Prognostic Assessment Of Transjugular Intrahepatic Portosystemic Shunt In Severe Cirrhotic Ascites

Background:Ascites is one of the common complications of decompensated cirrhosis.Its appearance suggests that the prognosis of patients is poor.The 5-year survival rate is about 50%,while the median survival period of patients is only 6 months when they progress to the stage of refractory ascites,but the survival rate varies greatly for individuals that might be caused by the heterogeneity of population.Therefore,it is of great significance for clinical practice to select a suitable prognosis model to guide treatment decision-making by risk stratification.Large volume paracenteses with intravenous albumin infusion and transjugular intrahepatic portosystemic shunt(TIPS)are commonly recommended for patients waiting for liver transplantation.Although there is no definite conclusion about the improvement of survival of these two treatments based on the previous randomized controlled study,meta-analyses show that after controlling the confounding factors such as liver transplantation,TIPS can significantly improve the transplant-free survival(TFS),especially in cirrhotic recurrent patients.Thus,a prudent optimal patient selection is determinant of survival beneficial effects of TIPS.Currently,several altemative risk stratification systems have been established for identifying candidates receiveing TIPS.Nevertheless,the research about true survival benefit provided by TIPS compared with the medical standard treatment in the different risk stratification models is still limited.Aims:1)A multicenter retrospective cohort study was conducted to evaluate and compare the performance of ten scores in predicting transplant-free survival after TIPS in severe cirrhotic ascites.2)A large,multicenter retrospective observation study was performed by risk stratification analysis to identify optimal patients who benefited the most from TIPS treatment(compared with standard treatment).Methods:1)The study population included 280 cirrhotic patients with severe ascites(including 123 cirrhotic patients with refractory ascites and 157 cirrhotic patients with recurrent ascites)who received TIPS in 9 university hospitals in China from March 2006 to December 2017.By the end of the follow-up on September 15,2019,the patients who lost the follow-up or not die were considered to have been deleted the data.Continuous variables are represented by mean ± standard deviation,and classified variables are represented by frequency and percentage.Survival curves was generated with Kaplan Meier analysis and was compared with log-rank test to evaluate the survival difference of three months,six months and one year.Cox regression analysis was used to estimate the relationship between risk ratio and other factors.Discriminatory ability of prognosis models was evaluated according to the area under the time-dependent ROC curve,c-index and likelihood ratio.At the same time,the ascites type,the etiology of cirrhosis and the type of TIPS stent were considered into subgroup analysis.The end point was 1-year survival without transplantation.The cutoff of i MELD score was selected according to the c-index of different percentiles.2)From March 2006 to December 2017,796 consecutive patients with severe cirrhotic ascites underwent TIPS or large volume paracenteses with intravenous albumin infusion in 10 centers across the country were included,consist of 280 patients receiving TIPS and 516 patients receiving medical standard treatment.The primary end point of the study was 1-year transplant-free survival,and the secondary end point was ascites recurrence,overt hepatic encephalopathy and other portal hypertension complications.Survival curves was generated with Kaplan Meier analysis and was compared with log-rank test.The Fine and Gray competing risk regression model was used to compare the outcomes of ascites recurrence and overt hepatic encephalopathy and death was used as the competitive event.Prognostic factors of survival were identified with Cox regression analysis.The treatment(TIPS vs.standard treatment),CTP grade,MELD score,MELD-Na score,i MELD score,platelet-bilirubin model and possible confounding factors were put into the regression model.The variables with P< 0.1 in the single factor analysis were included in the multivariate analysis,95% CI of the risk ratio(HR)was calculated and P < 0.05 in the multivariate analysis was regarded as statistically significant,also,propensity matching analysis was used to further control the confounding variables.Results:1)The median follow-up time of the patients was 22.9 months(range: 6.1-24.0 months).The transplant-free survival rate of 3-month,6-month and 12-month after TIPS treatment were 84.3%,75.4% and 65.7%,respectively.CTP score,MELD score,MELD-Na score,i MELD score,MESO score,Refit MELD score and Refit MELD-Na score were independent predictors of survival.According to the comprehensive evaluation of AUROC,c-index and likelihood ratio,i MELD score showed the best survival prediction ability in the short-term,long-term and subgroup analysis of different ascites types(refractory ascites/recurrent ascites),different etiology(hepatitis B/non-hepatitis B)and main stent types(covered stent).Taking 1-year transplant-free survival as end point,the corresponding i MELD scores of 32 and 38 in 30% and 70%percentages were selected as the cutoff values of risk stratification.Thus,the overall population was divided into three strata,the first stratum was i MELD < 32;the second stratum was i MELD ≥ 32 but < 38;the third stratum was i MELD ≥ 38.2)Toal of 796 patients with cirrhotic severe ascites were included in the study.The median follow-up time in TIPS treatment group was 22.9 months(range: 6.1-24.0months),and that in medical standard treatment group was 20.3 months(range: 4.6-23.1months).Compared with the standard treatment group,TIPS significantly improved the 1-year transplantation-free survival rate(65.7% vs.50.7%,P < 0.001).Kaplan-Meier analysis showed that the survival rate was significantly different in the risk stratification of ascites type,CTP score classification,MELD score,MELD-Na score and i MELD score;after adjusting the confounding factors,TIPS is an independent factor to improve the 1-year transplant-free survival among patients with recurrent ascites(HR = 0.59;95% CI: 0.40-0.90 P = 0.013)、CTP B classification(HR= 0.65;95% CI: 0.46-0.90;P = 0.01)、MELD≤15(HR=0.66;95% CI:0.45-0.96;P=0.029)、MELD-Na≤18(HR=0.60;95% CI:0.41-0.87;P=0.007)、i MELD<32(HR=0.47;95% CI:0.24-0.91;P=0.024).Platelet-bilirubin model can not help to guide the treatment decisions.In addition,TIPS significantly reduced the recurrence of ascites,but increased the occurrence of overt hepatic encephalopathy.CONCLUSIONS:1)The i MELD score proved to be the best prognostic model in predicting TFS in patients with severe cirrhotic ascites receiving TIPS.Meanwhile,the model could stratify patients in three stratums i MELD<32/32-38/≥38(low risk / medium risk / high risk)to help guiding clinical practice.2)Patients with recurrent ascites,CTP B classification,MELD≤15,MELD-Na≤18 and i MELD < 32 had a higher 1-year transplant-free survival after TIPS treatment in cirrhotic severe ascites,and they could benefit most from TIPS.Whereas platelet-bilirubin model can not guide the treatment decision-making.