Liver Fibrosis Scores (FIB-4) Outperform Metabolic Indices in Predicting All-Cause Mortality and CVD Mortality Among Patients with MASLD: A Nationally Representative Cohort Study Based on NHANES 2011-2018

Abstract Background In metabolic dysfunction–associated steatotic liver disease (MASLD), both fibrosis burden and cardiometabolic dysregulation may contribute to adverse outcomes. However, the comparative prognostic value of a fibrosis-based index versus lipid- and insulin-resistance–related indices for near-term mortality risk stratification remains unclear. Methods We analysed adults with MASLD from the National Health and Nutrition Examination Survey (NHANES) 2011–2018. MASLD was defined by hepatic steatosis (Fatty Liver Index ≥ 60) plus at least one cardiometabolic risk factor. Mortality outcomes (all-cause mortality [ACM] and cardiovascular disease mortality [CVDM]) were ascertained via linkage to the National Death Index through 31 December 2019. Baseline fibrosis-4 index (FIB-4), atherogenic index of plasma (AIP), and triglyceride–glucose (TyG) index were evaluated as continuous (per 1-SD) and categorical measures (FIB-4 risk categories; tertiles for AIP and TyG). Associations with ACM and CVDM were assessed using survey-weighted multivariable Cox proportional hazards models. Restricted cubic splines examined potential non-linear relationships. Predictive discrimination at 5 years was evaluated using time-dependent ROC analyses (with competing-risk methods for CVDM). Results The analytic sample included 4,037 participants with MASLD (median follow-up ≈ 4.5 years), during which 222 ACM events and 66 CVDM events occurred. In fully adjusted, age-stratified Cox models, higher FIB-4 was consistently associated with increased ACM risk (per 1-SD: HR 1.24, 95% CI 1.09–1.41 for < 65 years; HR 1.34, 95% CI 1.20–1.50 for ≥ 65 years). Participants with high FIB-4 had markedly elevated ACM risk compared with low-risk counterparts (HR = 9.46, 95% CI = 2.58–34.60 for < 65 years; HR = 3.35, 95% CI = 2.14–5.26 for ≥ 65 years). For CVDM, FIB-4 was strongly associated with risk among participants aged ≥ 65 years (per 1-SD: HR = 1.37, 95% CI = 1.19–1.59; high-risk vs low-risk: HR = 5.59, 95% CI = 2.29–13.68), whereas AIP and TyG were not significant in this age stratum. FIB-4 demonstrated superior discrimination for 5-year mortality (ACM-AUC 0.763; CVDM-AUC 0.785) compared with AIP (ACM AUC 0.494; CVDM AUC 0.505) and TyG (ACM-AUC 0.522; CVDM-AUC 0.557). Conclusions In a contemporary, nationally representative US MASLD cohort with relatively short-term follow-up, baseline FIB-4 substantially outperformed AIP and TyG in stratifying and discriminating both all-cause and cardiovascular mortality risk, supporting fibrosis-oriented prognostic assessment for near-term risk evaluation in MASLD.