BACKGROUND AND AIMS: Portal hypertension is the main pathophysiological driver of decompensation in patients with liver cirrhosis. Epithelial cell death markers, m30 and m65, correlate with hepatic injury and predict outcomes across various stages of liver disease. We aim (i) to evaluate whether portal hypertension itself contributes to liver outcome-relevant epithelial injury, and (ii) to analyse the capacity of m30/m65 to predict outcome in patients receiving a transjugular intrahepatic portosystemic shunt (TIPS) for refractory ascites. METHODS: Sixty-six patients undergoing TIPS placement for refractory ascites and 20 patients with compensated cirrhosis as controls were prospectively enrolled in this monocentric cohort study. Epithelial cell death markers were analysed pre-TIPS, as well as 1-3 and 6-9 months post-TIPS. The capacity of baseline levels of m30/m65 in predicting six-month transplant-free survival rates was analysed by multivariable Cox proportional hazards regression. RESULTS: Levels of m30 and m65 were higher in patients with decompensated cirrhosis (pre-TIPS) compared with compensated cirrhosis (controls). Following correction of portal hypertension by TIPS and recompensation, both markers decreased over time, reaching levels comparable to patients with compensated cirrhosis. On multivariable analysis, pre-TIPS baseline levels of m30 and m65 were not predictive for six-month survival. CONCLUSION: Correction of portal hypertension via TIPS reduces levels of epithelial cell death markers, indicating that portal hypertension is a driver of outcome-relevant, hepatic cell death in patients with decompensated cirrhosis. Baseline m30/m65 values do not affect six-month survival rates, which suggests that TIPS placement overcomes the unfavourable spontaneous prognosis otherwise indicated by elevated baseline m30/65 levels.