Abstract
PURPOSE: This study aims to evaluate the predictive value of the renal resistive index (RRI) and β2-microglobulin (β2-MG) for acute kidney injury (AKI) in urosepsis patients and to develop a clinical prediction model for AKI risk. METHODS: Data from 108 urosepsis patients at Tongji Hospital were analyzed. Patients were divided into AKI (67 patients) and non-AKI (41 patients) groups based on KDIGO guidelines. Univariate analysis identified potential AKI risk factors, which were further assessed using multivariate logistic regression. A nomogram was constructed based on significant predictors, with internal validation via the bootstrap method. The model's accuracy and clinical utility were evaluated using ROC curves and Decision Curve Analysis (DCA). RESULTS: Multivariate analysis identified RRI, β2-MG, procalcitonin (PCT), and serum creatinine (Scr) as independent AKI risk factors. The combined predictive indicators yielded an AUC of 0.879, outperforming individual markers (P < 0.05). The prediction model achieved an AUC of 0.949, with high sensitivity (92.5%) and specificity (82.9%). Further analysis revealed that RRI, β2-MG, PCT, and APACHE II scores were independent predictors of poor prognosis in urosepsis-related AKI, with combined RRI and β2-MG predictions showing superior performance. CONCLUSION: Elevated RRI, β2-MG, PCT, and Scr levels are independent predictors of AKI in urosepsis. RRI, β2-MG, PCT, and APACHE II scores also predict poor prognosis in urosepsis-related AKI. The nomogram combining these factors demonstrates high predictive accuracy and clinical applicability.