|
| ICU患者发生产ESBLs革兰阴性杆菌
感染的危险因素分析及预测模型的建立 |
| Analysis of risk factors for ESBLs Gram-negative bacilli infection in ICU patients and establishment of predictive models |
| |
| DOI：10.3969/j.issn.1007-8134.2022.04.009 |
| 中文关键词: 重症监护室 超广谱β-内酰胺酶 大肠埃希菌 肺炎克雷伯菌 感染 危险因素 列线图 预测模型 |
| 英文关键词: ICU extended spectrum beta-lactamases Escherichia coli Klebsiella pneumoniae infection risk factor nomogram prediction model |
| 基金项目:2018年度安徽高校自然科学研究重点项目（KJ2018C0318） |
|
| 摘要点击次数: 241 |
| 全文下载次数: 265 |
| 中文摘要: |
| [摘要] 目的 探讨ICU患者发生产超广谱β-内酰胺酶（extended spectrum beta-lactamases, ESBLs）革兰阴性杆菌感染的危险因素，并构建相关预测模型。方法 选取2017年5月—2021年4月我院ICU发生大肠埃希菌或肺炎克雷伯菌感染的189例患者作为研究对象，收集患者的临床资料，使用单因素分析、LASSO回归和多因素Logistic回归分析ICU患者30 d内发生产ESBLs革兰阴性杆菌感染的危险因素，并据此建立列线图预测模型。结果 急性生理与慢性健康评分≥16分、留置尿管时长≥7 d、抑酸剂使用时长≥3 d、第三代头孢菌素使用时长≥3 d、抗菌药物联用时长≥3 d和ICU住院时间≥15 d是ICU患者30 d内发生产ESBLs革兰阴性杆菌感染的危险因素（P均＜0.05）。依此建立预测ICU患者30 d内发生产ESBLs革兰阴性杆菌感染的列线图风险模型，模型验证结果显示C-index为0.795，校正曲线趋近于理想曲线，AUC为0.807（95%CI：0.775～0.839），在2%～81%预测范围内，列线图净获益。结论 ICU患者30 d内发生产ESBLs革兰阴性杆菌感染的危险因素包括APACHEⅡ评分≥16分、留置尿管时长≥7 d、抑酸剂使用时长≥3 d、第三代头孢菌素使用时长≥3 d、抗菌药物联用时长≥3 d和ICU住院时间≥15 d，据此构建的列线图模型能有效预测ICU患者30 d内发生产ESBLs革兰阴性杆菌感染的风险概率，具有一定的临床价值。 |
| 英文摘要: |
| [Abstract] Objective To investigate the risk factors and construct the prediction models of ESBLs Gram-negative bacilli infection in ICU patients. Methods From May 2017 to April 2021, 189 patients with Escherichia coli or Klebsiella pneumoniae infection in the ICU of our hospital were enrolled in the present research. The clinical data of those patients were collected. The risk factors of ESBLs Gram-negative bacilli infection within 30 days in ICU patients were analyzed by univariate analysis, Lasso regression and multivariate Logistic regression, and a nomogram prediction model was established based on those results. Results APACHEⅡscore≥ 16, indwelling catheter duration≥7 days, acid inhibitor duration≥3 days, the duration of the third generation cephalosporins≥3 days, the duration of combined antibiotics ≥ 3 days and the length of stay in ICU ≥ 15 days were independent risk factors for ESBLs Gram-negative bacilli infection in ICU patients within 30 days (P＜0.05). Based on this, a nomogram prediction model for predicting ESBLs Gram-negative bacilli infection within 30 days in ICU patients was established. The model verification results showed that the C-index was 0.795, the calibration curve was close to the ideal curve, and the AUC was 0.807 (95%CI: 0.775-0.839). Within the prediction range of 2%-81%, the nomogram had a net benefit. Conclusions The risk factors for ICU patients to develop ESBLs Gram-negative bacilli infection within 30 days include APACHE Ⅱ score≥16, indwelling catheter duration≥7 days, acid inhibitor duration≥ 3 days, the duration of the third generation cephalosporins≥3 days, the duration of combined antibiotics≥3 days and the length of stay in ICU≥15 days. The nomogram model has efficacy and certain clinical value in predict the risk probability of ESBLs Gram-negative bacilli infection in ICU patients within 30 days. |
|
HTML
查看全文
下载PDF阅读器 |
| 关闭 |