Prognostic Value of the Advanced Lung Cancer Inflammation Index for Predicting in-Hospital Mortality in Intensive Care Unit Patients with Septic Shock: A Retrospective Cohort Study.

Background: The Advanced Lung Cancer Inflammation Index (ALI), initially developed for advanced non-small cell lung cancer, has shown prognostic value across cardiovascular and oncologic conditions. Although ALI was originally developed for lung cancer prognosis, this study evaluates its applicability in general septic shock patients without active malignancy. However, its role in predicting outcomes in septic shock remains unclear. Conventional scoring systems like APACHE II and SOFA are limited by complexity and insufficient assessment of inflammatory and nutritional status. Methods: We conducted a retrospective cohort study of septic shock patients admitted to a tertiary medical surgical intensive care unit (ICU) between January 2019 and December 2023. ALI was calculated as body mass index × albumin / neutrophil-to-lymphocyte ratio. Patients from 2019-2021 formed the training set, while those from 2022-2023 served as the validation set. Bootstrap validation (1000 iterations) assessed model stability. The primary outcome was in-hospital mortality. Multivariable logistic regression, ROC curves, calibration analysis, and decision curve analysis (DCA) evaluated ALI's prognostic performance compared with APACHE II and SOFA scores. Results: A total of 1247 patients were included, with a mean age of 67.3±14.2 years; 58.2% were male, and the overall in-hospital mortality rate was 34.7%. In-hospital mortality was highest in the Low ALI group (<18.5; 48.8%) and decreased progressively across the Intermediate ALI (18.5-39.2; 32.3%) and High ALI (>39.2; 23.1%) groups. Low ALI (<18.5) was independently associated with increased in-hospital mortality compared with ALI ≥18.5, comprising the combined Intermediate and High ALI groups as the reference category (adjusted OR=2.34, 95% CI=1.67-3.28, P<0.001). In the validation cohort, ALI showed acceptable discrimination (AUC=0.70) and calibration (Hosmer-Lemeshow P=0.35). ALI demonstrated discrimination comparable to APACHE II (AUC=0.69) and SOFA (AUC=0.68), with no statistically significant differences according to the DeLong test. Decision curve analysis indicated comparable clinical utility between ALI, APACHE II, and SOFA across the evaluated threshold probability ranges. Conclusion: ALI is a simple, independent predictor of in-hospital mortality in septic shock. In this study, ALI demonstrated comparable discrimination to traditional scoring systems such as APACHE II and SOFA, while offering the advantage of a simpler calculation based on routinely available clinical parameters. Prospective multicenter studies are needed to confirm these findings and evaluate dynamic ALI monitoring and ALI-guided treatment strategies.