Steatosis-Dependent Concordance in Fibrosis Assessment Between Multiparametric Magnetic Resonance Imaging and Transient Elastography in Asian Chronic Hepatitis B Patients

Abstract

Background: Multiparametric magnetic resonance imaging (MRI) is an emerging non-invasive instrument for the assessment of liver fibrosis and steatosis. Its utilization in patients with chronic hepatitis B (CHB) has not been well investigated . Methods: We prospectively recruited consecutive clinically-stable chronic hepatitis B (CHB) patients with advanced fibrosis, defined via transient elastography (TE) (Fibroscan, Echosens) according to EASL recommendations (liver stiffness (LS)>9kPa among normal alanine aminotransferase). Steatosis via TE was quantified via controlled attenuation parameter (CAP) measurements . Multiparametric MRI (LiverMultiScanTM [LMS], Perspectum Diagnostics) examination was performed, in which liver fibrosis and steatosis were quantified by corrected T1 (cT1) and proton density fat fraction (PDFF) respectively . The cut-off for advanced liver fibrosis via LMS was cT1>794ms. Anthropometry, liver biochemistry, metabolic and virologic parameters were collected to investigate their potential effects on the concordance between LMS and TE . Results: 131 CHB patients (mean age 59 .5±8 .6 years, 78 .6% male) with advanced liver fibrosis (median LS 13.8 kPa, IQR 11.0 – 19.4) were recruited . Mean body mass index was 25 .6±4 .1 kg/m2 . 117 (89 .3%) were on nucleoside analogue treatment for a median duration of 8.3 (IQR 4.4-10.1) years. The median cT1 was 767 (IQR 695 – 853) ms. 57 (43.5%) had a concordant fibrosis assessment, with both TE and LMS demonstrating advanced fibrosis; correlation between cT1 and LS was mild (r=0.257, p=0.003). Patients with severe steatosis, defined as CAP ≥280 dBm-1, had increased concordance when compared to patients with nil to moderate steatosis (60 .0% vs . 29 .6%, p=0 .001) . Patients with concordant results had a mean CAP significantly higher than patients with discordant results (298 .3±67 .9 vs . 256 .7±50 .3, p<0 .001) . Multivariate analysis found a high CAP to be the only factor independently associated with a concordant result (p=0 .008, OR 1 .014, 95%CI 1 .004-1 .024) . For the measurement of liver steatosis, PDFF correlated well with CAP (r=0 .662, p<0 .001); the AUROC of CAP in predicting abnormal fat via PDFF was 0 .815 . Conclusion: The concordance of LMS with TE in liver fibrosis assessment is steatotic-dependent. LMS may complement TE in the non-invasive assessment of liver fibrosis in steatotic CHB. [https://aasldpubs.onlinelibrary.wiley.com/doi/10.1002/hep.30941]