四种评分模型对慢加急性乙型肝炎肝衰竭患者短期预后的评价

Evaluation of short-term prognosis in patients with acute-on-chronic hepatitis B liver failure by four scoring models

目的比较终末期肝病模型(MELD)、MELD-Na、慢性重型肝炎预后指数(PI)和肝移植标准(LTS)模型对慢加急性乙型肝炎肝衰竭患者短期预后的预测价值。方法在138例慢加急性乙型肝炎肝衰竭患者入院24小时内进行MELD、MELD-Na、PI和LTS评分,并随访3个月。应用受试者工作特征曲线(ROC)下面积(AUC)判断四个模型的预测能力。结果在观察期内与肝病有关的死亡患者72例,生存者66例。死亡组LTS、MELD-Na、MELD和PI平均值明显高于生存组(P<0.01),四个模型的AUC分别为0.860、0.801、0.749、和0.749,差异无统计学意义;四个模型预测的正确率分别为82.61%、76.81%、75.36%和73.91%,差异无统计学意义。结论 4种模型对慢加急性乙型肝炎肝衰竭患者短期预后均有较好的预测价值。

Objective To compare the model for end-stage liver disease （MELD）,MELD-Na,prognostic index（PI）and liver transplantation standard （LTS）in predicting the short-term prognosis of patients with acute- on-chronic hepatitis B liver failure （ACLF）. Methods 138 patients with ACLF were evaluated by MELD, MELD-Na,PI and LTS within 24 hours after admission. The area under curve（AUC） from the receiver operating characteristic curve （ROC）was used to determine the predictive abihty of each model. Results 72 patients with ACLF died,and 66 survived liver failure within three month follow-up. The average scores of LTS, MELD-Na, MELD and PI in the death group were greater than that in the survival group （P〈0.01）;The AUCs of the four models were 0.860,0.801,0.749 and 0.749,respectively and there was no significant difference between them; The total predictive accuracy of the four models were 82.61%,76.81%,75.36% and 73.91%,respectively（P〉0.05）. Conclusions MELD,MELD-Na,PI and LTS scoring models have the same efficacy in predicting short-term prognosis of patients with ACLF.