatic profile of patients with liver cirrhosis and PVT, characterized by increased FII, FVIII and FvW and decreased anticoagulant factors (PC, PS, ATIII).

Conclusion

The study included 730 patients treated with DAAs, of which 35 were diagnosed with non-malignant PVT after-SVR (15 men and 20 women, mean age 57.86 ± 7,068 years), corresponding to a prevalence of 4.8%. The mean time from SVR to complication was 290.00 ± 116.639 days. Most patients with nontumoral PVT received LED/SOF (71.4%), while the rest received PrOD (28.6%). During the study, an improvement in the Child-Pugh and MELD score was observed at the SVR. The evolution changes slightly at the 48-week assessment, with a slight increase in the proportion of patients in the Child B class and MELD ≥ 15. The pro- and anticoagulant factors evaluated reflect the classic hemostatic profile of patients with liver cirrhosis and PVT, characterized by increased FII, FVIII and FvW and decreased anticoagulant factors (PC, PS, ATIII).

Introduction

Cirrhosis is a condition of hypercoagulability that increases the risk of portal vein thrombosis (PVT) contrasting with thrombocytopenia related to hypersplenism. In this context, we propose to determine the profile of platelet parameters at the diagnosis of cirrhosis and to study their relevance in predicting PVT in this population.

Aims & Methods

This is a retrospective case-control study involving cirrhotic patients diagnosed between 2000 and 2015, at the Gastroenterology department of La Rabta Hospital. Epidemiological, clinical, and biological data were collected from medical records. Platelet parameters, including platelet count (PC), mean platelet volume (MPV), and platelet distribution width (PDW) were recorded at diagnosis. The MPV/PC, PC/lymphocytes, and PC/splenic index ratios were calculated. The occurrence of portal vein thrombosis, documented by abdominal Doppler ultrasound, was recorded. Statistical analysis was performed using SPSS 23.0 software.

Results

A total of 103 patients were included in our study. The gender ratio was 0.5, and the median age at diagnosis of cirrhosis was 59 years [14-80]. The etiological distribution of cirrhosis was as follows: post-hepatitis C cirrhosis (n=47), post-hepatitis B cirrhosis (n=35), cirrhosis due to other etiologies (n=10), and undetermined for 11 patients.
The median platelet count (PC) for the entire population was 85 G/L [16-480], with thrombocytopenia in 82% of cases. Portal vein thrombosis (PVT) was diagnosed in more than 1/5 of cases (n=23). The splenic index was significantly higher in the "cirrhosis with PVT" group (15.865 cm vs. 14.646 cm in the "cirrhosis without PVT" group; p=0.04).
However, no statistically significant differences were found between the two groups regarding platelet parameters such as PC, mean platelet volume (MPV), platelet distribution width (PDW), and MPV/PC ratio (respectively: 92 G/L vs. 83 G/L, p=0.9; 10.9 fl vs. 11.35 fl, p=0.2; 15.5% vs. 15.1%, p=0.2, and 0.12 fl/G/L vs. 0.11 fl/G/L, p=0.9). Nevertheless, the PC/lymphocytes ratio showed a modest discriminating power between the two groups, with an area under the curve (AUC) of 0.6; 95% CI: [0.475-0.748].
In addition, 19% of patients developed hepatocellular carcinoma (HCC) during the course of their disease. Analysis showed that the presence of HCC increases the risk of PVT occurrence (odds ratio 3, 95% CI: [1.052-8.679], p=0.03).

Conclusion

Platelets are key players in the inflammatory response found in several pathologies. In our study, the predictive power of platelet parameters for portal vein thrombosis could not be demonstrated due to the presence of other likely confounding factors (infection at diagnosis, used automated methods...). A larger prospective study would be desirable.

Introduction

Alcohol related liver disease, and particularly alcohol related hepatitis, is characterised by specific hepatic and systemic histo- and physiopathological changes (inflammation, ballooning, fibrosis, cholestasis, etc) [1]. On the other hand, increased HVPG was demonstrated to predict a worse outcome in AH [2]. In this study, we aimed to detect which specific AH changes are associated with an increased HVPG. Secondly, we analyzed the interrelation between canalicular cholestasis, systemic inflammation markers, and sepsis.

Aims & Methods

In this study, we aimed to detect which specific AH changes are associated with an increased HVPG. Secondly, we analyzed the interrelation between canalicular cholestasis, systemic inflammation markers, and sepsis.
We retrospectively re-evaluated the transjugular liver biopsy (TJLB) specimens of 90 patients with alcohol-related liver disease between 2016 and 2020 and rescored them using the SALVE scoring/grading system.
Serum from the biobank was used for ELISA tests for cytokeratin18 (CK18), heat shock protein 90 (Hsp90), toll-like receptor 4 (TLR4), tumor necrosis factor-alpha (TNFα), and lipopolysaccharide (LPS).
We analyzed the correlation between HVPG, systemic inflammation, infection, and the SALVE grading/staging criteria.

Results

Of the 90 enrolled patients, 81 went through TJLB and HVPG measurement. The gender ratio was 24 (29.6%) female to 57 (70.4%) male. The age range was 28-73 (median of 45). Median MELD score and MaddreyDF scores were 22 (6.61 SD, 95% CI) and 82.5 (37.57 SD, 95% CI), respectively. Median HVPG was 20 mmHg (4.9 SD, 95%CI). The biopsy length had a median of 20 mm (range 8-80 mm).
In the univariate analysis, ductular cholestasis was correlated with the prognostic scores (MELD, Maddrey DF, Child-Pugh), which was confirmed at the multivariate analysis (p<0.05). The canalicular type of cholestasis was correlated in both univariate and multivariate analysis with the scores, but also with the number of leucocytes, neutrophils, platelets, the C-reactive protein (CRP), bilirubin level, and HVPG ( p<0.05).
HVPG was correlated with canalicular cholestasis, but not with the ductular type. There was a significant difference in the mean HVPG value between patients without canalicular cholestasis (M=19.14; SD=4.79) and those with canalicular cholestasis (M=21.61; SD=4.84); t(80)=2.31, p=0.023.
One-way ANOVA revealed a statistically significant difference in the HVPG value between the non-cirrhotic and the 3 cirrhotic stages, i.e., SALVE fibrosis stage (SFS) 4A, 4B, and 4C (F(3, 78)=3.99; p=0.01). HVPG did not show a correlation with other histological criteria, such as steatosis, ballooning, and neutrophils.
The systemic inflammation markers CK18, Hsp90, TLR4, TNFα, and LPS did not correlate with cholestasis in the univariate analysis.
Patients who had canalicular cholestasis on histology also had a higher probability of being infected (Pearson Chi square p=0.041) than those with the ductular type of cholestasis (p=0.611).
Overall survival was not influenced by the presence of cholestasis or the canalicular/ductular type (Chi square test p=0.476).

Conclusion

From the histological SALVE criteria, HVPG is correlated with cholestasis, particularly the canalicular type, and the SFS 4A-C.
Cholestasis was found to be an independent predictor of infection in these patients, although, in this vicious circle, infections and sepsis are usually the cause of cholestasis.

References

[1] Lackner C, Stauber RE, Davies S, Denk H, Dienes HP, Gnemmi V, Guido M, Miquel R, Paradis V, Schirmacher P, Terracciano L, Berghold A, Pregartner G, Binder L, Douschan P, Rainer F, Sygulla S, Jager M, Rautou PE, Bumbu A, Horhat A, Rusu I, Stefanescu H, Detlefsen S, Krag A, Thiele M, Cortez-Pinto H, Moreno C, Gouw ASH, Tiniakos DG. Development and prognostic relevance of a histologic grading and staging system for alcohol-related liver disease. J Hepatol. 2021 Oct;75(4):810-819. doi:10.1016/j.jhep.2021.05.029.
[2] Rincon D, Lo Iacono O, Ripoll C, Gomez-Camarero J, Salcedo M, Catalina MV, Hernando A, Clemente G, Matilla A, Nuñez O, Bañares R. Prognostic value of hepatic venous pressure gradient for in-hospital mortality of patients with severe acute alcoholic hepatitis. Aliment Pharmacol Ther. 2007 Apr 1;25(7):841-8. doi: 10.1111/j.1365-2036.2007.03258.x.

Disclosure

The authors have nothing to declare.

Introduction

Transjugular intrahepatic portosystemic shunt (TIPS) is now an effective and standard procedure to prevent variceal bleeding and recurrent ascites in decompensated cirrhosis (1). Moreover, there is increasing evidence (2, 3) showing that TIPS is feasible in patients with partial or occlusive portal vein thrombosis (PVT). However, thrombosis may develop at intrahepatic branches of the portal vein after TIPS, even in patients without PVT before the procedure. Our retrospective study showed that warfarin could prevent portal branch PVT and might improve liver function among cirrhosis patients receiving TIPS (4). Thus, to obtain clinical evidence with higher quality, a prospective study on post-TIPS de novo PVT regarding the incidence, risk factors, and impact on liver function is necessary.

Aims & Methods

Aim: We aimed to assess the incidence and risk factors for thrombosis developed at branches of the portal vein after TIPS in cirrhosis patients without preexisting PVT. In addition, we aimed to evaluate the influence of post-TIPS PVT on liver function and cirrhosis-related events.
Methods: A total of 70 decompensated cirrhosis patients who received TIPS without PVT were included. Routine laboratory and radiological examinations and assessment of cirrhosis-related complications proceed at 1, 3, 6 and 12 months after TIPS. For the primary outcome, the presence of post-TIPS PVT was assessed by doppler ultrasound showing that the blood flow signal partially or completely disappeared or by contrast-enhanced computed tomography showing filling defects or occlusion in the portal vein branches. For the liver function and complications comparison within 1 year after TIPS (secondary outcome), a 1:1 match using propensity score matching (PSM) based on age, sex and preoperative Child‒Pugh score was conducted between patients with and without post-TIPS PVT.

Results

Among 70 patients, 34 (48.6%) were Child‒Pugh class A, and the median Child‒Pugh score was 7 (5-8). The most common indication for TIPS was variceal bleeding (76.8%), followed by refractory ascites (21.4%). During the follow-up, 25 (35.7%) patients developed post-TIPS PVT, with incidences of 8.0%, 16.0% and 26.8% at 1, 2 and 3 months after TIPS, respectively. Univariate analysis showed that lower serum albumin (p = 0.023), Child‒Pugh B/C class (p = 0.029) and stenting at the left branch of the portal vein (p = 0.020) were significantly associated with post-TIPS PVT. Multivariate analysis showed that lower serum albumin (OR 0.87, p = 0.022) and stenting at the left branch of the portal vein (OR 6.56, p = 0.020) were independent risk factors. Twenty four of 25 patients with post-TIPS PVT were matched with 24 of 45 patients without post-TIPS PVT. Preoperative clinical characteristics of the matched patients were compared, and no significant differences were shown between the two groups. However, the Child‒Pugh score was significantly higher in the PVT group at 1 year after TIPS compared with non-PVT group [7 (6-8) vs 8 (7-9), p = 0.044]. In addition, the incidence of overt hepatic encephalopathy (50.0% vs 20.8%, p = 0.035) and the presence of ascites (54.2% vs 25.0%, p = 0.039) at 1 year after TIPS were significantly higher in the PVT group.

Conclusion

PVT may develop at intrahepatic branches of the portal vein early in the post-TIPS period in cirrhosis, leading to further liver damage and cirrhosis-related complications. Lower serum albumin levels and stenting at the left branch of the portal vein were independent risk factors for de novo PVT within 1 year after TIPS.

References

1.de Franchis R, Bosch J, Garcia-Tsao G, Reiberger T, Ripoll C; Baveno VII Faculty. Baveno VII - Renewing consensus in portal hypertension. J Hepatol. 2022 Apr;76(4):959-974.
2.Lv Y, Bai W, Li K, Wang Z, Guo W, Luo B, Wang J, Wang Q, Wang E, Xia D, Li X, Yuan J, Han N, Niu J, Yin Z, Fan D, Han G. Anticoagulation and Transjugular Intrahepatic Portosystemic Shunt for the Management of Portal Vein Thrombosis in Cirrhosis: A Prospective Observational Study. Am J Gastroenterol. 2021 Jul 1;116(7):1447-1464.
3.Guo DF, Fan LW, Le Q, Huang CB. Transjugular intrahepatic portosystemic shunt for the prevention of rebleeding in patients with cirrhosis and portal vein thrombosis: Systematic review and meta-analysis. Front Pharmacol. 2022 Aug 16;13:968988.
4.Zhang L, Huan H, Tong H, Wei B, Wang Z, Liu C, Wu H. Warfarin prevented de novo portal vein thrombosis after transjugular intrahepatic portosystemic shunt: A retrospective study. Medicine (Baltimore). 2020 Jan;99(2):e18737.

Introduction

The gut-liver axis, the anatomical and physiological connection between the gut and liver, is hypothesized to contribute to the development of portal vein thrombosis (PVT) in patients with end-stage liver disease (ESLD). It has been proposed that the portal vein in patients with ESLD is an inflammatory vascular bed due to the translocation of endotoxins and proinflammatory cytokines from the gut. We hypothesized that a pro-inflammatory gut microbiome in ESLD patients, and elevated levels of the gut microbiota-derived metabolite trimethylamine N-oxide (TMAO) which is an established driver of thrombosis, may contribute to the development of PVT.

Aims & Methods

In this cross-sectional study we investigated whether differences in gut microbiome diversity, bacterial species, metabolic pathway abundances, and TMAO blood levels are associated with presence of PVT in ESLD patients. Fecal samples, serum samples and data from patients with ESLD, as well as living kidney transplant donors (‘healthy controls’), were collected from the TransplantLines Biobank and Cohort Study (NCT02811835). PVT was defined as presence of a thrombus in the portal vein assessed by either radiology or during surgery within a year prior to or after fecal sample collection. We analyzed the fecal samples using Shotgun Metagenomic Sequencing to assess and compare microbiome diversity, microbial taxonomy, and metabolic pathways abundances. TMAO levels were measured in serum samples using a Vantera® Clinical Analyzer (LabCorp, Morrisville, NC, USA).

Results

A total of 102 ESLD patients, of which 23 had PVT, and 246 healthy controls were included. We found no significant difference in microbiome diversity between patients with PVT (median Shannon Diversity Index (SDI): 2.08), and without PVT (median SDI: 1.84; P=0.18). Both ESLD groups had lower microbiome diversity compared to healthy controls (median SDI: 2.55; both P<0.05). A Beta Diversity Analysis confirmed that the gut microbiome of ESLD patients was markedly distinct from that of healthy controls. After false discovery rate correction, Bacteroides fragilis and three species from the family Clostridiales were increased in abundance in ESLD patients with PVT compared to those without PVT. No altered metabolic pathways abundances were observed between both ESLD groups. Overall, altered abundances of 29% (n=49) and 51% (n=86) bacterial species, and 18% (n=96) and 35% (n=191) metabolic pathways were observed between patients with and without PVT, respectively, when compared to healthy controls. TMAO levels were higher in both ESLD patients with (3.30µm) and without PVT (3.00µm) compared to healthy controls (2.70µm) (P = 0.13 and P = 0.84 respectively), with a trend towards higher levels in patients with PVT (P = 0.19).

Conclusion

Although we observed a profound difference in gut microbiota composition between patients with ESLD and healthy controls, our analysis revealed minimal differences in the gut microbiome composition between ESLD patients with or without PVT. These results indicate that a gut microbiome-derived pro-inflammatory state was not associated with the presence of PVT in these patients.

Introduction

The clinical implications of portal vein thrombosis (PVT) on the prognosis of acute variceal bleeding (AVB) in cirrhosis remain under debate. Most studies have not been specifically designed to assess the severity of PVT.

Aims & Methods

This study aimed to evaluate the severity of PVT and its impact on outcomes in patients with cirrhosis and AVB. From January 2016 to December 2016, a total of 1541 consecutive cirrhotic patients admitted for AVB, with (n=432; 28.1%) or without PVT (n=1109; 71.9%), were included retrospectively. The degree of luminal obstruction was used to classify the grades of PVT according to Baveno VII criteria. The primary endpoint was 6-week mortality.

Results

During the follow-up period, 179 patients (11.6%) died within 6 weeks. Patients with PVT had a higher risk of 6-week mortality compared to those without PVT (18.3% vs 8.9%; HR 2.12, 95% CI 1.52–2.94, p <0.001). Among them, those with malignant main PVT had a significantly elevated risk (39.3% vs 9.2%; HR 4.86, 95% CI 2.74–8.62, p <0.001). In the non-malignant PVT cohort, only completely occlusive main PVT was associated with increased 6-week mortality (completely occlusive 26.1% vs cavernous transformation 9.7% vs partially occlusive 6.5% vs minimally occlusive 5.9% vs without PVT 9.2%, p = 0.023). These results were consistent across various relevant subgroups.

Conclusion

Our study demonstrates that PVT is linked to a poorer outcome in cirrhotic patients with AVB, particularly in cases of malignant main PVT and completely occlusive main PVT.

Disclosure

None of the authors have any conflict of interest.