Methods. Randomised controlled trials comparing carvedilol vs. propranolol for portal hypertension in cirrhotic patients and esophageal varices with or without bleed history were included. The outcomes are expressed as odds ratio (〇R), difference of means (DM) and confidence interval. Results. The search identified 14 citations, and 4 randomized controlled comparisons met the eligible criteria. The

trials were conducted in Spain, India and Denmark, included a total of 161 patients, 82 underwent to carvedilol (6.5-50 mg/d) and 79 to propranolol (10-320 mg/d). Carvedilol was superior to get HVPG decrease ≥ 20% from baseline value or to 12 mmHg (OR 2.92; 95%CI 1.26-6.74) (Figure). The CH5424802 magnitude of reduction of HVPG was greater with carvedilol Doxorubicin purchase (DM −2.22; 95%CI −2.82 to −1.60

mmHg). The rate of orthostatic or symptomatic hypotension was no different (OR 1.6; 95%CI 0.644.02). Renal function, including glomerular filtration rate, serum creatinine and plasma renin activity were not different between the treatments. Adverse events leading to withdrawal occurred with the same freguency (OR 0.52; 95% Cl 0.18-1.54). Finally there was no difference about variceal bleeding or mortality. Conclusions. This systematic review and meta-analysis showed that carvedilol is more effective than propranolol for hemodynamic response of portal hypertension in cirrhotic patients and there are no important differences about adverse effects. Figure 1. HVPG decreases ≥ 20% from baseline value o; to ≤ 12 mmHg. Disclosures: The following people have nothing to disclose: Nancy E. Aguilar-Olivos, Nahum Mendez-Sanchez, Misael N. Uribe-Esguivel, Norberto C. Chavez-Tapia Background: Transjugular intrahepatic

portosystemic shunt (TIPS) remains an important treatment modality in patients experiencing severe complications of portal hypertension. The Model for End Stage Liver Disease (MELD) was originally created to predict medchemexpress survival in patients undergoing the procedure in the 1990s. However, the model may not be optimal in more recent patients, because of changes in patient mix, indications, and management for patients undergoing the procedure. Aims: We update the prediction model for cirrhotic patients undergoing the TIPS procedure and assess its generalizability in an independent cohort. Methods: In developing an updated model, a prospective database tracking patients undergoing interventional radiological procedures was gueried to identify all patients who had TIPS up to 2008. Medical records were reviewed to extract further clinical and laboratory data and to exclude patients who had emergency TIPS. Cox proportional hazards regression models were developed to predict 90-day mortality. In validating this updated model, we obtained a data set derived from another US medical center, which was used in a prior publication (Clin Gastroenterol Hepatol.2009;7: 1236). Observed versus expected survival was compared.