肝硬化门静脉高压症程度的临床定量诊断研究

发布时间：2018-04-25 13:25

本文选题：肝硬化 + 组织学分级　；参考：《南方医科大学》2015年硕士论文

【摘要】：研究背景与目的：在国内,肝硬化在组织学上是纤维化的终末阶段(S4),临床上又依据是否有肝功能减退、门静脉高压症状及各种并发症,将肝硬化分为代偿期肝硬化与失代偿期肝硬化。随着对肝硬化研究的深入以及抗纤维化治疗的发展,人们发现肝硬化患者的疾病严重程度存在较大的差别,从毫无症状,到出现腹水、食管胃底静脉曲张或破裂出血、肝性脑病甚至死亡。但在组织学上肝硬化仅笼统地归为一个级别,我们亟须对肝硬化进行进一步分级,以指导对患者病情的评估与临床干预的选择。近年来国外学者在肝纤维化METAVIR分级系统的基础上制定了Laennec肝纤维化分级系统,并证实了该分级系统能够较好地预测肝硬化患者的预后。FibroScan是基于超声技术基础上的快速诊断肝纤维化严重程度的方法,不必对患者进行肝脏穿刺活检即可快速评估纤维化严重程度,且具有一定的准确度,因此近些年对此项技术进行了大量研究与推广应用,并证明了此项技术对于诊断中重度纤维化具有较高临床应用价值。那么我们是否可以应用FibroScan技术对肝硬化的严重程度进行进一步的评估,所测得的FibroScan值与肝硬化的Laennec分级之间是否具有相关性,此类文献在国内报道较少。本研究旨在对肝硬化组织学分级与临床肝脏功能积分(Child-Pugh评分、MELD评分)及肝脏FibroScan值的相关性进行初步探讨。肝硬化门静脉高压症是由肝硬化引起的门静脉血流受阻、血液瘀滞,进而导致门静脉系统压力升高的一类疾病的统称。临床表现主要有脾大、脾功能亢进、食管胃底静脉曲张、腹水等。目前临床上有各种各样的方法来评估门静脉压力,包括有创测定法(门静脉导管术法、B超引导下门静脉穿刺法、术中直接测定等)和无创测定法(通过多普勒超声、CT、MRI、放射性核素等测定血流动力学指标来评估门静脉压力)。虽然公认的判断PHT的“金标准”为肝静脉压力梯度(hepatic vein pressure gradient, HVPG),但对于需要手术治疗的患者,术中直接测定门静脉压力因操作简单、准确,且可动态观察压力的变化与手术治疗的效果,具有不可替代的地位。用上述各种方法测定的门静脉压力可以用来指导临床干预和手术方式的选择,并对患者的预后进行初步评估。但患者门静脉压力与各种临床指标之间是否存在相关性,相关程度如何,国内文献报道较少。本文通过对开腹手术中直接测定的门静脉压力与患者临床资料进行回顾性分析,以期对二者的相关性进行初步探讨。方法：1、收集2012年1月至2014年1月在南方医科大学南方医院住院并经肝脏穿刺活检证实为肝硬化的病例,排除肝癌、心脑血管疾病、急性感染、肾病等其他疾病后共60例,收集的临床资料包括白蛋白(ALB)、凝血酶原时间(PT)、凝血酶原时间国际标准化比值(PT-INR)、总胆红素(TBIL)、肌酐(Cr)以及患者肝性脑病及腹水的严重程度等。所有病例临床资料的收集均在肝脏穿刺前一天完成。运用Laennec分级系统对病理切片进行组织学分级,分析肝硬化不同分级间FibroScan测定值是否存在差异、FibroScan测定值用来判断肝硬化严重程度的效果如何以及肝脏功能评分与肝硬化程度的相关性。2、收集2012年1月至2014年6月在南方医科大学南方医院肝胆外科住院并诊断为肝硬化门静脉高压症患者的临床资料,所有患者均行开腹手术且术中直接测定门静脉自由压(free portal pressure, FPP),并排除肝癌、心脑血管疾病、急性感染、肾病等其他疾病后共50例。记录患者术中测得的门静脉自由压(FPP),依据门静脉压力的高低,以30 cmH20为分界点,将患者分为Ⅰ、Ⅱ两组,比较两组间的临床指标是否有差异,并对FPP与临床指标进行相关性分析与多元回归分析。3、统计学处理：计量资料满足正态分布时以均数±标准差(x±s)表示,否则用中位数(极小值～极大值)表示,对成组设计的两个样本均数的比较采用独立样本t检验,对成组设计的多个样本均数的比较采用单因素方差分析,计数资料采用X2检验；双变量正态分布资料相关分析采用Pearson相关分析,非双变量正态分布资料采用Spearman相关分析；绘制受试者工作特征(receiver operating characteristic, ROC)曲线,以敏感度与特异度之和的最大值所对应的肝脏弹性值(liver stiffness measurement, LSM)作为最佳界值。各临床指标对门静脉压力的影响作用采用逐步多元回归分析。P0.05代表有统计学意义。所有统计学资料采用SPSS21.0统计学软件处理。结果：1.组织学分组为肝硬化轻度组27例(45%)、中度组21例(35%)、重度组12例(20%)。肝硬化轻、中、重度组的肝脏FibroScan测定值分别为14.90±5.54、25.23±10.11、33.99±13.55 (Kpa),三组间的差异具有显著性(P0.05)；组内比较后发现轻度组与中度组(P=0.001)、轻度组与重度组(P＝0.001)均值的差异具有统计学意义,中度组与重度组均值的差异无统计学意义(P=0.181)。肝脏FibroScan测定值用于判断中重度肝硬化和重度肝硬化的ROC曲线下面积分别为0.908、0.865。选肝脏弹性值LSM=16.05Kpa为界点时,诊断中重度肝硬化的灵敏度为97%,阴性预测值(NPV)为95%,阴性拟然比(NLR)为0.04,特异度为70%,阳性预测值(PPV)为80%,阳性拟然比(PLR)为3.28,正确率为85%。选LSM=21.10Kpa为诊断界点时,诊断重度肝硬化的灵敏度为92%,阴性预测值(NPV)为97%,阴性拟然比(NLR)为0.12,特异度为71%,阳性预测值(PPV)为44%,阳性拟然比(PLR)为3.14,正确率为75%。肝硬化轻、中、重度组的Child-Pugh评分分别为5.61±1.14、6.05±1.21、5.74±0.93,三组间的差异无显著性(P0.05)；肝硬化轻、中、重度组的MELD评分分别为5.90±4.22、7.14±5.33、7.03±5.13,三组间的差异无显著性(P0.05)。2.门静脉压力不同分组间有显著差异的指标有淋巴细胞百分数(L)、中性粒细胞百分数(N)、总胆红素(TBIL)、直接胆红素(DBIL)、凝血酶原活动度(PTA)、脾脏长、脾脏宽、CHILD评分,共9项指标；随着FPP增大, N、TBIL、DBIL、脾脏长、脾脏宽、CHILD评分逐渐增大,L、PTA逐渐减小。FPP与L、谷丙转氨酶(ALT)、谷草转氨酶(AST)、球蛋白(GLB)、A/G、DBIL、PTA、血浆纤维蛋白原测定(FIB)、脾脏长、脾脏宽、门静脉内径、CHILD评分及腹水量存在显著相关关系,其中与L、A/G、PTA、FIB存在负相关关系,与其余指标存在正相关关系,但与这些指标的相关关系并不密切(相关系数小于0.5)。以门静脉自由压(FPP)为因变量,以各项临床指标为自变量,采用逐步回归的方法进行多元线性回归分析,最终进入回归方程的自变量为腹水量、门静脉内径、GLB、L,说明腹水量、门静脉内径、GLB、L与FPP存在线性关系,建立的回归方程为FPP=0.366×门静脉内径+0.425×腹水量+0.375×球蛋白—0.300×淋巴细胞百分数。此模型的复相关系数R=0.698,决定系数R2=0.487,调整的R2=0.435,说明这四个变量可以解释因变量FPP 43.5%的变化。另外,根据各个变量标准化回归系数的大小可以看出,四个变量对FPP影响的大小依次为腹水量、球蛋白、门静脉内径和淋巴细胞百分数。结论：1.肝硬化轻、中、重度组的肝脏FibroScan测定值之间存在显著差异,肝脏FibroScan测定值用于判断中重度肝硬化和重度肝硬化的ROC曲线下面积分别为0.908、0.865,从而说明FibroScan在诊断肝硬化严重程度方面有一定的应用价值；肝硬化程度与肝脏功能评分无显著相关性。2.肝硬化门静脉高压症患者I、II两组之间有多个临床指标存在显著性差异,门静脉压力与L、ALT, AST, GLB、A/G, DBIL、PTA、FIB、脾脏长、脾脏宽、门静脉内径、CHILD评分及腹水量存在显著相关关系,并可以通过基于腹水量、球蛋白、门静脉内径和淋巴细胞百分数的多元线性回归模型对门静脉压力进行初步评估。
[Abstract]:Research background and purpose: in China, liver cirrhosis is histologically the final stage of fibrosis (S4). The liver cirrhosis is divided into compensatory cirrhosis and decompensated cirrhosis depending on whether the liver function is hypofunction, the symptoms of portal hypertension and all kinds of complications. With the deep study of cirrhosis and the development of anti fibrosis treatment There is a large difference in the severity of the disease in patients with liver cirrhosis, from asymptomatic, to ascites, esophageal varices or ruptured bleeding, and hepatic encephalopathy and even death. However, in histologically, the liver cirrhosis is only in a general level, and we are urgently required to further classify cirrhosis of the liver to guide the patient's condition. In recent years, foreign scholars have developed a Laennec classification system for liver fibrosis based on the METAVIR classification system of liver fibrosis, and confirmed that the classification system can predict the prognosis of liver cirrhosis patients better than that of.FibroScan, which is based on the rapid diagnosis of the severity of liver fibrosis based on ultrasound technology. The method, without liver biopsy, can quickly evaluate the severity of fibrosis and has a certain degree of accuracy. Therefore, this technique has been extensively studied and applied in recent years. It has been proved that this technique has a high clinical application value for the diagnosis of medium and severe fibrosis. Then whether we can apply FibroS or not The severity of liver cirrhosis was further assessed by can technique. The correlation between the measured FibroScan values and the Laennec classification of liver cirrhosis was found in the domestic report. The purpose of this study was to evaluate the histological grading of liver cirrhosis with the clinical liver function score (Child-Pugh score, MELD score) and the FibroScan value of the liver. The clinical manifestation of portal hypertension caused by cirrhosis is a general name for a class of diseases, such as splenomegaly, hypersplenism, varicose esophagogastric vein, and ascites. Evaluation of portal pressure, including invasive measurement (portal venous catheterization, B-ultrasound guided portal vein puncture, intraoperative direct measurement, etc.) and noninvasive measurement (using Doppler ultrasound, CT, MRI, radionuclides, and other hemodynamic indicators to assess portal pressure). Although it is recognized that the "gold standard" of PHT is the hepatic vein pressure The gradient (hepatic vein pressure gradient, HVPG), but for patients who need surgery, the direct measurement of portal pressure in the operation is simple, accurate, and can dynamically observe the changes of pressure and the effect of the operation. The portal pressure measured by these methods can be used to guide the clinical intervention. The selection of surgical methods and the prognosis of patients were evaluated preliminarily. However, there was a correlation between the portal pressure of the patients and the various clinical indicators, and the related degree was less reported in the domestic literature. Methods: 1. 1. A total of 60 cases of liver cirrhosis, including liver cancer, cardio cerebral vascular disease, acute infection and kidney disease, were collected from January 2012 to January 2014 at the Southern Hospital of Southern Medical University and were confirmed by liver biopsy. The clinical data included albumin (ALB), prothrombin, and prothrombin. Time (PT), prothrombin time international normalized ratio (PT-INR), total bilirubin (TBIL), creatinine (Cr), and the severity of hepatic encephalopathy and ascites. All cases were collected on the day before the liver puncture. The pathological sections were classified by the Laennec grading system, and the different grades of liver cirrhosis were analyzed. Whether there is a difference in the value of FibroScan determination, how the value of FibroScan is used to determine the severity of liver cirrhosis and the correlation between the liver function score and the degree of liver cirrhosis,.2, which was hospitalized in the Department of hepatobiliary surgery, Southern Hospital of Southern Medical University from January 2012 to June 2014, and was diagnosed as the clinical management of patients with cirrhosis of the portal hypertension. Materials, all patients were operated on open abdominal surgery (free portal pressure, FPP), and 50 cases of liver cancer, cardio cerebral vascular disease, acute infection, kidney disease and other diseases were excluded. The free pressure of portal vein (FPP) was recorded during the operation, and the level of portal vein pressure was 30 cmH20 as demarcation point. Divided into group I, group II and two groups, compare the difference between the clinical indexes between the two groups, and analyze the correlation between the FPP and the clinical indexes and the multivariate regression analysis.3. Statistical processing: when the measurement data satisfies the normal distribution, the mean number + standard deviation (x + s) is expressed, otherwise, the median (minimum value) is expressed, and the two samples are designed for the group. The comparison of the average number was compared with the independent sample t test. The single factor variance analysis was used for the comparison of the number of samples in the group design, and the count data was tested by X2 test. The correlation analysis of the bivariate normal distribution data was analyzed by Pearson correlation analysis, and the non bivariate normal distribution data were analyzed by Spearman correlation analysis, and the work characteristics of the subjects were drawn (rece Iver operating characteristic, ROC) curve, as the best value of the liver elasticity (liver stiffness measurement, LSM) corresponding to the maximum value of the sensitivity and the sum of the specificity. The effect of each clinical index on the portal pressure by stepwise multivariate regression analysis is statistically significant. All statistical data are collected. SPSS21.0 statistical software was used. Results: 1. histology groups were divided into 27 cases of liver cirrhosis (45%), 21 cases in moderate group (35%) and 12 cases in severe group (20%). The liver FibroScan values of liver cirrhosis were 14.90 + 5.54,25.23 + 10.11,33.99 + 13.55 (Kpa), respectively, and the difference between the three groups was significant (P0.05). The difference between the mild group and the moderate group (P=0.001), the mean value of the mild group and the severe group (P = 0.001) was statistically significant. The difference between the moderate and severe group was not statistically significant (P=0.181). The area of the liver FibroScan determination under the ROC curve of the moderate and severe cirrhosis and severe cirrhosis was 0.908,0.865. selection of the liver elasticity value respectively. When LSM=16.05Kpa was the boundary point, the sensitivity of the diagnosis of moderate and severe cirrhosis was 97%, the negative predictive value (NPV) was 95%, the negative pseudo natural ratio (NLR) was 0.04, the specificity was 70%, the positive predictive value (PPV) was 80%, the positive pseudo ratio (PLR) was 3.28, and the correct rate was 85%. selected LSM=21.10Kpa as the diagnostic point, the sensitivity of the diagnosis of severe cirrhosis was 92%, negative predictive value. (NPV) 97%, the negative pseudo natural ratio (NLR) was 0.12, the specificity was 71%, the positive predictive value (PPV) was 44%, the positive pseudo ratio (PLR) was 3.14, the correct rate was 75%. liver cirrhosis, the Child-Pugh score in the moderate and severe group was 5.61 + 1.14,6.05 + 1.21,5.74 + 0.93 respectively, and there was no significant difference between the three groups (P0.05); the MELD scores of the liver cirrhosis light, moderate and severe groups were respectively The difference between the three groups was 5.90 + 4.22,7.14 + 5.33,7.03 + 5.13. There was no significant difference between the three groups: the percentage of lymphocyte (L), the percentage of neutrophils (N), the total bilirubin (TBIL), the direct bilirubin (DBIL), the prothrombin activity (PTA), the spleen length, the spleen width, the CHILD score, and the 9 indexes. As FPP increased, N, TBIL, DBIL, spleen were long, spleen was wide, CHILD score increased gradually, L, PTA gradually reduced.FPP and L, glutamic pyruvic transaminase (ALT), gluten aminotransferase (AST), fibrinogen (GLB), spleen length, spleen width, portal vein diameter, score and ascites There is a negative correlation between A/G, PTA and FIB, and there is a positive correlation with the other indicators, but the correlation is not close to these indexes (the correlation coefficient is less than 0.5). With the free pressure of the portal vein (FPP) as the dependent variable, the multiple linear regression analysis is carried out by the stepwise regression method, and the regression equation is finally entered into the regression equation. The independent variable is the quantity of ascites, the diameter of the portal vein, GLB, L, indicating the linear relationship between the amount of ascites, the internal diameter of the portal vein, the GLB, the L and FPP. The regression equation is the +0.425 x ascites of the portal vein, +0.375 * globulin - 0.300 * lymphocyte percentage. The complex correlation coefficient R=0.698, the determining coefficient R2=0.487, R2=0.435 of the adjustment. These four variables can explain the change of the dependent variable FPP 43.5%. In addition, according to the size of the normalized regression coefficient of each variable, the size of the four variables affecting the FPP is ascites, globulin, portal vein and lymphocyte percentage. Conclusion: 1. the liver FibroScan values in the liver of the 1. liver are light, medium and severe. There were significant differences. The area of the liver FibroScan determination value used to determine the area under the ROC curve of moderate to severe cirrhosis and severe cirrhosis was 0.908,0.865, indicating that FibroScan had a certain value in the diagnosis of liver cirrhosis, and there was no significant correlation between the degree of liver cirrhosis and the evaluation of liver function,.2., the portal hypertension of cirrhosis. There were significant differences in multiple clinical indicators between the I and II two groups. Portal vein pressure was correlated with L, ALT, AST, GLB, A/G, DBIL, PTA, FIB, spleen length, spleen width, portal vein diameter, CHILD score and ascites, and could be multilinear through the amount of ascites, globulin, portal vein and lymphocyte percentage. The regression model was used to evaluate the pressure of the portal vein.

【学位授予单位】：南方医科大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：R657.34

【参考文献】