目的:结合终末期肝病评估模式(model for end-stage liver disease,MELD)评分评价血清降钙素原(procalcitonin,PCT)、内毒素和临床常用炎症指标对慢性重型肝炎预后的预测价值.方法:选择2011年5月至2011年12月在中南大学湘雅医院感染科住院的慢性重型肝炎患者124例,在入院24 h内测定其炎症指标、肝肾功能,进行MELD评分,并留取患者当日血清,集中测定内毒素、PCT和C-反应蛋白(C-reactin protein,CRP).在所有病例出院后随访3个月,并根据他们的转归分为生存组(n=58)和死亡组(n=66)进行比较分析.结果:1)124例患者中有66例死亡,58例生存.单因素分析表明:年龄、MELD评分、白细胞(white blood cell,WBC)计数、中性粒细胞(polymorphonuclear,PMN)计数、CRP和PCT在两组之间的差异有统计学意义(P＜0.05);二分类Logistic回归分析显示:年龄、MELD评分和PCT与预测慢性重型肝炎死亡与否存在显著性的相关关系,相对危险度(OR值)分别为1.07,1.42和1.02(P＜0.05),并且可以预测慢性重型肝炎患者的死亡概率.2)MELD评分与患者预后密切相关,年龄和MELD评分呈正相关,Pearson相关系数为0.21(P＜0.05),PCT和MELD评分呈正相关,Spearman相关系数为0.54(P＜0.01).3)受试者工作特征曲线(receiver operation characteristic curve,ROC)分析表明:MELD评分和PCT的ROC曲线下面积(area under the curve,AUC)较高,分别为0.91和0.77(P＜0.01).当MELD评分≥30.09分时,预测患者死亡正确率最高(82.26％),而单独使用PCT及联合PCT和MELD评分预测患者死亡正确率仅有75.00％.但MELD评分联合PCT评价患者预后的特异性高达100.00％,阳性预测值高达1.00.结论:内毒素和常用炎症指标(WBC,PMN,CRP)不是预测慢性重型肝炎预后的可靠指标.MELD评分与慢性重型肝炎预后明显相关,PCT和年龄均与MELD评分呈正相关.结合MELD评分、PCT和年龄可以预测慢性重型肝炎患者3个月内的死亡概率.PCT和MELD评分均对预测慢性重型肝炎患者预后有价值,但PCT不如MELD评分的预测价值大.MELD评分联合PCT可提高对患者预后预测的特异性.%Objective:To evaluate the mid-term prognostic value of procalcitonin (PCT),endotoxin and common inflammatory markers combining the model for end-stage liver disease (MELD) score in patients with chronic severe hepatitis.Methods:A total of 124 chronic severe hepatitis patients were enrolled,who were hospitalized in the Department of Infectious Diseases,Xiangya Hospital,Central South University from May 2011 to December 2011.Indexes of inflammation,liver and kidney function tests and MELD were determined within 24 h after the admission,and blood samples were collected for measurement of endotoxin,procalcitonin (PCT),and C-reactin protein (CRP).The outcome was confirmed after discharge follow-up at the end of the 3rd month.According to the outcome,the 124 patients were divided into a survival group (n=58) and a death group(n=66).Results:1) Of the 124 patients,66 died and 58 survived,with statistical difference in age,MELD score,white blood cell (WBC),polymorphonuclear (PMN),CRP and PCT by single factor analysis between the 2 groups(P＜0.05).Binary logistic regression analysis indicated that age,MELD scores and PCT were highly correlated with the outcome (OR=1.07,1.42 and 1.02 respectively,P＜0.05),which could be used to predict the 3 month mid-term mortality of chronic severe hepatitis.2)There was significant correlation between the MELD scores and the mid-term mortality.Age was positively correlated with the MELD score,and Pearson's correlation coefficient was 0.21 (P＜0.05).PCT was also positively correlated with the MELD,and Spearman's correlation coefficient was 0.54 (P＜0.01).3)According to the receiver operation characteristic (ROC) curve analysis,the area under the curve (AUC) of MELD score and PCT were 0.91 and 0.77 respectively,higher than those of other indexes (P＜0.01).When the MELD score was up to 30.09 or higher,the predicted mortality risk among these tested patients was the highest(82.26％).The mortality risk predicted by PCT combining MELD score and PCT alone was lower than by MELD score alone (75.00％),but the specificity of MELD score combining PCT was 100％,and the positive prediction value was 1.00.Conclusion:Endotoxin and common inflammatory markers (WBC,PMN,and CRP) are not reliable indicators to predict the prognosis in patients with chronic-severe hepatitis.MELD score is significantly correlated with the outcome of mid-term chronic severe hepatitis,PCT and age are both positively correlated with the MELD score.PCT and age combining MELD score can be used to predict the 3 month mid-term mortality of chronic severe hepatitis.MELD score has better prognostic value than PCT.MELD score combining PCT can improve the specificity of prediction.
展开▼
