MELD和CTP评分系统在慢性重型乙型肝炎患者短期预后评估中的价值

目的应用终末期肝病模型（MELD）和Child—Turcotte—Pugh（CTP）评分系统预测慢性重型乙型肝炎患者的短期预后，并探讨两者的临床应用价值。方法回顾性分析115例慢性重型乙型肝炎患者的资料，根据治疗3个月时的生存情况将患者分为生存组和死亡组，并用MELD和CTP评分系统进行评分。应用受试者工作特征曲线评价每一模型的预测价值，并用K—M生存曲线分析不同预测模型的差异。结果死亡组CTP和MELD平均分值明显高于生存组（t值分别为4．891和3．949，P值均〈0．05），2个评分系统显著相关（y=0．500，P=0．000）。MELD和CTP分值能够较好地预测慢性重型乙型肝炎患者3个月内的病死率，C—statistic分别为0．765和0．834，两者的预测能力比较差异无统计学意义（Z=1．516，P〉0．05）。CTP评分〈10．5分和MELD评分〈27．5分患者的生存时间分别较CTP评分〉10．5分和MELD评分〉27．5分患者长（Z值分别为17．88和25．28，P值均为0.000），生存率也显著增高，差异有统计学意义（X^2值分别为16．88和31．59，P值均为0，000）：结论MELD和CTP评分系统在预测慢性重型乙型肝炎患者短期预后方面效果相近，使用时可结合临床资料，以提高预测的准确度。

MELD and CTP scores in predicting short-term prognosis in patients with severe chronic hepatitis B

Objective To apply the model for end-stage liver disease （MELD） and the Child- Turcotte-Pugh （CTP） in predicting short-term prognosis of patients with severe chronic hepatitis B, and to evaluate their clinical value. Methods Data of 115 patients with severe chronic hepatitis B were retrospectively analyzed and all patients were divided into survival group and fatal group by survival status at 3rd month of the treatment. The scores of MELD and CTP were obtained. The accuracies of MELD and CTP were evaluated by the receiver operating characteristic （ROC） curve. The difference between two systems was analyzed b） Kaplan-Meier survival curve. Results MELD and CTP scores in fatal group were significantly higher than those in the survival group （t = 4. 891 and 3. 949, P 〈 0. 05 ）, and two systems were significantly correlated （γ = 0. 500, P = 0. 000 ）. MELD and CTP scores were good for predicting the fatality rate of patients with severe chronic hepatitis B within 3 months （C-statistic： 0. 765 and 0. 834, respectively）, and there was no significant difference between the two svstems （Z = 1. 516, P 〉 0. 05 ）. Patients with CTP 〈 10. 5 or MELD 〈27. 5 had longer survival time （Z = 17. 88 and 25.28, P =0. 000） and higher survival rates （X^2 = 16. 88 and 31.59, P = 0. 000） than those with CTP 〉 10. 5 or MELD 〉 27.5. Conclusion MELD and CTP scores have similar predictive value for short-term prognosis of patients with severe chronic hepatitis B, and clinical data should be contained to better predict the short-term prognosis.