Abstract: Objective  To investigate risk factors for exit site infection (ESI) in peritoneal dialysis (PD) patients, develop a predictive model for early identification of high-risk individuals, and guide personalized preventive strategies. Methods PD patients regularly followed at the peritoneal dialysis clinic of the Second Affiliated Hospital of Harbin Medical University from October 2023 to November 2024 were enrolled. Based on the 2023 International Society for Peritoneal Dialysis (ISPD) diagnostic criteria for ESI, patients were categorized into ESI and non-ESI groups. Binary logistic regression was used to identify ESI risk factors, followed by the construction of a nomogram model. Model performance was evaluated through predictive accuracy and clinical applicability. Results  Among 245 patients, 37 were classified into the ESI group and 208 into the non-ESI group. Significant differences were observed between groups in age (Z =-4.199, P ＜0.001), dialysis vintage (Z =-4.908, P ＜0.001), hemoglobin (Z =-4.445, P ＜0.001), serum albumin (Z =-5.271, P ＜0.001), external cuff extrusion (χ²=27.038, P ＜0.001), and history of catheter traction (χ²=24.797, P ＜0.001). Binary logistic regression identified. Based on binary logistic regression analysis, age (OR=0.796, 95% CI: 0.705～0.899, P＜0.001), dialysis vintage (OR=1.036, 95% CI: 1.010～1.064, P=0.007), hemoglobin (OR=0.951, 95% CI:0.914～0.990, P=0.013), serum albumin (OR=0.796, 95% CI: 0.705～0.899, P＜0.001), external cuff extrusion (OR=6.636, 95% CI:1.156～38.114, P=0.034), and history of catheter traction (OR=6.530, 95% CI: 1.275～33.454, P=0.024) were identified as independent risk factors for peritoneal dialysis catheter exit-site infection. The nomogram model demonstrated excellent predictive performance (AUC= 0.935, 95% CI: 0.8916～0.978). Hosmer-Lemeshow goodness-of-fit test (χ²=12.712, P=0.122) and decision curve analysis confirmed robust calibration and clinical utility.  Conclusion Age, dialysis vintage, hemoglobin, serum albumin, external cuff distance, external cuff extrusion, and catheter traction are independent risk factors for ESI. The developed predictive model effectively stratifies ESI risk in PD patients, providing a scientific basis for early intervention and personalized management.

Key words: Peritoneal dialysis, Peritoneal dialysis catheter exit-site infection, Risk factors, Predictive value