Different scoring systems in predicting survival in Chinese patients with liver cirrhosis undergoing transjugular intrahepatic portosystemic shunt

Transjugular intrahepatic portosystemic shunt (TIPS) is an established minimal-invasive procedure to treat complications of portal hypertension, and several scoring systems have been used to help choose suitable patients. However, its accuracy remains controversial.To compare the performance of the Child-Turcotte-Pugh (CTP) classification system, model for end-stage liver disease (MELD) score, Emory score, Bonn TIPS early mortality (BOTEM) score, and serum bilirubin and platelet count (SB/PLT model) in predicting survival in Chinese patients with liver cirrhosis undergoing TIPS.The clinical data of patients undergoing TIPS in our department were retrospectively analyzed to compare the five scoring systems on the basis of survival after TIPS.A cohort of 159 patients was analyzed. The survival curves showed a statistical significance between classification B and C of CTP (χ=9.451, P=0.002), between MELD less than 10 and MELD at least 10 (χ=10.099, P=0.001), and between low-risk and moderate-risk groups of the Emory score (χ=4.656, P=0.031), indicating a better discriminatory ability. By ROC curves and a logistic regression model, the MELD score and the CTP system had better power to predict 3-, 12-, and 24-month survival. The MELD score and the CTP classification system had smaller values of -2 Ln(L), Akaike Information criterion, and Schwarz-Bayesian criterion, respectively.The MELD score and the CTP classification system provide better prognostic stratification for a cohort of Chinese patients with advanced cirrhosis undergoing TIPS. However, the MELD score is not significantly superior to the CTP system.