Clinical Study Of Prognosis And Survival Time For Liver Failure

Objective: To discuss independent risk factors of prognosis and survival time in patients with liver failure. To evaluate the diagnostic value of MELD-Na scoring system, MELD scoring system and Child-Pugh scoring system for liver failure. To study the influence of different complications and treatment methods on survival time of liver liver failure. To analyze the prognosis of patients with different liver failure.Methods: Retrospective analysis was applied in studying the clinical aspects of 297 patients with liver failure using artificial liver trentment between May 2002 and March 2007 in Tianjin third centure hospital. Data were analyzed with SPSS 13.0 statistic software, t test and rank test was used on quantitative data , chi-square test was used on qualitative data, then Logistic and Cox regression analysis was used for selecting the independent risk factors influencing the prognosis and survival time respectively. The survival curves were plotted with Kaplan-Meier method with differences in survival estimated by Log rank test. To evaluate the diagnostic value of MELD-Na scoring system, MELD scoring system and Child-Pugh scoring system using ROC curves. According to independent risk factors from Logistic and Cox regression model, we can establish prognositic model for our patients with liver failure. The cumulative survival rate of 3-month,6-month, 1-year, 2-year, 3-year, 4-year in these patients was compared according to liver failure type.Result: 1. Independent risk factors (T<0.05) using Logistic regression analysis were: Child-Pugh score, MELD-Na score, tyrosine, bilirubin separation ALT, liver cancer, liver failure type, indirect bilirubin and leukocyte. Independent risk factors (P<0.05) using Cox regression analysis were: Child-Pugh score, MELD-Na score, bilirubin separation ALT, liver cancer, artificial liver times, methionine, tyrosine and liver failure type. 2. Drawed survival curves of 297 patients, median survival time was 120 days, and effective rate of treatment was 52.5%. We found the P value of survival curves of different sex, family history , whether use hepatocyte growth-promotting factors or not, whether use thymosin or not and whether use albumin or not was 0.137, 0.854, 0.371, 0.635, 0.290 respectively (P value more than 0.05), but the P value of bilirubin separation ALT, primary peritonitis, liver encephalopathy, liver cancer, ascites, hepatorenal syndrome, gastrointestinal bleeding, whether use prostaglandin E1 or not and liver failure type less than 0.05. The P value of curves of different age ( age<50 and age>50) was less than 0.05. 3. The P value of HBV DNA and whether use anti-virues drugs or not in patients with liver failure because of hepatitis B were 0.867, 0.486 respectively. 4. The AUC values generated by the ROC curves for Child-Pugh score was 0.794, higher than MELD-Na score and MELD score respectively. The cutoff scores of three systems were 10.5, 24.8, 26.4 respectively, which could discriminate high and low mortality accurately. The AUC values generated by our study ROC curves for the LOG model (bilirubin separation ALT , Child-Pugh score, IBIL, serum sodium, tyrosine )and COX model (bilirubin separation ALT , Child-Pugh score, IBIL, serum sodium, tyrosine , age, omithine) were 0.867, 0.845 respectively in predicting chronic liver failure patients' prognosis, which prognostic ability were better than MELD-Na score system and MELD score system. The cutoff scores of the two system were 7.5, 10.5 respectively. 5. The overall effective rate of treatment was 52.5%. Effective rate of treatment of acute liver failure was zero, of subacute liver failure was 63.8%, of acute on chronic liver failure was 64.1%, of chronic liver failure was 46%. The cumulative survival rate of 1-month, 3-month, 6-month, 1-year, 2-year, 3-year, 4-year was 71.04%, 58.59%, 45.79%, 36.36%, 21.21%, 10.78%, 5.39%, respectively.Conclusion: 1. Child-Pugh score, MELD-Na score, tyrosine, bilirubin separation ALT, liver cancer, liver failure type, indirect bilirubin and leukocyte were independent risk factors affecting prognosis. 2. Child-Pugh score, MELD-Na score, bilirubin separation ALT, liver cancer, artificial liver times, methionine, tyrosine, liver failure type affecting survival time. 3. The factors of HBV DNA and the effect of the anti-virus drugs had no significant influence on prognosis and survival time for hepatitis B induced to liver failure. 4. MELD-Na score system, Child-Pugh score system and MELD score system can predict the prognosis of liver failure. Their best cutoff value was 26.4, 10.5, 24.8 respectively. The Child-Pugh score system was the best . LOG model and COX model in our study had better predictability. 5. The survival curves according to age, bilirubin separation ALT, primary peritonitis, liver encephalopathy, liver cancer, ascites, hepatorenal syndrome, gastrointestinal bleeding, whether use prostaglandin El or not and liver failure type had significant differences.6.The prognosis of liver failure was poor, especially acute liver failure. Liver transplantation should be performed for acute liver failure patients.