Objective To assess the application of a new scoring system for severity evaluation of acute-on-chronic liver failure induced by hepatitis B.Methods A total of 399 patients (203 survivals and 196 deaths) with acute-on-chronic liver failure induced by hepatitis B were collected from the Third Affiliated Hospital of Sun Yat-sen University during January 2003 and June 2008.All patients were graded with the new scoring system and model for end-stage liver disease (MELD) at critical stage (survivals) or terminal stage (deaths).The survival rates and fatality rates of patients who were graded by two scoring systems were analyzed and compared.Results With MELD system,the fatality rate was 11.89％ (17/143) in patients with scores of 15-26,64.68％ (141/218) with scores of 27-48,and 100％ (38/38) with scores of 49-69.No score range with fatality rate of 0 was found.While with the new scoring system,the survival rate was 99.2％ (126/127) when the severity scores were between 2 to 8,and patients with scores 2,3,4,5,6 and 8 were all survived; the fatality rates were gradually raised from 4.2％ (1/24) with scores of 9-17 to 100％ (82/82) with scores of 18 and above.Conclusion The new scoring system is more objective,simple and sensitive than MELD system,which can be used for severity evaluation of acute-onchronic liver failure induced by hepatitis B.

Objective To establish a scoring system for evaluating the severity of hepatitis B patients with acute-on-chronic liver failure and to compare the validity of this system with model for end-stage liver disease (MELD). Methods MELD score was used in hepatitis B patients with acuteon-chronic liver failure who were divided into survival group (203 cases) and death group (196 cases).Seven clinical relative indices, including prothrombin activity, serum creatinine, hepatic encephalopathy, accompanying infections, serum total bilirubin, the dimension of liver, the amount of ascites, were selected for evaluating the severity. Each index was graded with 1 to 4 points based on the severity. Then the total score was counted by adding up scores of each index. T test and area under receiver operating characteristic (ROC) were used to evaluate the difference and similarity of the two systems. Results According to the new scoring system, the total score was 8. 07±3. 14 in the survival group and 16. 91 ±3. 54 in the death group. There was a statistically significant difference between these two groups (t = 26.125. P<0.01). In 81.32% of survival patients, their scores ranged from 3.91 to 12.23, while in 81.32% of dead patients, their scores ranged from 12.23 to 21.60. The two ranges overlapped at 12.23. According to the MELD system, the total score was 26. 43 ±5. 58 in the survival group and 40. 16 ±10. 22 in the death group. The difference between the two groups was statistically different (t = 16. 566, P<0. 01). In 61.02% of survival patients, the MELD scores ranged from 21. 49 to 31. 19, while in 61. 02% of the dead patients, the MELD scores ranged from 31. 19 to 48. 94. The two ranges overlapped at 31.19. The areas under ROC of the new scoring system and MELD system were 0.960 (95% CI: 0. 944-0. 977) and 0.886 (95% C/;0. 852 - 0. 920). No overlap was found in these two 95%CJ and there was a statistically significant difference. Conclusions The new scoring system is applicable for evaluating the severity and prognosis of acute-on-chronic liver failure in hepatitis B patients. The sensitivity of this new scoring system is approximate to the MELD system.