The clinical value of the SII for predicting the development of urosepsis after percutaneous nephrolithotripsy

Purpose Based on accumulating evidence, biomarkers related to the inflammatory response are closely associated with tumors. However, there are fewer studies related to urosepsis. The aim of this research was to investigate the importance of the SII as a predictor of the development of urosepsis after percutaneous nephrolithotripsy, utilizing a retrospective research design. Materials and Methods This study encompassed a cohort of 639 individuals diagnosed with kidney stones between January 2019 and August 2022. The patients were categorized into a modeling group consisting of 439 individuals and a validation group comprising 200 individuals, following a ratio of 7:3. R software was used to perform multivariate logistic regression analysis after screening with LASSO regression. The risk line graph model, ROC curve, calibration curve, and decision curve of the modeling group were drawn and visualized using R statistical software. These findings were also drawn and verified in the validation cohort. Results In a cohort of 439 patients, the prevalence of urosepsis was found to be 9.11% (40/439). Subsequently, a multivariate logistic regression analysis was conducted following a screening process utilizing LASSO regression. Our results suggested four risk factors for PCNL-US, namely, positive urinary nitrite (OR = 3.176, 95%CI: 1.390–7.097, P < 0.001), preoperative fever (OR = 2.762, 95%CI: 1.021–7.104, P = 0.039), positive urine culture (OR = 2.447, 95%CI: 1.077–5.476, P = 0.030), and high preoperative SII (OR = 4.943, 95%CI: 2.323–10.776, P < 0.001). According to above four factors, we constructed a column-line graph prediction model of risk factors for PCNL-US. The area under the ROC curve (AUC) of the modeling group was 0.818 (95% CI: 0.739–0.898). The area under the ROC curve (AUC) of the validation group was 0.794 (95% CI: 0.679–0.909). The Hosmer-Lemeshow test was greater than 0.05 in both groups, indicating a good calibration curve and good clinical decision-making performance. Conclusions This study suggested that positive urinary nitrite levels, preoperative fever, and positive urine culture are risk factors for PCNL-US. Additionally, a high preoperative SII level is recognized as a separate risk factor for the occurrence of urosepsis. The clinical prediction model constructed based on these four risk factors may serve as a reference for preventing the occurrence of PCNL-US.