Prognostic value of ALBI score for all-cause mortality in metabolic associated fatty liver disease patients: a cohort study from NHANES 2003-2018.

BACKGROUND &

aimsMetabolic Associated Fatty Liver Disease (MAFLD) is a prevalent chronic liver disorder with severe potential outcomes. While the albumin-bilirubin (ALBI) score demonstrates prognostic utility in other chronic liver diseases, its specific role and predictive performance in MAFLD patients, particularly regarding all-cause mortality, remain incompletely understood. This study aims to investigate the association between ALBI scores and all-cause mortality in individuals with MAFLD and to evaluate its prognostic potential using large-scale NHANES data.

methodsDrawing on data from the population-based National Health and Nutrition Examination Survey (NHANES) conducted between 2003 and 2018, we employed weighted multivariable Cox proportional hazards regression to assess the relationship between ALBI scores and all-cause mortality in a cohort of 5,666 MAFLD patients. ALBI scores were calculated and categorized into tertile (Q1: ALBI < -2.96; -2.96 ≤ ALBI < -2.70; Q3: ≥ -2.70). MAFLD was diagnosed using the U.S. Fatty Liver Index (US-FLI) with a cutoff score of ≥ 30, while excluding other chronic liver diseases. Statistical analyses incorporated NHANES sampling weights and adjusted for potential confounders using multivariable Cox regression models. Additionally, we conducted threshold effect analysis to identify potential inflection points in the ALBI-mortality relationship and used Kaplan-Meier survival analysis to visualize survival differences across ALBI tertiles.

resultsIn this cohort study of 5,666 MAFLD patients, 1,093 (19.29%) experienced mortality during a median follow-up of 8.5 years. Following adjustment for confounding factors, elevated ALBI scores demonstrated a significant correlation with an increased risk of death from any cause (p < 0.001). The hazard ratios (HR) for mortality across ALBI tertile (Q1-Q3) were 1.00 (reference), 1.32 (1.05–1.65), and 1.74 (1.42–2.12), respectively. Each 1-unit increase in ALBI score was associated with a 193% higher risk of death (HR: 2.93, 95% CI: 2.02–4.24). Threshold effect analysis identified an inflection point at ALBI = -2.69, using piecewise Cox regression, mortality risk increased sharply above this threshold (HR = 4.86, 95% CI: 3.32–7.11, p < 0.0001). ROC curve analysis showed AUC values of 0.715, 0.646, and 0.652 for 1-, 2-, and 3-year mortality, respectively, with calibration curves indicating strong agreement between predicted and actual probabilities.

conclusionOur study demonstrates that ALBI scores are a moderate predictor of all-cause mortality in MAFLD patients, particularly at ALBI ≥ -2.69, with an AUC of 0.715 for 1-year mortality. These findings highlight the potential of ALBI scores to identify high-risk patients early, supporting their use in clinical prognostic assessments. Future research should validate these results in diverse populations.