INTRODUCTION: Recently, diffusion-weighted (DW) imaging–based elastography was proposed for noninvasive liver staging by converting a shifted apparent diffusion coefficient to tissue elasticity based on two b values (200 and 1500sec/mm²) [1-2]. The purpose of this study was to compare the diagnostic performance of DWI-based virtual MRE and US shear-wave elastography (SWE) for staging liver fibrosis, with histopathological findings as the reference standard.
METHODS: This retrospective study included 124 patients (90 men and 34 women; mean age, 55.3±11.2 years; age range, 26–84 years) who underwent multiple b-values DW MRI with a 1.5T MR scanner (MAGNETOM Aera, Siemens Healthcare, Erlangen, Germany) and SWE. Shifted apparent diffusion coefficient was calculated from DW MRI (b=200 and 1500 sec/mm²) and converted to DW MRI–based virtual shear modulus (µ_Diff) by using an in-house developed program based on MATLAB (Mathworks, Natick, Mass) according to the method proposed in the previous studies [1-2]. µ_Diff and liver stiffness (LS) of SWE were measured on the right lobe of the liver. Liver fibrosis stage was histologically determined according to Scheuer scoring system: S0 (n=6), S1 (n=10), S2 (n=18), S3 (n=21) and S4 (n=69). The µ_Diff and SWE were compared with fibrosis stage by Spearman's rank correlation coefficient. Receiver operating characteristic (ROC) curves for µ_Diff and LS were built and compared for diagnostic performance in staging liver fibrosis.
RESULTS: The representative image of DW MRI-based virtual shear modulus is shown in Fig.1. The µ_Diff and LS showed significant correlations with liver fibrosis stage (rho = 0.46, P < 0.001; rho = 0.76, P < 0.001). The ROC curves of µ_Diff and LS for the diagnosis of significant fibrosis and advanced fibrosis are shown in the Figs 2, and the areas under the curves of µ_Diff and LS were 0.79 (95% CI: 0.68-0.90) and 0.88 (95% CI: 0.82-0.94) for significant fibrosis, and 0.87 (95%CI: 0.80-0.93) and 0.96 (95%CI: 0.93-0.99) for advanced fibrosis, respectively. The area under the curve of LS was significantly higher than that of µ_Diff for advanced fibrosis (P < 0.05), but there was no significant difference between µ_Diff and LS for significant fibrosis (P > 0.05).
DISCUSSION: Diffusion-weighted (DW) imaging explores the increased hindrance of diffusion in water protons caused by the increased connective tissue in fibrotic liver parenchyma. [1-2]. Previous studies have demonstrated significant correlation between µ_Diff and MR elastography [1-2]. SWE enables assessment of liver fibrosis by measuring liver stiffness.
CONCLUSION: SWE is superior to DW imaging–based elastography for predicting liver fibrosis, especially for advanced fibrosis.

Acknowledgements

This work was supported by the Youth Project of National Natural Science Foundation of China (Grant No. 82202112), and Youth Project of Natural Science Foundation of Fujian Province (Grant No. 2023J05293)