The AUROC for liver Vs was 0.708 (the cut off value, 2.09 m/s), and its diagnostic ability was lower than the one of spleen Vs. The AUROC for platelet, SI, and APRI were 0.65-0.75, but the AUROC for hyaluronic acid was 0.796, and showed a better diagnostic ability. The cut off for hyaluronic acid was 124 ng/ml with sensitivity 100%, specificity 57%, and it suggested that this parameter could be useful for screening. Conclusion: Spleen and liver Vs increased with the development of esophageal and gastric varices. Spleen Vs was useful in distinguishing F2 and above esophageal and gastric

varices from the one of F0-1. Disclosures: The following people have nothing to disclose: Hiroko Iijima, Tomoko Aoki, Chikage Nakano, Kenji Hashimoto, Akio Ishii, Tomoyuki Takashima, Nobuhiro Aizawa, Naoto Ikeda, Yoshiyuki Sakai, Hironori Tanaka, Yoshinori Iwata,

Hirayuki Enomoto, Masaki Saito, Shuhei Nishiguchi Introduction: PHT-related bleeding is a frequent Cabozantinib and severe complication of cirrhosis. A recent RCT suggested that early-TIPS placement within 72 hours improved prognosis in high-risk patients, defined as variceal bleeding in Child B patients+ac-tive bleeding or Child C patients. The latest consensus meeting on PHT recommended to consider early-TIPS in this subset of patients. Whether this therapeutic approach would be feasible in real-life setting is unclear. Aims : To determine in a national prospective multicentric observational study (1) the proportion of high-risk

patients eligible to an early-TIPS among cirrhotic patients admitted for variceal bleeding; (2) the proportion of high-risk Deforolimus concentration patients who finally underwent early-TIPS placement; (3) the improvement of survival associated with early-TIPS placement. Material et Methods: All French centres recruiting gastrointestinal bleeding were invited to participate. All patients with cirrhosis and PHT-related bleeding were included. Results: 914 patients were included between 04/2012 and 05/2013 in 59 centres Tideglusib (28 university and 31 general-hospitals). Patients’ characteristics: male gender:76.5%; age:59.5±12.1 yrs; aetiologies: alcohol:77%/HCV 14%/other:9%; source of bleeding : OV/GV/other:82/11/7%; active bleeding at endoscopy:38.3%). Distribution of Child-Pugh class was: Child A 20.1%/B 44.5%/C 35.4%. Overall, 439 patients displayed Child-Pugh C or Child-Pugh B class with active bleeding. After excluding patients older than 75, patients with HCC, Child-Pugh C14-15 or with other source of bleeding than OV/GV and patients with serum creatinin>265L mol/l, 232 (25.4%) patients could be considered as eligible for an early-TIPS. In the whole population, 76 patients underwent TIPS placement between admission and Day-3 (44 for uncontrolled bleeding and 32 early-TIPS). Among eligible high-risk patients, only 22 patients underwent early-TIPS (9.4%), 92% of them being indicated in university hospitals. Mortality at 6-week was of 15%. In high risk patients, mortality was of 7.