Bleeding risk of variceal band ligation in extrahepatic portal vein obstruction is not increased by oral anticoagulation.

BACKGROUND AND OBJECTIVES: Noncirrhotic nontumoral extrahepatic portal vein obstruction (EHPVO) is the second leading cause of portal hypertension (PHT) and is mainly related to prothrombotic disorders. Patients with EHPVO often require prolonged oral anticoagulation therapy (OAT) together with variceal band ligation (VBL) to prevent thrombosis recurrence and PHT-related bleeding, respectively. The benefit-risk balance of VBL in this context remains unknown. We aimed to assess upper gastrointestinal bleeding (UGB) risk and variceal eradication efficacy in EHPVO patients undergoing a VBL program without stopping OAT. PATIENTS AND METHODS: All patients with EHPVO treated (group A) or not (group B) with OAT and undergoing the VBL program were included between 2001 and 2010 in two tertiary French liver centers. We compared the incidence, source, and severity of UGB and variceal eradication efficacy. All EHPVO patients were then matched 1 : 1 with compensated cirrhotic patients with PHT not receiving OAT (group C) to compare UGB incidence and VBL efficacy. RESULTS: Forty-three EHPVO patients (30 with and 13 without OAT) and 43 cirrhotic patients were included for a total of 471 VBL sessions. The incidence of UGB was similar between group A (nine episodes/121 sessions) and group B (6/130), and tended to be higher in EHPVO patients (group A and B) than in cirrhotic patients (2/220). There was no difference between groups when considering bleeding source or severity and variceal eradication efficacy (84%). CONCLUSION: VBL can be performed safely and efficiently without stopping anticoagulation therapy in EHPVO patients.

Central obesity is associated with non-cirrhotic portal vein thrombosis.

BACKGROUND & AIMS: 30-40% of portal vein thrombosis (PVT) remains of unknown origin. An association between metabolic syndrome (MetS) and peripheral vein thrombosis has been reported but not with PVT, to date. The aim of this study was to investigate the association between MetS and PVT. METHODS: Between 2003 and 2014, all consecutive patients with non-cirrhotic PVT were prospectively included. Patient's characteristics and risks factors were recorded at the time of inclusion. Controls were selected by random in the general population and were matched 1/1 according to age and sex. RESULTS: Seventy-nine patients with PVT were included: 40 present with at least one risk factor for PVT (SPVT) and 39 were found to be idiopathic (IPVT). The prevalence of MetS was 25.6% in SPVT group vs. 47.4% in IPVT group and 17.9% in controls from the general population (C-IPVT: p=0.01). The waist circumference and body mass index were higher in the IPVT group than in the SPVT group (105 vs. 93cm, p=0.004 and 29.4 vs. 25.0kg/m(2), p=0.004) and in the C-IPVT group (105 vs. 92cm, p=0.001 and 29.4 vs. 25.8kg/m(2), p=0.003). Overweight was observed in 82.0% of patients in the IPVT group vs. 44% in the SPVT group (p=0.002) and 51% in the C-IPVT group (p=0.01). The mean visceral fat area was higher in IPVT than in SPVT (18,223mm(2)vs. 12,690mm(2), p=0.02). In multivariate analyses, an increase in waist circumference was the strongest parameter associated with idiopathic PVT. CONCLUSION: Central obesity is associated with PVT and could become one of the main risk factors for digestive thromboses.

A French national survey on the use of antibiotic prophylaxis in cirrhotic patients.

BACKGROUND: Guidelines recommend antibiotic prophylaxis (AP) in well-selected groups of cirrhotic patients, but the impact of these recommendations has not been assessed in France. AIM: To evaluate AP prescription tendencies for gastrointestinal bleeding, and primary and secondary prophylaxis of spontaneous bacterial peritonitis (SBP). METHODS: Practitioners (n = 1,159) working in general hospitals (GH) or in university hospitals (UH) received a self-administered questionnaire. RESULTS: Three hundred and eighty-nine (33.6%; GH 35% and UH 30.4%) practitioners responded. AP was prescribed by 97.7%, 72.3% and 94.8% of practitioners, without significant differences between UH and GH, respectively, for gastrointestinal bleeding (quinolones 48.2%, third-generation cephalosporins 27.7% and amoxicillin-clavulanic acid 22.2%), primary (quinolones 97.2%) and secondary prophylaxis of SBP (quinolones 99%). For gastrointestinal bleeding, ofloxacin (47.6%) and norfloxacin (37.4%) were the main quinolones prescribed, and ceftriaxone (77%) was the main third-generation cephalosporin prescribed. The principal reasons for prescribing AP were a decrease in bacterial infection (88.9% for gastrointestinal bleeding, 91.3% for primary and 94.3% for secondary prophylaxis of SBP), a recommendation by a consensus conference (83%, 38% and 74.4% respectively) and an improvement in survival (72.8%, 41.3% and 57.7% respectively). Only 31.7% of practitioners (39.6% for UH vs. 28.6% for GH; P = 0.038) believed that AP may reduce the risk of bleeding recurrence. Reported side effects (28%) of AP mainly concerned the risk of quinolone resistance (62% of cases). CONCLUSION: Antibiotic prophylaxis is well-recognized by French practitioners, but its routine use depends on the expertise of practitioners. Quinolones remain the main antibiotic class prescribed irrespective of the type of prophylaxis.

The Role of the 3' Untranslated Region in the Post-Transcriptional Regulation of KLF6 Gene Expression in Hepatocellular Carcinoma.

KLF6 is ubiquitously expressed in human tissues and regulates many pathways such as differentiation, development, cellular proliferation, growth-related signal transduction, and apoptosis. We previously demonstrated that KLF6 expression is altered during liver carcinogenesis. More importantly, KLF6 invalidation results in cell cycle progression inhibition and apoptosis of liver cancer cells. On the other hand, enforced expression of KLF6 variant 2 (SV2) induces cancer cell death by apoptosis. Thus, we and others demonstrated that KLF6 and its splicing variants play a critical role in liver cancer. However, little is known on the mechanisms governing KLF6 expression in HCC. In the present work, we asked whether the 3' untranslated region (3'UTR) of the KLF6 mRNA may be responsible for regulation of KLF6 expression in HCC. We found that KLF6 mRNA stability was altered in liver-derived cell lines as compared to cervical cancer-derived cell lines and human embryonic fibroblasts. Interestingly, KLF6 mRNA was highly unstable in liver cancer-derived cell lines as compared to normal hepatocytes. We next cloned the KLF6 mRNA 3'UTR into luciferase-expressing vectors and found that gene expression and activity were strongly impaired in all liver-derived cell lines tested. In addition, we found that most the KLF6 3'UTR destabilisation activity resides between nt 1,835 and nt 2,615 of the KLF6 gene. Taken together, we provide the first steps towards better understanding of the regulation of KLF6 expression in HCC. Further work is needed to identify the factors that bind to KLF6 3'UTR to regulate its expression in liver cancer-derived cell lines.

Tolerance of liver biopsy in a tertiary care center: comparison of the percutaneous and the transvenous route in 143 prospectively followed patients.

BACKGROUND: Transjugular liver biopsy (TJLB) is usually performed when a percutaneous liver biopsy (PLB) is contraindicated. TJLB is an invasive procedure and the patient's tolerance may be variable. AIM: To compare patient tolerance and quality of the biopsy sample between PLB and TJLB. PATIENTS AND METHODS: A total of 143 patients underwent a liver biopsy; of these, 75 underwent TJLB and 68 underwent PLB. To evaluate patient tolerance, we used a visual analog scale that scored the intensity of the symptoms. The length of the biopsy sample and the total number of portal tracts per biopsy were also determined for assessment of biopsy quality. RESULTS: The biopsy sample length was similar in both groups (18.88 ± 8.83 mm on PLB vs. 18.26 ± 10.30 mm on TJLB). No differences were found in the number of portal tracts between the two groups (10.43 ± 8.25 on TJLB vs. 12 ± 10.09 on PLB). Fewer complications were observed in the TJLB group compared with the PLB group (P=0.002).Further, higher degree of pain was reported by patients who underwent PLB compared with patients who underwent TJLB (3.18 ± 3.17 vs. 1.19 ± 2.07); as such, there was a greater need for analgesics on PLB. CONCLUSION: TJLB and PLB techniques provide similar quality of tissue samples; however, TJLB is less painful and therefore better tolerated by patients.