Prognostic Value of Albumin-Bilirubin Score in Acute Pancreatitis: Insights From the MIMIC-IV Database

BACKGROUND This study examined the association of the albumin-bilirubin (ALBI) score with short-term and long-term prognoses in individuals with acute pancreatitis (AP), using the MIMIC-IV database. MATERIAL AND METHODS A retrospective cohort study was conducted using MIMIC-IV (V3.0). ALBI score was calculated based on albumin and bilirubin serum levels. Patients were divided into 3 groups according to threshold values of ALBI at -2.60 and -1.39. Primary and secondary outcomes were 30-day and 1-year all-cause mortality, respectively. A multivariable Cox proportional hazards model assessed the association between ALBI and mortality risk. Restricted cubic spline analysis was used to explore their nonlinear relationship. Kaplan-Meier curves compared survival, and subgroup analyses evaluated interaction effects. RESULTS A total of 1953 AP patients were included. Results indicated each unit increase in ALBI score was associated with a higher risk of mortality (30-day: HR, 1.37; 95% CI, 1.16-1.61; P<0.001; 1-year: HR, 1.17; 95% CI, 1.03-1.33; P=0.018). Kaplan-Meier curves showed significant differences in survival (log-rank P<0.001). Restricted cubic spline analysis identified a nonlinear association between ALBI score and risk of mortality, with a turning point at -2.3, beyond which the risk of mortality increased sharply. Subgroup analysis indicated stronger associations in patients not receiving carbapenems or fluoroquinolones or requiring mechanical ventilation within the first 24 hours of ICU admission, suggesting ALBI may reflect risk of mortality in these patients, with no significant association observed in those receiving these treatments. CONCLUSIONS ALBI score correlates with AP prognosis, supporting its use in early risk stratification and personalized management.