Gd-EOB-DTPA-enhanced T1ρ imaging vs diffusion metrics for assessment liver inflammation and early stage fibrosis of nonalcoholic steatohepatitis in rabbits

To assess the value of T1 rho,T1 rho on hepatobiliary phase (HBP) and diffusion metrics in staging of Non-alcoholic fatty liver disease (NAFLD) activity scores, inflammation, fibrosis in NASH rabbits model. Non-alcoholic steatohepatitis (NASH) rabbits model was induced by feeding a varied duration of high-fat, high-cholesterol diet. T1 rho,T1 rho (HBP) 20 min after administration of Gd-EOB-DTPA, and Intravoxel incoherent motion imaging (IVIM) diffusion-weighted imaging were performed on a 3.0 T magnetic resonance (MR) imaging unit. The diagnostic value of each parameter for NAS, inflammation and fibrosis severity were determined. T1 rho (r = 0.658) and T1 rho (HBP) (r = 0.750) have strong association with NASH overall activity, T1 rho (HBP) is strongly relevant to inflammation stage (r = 0.812). There was negative association between f and inflammation (r = -0.480), whilst no significant relation between other three parameters (apparent diffusion coefficient (ADC), pseudo-diffusion coefficient (D*) and true diffusion coefficient (D)) and inflammation or overall activity. The areas under the receiver operating characteristic curves (AUCs) of f, ADC, T1 rho and T1 rho-HBP were 0.871, 0.728, 0.849 and 0.949 for differentiating NASH; 0.731, 0.552, 0.925 and 0.922 for G2-3 inflammation; and 0.767, 0.625, 0.816, and 0.882 for S1-2 fibrosis. Comparison of ROC curve showed T1 rho (HBP) had an optimal diagnostic performance for NASH [T1 rho (HBP) vs ADC, AUC:0.949 vs 0.728, P = 0.043], inflammation [T1 rho (HBP) vs ADC, AUC:0.922 vs 0.552, P = 0.003], fibrosis [T1 rho (HBP) vs ADC, AUC:0.882 vs 0.625, P = 0.046]. The combination of T1 rho (HBP) + perfusion fraction (f) showed highest diagnostic value for NASH (AUC:0.971), inflammation (AUC:0.935). Among T1 rho imaging and IVIM diffusion metrics, combination of T1rho (HBP) + f was found to be superior noninvasive imaging biomarker for NASH activity assessment.