摘要
目的构建急性肝衰竭（acute liver failure,ALF）短期预后模型并与MELD、MELD-Na、IMELD和UKELD评分进行对比,分析其有效性。方法筛选234例ALF患者,采用回顾性病例对照研究,根据预后分为生存组与死亡组,比较两组间达到ALF诊断标准时的实验室相关指标及并发症发生情况。应用SPSS 19.0软件进行统计分析,Logistic多因素回归分析法对以上指标构建预测ALF患者短期死亡危险性的模型,通过受试者工作特征曲线法（ROC）检验模型预测有效性。结果 ALF 3个月死亡率为76.50%,筛选出与ALF预后相关变量分别是达到ALF诊断标准时的总胆红素（TB）（OR1.077,95%CI 1.028~1.130,P=0.002）,国际标准化比率（INR）（OR 2.555,95%CI 1.561~4.182,P=0.000）,血钠（Na＋）（OR 0.938,95%CI 0.883~0.996,P=0.037）,腹水（OR 2.804,95%CI 1.140~6.892,P=0.025）和感染（OR 5.348,95%CI 2.268~12.610,P=0.000）。建立ALF短期预后模型（model for acute liver failure,MALF）为：P=1/[1＋exp（-y）],y=4.448＋0.074×TB＋0.938×INR-0.064×Na＋＋1.031×腹水＋1.677×感染。模型敏感度为98.03%,特异度为97.26%。MALF曲线下面积（area under the curve of ROC,AUC）为0.952（95%CI 0.907~0.997）,高于MELD 0.783（95%CI0.720~0.845）、MELD-Na 0.825（95%CI 0.762~0.887）、IMELD 0.793（95%CI 0.726~0.860）、UKELD 0.841（95%CI 0.781~0.901）,均具有统计学差异（P〈0.001）。结论构建模型MALF具有较高的敏感度和特异性,对ALF患者短期死亡危险性评估有较好的临床应用价值,预测能力优于MELD、MELD-Na、IMELD和UKELD评分。
Objective To establish a new prognostic model of 3-month mortality in patients with acute liver failure（ ALF） and compare the values of our new model with the model for end-stage liver disease（ MELD）,MELD with incorporation of sodium（ MELD-Na）,integrated MELD（ IMELD） and United Kingdom model for end-stage liver disease（ UKELD） in predicting the prognosis of ALF. Methods In a retrospective case-control study,234 patients with ALF hospitalized in our institute from January 2000 to December 2012 were retrospectively recruited in the study,and divided into 2 groups according to their prognosis,that is,the survival and the dead. Their laboratory parameters and complications were compared.SPSS 19. 0 software was used for statistical analysis. Multivariate logistic regression analysis was used to establish the model. The predictive accuracy of models was compared using the area under the receiver operating characteristic（ ROC） curve. Results The 3-month mortality in ALF was 76. 50%. The factors predicting prognosis were total bilirubin（ TB,OR = 1. 077,95% CI = 1. 028 ~ 1. 130,P = 0. 002）,international normalized ratio（ INR,OR = 2. 555,95% CI = 1. 561 ~ 4. 182,P = 0. 000）,serum Na＋（ OR =0. 938,95% CI = 0. 883 ~ 0. 996,P = 0. 037）,ascites（ OR = 2. 804,95% CI = 1. 140 ~ 6. 892,P = 0. 025）and infections（ OR = 5. 348,95% CI = 2. 268 ~ 12. 610,P = 0. 000） when the indexes reached the criterion of ALF on admission. Our short-term prognosis model for ALF（ MALF） is： P = 1 /[1 ＋ exp（- y） ],y =4. 448 ＋ 0. 074 × TB ＋ 0. 938 × INR- 0. 064 × Na＋＋ 1. 031 × Ascites ＋ 1. 677 × Infections. MALF had a sensitivity of 98. 03% and a specificity of 97. 26%. MALF had a higher value of AUC（ 0. 952,95% CI =0. 907 ~ 0. 997）,when compared to MELD（ 0. 783,95% CI = 0. 720 ~ 0. 845）,MELD-Na（ 0. 825,95%CI = 0. 762 ~ 0. 887）,IMELD（ 0. 793,95% CI = 0. 726 ~ 0. 860）,and UKELD（ 0. 841,95% CI = 0. 781 ~0. 901）,and their differences were all statistically significan
出处
《第三军医大学学报》
CAS
CSCD
北大核心
2015年第18期1864-1869,共6页
Journal of Third Military Medical University
基金
国家自然科学基金面上项目(81270525)~~
