HBV感染所致重型肝炎肝衰竭影响肝移植预后的危险因素及肝移植手术时机选择的临床研究

发布时间：2018-06-21 03:42

本文选题：重型肝炎 + 肝衰竭　；参考：《第三军医大学》2013年硕士论文

【摘要】：背景和目的肝衰竭(liver failure，LF)是由多种病因引起的严重肝脏损害，以黄疸、肝性脑病、凝血机制障碍及肝肾综合征等为主要临床表现的严重肝病症候群。在我国LF的主要病因是病毒性肝炎，尤其是乙型病毒性肝炎感染所致的重型肝炎肝衰竭(severe hepatitisB-induced liver failure，SHBLF)；在欧美等西方国家，LF主要病因是药物(尤其是对乙酰氨基酚)及酒精性肝损害。SHBLF病情危重，预后凶险，病死率高，内科治疗生存率不足30%，而施行肝脏移植(liver transplantation，LTx)后1个月和1年的生存率分别为90%和70%，LTx已被国际上公认为治疗SHBLF最有效的方法。世界范围内的供肝短缺对LTx的发展是一项极大的挑战，且SHBLF患者病情进展迅速，LTx的手术时机稍纵即逝；因此如何及时、准确、客观的评估LF的严重程度并预测其行LTx的预后，据此选择最佳的手术时机，合理化分配供肝，对挽救患者生命、提高其生存率至关重要。目前国外多采用英国皇家学院(King’s College Hospital，KCH)标准、终末期肝病模型(model for end-stage liver disease，MELD)评分等评估系统进行LF患者行LTx手术时机的选择、受体的筛选及预后的预测；但不论是在受体病因及分类的构成方面，还是在患者的LTx指征与手术时机的选择方面，我国都与欧美等西方国家有很大的不同，且KCH标准及MELD评分对预后的预测效能尚存在争议。本课题选择我中心近10年来因SHBLF行LTx的受者为研究对象，对其临床病历及术后随访资料进行回顾性研究，利用生存分析、受试者工作特征曲线等统计学方法确定影响预后的危险因素，建立预后评估模型，评估其预测能力，探讨行LTx的最佳手术时机，为我国SHBLF肝移植受体的选择及手术时机的决断提供科学依据。方法 1．研究对象：根据相关纳入、排除及诊断标准，筛选出第三军医大学第一附属医院肝移植中心自1999年1月至2010年12月收治的病因为SHBLF而行LTx的受者共98例，收集整理在院及术后随访数据，建立数据资料库。随访截止日期为2011年12月31日。 2．根据文献资料结合临床经验，提出可能与SHBLF行LTx预后相关的因素；先利用单因素Cox比例风险回归模型(简称Cox模型)分析初筛，再行多因素Cox模型逐步回归分析得到影响预后的主要变量。 3．利用筛选出的变量及其回归系数构建HBV感染所致重型肝炎肝衰竭肝移植预后评估模型(severe hepatitis B-induced liver failure prognosis model，SHBLFPM)。 4．利用受试者工作特征(receiver-operating-characteristic，ROC)曲线下面积(theareas under the curves，AUCs)的比较评估SHBLFPM、KCH标准及MELD评分的预测能力。 5．根据SHBLF的ROC曲线最佳诊断阈值(cut-off value)分组，绘制两组Kaplan-Meier生存曲线，并用log-rank检验比较两组差异，探讨最佳手术时机。 6．本研究中的计量资料用均数±标准差(mean±SD)表示，，除三个模型AUCs的比较使用MedCalc统计软件进行外，所有统计采用SPSS13.0统计软件处理，P0.05被认为差异有统计学意义。结果 1．受者一般情况及总体生存分析：入选98例SHBLF受者，其中男性88例，女性10例，平均年龄42±9岁。到随访截止日时，无一例患者失访，其中64例存活，34例死亡，总死亡率为34.7%；LTx术后1个月、6个月和1年受体的生存率分别为76.5%、70.4%和67.3%。 2．HBV所致重型肝炎肝衰竭肝移植预后评估模型的构建。 (1)单因素Cox模型初筛结果：共8个因素对SHBLF移植受体预后的影响有统计学意义，分别是：年龄(AGE，P=0.036)、肝肾综合征(HRS, P=0.003)、肝性脑病(HE, P=0.016)、白细胞(WBC, P=0.037)、总胆红素(TBiL, P0.001)、凝血酶原时间国际标准化比值(INR,P0.001)、肌酐(Cr, P=0.008)及尿素氮(UN, P=0.032)。 (2)多因素Cox模型逐步回归结果：最终有4个变量进入方程，分别是：AGE(P=0.017)、HE(P=0.013)、TBiL(P0.001)及INR(P=0.001)。 (3)利用筛选出的变量及其回归系数，构建得到HBV感染所致重型肝炎肝衰竭肝移植预后评估模型(SHBLFPM)： SHBLFPM=1.806×loge[AGE(岁)]+2.221×loge[TBiL(μmol/L)]+1.951×logeINR+0.318×HE(无肝性脑病时取0；肝性脑病Ⅰ-Ⅳ级分别对应1-4)。 3．新模型SHBLFPM预测效能的评估 (1)SHBLFPM、 KCH标准及MELD评分ROC曲线下面积(AUC)的比较：SHBLFPM、KCH标准及MELD评分的AUC分别为0.881、0.596及0.783；SHBLFPM的预测效能明显优于KCH标准及MELD评分，MELD评分有一定的临床应用价值，而KCH标准预测价值差。 (2)SHBLFPM评分在23.57时有最好的预测效能；根据此值将受体分为高分组和低分组，比较两组生存时间及生存曲线，低分组预后明显好于高分组(P0.001)。结论 1．AGE、HE、TBiL及INR是影响SHBLF受体行LTx预后的主要危险因素。 2．根据我中心资料构建得到HBV感染所致重型肝炎肝衰竭肝移植预后评估模型SHBLFPM=1.806×loge[AGE(岁)]+2.221×loge[TBiL(μmol/L)]+1.951×logeINR+0.318×HE(无肝性脑病时取0；肝性脑病Ⅰ-Ⅳ级分别对应1-4)。 3．新模型SHBLFPM能准确预测SHBLF受体行LTx的预后，其预测效能明显优于KCH标准及MELD评分，MELD评分有一定的临床应用价值，而KCH标准预测价值差。 4．术前SHBLFPM评分≤23.57组行LTx的预后明显好于术前SHBLFPM评分23.57组。 5．对术前SHBLFPM评分≤23.57的受者，在积极内科支持治疗的基础上择期行LTx；对术前SHBLFPM评分23.57的受者，应在严密监护及强化支持治疗的基础上急诊行LTx。
[Abstract]:Background and purpose
Liver failure (LF) is a serious liver damage caused by a variety of causes, with jaundice, hepatic encephalopathy, coagulation mechanism and hepatorenal syndrome as the major symptoms of liver disease. In China, the main cause of LF is viral hepatitis, especially severe hepatitis B caused by hepatitis B virus infection (severe HepatitisB-induced liver failure, SHBLF); in western countries such as Europe and America, the main cause of LF is drug (especially acetaminophen) and alcoholic liver damage,.SHBLF is critical, the prognosis is dangerous, the mortality rate is high, the survival rate of medical treatment is less than 30%, and the survival rate of 1 months and 1 years after the liver transplantation (liver transplantation, LTx). For 90% and 70%, LTx has been internationally recognized as the most effective way to treat SHBLF.
The shortage of donor liver in the world is a great challenge for the development of LTx, and the patient's condition of SHBLF is progressing rapidly and the timing of LTx's operation is fleeting. Therefore, how to evaluate the severity of LF and predict the prognosis of LTx in a timely, accurate and objective way, so as to choose the best operation time, rationalize the distribution of donor liver, and save the life of the patient. It is very important to improve the survival rate. At present, many foreign countries adopt the King 's College Hospital (KCH) standard, the end stage liver disease model (model for end-stage liver disease, MELD) scoring system for the selection of the timing of the operation, the selection of the recipient and the prognosis, but whether it is in the receptor cause and the prognosis. The composition of the classification, or indications and timing of surgery in patients with LTx in our country are very different from the western countries, and the forecasting performance of KCH standard and MELD score for the prognosis is still controversial.
This topic selected our center for the last 10 years for the study of SHBLF line LTx recipients. The clinical records and postoperative follow-up data were reviewed. The risk factors affecting the prognosis were determined by the survival analysis, the working characteristic curve of the subjects and other statistical methods. The prognosis evaluation model was established, the prediction ability was evaluated, and the most LTx was discussed. The best timing of operation is to provide scientific evidence for the selection of SHBLF liver transplant recipients and the timing of operation.
Method
1. research object: according to the inclusion and exclusion criteria, selected transplantation center, the First Affiliated Hospital of Third Military Medical University from January 1999 to December 2010 were liver disease due to SHBLF and LTx in 98 cases, collected in the hospital and postoperative follow-up data, establishment of data database. The deadline for follow-up in December 31, 2011.
2. according to the literature and clinical experience, the factors that may be related to the prognosis of SHBLF LTx were proposed. First, a single factor Cox proportional risk regression model (Cox model) was used to analyze the initial screening, and the stepwise regression analysis of the multi factor Cox model was used to get the main prognostic variables.
3. variables were selected by regression coefficient and construction of HBV infection in liver failure caused by severe hepatitis (severe hepatitis B-induced transplantation evaluation model of liver failure prognosis model, SHBLFPM).
4. using the receiver operating characteristic (receiver-operating-characteristic, ROC) and area under the curve (theareas under the curves, AUCs) of the evaluation and comparison of SHBLFPM, KCH and MELD score standard prediction ability.
5. according to the ROC curve of the best diagnostic threshold SHBLF (cut-off value) group, two Kaplan-Meier group draw survival curves, and use log-rank test to compare the difference between the two groups, to explore the optimal timing of surgery.
The mean and standard deviation of measurement data used in this study 6. (mean + SD) said, in addition to the three AUCs model compared with statistical software MedCalc, SPSS13.0 statistical software was used for statistical processing of all, P0.05 was considered statistically significant.
Result
1. the general situation and overall survival analysis: 98 cases of SHBLF recipients were selected, including 88 men and 10 women, with an average age of 42 9 years. When the follow-up deadline, none of the patients were lost, 64 of them survived, 34 died, and the total mortality was 34.7%; 1 months after LTx, the survival rates of 1 and 1 years were 76.5%, 70.4% and 67.3%., respectively.
Construction of a prognostic model for severe hepatitis liver failure caused by 2.HBV.
(1) preliminary screening results of single factor Cox model: a total of 8 factors have significant effects on the prognosis of SHBLF transplant recipients: age (AGE, P=0.036), HRS (P=0.003), hepatic encephalopathy (HE, P=0.016), leukocyte (WBC, P=0.037), total bilirubin (TBiL, P0.001), Prothrombin time international standardization ratio Anhydride (Cr, P=0.008) and urea nitrogen (UN, P=0.032).
(2) stepwise regression results of multivariate Cox models: finally, 4 variables entered the equation, namely: AGE (P=0.017), HE (P=0.013), TBiL (P0.001) and INR (P=0.001).
(3) using the selected variables and their regression coefficients, we constructed a liver transplantation prognosis assessment model for severe hepatitis caused by HBV infection (SHBLFPM):
SHBLFPM=1.806 * loge[AGE (age)]+2.221 x loge[TBiL (U mol/L)]+1.951 * logeINR+0.318 * HE (0 without hepatic encephalopathy); hepatic encephalopathy I - IV corresponding to 1-4 respectively.
Evaluation of predictive effectiveness of 3. new model SHBLFPM
(1) the comparison of area (AUC) under SHBLFPM, KCH standard and MELD score ROC curve: SHBLFPM, KCH standard and MELD score AUC are 0.881,0.596 and 0.783 respectively; SHBLFPM predictive efficiency is obviously superior to KCH standard and score score.
(2) the SHBLFPM score is the best forecast performance in 23.57; according to this value the receptors are divided into high and low groups, compared two groups of survival time and survival curve, low packet prognosis is significantly better than the high score group (P0.001).
conclusion
1.AGE, HE, TBiL and INR are the major risk factors affecting the prognosis of SHBLF receptor LTx.
2. according to the data of our center, the prognosis evaluation model of liver failure of severe hepatitis caused by HBV infection was established, SHBLFPM=1.806 x loge[AGE (year old)]+2.221 x loge[TBiL (mu mol/L)]+1.951 x logeINR+0.318 x HE (0 in the absence of hepatic encephalopathy), and 1-4 of hepatic encephalopathy grade I - IV respectively.
3. new SHBLFPM model can accurately predict the prognosis of SHBLF receptor for LTx, its forecast performance is better than the standard KCH and MELD score, MELD score has certain clinical application value, and the predictive value of KCH standard deviation.
4. the prognosis of LTx before operation in group SHBLFPM was significantly better than that in group SHBLFPM before operation (23.57) in group SHBLFPM.
5. of the preoperative SHBLFPM score less than 23.57 of the subjects in support of active medical treatment on the basis of elective LTx; the preoperative SHBLFPM score by 23.57, should be in strict monitoring and strengthening support treatment on the basis of emergency LTx.
【学位授予单位】：第三军医大学
【学位级别】：硕士
【学位授予年份】：2013
【分类号】：R575.3

【参考文献】