Prognostic value of the advanced lung cancer inflammation index for 28 day mortality in sepsis associated acute kidney injury

Sepsis-associated acute kidney injury (SA-AKI) is highly heterogeneous in critically ill patients. Reliable tools that integrate multiple indicators to predict outcomes remain limited. The advanced lung cancer inflammation index (ALI), a composite marker reflecting both inflammation and nutritional status, has demonstrated prognostic value in oncology and chronic diseases. However, evidence for its predictive performance and clinical utility in SA-AKI is scarce and requires further validation. This study analyzed 5,565 SA-AKI patients from the MIMIC-IV 3.1 database. Patients were divided into quartiles based on the ALI. The primary outcomes were 28-day and 90-day all-cause mortality. Multivariate Cox proportional hazards models, Kaplan–Meier survival analysis, and restricted cubic spline models were used to explore associations between the ALI and patient outcomes. Subgroup and sensitivity analyses were performed to test consistency and robustness. ROC analysis was used to compare the discriminatory performance of ALI with other indices and assess its incremental value in existing models. Patients in the highest ALI quartile had significantly lower risks of 28-day (HR = 0.72, p < 0.0001) and 90-day (HR = 0.74, p < 0.0001) mortality than did those in the lowest quartile. Survival curves indicated a strong positive association between the ALI and cumulative survival (log-rank p < 0.0001), and RCS analysis suggested a nonlinear trend. Subgroup analyses revealed no substantial heterogeneity, and sensitivity analyses supported the stability of these associations. ROC analysis demonstrated ALI had superior discrimination for 28-day mortality and provided incremental prognostic value when integrated into the SOFA score. ALI is independently associated with 28-day and 90-day mortality in SA-AKI. It offers incremental prognostic value beyond established scores and may serve as a supplementary risk stratification tool, but validation in prospective multicenter cohorts is warranted.