Quantitative Diagnosis And Clinical Research On The Severity Of Patients With Liver Cirrhosis And Portal Hypertension

Background & ObjectiveIn our country, liver cirrhosis is the advanced stage of fibrosis in histology. According to the presence of hypohepatia, portal hypertension and other complications, the liver cirrhosis can be clinically divided into compensated stage and decompensated stage. As the intensive study of liver cirrhosis and anti-fibrosis treatment, we found that the disease severity of liver cirrhosis patients vary greatly, from no symptoms to appear ascites, esophageal and gastric varices bleeding, hepatic encephalopathy and even death. Meanwhile, liver cirrhosis histologically considered as only one grade generally. So We eagerly need a histological sub-classification of cirrhosis to guide the disease evaluation and clinical intervention. In recent years, foreign scholars have developed the Laennec system on the basis of METAVIR system, and proved that the classification system can primely predict the prognosis of liver cirrhosis patients.FibroScan is a method based on ultrasonic technology to rapidly diagnosis the liver fibrosis. This technology can quickly estimate the severity of fibrosis without liver biopsy and has relatively high precision. It’s clinical application value has been demonstrated by numerous researches. And it is proved that the accuracy of FibroScan to diagnosis S3 and S4 stage of liver fibrosis is higher than SI and S2 stage. But whether we can use FibroScan to further classify the liver cirrhosis, the correlation between liver stiffness and the Laennec classification of cirrhosis, have not yet been reported in the literature. The purpose of this research is to discuss the correlation between liver functional scores(Child-Pugh score, MELD score), FibroScan measurement and histological subclassification of cirrhosis.Portal hypertension (PHT) is a disease with high pressure of portal vein and its branches due to blood flow-blocking and hyperviscosity, which is caused by different causes. The principal clinical symptoms include splenomegaly, hypersplenism, esophageal and gastric varices, ascites and so on. In our country, the main cause of the portal hypertension is liver cirrhosis associated with chronic viral hepatitis. Currently, there are various methods to assess the portal vein pressure in clinical, including invasive methods (catheterization, percutaneous portal vein puncture guided by color Doppler ultrasound., measuring during the operation directly, etc) and noninvasive methods (hemodynamic parameters are obtained by color Doppler ultrasound, computed tomography, magnetic resonance imaging, radionuclide imaging, etc). Though we recognized the hepatic vein pressure gradient (HVPG) as "golden standard" to diagnose PHT, detecting portal vein pressures during the operation directly, because of its good accuracy and feasibility, plays an unparalleled role for patients who need surgery. This method can also observe the change of portal vein pressure and the effect of operation dynamically. Portal vein pressure can be used to guide clinical intervention and the selection of the different operative methods., and also evaluate the prognosis. But whether there is a significant correlation between the portal vein pressure and various clinical indexes is less reported in domestic. This paper retrospectively analyzed the correlation between portal vein pressure measured during the operation directly and clinical data, in order to preliminary explain the relationship between them.Methods1. Biopsy-proven cirrhosis patients were obtained from the Department of Hepatobiliary Surgery of Nanfang Hospital during the period of January 2012 to January 2014. In addition, patients with liver cancer, cardiovascular disease, acute infection, kidney disease and other diseases were excluded.60 cases were enrolled in this study finally. The clinical datas including albumin (ALB), prothrombin time (PT), prothrombin time international standardization ratio (PT-INR), total bilirubin (TBIL), creatinine (Cr) and the severity of hepatic encephalopathy and ascites were collected on the day before the liver biopsy. Histology of cirrhosis was blindly subclassified using the Laennec fibrosis scoring system semi-quantitatively without knowledge of the clinical dates. Our purpose was to analyze the difference of FibroScan measurements between patients of different stages of liver cirrhosis, feasibility of using FibroScan measurements to evaluate the severity of liver cirrhosis and the correlation between liver function scores and degrees of liver cirrhosis.2. Liver cirrhosis patients with portal hypertension were obtained from the Department of Hepatobiliary Surgery of Nanfang Hospital during the period of January 2012 to June 2014. Patients with liver cancer, cardiovascular disease, acute infection, kidney disease and other diseases were excluded. All cases were performed laparotomy and the free portal vein pressure(FPP) was measured directly during operations.50 cases were enrolled in this study finally. The patients were divided into two groups based on the the cut-off point of 30 cm H20 of portal vein pressure. The clinical states between the two groups was compared and correlation analysis and multiple regression analysis between FPP and clinical states was also researched.3. Statistical analysis:The results were expressed as means±standards when satisfied a normal distribution, or expressed as median (range). Student’test was used to compare the means of twolindependent samples.One-way ANOVA was used to deal with the means of multiple independent samples. Calculators information was analyzed by chi-square test. Pearson’s correlation was used to analyze the dates when materials met the bivariate normal distribution, or Spearman’s correlation was used. The receiver-operating characteristic curve was drawed. When maximal sensitivity and specificity were required, the liver stiffness measurement (LSM) was the best cutoff value. p value<0.05 was considered to indicate a significant difference. The effect of clinical indexes on the portal vein pressure was explained by stepwise multiple regression analysis. Statistical analysis was performed with SPSS software for Windows (version 21.0; SPSS, Chicago, IL).Result1. Three groups of mild cirrhosis, moderate cirrhosis and severe cirrhosis were identified in 27(45%),21(35%),12(20%) patients, respectively. FibroScan values of the three groups were 14.90±5.54,25.23±10.11,33.99±13.55 (Kpa), respectively. The difference among the three groups was significant (P<0.05). After comparison inter-groups, there were significant differences between mild cirrhosis and moderate cirrhosis (P=0.001) and also between mild cirrhosis and severe cirrhosis (P=0.001), but no significant differences between moderate cirrhosis and severe cirrhosis (P=0.181). The areas under the receiver operating characteristic(ROC) curve of FibroScan were 0.908 for moderate to severe cirrhosis, and 0.865 for severe cirrhosis. When we chose LSM= 16.05 Kpa as cutoff value, FibroScan used to diagnose moderate and severe cirrhosis had a sensitivity of 97%, a NPV of 95%, a NLR of 0.04, a specificity of 70%, a PPV of 80%, a PLR of 3.28, and a diagnostic accuracy of 85%. When we chose LSM=21.10 Kpa as cutoff value, FibroScan used to diagnose severe cirrhosis had a sensitivity of 92%, a NPV of 97%, a NLR of 0.12, a specificity of 71%, a PPV of 44%, a PLR of 3.14, and a diagnostic accuracy of 75%. Child-Pugh scores of the three groups were 5.61±1.14、6.05±1.21、 5.74±0.93, respectively. MELD scores of the three groups were 5.90±4.22、7.14± 5.33、7.03±5.13, respectively. The difference of Child-Pugh scores and MELD scores among the three groups was not significant(P>0.05).2. Neutrophile granulocyte percentage (N), total bilirubin (TBIL), direct bilirubin (DBIL), spleen diameter, spleen width, Child-Pugh score remarkably increased and lymphocyte percentage (L), prothrombin time activity (PTA) declined with increases of FPP. The difference of these indexes among the two groups was significant (P<0.05). Alanine transaminase (ALT), aspertate aminotransferase (AST), globulin (GLB), direct bilirubin (DBIL), spleen diameter, spleen width, portal vein diameter, Child-Pugh score, volume of ascites were significantly positively correlated with FPP. Lymphocyte percentage (L), albumin/globulin (A/G), prothrombin time activity (PTA), fibfinogen (FIB) were significantly negatively correlated with FPP. But the relationship was not close (the correlation coefficient was less than 0.5). The influence of these clinical indexes on FPP was studied by multivariate regression analysis. Finally, the independent variables entered the regression equation were volume of ascites, portal vein diameter, GLB and L, proved that there was a linear relationship between these four indexes and FPP. The regression equation was:FPP= 0.366 x portal vein diameter+0.425 x volume of ascites+0.375 x GLB-0.300 x L. Multiple correlation coefficient (R), coefficient of determination (R2) and adjusted R square was 0.698,0.487,0.435, respectively. This suggested that these four indexes could explain 43.5% of the change of FPP. The influence degree of each factors on FPP was volume of ascites, GLB, portal vein diameter and L according to the size of adjusted R square.Conclusions1. There was significant difference of FibroScan values among the three groups of mild, moderate, and severe liver cirrhosis. The areas under the receiver operating characteristic(ROC) curve of FibroScan were 0.908 for moderate to severe cirrhosis, and 0.865 for severe cirrhosis. So FibroScan is a valuable examination for liver cirrhosis. There is no correlation between liver functional scores and histological subclassification of cirrhosis.2. Some clinical indexes existed significant differences between two groups of cirrhotic patients with portal hypertension. Based on four indexes of ascites, GLB, portal vein diameter and L, we could establish a multiple linear regression model to evaluate the portal vein pressure.