DEVELOPMENT AND VALIDATION OF A PREDICTION MODEL FOR SEPTIC SHOCK-ASSOCIATED ACUTE KIDNEY INJURY: A MULTICENTER STUDY USING NOMOGRAM MODELING

Abstract

Background: Septic shock-associated acute kidney injury (SS-AKI) is a severe complication with high mortality. This study aimed to investigate the risk factors associated with AKI in patients with septic shock and establish a nomogram to predict its occurrence. Methods: Patients with septic shock were categorized based on the development of AKI. A binary logistic regression was used to identify significant risk factors, which were then incorporated into a nomogram. The performance of the nomogram was evaluated using receiver operating characteristic curve analysis, calibration curve, and decision curve analysis. A validation set was used to assess the model's generalizability. Results: Of the 507 septic shock patients enrolled in this study, 174 (34.3%) developed AKI. The dataset was randomly partitioned into a training set (n = 355) and a validation set (n = 152) at a ratio of 7:3. The predictive factors incorporated into the nomogram included chronic kidney disease, diuretic administration, deresuscitation during vasopressor administration, mechanical ventilation, source control failure, restrictive fluid resuscitation, and Sequential Organ Failure Assessment scores. The developed nomogram demonstrated excellent performance in predicting the risk of AKI in patients with septic shock. The model achieved an area under the receiver operating characteristic curve of 0.788 (95% confidence interval, 0.737-0.839) in the training set and 0.770 (95% confidence interval, 0.693-0.846) in the validation set, indicating strong discriminatory ability. The calibration curve analysis, using the Hosmer-Lemeshow test, indicated good agreement between the predicted and observed probabilities of AKI in both the training set ( P = 0.468) and the validation set ( P = 0.396). The decision curve analysis further indicated that the nomogram demonstrated substantial clinical utility in both the training set (0.09-0.87) and the validation set (0.11-0.64). Conclusions: The nomogram serves as an invaluable tool for clinicians to assess the risk of AKI in patients experiencing septic shock and facilitates timely intervention.

Keywords: AKI—acute kidney injury; APACHE II—Acute Physiology and Chronic Health Evaluation II; AUC—area under the curve; Acute kidney injury; CKD—chronic kidney disease; CRRT—continuous renal replacement therapy; DCA—decision curve analysis; EMRS—electronic medical record system; KDIGO—Kidney Disease Improving Global Outcomes; ML—machine learning; MV—mechanical ventilation; RFR—restricted fluid resuscitation; ROC—receiver operating characteristic; SA-AKI—sepsis-associated acute kidney injury; SOFA—Sequential Organ Failure Assessment; SS-AKI—septic shock–associated acute kidney injury; deresuscitation; restricted fluid resuscitation; septic shock; source control.