Early-TIPS improves survival in high-risk variceal bleeders. Results of a multicenter variceal bleeding observational study

Background: Despite gold standard therapy, about 20% of patients with acute variceal bleeding (AVB) die during the first 6 weeks. Early-TIPS (placed within 72 hours from admission) has been shown to improve survival in patients with AVB and high risk of treatment failure (defined as Child C-Pugh score up to 13 (CP-C) or Child B plus active bleeding at endoscopy (CPB-AB)). However, studies evaluating its benefit included a relatively small number of patients. In addition, how much CPB-AB are really high-risk has been questioned. The present study aimed to evaluate the role of early TIPS in a large number of CP-C and CPB-AB patients with AVB admitted to referral centers worldwide. Methods: Multicenter, international, observational study including 2168 patients from 34 centers between April 2013 and April 2015. Patient’s data were recorded and centralized on ReDCap (Research Electronic Data Capture) and centralized by the study-coordinating center. Patients were managed according to current guidelines. Placement of early-TIPS was based on individual center policy. Patients were followed up to 1 year, death or liver transplantation. Multiple excellence control procedures were used to optimize data quality. Additionally, data were regularly monitored by a data validation committee for detection of inconsistencies or errors followed by queries to local investigators requiring resolution; and further validated all reported outcomes. Results: 671/2168 patients (30.9%) were high-risk and had no exclusion criteria (HCC beyond Milan, age >75, creatinine ≥3mg/dl, PVT or Child>13); 434 (65%) CP-C <14 points and 237 (35%) CPB-AB. 589 patients were treated with pharmacological + endoscopic therapy (Drug+Endo) and 82 with Early-TIPS. There were no major differences among groups. The 6-week and 1-year actuarial probabilities of remaining free of rebleeding were significantly higher in the Early-TIPS than in the Drug+Endo group (93% vs 81%, and 87% vs 68%; P=0.002). The benefit of early-TIPS was observed in CPB-AB and CPC. Survival was also Higher in the early-TIPS group (6w-survival: 87% vs 77%; 1-y: 70% vs 62%; P=0.08). However, a survival benefit was only observed in CP-C (6w: 85% vs 69%; 1y: 66% vs 53%; P=0.034), not in the CPB-AB group. These results were confirmed using models adjusted for baseline covariates as well as propensity score. Conclusions: The present study confirms in a large group of high-risk patients that the use of early-TIPS reduces rebleeding risk. Early-TIPS also improves survival in Child C patients.