Feasibility Of Contrast-enhanced Ultrasonography In The Assessment Of Liver Fibrosis And Selection Of Quantitative Parameters

Chronic hepatitis B can lead to liver fibrosis or even cirrhosis, but liver damage can be reversed at the stage of fibrosis with early diagnosis and effective intervention. Therefore, appropriate staging of liver fibrosis is essential for improving the prognosis of patients with chronic hepatitis. At present, liver biopsy is the gold standard for diagnosing and staging fibrosis. However, it is an invasive procedure associated with many potential complications and it is not suitable for routine clinical follow-up. In addition, sampling errors occur because of inadequate liver specimens and intra-and interpathologist variability. Therefore, noninvasive, safe, and reliable methods to assess the severity of liver fibrosis are urgently needed.With the development of liver fibrosis, both intrahepatic and extrahepatic hemodynamics are changed. Development of the contrast-enhanced ultrasonography (CEUS) technique provides a new way to assess hemodynamic changes of the liver. The purpose of this study was to investigate the use of CEUS with quantitative measurement of intrahepatic circulation times to assess liver fibrosis in patients with chronic hepatitis and to correlate CEUS parameters and findings to variable stages of liver fibrosis.Part one:Quantitative assessment of liver fibrosis by contrast-enhanced ultrasonography with different regions of interestObjective To investigate the difference of quantitative parameters on CEUS with various shapes and positions of regions of interest (ROI) in the patients with hepatic fibrosis.Methods Thirty cases of hepatic fibrosis diagnosed pathologically were enrolled in the study. The dynamic images of CEUS stored in hard disk were analyzed off-line with the quantitative software. Three kinds of ROIs with elliptic, rectangular and irregular shapes were drawn separately in the same four positions:peripheral liver parenchyma, central liver parenchyma, main portal vein and right hepatic vein. The time intensity curve (TIC) and some quantitative parameters including increased signal intensity (ISI), decent curvature (a2), up slope rate (a3), arrival time (AT), time to peak (TTP), peak intensity (PI), area under curve (AUC) and mean transit time (MTT) were achieved. The quantitative parameters were compared among different ROIs and between peripheral and central liver parenchyma.Results There were no statistical differences among three shapes of ROIs in all quantitative parameters of ISI, a2, a3, AT, TTP, PI, AUC and MTT, respectively (P>0.05).The parameters of ISI, a2, AT, PI and AUC were significantly different between peripheral and central liver parenchyma (P<0.05).Conclusion The quantitative parameters of CEUS with different shapes of ROIs in hepatic fibrosis are compatible with high quality. The microcirculation perfusion is different between peripheral and central liver parenchyma.Part two:Screening of quantitative parameters on contrast-enhanced ultrasonography in assessing liver fibrosisObjective To investigate the correlations between quantitative parameters of CEUS and histopathological stage of liver fibrosis and select parameters with better correlation.Methods Seventy-five cases of hepatic fibrosis diagnosed pathologically were examined with CEUS. The dynamic images of CEUS were analyzed off-line with quantitative software. ROI were drawn separately in HA, PA, PV and HV. The TIC and some quantitative parameters, including AT, TTP, MTT, PI, ISI, AUC, decent curvature and up slope rate, were acquired. Intrahepatic transit time of HA-HVTT, PV-HVTT and PA-HVTT were calculated. All the parameters were compared with histopathologic stage.Results All cases were grouped into mild fibrosis (S0 and S1,30 cases), moderate fibrosis (S2 and S3,25 cases) and severe fibrosis (S4,20 cases) according pathologic stage. HV-AT shorted with the fibrosis progression, and changed significantly between mild fibrosis and severe fibrosis groups (P<0.05). HA-HVTT, PV-HVTT and PA-HVTT decreased with the development of liver fibrosis, and changes were significantly correlated with liver fibrosis severity (r=-0.6012, P<0.0001; r=-0.6570, P<0.0001; r=-0.5952, P<0.0001). HA-HVTT and PV-HVTT were significant differences between any paired groups (P<0.05). Receive operating characteristic curve analysis showed that the area under curve of HA-HVTT, PV-HVTT and PA-HVTT were 0.810,0.852 and 0.828 (S≥S2) and 0.858,0.897 and 0.814 (S=S4), respectively. The parameters of PI and ISI obtained from TIC of portal vein reduced with the progression of fibrosis, too (r=-0.3857, P=0.0029; r=-0.3886, P=0.0032). Other parameters showed no correlation to the severity of liver fibrosis (P>0.05). Conclusions Quantitative parameters of CEUS reflect the stage of liver fibrosis mediately. HV-AT, HA-HVTT, PV-HVTT, PA-HVTT, PI and ISI of PV are correlated with the stage of liver fibrosis.Part three:Non-invasive assessment of liver fibrosis with the quantitative parameters of contrast-enhanced ultrasonography and FibroIndexObjective To compare the quantitative parameters of CEUS and FibroIndex in evaluating the stage of liver fibrosis noninvasively.Methods One hundred and twenty-two patients with chronic hepatitis B (77 patients with hepatic fibrosis diagnosed pathologically and 45 patients with clinical evidence of cirrhosis) grouped into mild fibrosis group (S0 and S1,36 cases), moderate fibrosis group (S2 and S3,24 cases) and cirrhosis group (S4 and clinically evident chirrosis,62 cases). All patients underwent CEUS and serologic examinations. The real-time CEUS dynamic images were off-line analysis and the parameters of HA-AT, HA-HVTT, PV-HVTT, PA-HVTT, PI, ISI of PV, and FibroIndex were calculated and compared with histopathologic stage of liver fibrosis.Results HV-AT, HA-HVTT, PV-HVTT and PA-HVTT were shortened, and were correlated with liver fibrosis severity (r=-0.3148, P=0.007; r=-0.5930, P<0.001; r=-0.8215, P<0.001; r=-0.4171, P<0.001); PI and ISI of portal vein were decreased, and had certain relationship with liver fibrosis (r=-0.5170, P<0.001; r=-0.5417, P<0.001); FibroIndex increased with the progression of fibrosis(r=0.5915, P<0.001). Area under ROC curves for HA-HVTT and PV-HVTT and FibroIndex were 0.891, 0.955,0.859 at fibrosis scores≥S2, and were 0.785,0.946,0.803 at fibrosis score≥S4, respectively. The area under the ROC curve of PV-HVTT was visibly higher than either HA-HVTT or FibroIndex (P<0.05).Conclusion Intrahepatic transit-times (HA-HVTT, PV-HVTT) and Fibrolndex reflect the stage of liver fibrosis objectively, which are potiential in evaluating the stage of liver fibrosis noninvasively. Among them, PV-HVTT has the highest diagnostic accuracy.