应用终末期肝病模型对慢性重型肝炎患者的预后评估

目的 探讨终末期肝病模型（MELD）在慢性重型肝炎患者的预后评估与治疗策略中的意义。方法 回顾性分析135例慢性重型肝炎患者的临床资料,计算患者入院时的MELD分值和接受治疗2周后的MELD分值与入院时MELD分值的差值（△MELD分值）,分析MELD分值和△MELD分值与患者3个月病死率的关系。结果 入院时MELD分值死亡组患者为37．00±6．50,存活组为25．80±5．20,死亡组患者入院时的MELD分值远高于存活组,两组间差异有统计学意义（x^2=72.00,P〈0．01）,接受治疗2周后△MELD分值死亡组患者为1．57±0．89,存活组为-0．99±0．73,两组间差异有统计学意义（x^2=56．35,P〈0．01）。患者入院时MELD分值和接受治疗2周后△MELD分值预测慢性重型肝炎患者3个月内病死率的c-统计值分别为0．90和0．76。MELD分值〈25组,患者3个月后病死率为2％,25≤MELD分值≤30组病死率为7％,300组和≤0组的病死率分别为51％和13％,两组间病死率差异有统计学意义（P〈0.01）。结论 MELD分值和△MELD分值越高,慢性重型肝炎患者的病死率越高,MELD能较准确地预测慢性重型肝炎患者的预后,可以结合患者入院时的MELD分值和接受治疗后的△MELD分值来决定慢性重型肝炎患者的治疗方案。

An evaluation of the prognosis of patients with chronic severe hepatitis using a model for end-stage liver disease

OBJECTIVE: To investigate the prognosis evaluation and treatment strategy of chronic severe hepatitis (CSH) patients using a model of end-stage liver disease (MELD). METHODS: The MELD scores of 135 CSH patients on the day of their admittance to our hospital and the DeltaMELD scores after two-weeks of medical treatment were retrospectively analyzed. They were also compared with the scores of the three-month mortality rate of the patients. RESULTS: The mean MELD score calculated on the first day of the patients who died after their admission to the hospital was 37.00+/-6.50, while that of the living group was 25.80+/-5.20. The difference was highly significant (chi(2)=72.00, P < 0.01). MELD score after two-weeks medical treatment of the patients who died was 1.57+/-0.89, while that of the living group was -0.99+/-0.73; the difference was also highly significant (chi(2)=56.35, P < 0.01). The area under the ROC curve of MELD score (c-statistic) was 0.90, while the c-statistic for DeltaMELD score was 0.76. On the first day of their admission, when the MELD score was < 25, the three-month mortality rate was 2%; when it was 25 or= 35, the three-month mortality rate was 81%; the differences between these groups were all highly significant (P less than 0.01). When MELD scores were above zero, the three-month mortality was 51%, and when DeltaMELD scores were less than or equal to zero, the three-month mortality rate was 13%. All the differences were highly significant (P < 0.01). CONCLUSION: A high MELD score and a high Delta MELD score herald high three-month mortality rates in patients with CSH. MELD is quite usable in assessing the prognosis in patients suffering CSH. The choice of treatment for the CSH patients could be made by integrating the MELD score calculated on the first day of being admitted to a hospital and the Delta MELD score after their medical treatment.