MELD相关评分体系对乙型肝炎相关慢加急性肝衰竭患者预后的预测价值  

MELD-based models predict the prognosis of patients with acute-on-chronic hepatitis B liver failure

作　　者：林贤丰[1] 李凌菲[1] 俞燮琰[1] 肖冬冬[1] 吴圣杰[1] 陈黎黎[2,3] 宋梅[2,3] 刘文悦[1] 施可庆[2,3] 陈永平[2,3] 郑明华[2,3] 

机构地区：[1]温州医学院第一临床学院,浙江省325000 [2]温州医学院附属第一医院感染内科 [3]温州医学院肝病研究所

出　　处：《中华临床医师杂志（电子版）》2012年第19期5853-5857,共5页Chinese Journal of Clinicians(Electronic Edition)

基　　金：国家科技部十二五"艾滋病和病毒性肝炎等重大传染病防治"科技重大专项(2012ZX10002004);浙江省医药卫生优秀青年科技人才专项基金计划(2010KYB070);浙江省大学生科技创新活动计划(新苗人才计划)(2011R413020);温州医学院学生科研项目(wyx201101010);温州医学院第一临床医学院学生科研项目(wyyxky2010-06)

摘　　要：目的比较分析MELD、MELD-Na、MELDNa、MESO、iMELD、UKELD六种MELD相关评分体系预测乙型肝炎相关慢加急性肝衰竭(ACHBLF)患者的预后。方法纳入温州医学院附属第一医院327例ACHBLF患者,分别计算各个MELD评分体系值,比较各评分系统在死亡组和生存组中的差异。通过受试者工作特征(ROC)曲线和曲线下面积(AUC)进行比较分析。结果死亡组MELD、MELD-Na、MELDNa、MESO、iMELD、UKELD值分别是29.2±5.6、37.1±12.6、31.1±5.1、26.4±13.8、57.6±23.3、72.8±27.5,虽高于生存组的27.1±6.1、36.3±13.8、29.2±5.8、27.0±16.7、56.1±29.8、70.2±21.4,但除MELD(P=0.02)和MELDNa(P=0.02)评分外,其余差异无统计学意义(P=0.374,P=0.745,P=0.639,P=0.375);各评分系统的AUC分别为0.630、0.556、0.609、0.537、0.555、0.530。结论除MELD和MELDNa评分可能有一定评判价值外,MELD相关评分体系总体上对预测ACHBLF的预后表现较差,无法对ACHBLF进行有效、准确的预测。Objective To compare and analyze the values of the model for end-stage liver disease(MELD)and five MELD-based models including MELD-Na,MELDNa,MESO,iMELD and UKELD,in predicting the prognosis of patients with acute-on-chronic hepatitis B liver failure(ACHBLF).Methods The clinical data of 327 patients with ACHBLF were enrolled into the study.The six models were applied respectively to evaluate the scores in survival group and death group.The capabilities of these six models to predict the prognosis of ACHBLF were compared by the receiver operating characteristic(ROC)curve and area under curve(AUC).Results The scores of death group evaluated by MELD,MELD-Na,MELDNa,MESO,iMELD,UKELD(29.2±5.6,37.1±12.6,31.1±5,26.4±13.8,57.6±23.3,72.8±27.5,respectively)were higher than those of survival group(27.1±6.1,36.3±13.8,29.2±5.8,27.0±16.7,56.1±29.8,70.2±21.4),but besides MELD and MELDNa,the other scores′ difference was not statistically significant(P=0.02,P=0.374,P=0.02,P=0.745,P=0.639,P=0.375).The values of AUC of these six models were 0.630,0.556,0.609,0.537,0.555 and 0.530,respectively.Conclusions Besides MELD and MELDNa models may have the value of judgement,the MELD and MELD-based models cannot overall predict the prognosis of ACHBLF very well;therefore,these six models cannot make effective and precise predication of ACHBLF.

关 键 词：肝功能衰竭 预后 曲线下面积 MELD相关评分体系 

分 类 号：R575.3[医药卫生—消化系统]