Revisiting Mac-2-Binding Protein Glycosylation Isomer (M2BPGi) for Diagnosing High-Risk Liver Fibrosis: A Stepwise Diagnostic Analysis

Abstract

Abstract* Background The level of liver fibrosis is the basis for the treatment of chronic hepatitis B (CHB), and it is necessary to adapt non-invasive liver fibrosis modalities. We aimed to investigate the use of M2BPGi as a single or combined diagnostic modality for liver fibrosis in CHB patients through a stepwise diagnostic analysis. Methods Cross-sectional data were taken from patients between October 2021 and August 2022. Demographic data, blood profile, liver function, and liver stiffness were measured in CHB patients over 18 years old, willing to take part in the research, and had complete data. APRI, FIB-4, and AAR were calculated using the well-known formulas. Serum M2BPGi-levels were converted into a cut-off index (COI). The patients were divided into low-risk (LR) and high-risk fibrosis (HR) groups. A cut-off for each predictor variable to differentiate between the LR and HR groups was determined. The obtained cut-off was assessed for its association with the grouping of liver elastography results. Models to diagnose the liver stiffness measurement (LSM) ≥8 kPa were created and compared through multivariate and ROC analyses. Results The number of patients that met the inclusion and exclusion criteria was 143 (HR = 65, LR = 78). The cut-off for diagnosing LSM ≥8kPa was 0.311, 0.742, 0.635, and 1.434 for APRI, FIB-4, AAR, and M2BPGi, respectively. This cut-off was significantly associated with the results of the HR and LR groupings. A multivariate analysis found that FIB4, AAR, and M2BPGi added significantly to the model. Statistically, the most optimal use of M2BPGi was combined with FIB-4, with an AUC of 0.835. Conclusions The optimal cut-off of M2BPGi for diagnosing high-risk liver fibrosis in this study was 1.434. M2BPGi should be used with FIB-4 as a diagnostic tool for diagnosing liver fibrosis, especially in the absence of a liver biopsy or elastography.