Analysis Of Risk Factors Of Rebleeding Within One Year After Endoscopic Treatment Of Liver Cirrhosis And Gastroesophageal Variceal Hemorrhage

Objective:1.Analyze the risk factors of rebleeding within one year after endoscopic treatment of the decompensated cirrhosis and gastroesophageal variceal bleeding,so manage the risk stratification for the patients.Make better secondary prevention measures as soon as possible and reduce rebleeding rate.2.Study which is better in self-built model,CTP,MELD for predicting the decompensated cirrhosis and gastroesophageal variceal rebleeding. Methods:from January 2008 to December 2014,a total of 469 patients diagnosed as liver cirhosis and gastroesophageal variceal bleeding with integrated medical datas and follow-up results in our hospital were retrospectively analyzed.The follow-up end was 1 year or bleeding within 1 year after endoscopic treatment.Then the patients were divided into bleeding group(145 cases) and no bleeding group(324 cases).Then compared general conditions,cirrhosis etiology,laboratory examinations,previous history of bleeding,hepatocellular carcinoma and portal vein thrombosis,ascites,oral nonselective ?-blocker,endoscopic variceal risk factors.Calculate CTP classification,CONUT,upper gastrointestinal bleeding Blatchford score,Rockall score,MELD and MELD-Na score,etc.To showe the independent risk factors in univariate analysis and multivariate unconditional Logistic regression analysis. Results:1.145 patients rebled after endoscopic treatments in one year,rebleeding rate was 30.9%.2.Univariate analysis:When compared the two groups in age,gender,etiology, previous history of bleeding,hepatocellular carcinoma and portal vein thrombosis, ALT,Cr,BUN,endoscopic variceal erosion and thrombosis,Blatchford score,there were no statistical differences.AST,TBIL,DBIL,ALB,endoscopic active bleeding,ascites, oral non-selective ?-blockers,CTP,CONUT,Rockall,MELD and MELD-Na score had statistical significance.3.Multivariate Logistic regression analysis:oral non-selective ?-blockers, ascites,CONUT, MELD were independent risk factors.4.Self-built model has no statistically significant difference compared with CTP and MELD model for predicting rebleeding within one year after endoscopic treatment of the decompensated cirrhosis and gastroesophageal variceal bleeding.But the AUC of the self-bulit is greater than the CTP classification and the MELD model. Conclusion:1.Endoscopic treament is a very effective secondary prevention of liver cirrhosis and gastroesophageal variceal hemorrhage.2.Rebleeding risk factors include AST,TBIL,DBIL,ALB,endoscopic active bleeding,ascites,oral non-selective ?-blockers,CTP,CONUT,Rockall,MELD and MELD-Na.Among them,oral non-selective ?-blockers is protective factors,ascites,CONUT and MELD are independent risk factors.3.The effect is the same in Self-built model, liver function CTP classification, MELD score for predicting rebleeding rate after endoscopic treatments of cirrhosis and EGVB.