Assessment Of Liver Fibrosis Using Variable Flip Angle T1 Mapping And Pharmacokinetic Parameters Of Dynamic Contrast-enhanced Magnetic Resonance Imaging

Part I Assessment of liver fibrosis by magnetic resonance imaging with variable flip angle Tl mappingPurpose:To evaluate the possibility of using a variable flip angle （VFA） T1 mapping technique to diagnose liver fibrosis.Materials and Methods:Liver fibrosis was induced in rabbits by repetitive administration of carbon tetrachloride （CCl4）. T1-weighted magnetic resonance imaging （MRI） was performed in 29 animals （liver fibrosis, n=18; control, n=11） using a series of nonenhanced liver acquisition volume acceleration （LAVA） with VFAs at 3.0 Tesla. Hepatic T1 relaxation times were measured via regions of interest, which were correlated with subsequent histologic confirmation. The results of T1 mapping in assessment of liver fibrosis were compared with those of apparent diffusion coefficient （ADC） values.Results:The mean T1 relaxation time of the control group was the lowest （250.07±88.12 ms）, followed by the non-advanced fibrosis group （387.83±166.58 ms） and the advanced fibrosis group （496.90±291.24 ms）. T1 relaxation time measurements differed significantly between the liver fibrosis group and control group （p< 0.05）, with a trend of increased mean T1 relaxation times as the fibrotic stage increased. Such significant differences were also observed between early-staged fibrotic livers and normal livers. In discriminating between the control group and liver fibrosis group, stage F0-1 （control and stage F1） and stage F2-3, stage F0-2 （control and stage F1-2） and stage F3, area under the receiver operating characteristic （ROC） curves were 0.803 （cutoff value 273.01 ms）,0.712 （cutoff value 371.54 ms）, and 0.696 （cutoff value 276.99 ms）, respectively. No difference was found between T1 relaxation times and ADC values in assessment of liver fibrosis.Conclusion:VFA T1 mapping may become a noninvasive imaging tool for the diagnosis of liver fibrosis.Part Ⅱ Assessment of liver fibrosis using pharmacokinetic parameters of dynamic contrast-enhanced magnetic resonance imagingPurpose:To evaluate the pharmacokinetic parameters of dynamic contrast-enhanced magnetic resonance imaging （DCE-MRI） in diagnosing and staging liver fibrosis in rabbits.Materials and methods:DCE-MRI with gadodiamide （Gd-DTPA-BMA） was performed on a 3.0 Tesla,60 cm bore MR scanner for rabbits with CCl4-induced liver fibrosis, and an untreated control group. Fibrosis was staged according to the METAVIR system:control （F0; n= 13）, nonadvanced fibrosis （F1-2; n=15）, and advanced fibrosis （F3-4; n=12）. The DCE-MRI parameters K^trans, kep, Ve, and vp were measured with a dual-input extended Tofts model. Receiver operating characteristic analyses were performed to assess the diagnostic performance of K^trans, Ve and vp in staging liver fibrosis.Results:Both K^trans and Ve decreased with increasing fibrosis stage. K^trans of the control group was significantly different from those of the overall fibrosis group, non-advanced group, and advanced group （p< 0.001 for all）. Significant differences were found between Ve of the control group and those of the overall fibrosis and advanced groups （p=0.019 and p=0.009, respectively）. For ktrans, the areas under the receiver operating characteristic curve （AUROCs） for discriminating the control group from the overall fibrosis and advanced fibrosis groups were 0.909 （95% confidence interval[CI], 0.809-1.000）, and 0.936 （95% CI,0.847-1.000）, respectively. For discriminating between the control and non-advanced fibrosis groups, the AUROC of K^trans was 0.887 （95% CI,0.762-1.000）. The AUROCs of the K^trans were higher than those of Ve and vp for discrimination of the control versus overall fibrosis groups, the control versus non-advanced fibrosis groups, and the control versus advanced fibrosis groups. Pharmacokinetic parameters were negatively correlated with fibrosis stage (K^trans, rho=-0.668, p<0.001; Ve, rho=-0.438, p= 0.005; vp, rho=-0.360, p= 0.023).Conclusion:Among pharmacokinetic parameters of DCE-MRI in our study, K^trans was an excellent predictor for differentiating fibrotic livers from normal livers, and differentiating normal livers from non-advanced or advanced fibrosis livers.