Construction and validation of the nomogram predictive model for post-percutaneous nephrolithotomy urinary sepsis

Area of Science:

• Urology
• Nephrology
• Infectious Disease

Background:

• Post-percutaneous nephrolithotomy (PCNL) sepsis is a significant complication.
• Early identification and intervention are crucial for managing urosepsis.
• Accurate predictive tools are needed to assess patient risk.

Purpose of the Study:

• To construct and validate a clinical nomogram for predicting post-PCNL sepsis.
• To aid urologists in early identification and intervention of urosepsis.
• To provide an evidence-based tool for risk assessment in PCNL patients.

Main Methods:

• Retrospective study of 946 patients undergoing PCNL for kidney or upper ureteral stones (Jan 2019-Sep 2022).
• Univariate and multivariate logistic regression analysis to identify risk factors.
• Nomogram construction and validation using ROC, calibration, and decision curve analysis.

Main Results:

• 69 patients (7.29%) developed post-PCNL urinary sepsis.
• Independent risk factors identified: positive urinary nitrite (OR=5.9), positive urine culture (OR=7.54), operative time ≥120 min (OR=20.93), and stone size ≥30 mm (OR=13.81).
• Nomogram showed high accuracy (AUC=0.909 in development, 0.922 in validation) with good calibration and clinical utility.

Conclusions:

• Positive preoperative urine culture, positive urinary nitrite, operative time ≥120 min, and stone size ≥30 mm are independent predictors of post-PCNL sepsis.
• The developed nomogram effectively assesses the risk of urinary sepsis post-PCNL.
• This tool can support clinical decision-making for preventing and managing post-PCNL urosepsis.