The prognostic value of the neutrophil‑to‑platelet ratio for predicting postoperative sepsis complications after ureteroscopic lithotripsy.

Background: Sepsis is a leading cause of mortality following ureteroscopic lithotripsy (URSL). Current diagnostic criteria, such as the Sequential Organ Failure Assessment (SOFA) score, are effective for identifying established sepsis but often fail to provide early warning before clinical deterioration. This creates a critical need for readily available predictive biomarkers. Given that a rise in neutrophils and a fall in platelets are central to sepsis pathophysiology, neutrophil-to-platelet ratio (NPR) represents a promising candidate for early risk stratification. This study aimed to evaluate the NPR for the early prediction of postoperative sepsis following URSL. Methods: We retrospectively analyzed data from 674 patients undergoing URSL (January-December 2024). After screening, 385 patients were included and classified into sepsis (n=40) and control (n=345) groups based on postoperative urine culture and SOFA score. Biomarkers measured 3-hour postoperatively were analyzed using logistic regression. Diagnostic performance was assessed by receiver operating characteristic (ROC) curves. Results: Multivariable analysis identified NPR, monocyte-to-lymphocyte ratio (MLR), and female gender as independent predictors of sepsis. Among these, NPR demonstrated the strongest association. The sepsis group exhibited a significantly higher NPR (0.081±0.051 vs. 0.020±0.013, P<0.01). NPR showed robust diagnostic efficacy with an area under the curve (AUC) of 0.89 [95% confidence interval (CI): 0.838-0.955], sensitivity of 80.0%, and specificity of 88.4% at an optimal threshold of >2.77×10-2. Conclusions: The NPR, obtained from a routine blood count 3-hour after URSL, is a strong independent predictor of subsequent sepsis, showing high diagnostic accuracy. These findings highlight its potential clinical utility for early risk stratification, pending validation in prospective multicenter cohorts.