| BackgroundDue to the inflammation and necrosis of liver cells stimulated by a variety of causative agents, a large number of collagen is produced by activated hepatic stellate cells and deposits in the liver cells, liver fibrosis and thus forms. If the fibrous tissue unceasingly proliferates and the synthesis of extracellular matrix is greater than degeneration, fibrous tissue will invade liver cells and destroy the normal liver structure, many nodules surrounded by fibrous tissue will form, thus, liver texture will become hard and auantic, and develop to liver cirrhosis. The liver fibrosis is the early stage of liver cirrhosis. Current study suggests that liver fibrosis and early liver cirrhosis can still be reversed, but advanced liver cirrhosis is difficult to be reversed. Therefore, in the progression of chronic viral hepatitis, accurately understanding the degree of the liver fibrosis is of great importance for the treatment and prognosis of the disease.At present, the methods of monitoring liver fibrosis can be mainly classified as invasive and noninvasive ways. Although pathological examination of liver biopsy is considered the gold standard for diagnosis of liver fibrosis and the important basis for clear diagnosis, measuring inflammation, fibrosis, and determining the medicine curative effect. However, it is a kind of invasive examination method, and diagnostic error may be caused because of the uneven distribution of liver fibrosis in the liver and the difference form specimens and observers. Therefore, exploring a non-invasive alternative way to liver biopsy becomes a top priority.Biochemical test and imaging examination are the main non-invasive diagnostic methods. Most serum markers can only reflect the liver fibrosis in hepatitis active status, but they can not reflect the liver fibrosis in resting stage, moreover, the same serum marker may have different values in different laboratories. The use of ultrasound examination in the diagnosis of decompensated cirrhosis has been widely accepted, but the usefulness in the diagnosis of liver fibrosis and early liver cirrhosis should be further explored. Some other scholars have done a lot of research work in the semi-quantitative diagnosis of liver fibrosis, however, these results seem to be limited to the significant difference between the lowest and highest stage. There are many studies about the blood flow velocity and per unit time flow rate of portal vein on liver fibrosis patient with CDFI, but they lack the assessment based on morphological changes, and various reports vary.In addition, some scholars did studies about the diagnosis of liver fibrosis with CT and MRI, and they have made certain progress. However, the use of CT, MRI is limited in the diagnosis of fibrosis because of radiation damage, respiratory motion artifacts and high costs.Transient elastography, without the interference of other factors, is a direct measurement of liver's physical parameters and a new noninvasive diagnostic method for diagnosing liver fibrosis. Although the results of transient elastography can response the severity of liver fibrosis, the reported data about its stage has a lot of overlap, and delimiting threshold will affect the diagnostic sensitivity or the specificity.In the past 20 years, elastography has been developed rapidly. This technology can offer the information of tissue hardness and make up for the shortage of medical imaging mode, so, it is valuable for clinical diagnosis and has broad application prospect. Currently, elastography has been well applied in the breast, thyroid and lymph nodes, and achieved good results. However, there are rare reports about the application in the diagnosis of liver fibrosis, especially in the quantitative analysis of tissue diffusion.In this study, patients with chronic hepatitis B were studied and the pathological results of liver needle biopsy were regarded as the gold diagnostic standard. The 5-point method was adopted in part I, according to the elastographic score of liver tissue on the patients with liver fibrosis, whether did the patient suffer form liver fibrosis was determined. In part II, the quantitative analysis of tissue diffusion was used to access the severity of liver fibrosis, according to the mean strain value and the area percentage of blue color, the clinical value of this techology in determining whether the patient suffered from liver fibrosis as well as its severity was evaluated.Objectives1 To preliminarily explore the value of the ultrasonic elastography in evaluating liver fibrosis.2 To evaluate the application value of quantitative analysis of tissue diffusion in the differential diagnosis of liver fibrosis.Materials and Methods1. Clinical dataPart I:120 patients with chronic hepatitis B of our hopital in June 2009～April 2010 were included, of which 77 males,43 females, aged from 12 to 78 years, mean (37.3±14.3) years. Another 40 patients were selected and classified in healthy control group, of which 22 males and 18 females, aged from 20 to 55 years, mean (38.5±9.9) years. Percutaneous liver biopsy was performed on all patients to get a final pathological diagnosis. Part II:120 patients with chronic hepatitis B of our hopital in June 2009～April 2010 were included, of which 70 males,50 females, aged from 14 to 68 years, mean (38.03±12.05) years. Another 30 patients were selected and classified in healthy control group, of which 18 males and 12 females, aged from 15 to 60 years, mean (36.1±13.7) years. Percutaneous liver biopsy was performed on all patients to get a final pathological diagnosis.2. Apparatus and methodsPart I:Hitachi Vision 900 ultrasoud system with ultrasonic elastography was used in this study. The frequency of the linear array probe is 6.0～13.0 MHz and the frequency of the convex array probe is 4.0～8.0 MHz. First, the gray scale ultrasonography with high frequency array probe was performed on all patients and the echo of liver capsule and parenchyma of the control group people were observed, then, ultrasonic elastography was performed on liver tissue of ROI. Each patient were effectively scanned and recorded for 5 times.Elastographic score is divided from 1 to 5.1 point was named when ROI is almost entirely green or with a little red; 2 pionit was named when ROI is almost entirely green and just with a little bule cords; 3 point was named when ROI area is main green and with some bule Cords; 4 point was named when ROI is almost mixed with half blue and half green; 5 point was named when ROI is main blue and with some green and red. And then ROC were drawn, and 2 point was regarded as the deviding point for the diagnosis of liver fibrosis in the diagnosis, patients was diagnosed without liver fibrosis when the elastographic score is 1 point, and with liver fibrosis when the score is= 2 point.In the scanning condition of Gray scale:the patient was regarded without liver fibrosis when the liver capsule is smooth and the echo of liver parenchyma is normal or slightly coarse; they were regarded with liver fibrosis when liver capsule is not smooth or even if it is smooth but the echo of liver parenchyma is significantly coarse.Joint diagnostic criteria:It bases on elastography, when the elastographic score was 1, the capsul was smooth, and the echo of liver parenchyma was normal or slightly coarse, we regared the patient had no liver fibrosis, even if the elastographic score was 2, but the capsul was smooth, and the echo of liver parenchyma was normal, we still regared the patient had no liver fibrosis. However, when the elastographic score was= 2, and the echo of liver parenchyma was coarse or with cord-like changes, liver fibrosis was diagnosed.Part II:Hitachi Vision Preirus ultrasoud system with ultrasonic elastography was used in this study. The frequency of the linear array probe is 3.0-7.0MHz. Strain Histogram Measurement software was used as well. Five elastograms were taken on each patient for the quantitative analysis of tissue diffuse, the strain mean value and the area percentage of the blue color were recorded to evaluate liver fibrosis.3. Statistic analysisPart I:SPSS 13.0 statistical software was used for statistical analysis. The pathological result of liver biopsy was regarded as golden diagnostic standard. According to the pathological stage of liver fibrosis, So meaned without liver fibrosis and S1-4 meaned with liver fibrosis, all the 120 patients were classified in two groups, the group with liver fibrosis and the group without liver fibrosis. Spearman correlation analysis was used in analyzing the correlation between the liver fibrosis pathology stages and the elastographic score of the liver tissue; Levene's Test for equality of variances and One-Way ANOVA were used to analysis the different elastographic scores among the three groups, if heterogeneity of variance occured, the approximate analysis of variance was adopted. The receive operation characteristic curve(ROC) in which the vertical axis was sensitivity and the abscissa was 1-specificity was drawed to calculate the diagnostic sensitivity and specificity when each point was taken as the deviding point. After the deviding point of with or without liver fibrosis was selected, The sensitivity, specificity and accuracy of diagnosing liver fibrosis with UE, gray scale ultrasound and their combination were separately calculated; The McNemar chi-square test of paired count data were compared between the two groups.α= 0.05 was selected as the significance level.Part II:SPSS 13.0 statistical software was used for statistical analysis. The pathological result of liver biopsy was regarded as golden diagnostic standard. variance test and t test were used to compare and analyze the mean of quantitative data; Wilcoxon non-parametric test of two independent samples was used to compare the mean strain value and the percentage of blue area between the groups with and without liver fibrosis, and ROC curve was used to select the best diagnostic deviding point, the value of area under the curve, sensitivity, specificity and accuracy were calculated. Homogeneity of variance test and one-way analysis of variance (One-Way ANOVA) were adopted to compare the mean strain value and the percentage of blue area among different pathological stages. Analysis of non-parametric and Spearman correlation coefficient test were used to analyze the correlation between the mean strain value and the percentage of blue area among different pathological stages. a= 0.05 was selected as the significance level.ResultsPart I:The Spearman correlation coefficient between the pathological stages and ultrasonic elastographic scores is r= 0.875 (P= 0.000), they are significant positive correlated.The elastographic scores of liver tissue in the normal control group, the group with fibrosis and the group without fibrosis were 1.05±0.22,2.95±1.25,1.18±0.45, respectively. In the homogeneity of variance test, F=50.269, P=0.000, it showed Heterogeneity of variance, therefore, we use approximate variance analysis. Dunnett's T3 test was adopted in multiple comparisons if it was with significant difference, the result showed that there was no significant difference between the control group and the group without fibrosis (P=0.311), but there was significant difference between control group and the group with fibrosis(P=0.000), it was significantly different between the groups of with and without fibrosis(P=0.000).The area under curve (AUC) in ROC was 0.915, standard error of the area was 0.027 and 95% CI was (0.807～0.955), P=0.000, it suggested that ultrasound can be used to evaluate liver fibrosis, the higher of the elastographic score, the higher of the likelihood of suffering from liver fibrosis. The best deviding point of identifing liver fibrosis was 2,(?)2 meaned with liver fibrosis,<2, that was 1, meaned without liver fibrosis, The sensitivity, specificity, and accuracy of ultrasonic elastography to diagnose liver fibrosis were:91.3%,85.0%,89.2%, respectively. The sensitivity, specificity, and accuracy of ultrasound elastography combined gray-scale ultrasound were:93.8%,95.0% and 94.2%, respectively.The difference of accuracy between ultrasonic elastography and gray-scale ultrasound was significant (chi-square value x2=38.52, P<0.05).PartⅡ:The correlation coefficient of strain mean and pathological stage of liver fibrosis was r=-0.928, P=0.000, they were with a significant negative correlation; The correlation coefficient of the percentage of blue area and pathological stage of liver fibrosis were r=0.902, P=0.000, they were with a significant positive correlation.With Wilcoxon rank sum test, there was statistically significant difference on strain mean between the groups with or without liver fibrosis (Z=-8.065, P=0.000). The percentage of blue area between the two groups was significantly different (Z=-7.777, P=0.000).Multiple comparison LSD (Least-significant Difference) showed that the strain mean and percentage of the blue area among different pathological stage of liver fibrosis were significantly different (P=0.000).Conclusions1. Ultrasonic elastography can effectively monitor the change in tissue hardness of live fibrosis, which is a useful exploration of evaluation for liver fibrosis in a non-invasive method.2.The quantitative analysis of tissue diffusion by untrasonic elastography can timely and accurately evaluate the severity of liver fibrosis, therefore, it is of positive significance in evaluating liver fibrosis, predicting the prognosis and guiding clinical diagnosis and treatment for patients. |
PDF Full Text Request