The Relationship Between Conventionally Obtained Serum-Based Liver Function Indices and Intravoxel Incoherent Motion Diffusion-Weighted Imaging and Magnetic Resonance Elastography in Patients With Hepatocellular Carcinoma

Abstract

Objectives To investigate the relationship between conventionally obtained serum-based biochemical indices and intravoxel incoherent motion imaging (IVIM) parameters compared with magnetic resonance elastography (MRE). Methods Patients with hepatocellular carcinoma who underwent ≥2 liver magnetic resonance imaging (MRI) scan, including IVIM and MRE, between 2017 and 2020 and biochemical testing within 1 week before or after MRI were included in this study. Biochemical tests were performed to determine the albumin-bilirubin (ALBI) score and modified ALBI (mALBI) grade, aspartate aminotransferase to platelet ratio index (APRI), and fibrosis-4 index (FIB-4). The diffusion coefficient (D), pseudo-diffusion coefficient (D*), fractional volume occupied by flowing spins (f), and apparent diffusion coefficient were calculated for IVIM. The correlations between (1) the imaging parameters and biochemical indices and (2) the changes in mALBI grades and imaging parameters were evaluated. Results This study included 98 scans of 40 patients (31 men; mean age, 67.7 years). The correlation analysis between the biochemical and IVIM parameters showed that ALBI score and D* had the best correlation (r = −0.3731, P < 0.001), and the correlation was higher than that with MRE (r = 0.3289, P < 0.001). However, among FIB-4, APRI, and MRI parameters, MRE outperformed IVIM parameters (MRE and FIB-4, r = 0.3775, P < 0.001; MRE and APRI, r = 0.4687, P < 0.001). There were significant differences in the changes in MRE among the 3 groups (improved, deteriorated, and unchanged mALBI groups) in the analysis of covariance (P = 0.0434). There were no significant changes in IVIM. Conclusions Intravoxel incoherent motion imaging has the potential to develop into a more readily obtainable method of liver function assessment.