[Analysis of prognosis on patients with severe viral hepatitis using the model for end-stage liver disease]

To study the practical use of the model for end-stage liver disease (MELD) in clinics and to assess its validity in predicting the prognoses of patients with severe viral hepatitis and to determine the validity of the model by the concordance (c)-statistic which is equivalent to the area under the receiver operating characteristic (ROC) curve. Also, the optimal cutoff values of MELD to discriminate between deceased and surviving patients were calculated using ROC curves.

121 patients were divided into a plasma exchange (PE) group and a non-plasma exchange (non-PE) group, and they were graded with the MELD formula. The death rate was observed within three months.

81 patients died within 3 months (35 cases in the PE group, 46 cases in the non-PE group). The mortality rates of patients in the PE group whose MELD scores were between 20 to 30 and 30 to 40 were 31.6% and 57.7% respectively, but in non-PE cases they were 67.6%, 81.3% respectively; there was a significant difference between the PE group and non-PE group. The mortality rates of patients whose MELD score was higher than 40 were 93.3% in the PE group and 100% in the non-PE group, however there was no significant difference between the two groups. The optimal cut-off values of MELD to predict the prognoses of patients were 30 in the PE group whose sensitivity, specificity and c-statistic were 80.0%, 52.0% and 0.777, but in the non-PE group they were 25, 82.6%, 86.7% and 0.869, respectively.

This study suggests that MELD scores can serve as an index of the severity of the disease of patients with severe viral hepatitis, and that the mortality rates of the patients increase with the increase of their MELD scores. MELD can accurately predict the short-term prognoses of patients with severe viral hepatitis.