A multimodal predictive model incorporating transcriptomic-guided blood biomarkers and clinical variables for sepsis-associated acute kidney injury

Sepsis-associated acute kidney injury (SA-AKI) is a major complication in the intensive care unit (ICU), and early risk stratification remains challenging. This study developed a multimodal predictive model integrating clinical variables with transcriptomic-guided blood biomarkers. Differentially expressed genes were identified from the Gene Expression Omnibus (GEO) database, and cluster of differentiation 177 (CD177) and interleukin-18 receptor 1 (IL18R1) were identified as immune-activation biomarkers associated with SA-AKI. In a retrospective cohort of 188 septic patients (89 SA-AKI), gene expression was quantified by quantitative polymerase chain reaction (qPCR) using blood samples collected within 24 h of ICU admission. Clinical predictors including the Sequential Organ Failure Assessment (SOFA) score, mean arterial pressure (MAP), blood urea nitrogen (BUN), C-reactive protein (CRP), and mechanical ventilation were incorporated using least absolute shrinkage and selection operator (LASSO) feature selection and logistic regression. Model performance was evaluated using the area under the receiver operating characteristic curve (AUC), calibration curves, and decision-curve analysis in an internal test cohort and an external cohort (n = 83). SOFA score, MAP, BUN, CRP, and CD177 were independent predictors of SA-AKI. The combined model showed excellent discrimination (AUC 0.989) with good calibration (Hosmer–Lemeshow p = 0.996; Brier score 0.040) and remained robust in external validation (AUC 0.909). Integrating immune biomarkers with routine clinical parameters may improve early prediction of SA-AKI, although multicenter prospective validation is required.