Portal hypertensive gastropathy (PHG) and gastric antral vascular ectasia (GAVE) as well as esophagogastric varices (EV) are crucial complications, which may provoke gastrointestinal bleeding, in patients with portal hypertension. Recently, the liver stiffness measured using ultrasound was shown to be associated with the extent of portal hypertension. Moreover, we previously reported that the spleen stiffness also showed association with hepatic venous pressure gradient (HVPG), which may reflect the risk for EV bleeding. Thus, in the present study, the significance of liver/spleen stiffness on possible risk for bleeding from hypertension-related gastrointestinal lesions (PHRGLs), including PHG and GAVE as well as EV, was investigated. A total of 92 patients with chronic liver diseases were enrolled, and the relation between PHRGLs and clinical factors such as hepatocellular carcinoma, liver function and liver/spleen stiffness were evaluated. EV, PHG, and GAVE were seen in 41.3%, 43.5%, and 9.8% of patients, respectively. Liver stiffness (mean±SD) was greater in patients with PHRGLs (2.114±0.060) than in those without PHRGLs (1.802±0.046) (p＜0.005). Spleen stiffness was also greater in the former patients (2.621±0.102) than in the latter patients (2.263±0.041) (p＜0.05). Multivariate analysis identified the spleen stiffness, peripheral platelet counts and serum albumin levels as significant factors responsible for development of PHRGLs. In conclusion, although both liver stiffness and spleen stiffness showed association with PHRGLs development, the spleen stiffness was the most useful factor predicting the occurrence of PHRGLs.

A Denver shunt, a peritoneo-venous shunt, is a devise to shunt ascites to the superior vena cava in patients showing refractory ascites. In the present study, efficacy and safety of a Denver shunt for patients with decompensated cirrhosis manifesting tolvaptan-resistant intractable ascites were evaluated. Subjects were 7 patients with cirrhosis in whom Denver shunt construction was done for intractable ascites in our institution from 2014 to 2018. Improvement of ascites-related symptoms was seen in 4 patients (57%), while decrease of ascites volume evaluated based on body weights and/or CT/MRI images was achieved in 5 patients (71%). Repeated drainage for ascites was discontinued in 2 patients (29%) and doses of diuretics were reduced in 5 patients (71%). Consequently, the overall response rate, fulfilling either of these criteria. was calculated as 86% In contrast, disseminated intravascular coagulation (DIC), wound divergence, spontaneous bacterial peritonitis, hepatic encephalopathy, and right heart failure occurred as adverse events following the procedure in 3, 1, 1, 1 and 1 patient(s), respectively, but the extents of DIC were none-severe in all patients and other events were attenuated following the conservative treatment. One patient died 30 days after the procedure due to liver failure without decrease of ascites. A Denver shunt was useful to attenuate refractory ascites in patients with decompensated cirrhosis even in those showing resistance against conventional therapies including tolvaptan.