Comparative Analysis of Functional and Morphological MRI Biomarkers in Assessing Cirrhosis Severity

Purpose The severity of liver cirrhosis is primarily assessed through biochemical tests. However, relying only on laboratory parameters limits the ability to evaluate hepatocellular function and structural damage. Imaging biomarkers obtained from liver MRI offer noninvasive information that could improve severity assessment. This study aims to explore the potential of functional and morphological imaging biomarkers in determining the severity of liver disease. Materials and Methods This retrospective single-center study involved patients with cirrhosis who underwent gadoxetic acid–enhanced MRI (GA-MRI). The Functional Liver Imaging Score (FLIS), which reflects hepatic function, was assessed on hepatobiliary-phase images. Liver Surface Nodularity (LSN), a measure of structural distortion, was evaluated on MRI using a commercial semi-automated tool. Model for End-stage Liver Disease (MELD-Na) scores were obtained from medical records and served as the clinical reference standard for cirrhosis severity in this study. The associations of FLIS and LSN with MELD-Na were analyzed through correlation analysis, regression modeling, and ROC curve analysis. Results A total of 165 patients (mean age, 59.2 ± 12.5 years; 112 women) were included in the study. FLIS demonstrated a moderate correlation with MELD-Na (ρ = −0.54, p < 0.001), whereas LSN showed a poor correlation (ρ = +0.23, p = 0.004). On ROC analysis, FLIS achieved an AUC of 0.78 (95% CI, 0.70–0.87) for detecting patients with MELD-Na ≥ 15, while LSN showed weaker performance with an AUC of 0.62 (95% CI, 0.52–0.71). The combined FLIS + LSN score showed a marginal improvement in discriminative ability (AUC 0.80 vs 0.78), driven primarily by increased sensitivity (81% vs 70%). Inter-reader agreement (0.86) and intra-reader reliability (ICC = 0.88) were good for FLIS scoring. Conclusion This study indicates that functional and morphological imaging biomarkers provide complementary prognostic information to biochemical scoring, particularly in identifying functional impairment not captured by laboratory tests. Prospective studies involving long-term clinical outcomes are needed to determine their role in risk stratification and transplant decision-making.