ATU-06 Does the timing of TIPSS in patients with acute oesophageal variceal bleeding alter patient outcome?

Introduction International guidelines recommend consideration of early-TIPSS ( Methods This is a single centre cohort study of prospectively collected data in patients undergoing covered TIPSS following urgent ( 72hrs. Primary outcome is all cause mortality. All cases were followed up from the date of OVB necessitating TIPSS insertion for a minimum of 4 years or until death/liver transplant. Results 281 patients were included in the study with 199 undergoing early-TIPSS and 82 having late-TIPSS. Analysis was carried out on 70 patients in each group following propensity score matching. For early and late TIPSS respectively, there were no statistical differences between age (51.2 vs 54.0), female sex (31.4% vs 35.7%), Childs Pugh (CP) score (10.3 vs 10.0), MELD (16.8 vs 18.1), rates of ascites (71.4% vs 65.7%) or encephalopathy (31.4% vs 28.6%) on admission, portal pressure gradient fall (15.1 vs 14.9), rates of post-TIPSS encephalopathy (28.6% vs 24.3%) and rebleeding (24.3% vs 20%). Using competing risk analysis, median follow-up days were as follows: Early TIPSS dead: 259, alive: 5503, liver transplant: 651; Late-TIPSS dead: 322, alive: 5134, liver transplant:172. Equality of cumulative incidence function across groups was non-significant and there was no statistical significance in the incidence of death between the two cohorts (figure 1). On multivariate competing risk regression analysis for the outcome of death, MELD, CP score, ascites and post-TIPSS rebleeding were all statistically significant predictors. Conclusions To our knowledge this is the largest real-world study, with the longest follow-up times, carried out on patients undergoing TIPSS following urgent endoscopic treatment for acute OVB. With advanced statistical analysis, our data suggest that timing of TIPSS does not alter survival in this group of patients.