Assessing the albumin–bilirubin score’s role in predicting the mortality rate of severe acute pancreatitis: A multicenter retrospective cohort study

The albumin-bilirubin (ALBI) score serves as an effective alternative for assessing patients with hepatocellular carcinoma, sepsis, and acute kidney injury. Its relevance to clinical outcomes in acute pancreatitis (AP) is not well defined. A retrospective multicenter analysis of critically ill patients with AP was conducted using the medical information mart for intensive care (MIMIC) IV (v3.1) and eICU collaborative research database (eICU-CRD) (v2.0) databases. The association between ALBI score and mortality was analyzed using Kaplan-Meier survival curves, smooth curve fitting, and both univariate and multivariate Cox regression analyses. A subgroup analysis was conducted to evaluate the robustness of the results. This study analyzed 529 AP patients from the MIMIC-IV database and 410 from the eICU-CRD database. ALBI scores were categorized into 2 groups: <-1.546 and ≥-1.546. The higher-ALBI-scores group was associated with higher in-hospital (adjusted hazard ratios [HRs]: 2.34, 95 % confidence intervals [CIs]: 1.45-3.79, P < .001) and intensive care unit mortality (adjusted, MIMIC-IV HR: 2.17, 95 % CI: 1.23-3.84, P = .008 vs eICU-CRD HR: 2.22, 95% CI: 1.08-4.56, P = .029) compared to the lower-ALBI-scores group. Curve-fitted analysis revealed a linear relationship between ALBI scores and mortality was observed (P for nonlinearity > .05). The ALBI score did not interact with most subgroups. In critically ill patients with AP, an elevated ALBI score is associated with worse clinical outcomes.