Construction and Validation of a Risk Prediction Model for Postoperative Urinary Tract Infection in Intracranial Hemorrhage Patients

Yaqiang Li^1,2*Ke Jiang³Zhongbo Sun³Yangyang Chen³

• ¹People's Republic of China, Anhui, China
• ²Department of Neurology, People's Hospital of Lixin country, Bozhou, China
• ³Department of Neurology, First Affiliated Hospital of Anhui University of Science and Technology (First People’s Hospital of Huainan), Huainan, China

Objectives: This study aimed to explore the risk factors for postoperative urinary tract infection (UTI) in patients with intracranial hemorrhage (ICH) and to establish and validate a nomogram that integrates predictive factors to estimate the likelihood of postoperative UTI following ICH. Methods:This study enrolled surgical patients with intracerebral hemorrhage (ICH) from January 2020 to November 2023, categorizing them into training and validation groups. Using multivariate logistic regression analysis, we identified significant predictors of postoperative UTI in the training group to include in the nomogram. To evaluate the discriminative ability of the nomogram, the area under the receiver operating characteristic curve was utilized. The calibration of the nomogram was examined using the Hosmer-Lemeshow goodness-of-fit test and calibration plots. Additionally, decision curve analysis was performed to determine the clinical utility of the nomogram. Results: A total of 436 patients with ICH were enrolled in the study. In the training cohort (n=305), 87 patients had UTI after intracerebral hemorrhage. Multivariate logistic regression analysis demonstrated neutrophil-to-lymphocyte ratio (NLR) (OR, 1.357; 95% CI, 1.162-1.899; P = 0.036), D-Dimer (OR, 3.050; 95% CI, 1.925-4.856; P = 0.023), tumor necrosis factor-α (TNF-α) (OR, 1.957; 95% CI, 1.670-2.378; P < 0.001), and age greater than or equal to 65 years (OR, 2.531; 95% CI, 1.765-3.625; P = 0.043) were independent predictors for postoperative UTI and constructed the nomogram. The nomogram demonstrated a high predictive capability with a C-index of 0.865 (95% CI, 0.796-0.935) in the training cohort and 0.867 (95% CI, 0.777-0.958) in the validation cohort. The Hosmer-Lemeshow goodness-of-fit assessment indicated a strong agreement between the predicted probabilities and the observed outcomes for both the training cohort (χ2 = 26.01, df = 8, P = 0.136) and the validation cohort (χ2 = 5.652, df = 8, P = 0.238). The decision curve analysis demonstrated that the nomogram was markedly effective for predicting UTI in the training cohort and was further validated in the subsequent cohort. Conclusions: This study presents a new and practical nomogram that uses NLR, D-Dimer, TNF-α, and age 65 years or older to effectively predict the risk of postoperative UTI in patients with ICH.