Scholarlink

<p>To establish a predictive model for urinary tract infections(UTIs) caused by extended spectrum β-lactamases (ESBLs)- producing <i>Escherichia coli(E.coli)</i>. A total of 558 hospitalized patients who were diagnosed with UTIs and had a urine culture of <i>E.coli</i> at our hospital between 2021 and 2024 were included. The patients were divided into a modeling set (<i>n</i> = 279) and a validation set (<i>n</i> = 279) using cluster sampling. Based on the modeling set, univariate and multivariate logistic regression analyses were used to screen risk factors for <i>E.coli</i> producing ESBLs, and a nomogram prediction model was established, then validated in the validation set. Receiver operating characteristic (ROC) curves were utilized to assess diagnostic performance, while calibration curves and Hosmer-Lemeshow test were employed to evaluate the consistency between predicted values and actual values according to the nomogram. The underlying diseases (OR = 1.79, 95%CI: 1.022–3.134), nosocomial infections (OR = 2.261, 95%CI: 1.017–5.026), previous detection of ESBLs-producing <i>Escherichia coli</i> (OR = 5.242, 95%CI: 1.371–20.043), and the number of hospitalizations within a year (OR = 1.958, 95%CI: 1.056–3.631) were identified as independent risk factors for UTIs caused by ESBLs-producing <i>E.coli</i>. The ROC curve of the prediction model shows that the area under the curve (AUC) for the modeling set is 0.734 (0.676–0.792), and the AUC for the validation set is 0.657 (0.594–0.720). The Hosmer-Lemeshow tests for the model and validation sets were <i>P</i> = 0.098 and <i>P</i> = 0.903, respectively. The nomogram model established based on the identified risk factors in this study demonstrates commendable predictive performance. It serves as a valuable reference for the rapid clinical identification of high-risk populations for UTIs caused by ESBLs-producing <i>E.coli</i> and can guide the optimization of empirical anti-infective treatment strategies.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Analysis of risk factors for urinary tract infections caused by extended spectrum β-Lactamases—producing Escherichia coli and establish the nomogram model

  • 期刊论文

  • 发表日期：

    2025年12月22日

查看全文
  • Qiang Feng,
  • Wenqiong He,
  • Qian Xu,
  • Weineng Cheng

摘要

To establish a predictive model for urinary tract infections(UTIs) caused by extended spectrum β-lactamases (ESBLs)- producing Escherichia coli(E.coli). A total of 558 hospitalized patients who were diagnosed with UTIs and had a urine culture of E.coli at our hospital between 2021 and 2024 were included. The patients were divided into a modeling set (n = 279) and a validation set (n = 279) using cluster sampling. Based on the modeling set, univariate and multivariate logistic regression analyses were used to screen risk factors for E.coli producing ESBLs, and a nomogram prediction model was established, then validated in the validation set. Receiver operating characteristic (ROC) curves were utilized to assess diagnostic performance, while calibration curves and Hosmer-Lemeshow test were employed to evaluate the consistency between predicted values and actual values according to the nomogram. The underlying diseases (OR = 1.79, 95%CI: 1.022–3.134), nosocomial infections (OR = 2.261, 95%CI: 1.017–5.026), previous detection of ESBLs-producing Escherichia coli (OR = 5.242, 95%CI: 1.371–20.043), and the number of hospitalizations within a year (OR = 1.958, 95%CI: 1.056–3.631) were identified as independent risk factors for UTIs caused by ESBLs-producing E.coli. The ROC curve of the prediction model shows that the area under the curve (AUC) for the modeling set is 0.734 (0.676–0.792), and the AUC for the validation set is 0.657 (0.594–0.720). The Hosmer-Lemeshow tests for the model and validation sets were P = 0.098 and P = 0.903, respectively. The nomogram model established based on the identified risk factors in this study demonstrates commendable predictive performance. It serves as a valuable reference for the rapid clinical identification of high-risk populations for UTIs caused by ESBLs-producing E.coli and can guide the optimization of empirical anti-infective treatment strategies.