Abstract
BACKGROUND: Acute kidney injury (AKI) after pancreaticoduodenectomy is common and early identification of such patients is critical. Inflammation contributes significantly to the onset of postoperative acute kidney injury. We aimed to construct and evaluate a predictive nomogram based on preoperative inflammatory indicators for postoperative AKI in patients undergoing pancreaticoduodenectomy. METHODS: In the current retrospective cohort study, we included 844 adult patients who underwent pancreaticoduodenectomy between December 2016 and June 2020. All enrolled patients were randomly assigned to the training and validation cohorts in a 7:3 ratio. We utilized least absolute shrinkage and selection operator (LASSO) regression for feature selection and multivariable logistic regression analyses to identify key risk factors in the training cohort. These selected factors were subsequently used to construct a nomogram. The nomogram's performance was assessed using various metrics such as the receiver operating characteristic (ROC) curve, calibration curves, Hosmer-Lemeshow goodness of fit, and decision curve analysis (DCA). RESULTS: In this cohort, AKI was observed in 98 out of 844 patients, representing an incidence rate of 11.6%. LASSO regression and multivariable logistic analysis showed that monocyte-to-lymphocyte ratio (MLR), red blood cell distribution width (RDW), and alkaline phosphatase (ALP) were independent influencing factors of postoperative AKI. The nomogram, which integrated the three identified factors, demonstrated an area under the curve (AUC) of 0.799 in both the training and validation cohorts, indicating moderate discriminative ability. The Hosmer-Lemeshow goodness of fit test and the calibration curve demonstrate good agreement between predicted and observed values. The DCA indicated a positive net clinical benefit. CONCLUSION: We developed and validated a nomogram based on preoperative MLR that could help identify individuals at risk of AKI following pancreaticoduodenectomy. This model may help clinicians optimize perioperative management for these patients.