UDFSSF Model is Superior to SOFA Score in Predicting 30-day Mortality of Septic Patients With Acute Kidney Injury

Abstract

Objective

While the sequential organ failure assessment (SOFA) score is primarily utilized for sepsis severity stratification, its predictive performance for sepsis-associated acute kidney injury (SA-AKI) prognosis remains suboptimal. This study specifically focused on identifying 30-day mortality risk factors in SA-AKI patients and developing a novel prognostic model to enhance short-term outcome prediction in this critically ill population.

Methods

389 SA-AKI patients were finally enrolled and categorized into survivors (n = 286) and non-survivors (n = 103) based on 30-day survival status. The least absolute shrinkage and selection operator (LASSO), and univariate and multivariate logistic regression analyses were utilized to examine the independent risk factors associated with 30-day mortality. Calibration and decision curve analysis (DCA) curves were employed to assess the model’s goodness of fit.

Results

Sequential organ failure assessment score, urinary albumin-to-creatinine ratio (uAlb/Cr), D-dimer (D-D), fibrinogen (FIB), systemic inflammation response index (SIRI), systemic immune-inflammation index (SII), and ferritin were independent risk factors for predicting 30-day mortality in SA-AKI patients. The uAlb/Cr + D-D + FIB + SIRI+SII + ferritin (UDFSSF) model showed an AUC of 0.975 (95% CI: 0.959-0.992), significantly outperforming the SOFA score (AUC: 0.716, 95% CI: 0.655-0.778).

Conclusion

Due to its multidimensional yet readily obtainable parameters, the UDFSSF model holds significant promise for clinical implementation as a superior forecasting tool for 30-day mortality of SA-AKI patients.