成人良性终末期肝病肝移植受体术中大量输血的预测因素分析

目的 分析成人良性终末期肝病肝移植受体术中发生大量输血的术前预测因子.方法 回顾性总结浙江大学医学院附属第一医院肝移植中心2003年1月至2007年12月268例成人终末期良性肝病(ESLD)肝移植受体的临床资料,对术中输血量≥12 U(MBT组)和12 U(NBT组)两组受体的术前各项指标进行比较,采用Logistic回归分析预测术中大量输血发生的预测因素;分析5组麻醉医生对术中输血量的影响.结果 268例受体术中5例未输注红细胞和血浆,其余98.13%的受体术中输注红细胞,全组平均(13±8)U.手术由同一外科团队完成,不同麻醉医生分组对输血量影响差异无统计学意义(P>0.05).术毕两组受体的血红蛋白值差异无统计学意义,P>0.05.单因素分析显示术中大量输血的术前预测因素为胆红素、肌酐、INR、白陶土部分凝血活酶时间、血小板、血红蛋白、总蛋白、MELD评分、是否重症肝炎以及术前腹水;多因素分析结果 显示独立预测因素为MELD评分和血红蛋白,预测评分=0.593-0.049×术前血红蛋白+0.137×术前MELD评分,其ROC曲线下而积为0.832.截断点为-0.67分(计算结果大于-0.67提示术中大量输血),其预测灵敏度为84.1%,特异度为71.2%.结论 以乙型肝炎为主要原发疾病的ESLD患者,术中发生大量输血的独立预测因素为术前MELD评分和血红蛋白值,其预测模型具有较好的敏感度和特异度.

Predictors of massive blood transfusion in liver transplantation for patients with benign end-stage liver disease

Objective To identify the predictors of massive blood transfusion in liver transplantation for patients with benign end-stage liver disease (ESLD). Methods The clinical data of 268 patients with ESLI) mainly caused by hepatitis B who were to receive cadaver liver transplantation performed by 5 groups of anesthesiologists respectively were retrospectively reviewed. The patients were divided into 2 groups according to the transfusion amount:massive blood transfusion (MBT) group (n=120) receiving the blood transfusion of more than 12 units of red blood cells (RBCs) during the operation,and non- massive blood transfusion (NBT) group (n=148). Univariate analysis and stepwise logistic regression were used. The influence of anesthesiologists on administration of blood cells was analyzed. Results During the liver transplantation, 98.13% of the patients received RBC transfusion of the mean dose of 13.32 units. Different group of anesthesiologists exerted no influence on the amount of RBC transfusion. Logistic regression showed that the preoperative MELD score, hemoglobin, platelet, interactional normalized ratio of prothrombin time, kaolin partial thromboplastin time, total protein, serum creatinine, total bilirubin, ascite, and severity of hepatitis were risk factors of MBT. Multivariable logistic regression showed that 2 independent predictors of MBT were identified: preoperative MELD score and hemoglobin. Accordingly, MBT during liver transplantation was predicted by 0.593 -0.049 × preoperative hemoglobin + 0.137 × preoperative MELD score, with a cut off value of -0.67, a sensitivity of 84.1% and a specify of 71.2%. The area under receiver operating characteristic curves of prognostic score was 0.832. Conclusion MBT during liver transplantation can be predicted by preoperative MELD score and hemoglobin value in patients with ESLD secondary to type B hepatids.