不同肝病基础的重症酒精性肝炎患者短期预后评估及其影响因素

Evaluation and influencing factors of the short-term prognosis of severe alcoholic hepatitis with different underlying liver diseases

目的探讨不同肝病基础的重症酒精性肝炎(AH)患者临床特征及其短期预后的评估和影响因素。方法回顾性分析天津市第三中心医院2004年8月—2018年8月收治的170例重症AH患者临床资料,按不同肝病基础分为A型(无肝硬化,n=27)、B型(代偿期肝硬化,n=52)和C型(失代偿期肝硬化,n=91)。计算Maddrey判别函数(MDF)评分、慢性肝衰竭序贯性器官衰竭评估(CLIF-SOFA)评分、终末期肝病模型(MELD)评分、ABIC评分(年龄、胆红素、国际标准化比值、肌酐)以及Glasgow酒精性肝炎评分(GAHS)。计量资料多组间比较采用方差分析或Kruskal-Wallis H检验;计数资料多组间比较采用χ^2检验。采用单因素和多因素Cox回归分析筛选影响重症AH患者短期预后的独立危险因素。应用Kaplan-Meier法绘制生存曲线,生存差异组间比较采用log-rank检验。受试者工作特征曲线计算各预测模型的曲线下面积(AUC)及95%CI、敏感度、特异度,并应用DeLong法进行比较。结果A、B、C型患者28 d生存率分别为88.9%、80.8%和51.6%,3组比较差异有统计学意义(χ^2=19.83,P<0.001)。MELD评分、MDF评分、GAHS评分、ABIC评分和CLIF-SOFA评分预测28 d病死率的AUC(95%CI)分别为0.584(0.493~0.676)、0.696(0.605~0.786)、0.644(0.554~0.735)、0.745(0.662~0.827)和0.795(0.726~0.863);CLIF-SOFA评分与MDF评分、MELD评分、GAHS评分相比,差异均有统计学意义(P值均<0.05);CLIF-SOFA评分预测28 d病死率的最佳阈值为8.50分,敏感度为79.0%,特异度为67.9%。发病时不同肝病基础(HR=2.296,95%CI:1.356~3.887,P=0.002)以及合并肝性脑病(HR=1.911,95%CI:1.059~3.449,P=0.031)是28 d预后的危险因素。结论不同肝病基础的重症AH患者具有不同的临床特征和短期预后,发病时不同肝病基础及合并肝性脑病与重症AH患者28 d预后密切相关。CLIF-SOFA评分能够较好地预测重症AH患者28 d预后。

Objective To investigate the clinical features of patients with severe alcoholic hepatitis(AH)with different underlying liver diseases and the influencing factors for short-term prognosis.Methods A retrospective analysis was performed for the clinical data of 170 patients with severe AH who were admitted to Tianjin Third Central Hospital from August 2004 to August 2018,and according to the underlying liver disease,they were divided into group A(27 patients without liver cirrhosis),group B(52 patients with compensated liver cirrhosis),and group C(91 patients with decompensated liver cirrhosis).Related scores were calculated,including Maddrey’s discriminant function(MDF)score,Chronic Liver Failure-Sequential Organ Failure Assessment(CLIF-SOFA)score,Model for End-Stage Liver Disease(MELD)score,age-bilirubin-international normalized ratio-creatinine(ABIC)score,and Glasgow alcoholic hepatitis score(GAHS).An analysis of variance or the Kruskal-Wallis H test was used for comparison of continuous data between multiple groups,and the chi-square test was used for comparison of categorical data between multiple groups.Univariate and multivariate Cox regression analyses were used to screen out the independent influencing factors for the short-term prognosis of patients with severe AH.The Kaplan-Meier method was used to plot survival curves,and the log-rank test was used for comparison of survival rate between groups.The receiver operating characteristic(ROC)curve was used to calculate the area under the ROC curve(AUC)and 95%confidence interval(CI),sensitivity,and specificity for each predictive model,and the DeLong method was used for comparison.Results The 28-day survival rates of patients in groups A,B,and C were 88.9%,80.8%,and 51.6%,respectively,with a significant difference between the three groups(χ^2=19.83,P<0.001).The AUCs(95%CI s)of MELD score,MDF score,GAHS score,ABIC score,and CLIF-SOFA score were 0.584(0.493-0.676),0.696(0.605-0.786),0.644(0.554-0.735),0.745(0.662-0.827),and 0.795(0.726-0.863),respectively,in predicting 28-day mortality rate,and there were significant differences between CLIF-SOFA score and MDF,MELD,and GAHS scores(all P<0.05);CLIF-SOFA score had a sensitivity of 79.0%and a specificity of 67.9%at the optimal cut-off value of 8.50 points in predicting 28-day mortality rate.Different underlying liver diseases(hazard ratio[HR]=2.296,95%CI:1.356-3.887,P=0.002)and hepatic encephalopathy(HR=1.911,95%CI:1.059-3.449,P=0.031)at disease onset were risk factors for 28-day prognosis.Conclusion Patients with severe AH with different underlying liver diseases have different clinical features and short-term prognoses.Different underlying liver diseases and hepatic encephalopathy at disease onset are closely associated with the 28-day prognosis of patients with severe AH.CLIF-SOFA score can predict the 28-day prognosis of patients with severe AH.