Abstract
BACKGROUND: Urinary tract infection (UTI) is a common disease in urology and often occurs in patients with urolithiasis. This study aimed to identify the risk factors for UTI in patients with ureterolithiasis complicated with hydronephrosis, and to construct a simple and practical nomogram to predict the incidence of UTI for patients. METHODS: A total of 383 patients were enrolled from September 2019 to June 2022. The results from univariate and multivariate logistic regression showed the risk factors for predicting UTI and a prediction model was constructed. Subsequently, the differentiation, calibration, and clinical applicability of the model were estimated by receiver operating characteristic (ROC) curve analysis, calibration curve, and decision curve analysis (DCA), respectively. RESULTS: The study included 72 (18.80%) patients with UTI. Multivariate logistic regression showed that tissue rim sign (P=0.04), positive urinary nitrite (P<0.001), and positive urinary leukocyte esterase (P=0.005) were independent predictive indexes of UTI for patients with ureterolithiasis complicated with hydronephrosis, and a nomogram was constructed in accordance with these indicators. The area under the ROC curve was 0.773, which indicated good prediction ability. The Hosmer-Lemeshow test (P=0.97) indicated that the model fitted well. The calibration curve and DCA showed good consistency and clinical applicability, respectively. CONCLUSIONS: The prediction model constructed with the risk factors including tissue rim sign, positive urinary nitrite, and positive urinary leukocyte esterase can better detect patients with UTI early and take timely intervention measures.