JMIR Publications

ABSTRACT

Background:

Sepsis-associated acute kidney injury (SA-AKI) is a frequent and life-threatening complication of sepsis. However, reliable prognostic biomarkers for early risk stratification remain limited.

Objective:

To identify distinct dynamic trajectories of lactate-to-albumin ratio (LAR) and evaluate their association with mortality in patients with SA-AKI.

Methods:

Data were extracted from the Medical Information Mart for Intensive Care IV (MIMIC-IV) database, including adult SA-AKI patients with at least three lactate and albumin measurements within the first 72 h of intensive care unit (ICU) admission. A multicenter external validation cohort was assembled from five tertiary hospitals in Beijing, China. Group-based trajectory modeling (GBTM) identified distinct LAR trajectories. The primary outcome was the 28-day mortality. Secondary outcomes included 90-day mortality and continuous renal replacement therapy (CRRT) initiation. Associations were assessed using Kaplan-Meier curves, multivariable Cox models, restricted cubic splines, and subgroup analyses. Sensitivity analyses using inverse probability of treatment weighting, weight truncation, and doubly robust estimation evaluated robustness.

Results:

Among the 615 patients in the primary cohort, three LAR trajectories were identified: Trajectory 1 (low-stable), Trajectory 2 (moderate rise-fall), and Trajectory 3 (high fall-rise). Trajectory 3 was associated with higher illness severity, more frequent CRRT initiation, and significantly increased 28-day mortality compared with Trajectory 1 (HR = 2.43, 95% CI 1.46-4.05; P <.001). Similar results were observed for the 90-day mortality risk. Multiple sensitivity analyses consistently confirmed the robustness of the association between trajectory three and increased 28-day mortality. In the external validation cohort (n = 508), three analogous trajectories were identified, and Trajectory 3 again demonstrated the highest 28-day mortality.

Conclusions:

Dynamic LAR trajectories are independently associated with 28-day and 90-day mortality in SA-AKI. The high fall-rise trajectory identifies a specific high-risk phenotype, suggesting that longitudinal LAR monitoring may enhance early risk stratification and inform clinical decision-making. Clinical Trial: This study has been registered with the Chinese Clinical Trial Registry (registration numbers: ChiCTR2300074175).