Abstract:

Objective　To identify the risk factors for methicillin-resistant Staphylococcus aureus (MRSA) infection in patients with cerebral infarction in Guangdong Provincial Second Hospital of Traditional Chinese Medicine and establish a clinical prediction model, which can provide references for the prevention and treatment of MRSA infection. Methods　A total of 277 patients diagnosed with cerebral infarction and Staphylococcus aureus infection who were hospitalized in Guangdong Provincial Second Hospital of Traditional Chinese Medicine from January 2021 to December 2023 were included in this retrospective analysis, including 188 males and 89 females who were 77 (29, 100) years old. According to whether they were methicillin-resistant, the patients were divided into a MRSA group and a methicillin-susceptible Staphylococcus aureus (MSSA) group. The data were compared between the two groups by χ² and rank sum tests. R 4.3.0 was used for the data statistics and visualization processing; the optimal subset regression analysis and multivariate logistic regression analysis were used to analyze the risk factors of the cerebral infarction patients complicated by MRSA infection; a corresponding nomogram model was established. The receiver operating characteristic curve (ROC) and DCA curve were drawn to evaluate the model. Results　The isolation rate of MRSA in Guangdong Provincial Second Hospital of Traditional Chinese Medicine was 71.84% (199/277). The multivariate logistic regression analysis showed that creatinine reduction (OR=0.995, 95%CI 0.992-0.999), using central venous catheterization (OR=2.390, 95%CI 1.262-4.223), history of hospitalization in respiratory department before infection (OR=4.683, 95%CI 1.377-15.929), and combined use of antibiotics (OR=2.270, 95%CI 1.231-4.187) were independent risk factors for MRSA infection in the patients. The area under the curve of the nomogram model based on the 6 variables obtained from the optimal subset regression analysis was 0.720; the sensitivity, specificity, and accuracy were 67.84%, 70.51%, and 68.59%, respectively. The diagnostic and predictive performance was acceptable. Conclusion　We should pay attention to the risk factors of MRSA infection in patients with cerebral infarction, and use the nomogram model to predict the occurrence of MRSA infection.

Key words:

Cerebral infarction, MRSA, Risk factors, Nomogram

摘要：

目的　明确广东省第二中医院脑梗死患者耐甲氧西林金黄色葡萄球菌（MRSA）感染的危险因素，并建立临床预测模型，为MRSA的感染防治提供参考。方法　选取2021年1月至2023年12月于广东省第二中医院住院并诊断为金黄色葡萄球菌感染的脑梗死患者277例进行回顾性研究，其中男188例，女89例，年龄77（29，100）岁。根据是否耐甲氧西林分为MRSA组和甲氧西林敏感金黄色葡萄球菌（MSSA）组。采用χ²检验、秩和检验比较两组间差异。使用R 4.3.0语言对数据进行统计和可视化处理，采用最优子集回归分析和多因素logistic回归分析脑梗死患者合并MRSA感染的危险因素，建立相应列线图模型。绘制受试者操作特征曲线（ROC）和DCA曲线对模型进行评价。结果　广东省第二中医院MRSA分离率为71.84%（199/277）。多因素logistic回归分析结果显示，肌酐降低（OR=0.995，95%CI 0.992～0.999）、使用中心静脉插管（OR=2.390，95%CI 1.262～4.223）、感染前呼吸科住院史（OR=4.683，95%CI 1.377～15.929）、联用抗菌药物（OR=2.270，95%CI 1.231～4.187）是脑梗死患者发生MRSA感染的独立危险因素。基于最优子集回归分析得到的6个变量建立的列线图模型的曲线下面积为0.720，灵敏度、特异度、准确度分别为67.84%、70.51%、68.59%，诊断及预测性能尚可。结论　临床上应关注脑梗死患者MRSA感染危险因素，并利用本列线图模型预测感染发生。

关键词:

脑梗死, MRSA, 危险因素, 列线图