【摘要】 目的：探究临床患者碳青霉烯类耐药肠杆菌目（CRE）细菌感染及感染患者死亡相关危险因素，并对CRE感染进行预测分析。方法：回顾性分析2021年4月—11月112例CRE感染患者的临床资料，以同期112例碳青霉烯类敏感肠杆菌目细菌（CSE）感染患者作为对照组，分析CRE菌株分布特征及耐药性，单因素分析住院患者CRE感染的危险因素及感染CRE患者死亡的危险因素。通过二元Logistic回归分析方法，分析患者CRE感染的独立危险因素及感染CRE患者死亡的独立危险因素。并将独立危险因素中的连续变量进行受试者工作特征（ROC）曲线分析，以评估该因素预测住院患者感染CRE的效能。结果：CRE感染多见于重症监护病房（ICU），以呼吸道标本为主，肺炎克雷伯菌最常见。单因素分析显示，泌尿系统疾病（χ~2=4.074,P<0.05）、实体恶性肿瘤（χ~2=11.687,P<0.05）、气管插管机械辅助通气（χ~2=32.749,P<0.05）、气管切开（χ~2=30.701,P<0.05）、留置静脉导管（χ~2=14.613,P<0.05）、留置动脉导管（χ~2=6.298,P<0.05）、尿管（χ~2=4.464,P<0.05）、胃管（χ~2=15.076,P<0.05）、碳青霉烯类药物暴露史（χ~2=39.703,P<0.05）、β内酰胺酶抑制剂暴露史（χ~2=58.810,P<0.05）、非碳青霉烯类药物联合用药（χ~2=39.938,P<0.05）、三四代头孢暴露史（χ~2=5.098,P<0.05）、入院时急性生理学及慢性健康状况评分系统Ⅱ（APACHEⅡ）评分（U=4 367.000,P<0.05）、APACHEⅡ评分≥11分（χ~2=4.210,P<0.05）、中性粒细胞百分数（t=12.648,P<0.05）、淋巴细胞百分数（U=4 205.000,P<0.05）、粒淋比（U=3 241.000,P<0.05）、凝血酶原时间（t=7.802,P<0.05）、活化部分凝血活酶时间（U=3 662.000,P<0.05）、D-二聚体（U=2 616.000,P<0.05）等与住院患者感染CRE相关。年龄（U=625.500,P<0.05）、年龄≥78岁（χ~2=17.824,P<0.05）、心脑血管疾病（χ~2=9.737,P<0.05）、留置静脉导管（χ~2=5.379,P<0.05）、尿素（U=606.000,P<0.05）、肌酐（U=788.000,P<0.05）、尿酸（U=714.500,P<0.05）等因素与CRE感染患者死亡相关。多因素分析显示，泌尿系统疾病（OR=14.222,95%CI:1.474～137.270）、留置静脉导管（OR=8.735,95%CI:1.620～47.083）、碳青霉烯类药物暴露史（OR=11.520,95%CI:2.781～47.716）、β内酰胺酶抑制剂暴露史（OR=5.763,95%CI:1.368～24.280）、非碳青霉烯类药物联合用药（OR=8.705,95%CI:2.114～35.841）、中性粒细胞百分数（OR=1.083,95%CI:1.026～1.142）、凝血酶原时间（OR=1.714,95%CI:1.109～2.648）为住院患者感染CRE的独立危险因素。年龄≥78岁（OR=26.831,95%CI:2.745～262.287）和患有心脑血管疾病（OR=28.427,95%CI:2.379～339.668）为CRE感染患者死亡的独立危险因素。将感染CRE的独立危险因素中的连续变量中性粒细胞百分数、凝血酶原时间进行ROC曲线分析，中性粒细胞百分数的曲线下面积（AUC）为0.645，高于凝血酶原时间的AUC(0.598)，最佳截断值为69.45，对应敏感性和特异性分别为75.5%和52.2%。结论：住院患者感染CRE的独立危险因素有：患有泌尿系统疾病、留置静脉导管、碳青霉烯类药物暴露史、β内酰胺酶抑制剂暴露史、非碳青霉烯类药物联合用药、中性粒细胞百分数和凝血酶原时间。CRE感染患者死亡的独立危险因素有：年龄≥78岁及患有心脑血管疾病。中性粒细胞百分数对住院患者感染CRE有一定的预测价值。更多还原

【Abstract】 Objective:To explore the risk factors of carbapenem-resistant Enterobacterales(CRE)infection in clinical patients and death of infected patients,and to analyze the predictive value of CRE infection. Methods:A retrospective method was used to select the clinical data of 112 cases with CRE from April to November 2021. And 112 cases with carbapenem-sensitive Enterobacterales(CSE)infection in the same period were taken as the control group. The distribution and drug resistance of CRE strains was analyzed. And the risk factors of CRE infection and death of infected patients were analyzed by univariate analysis. The independent risk factors of CRE infection and death of infected patients were analyzed by binary Logistic regression analysis. The continuous variables among independent risk factors were analyzed using the receiver operating characteristic(ROC)curve to evaluate the effectiveness of this factor in predicting CRE infection in hospitalized patients. Results:CRE infection was common in intensive care unit(ICU),mainly in respiratory tract specimens,and Klebsiella pneumoniae was the most common pathogen. A univariate analysis showed that,urinary system diseases （χ~2=4.074,P<0.05）,solid malignancy（χ~2=11.687,P<0.05）,endotracheal intubation and mechanical ventilation（χ~2=32.749,P<0.05）,tracheotomy（χ~2=30.701,P < 0. 05）,indwelling venous catheter（χ~2= 14. 613,P < 0. 05）,indwelling arterial catheter（χ~2= 6. 298,P < 0. 05）,urinary catheter（χ~2= 4.464,P<0.05）,gastric tube（χ~2=15.076,P<0.05）,exposure history of Carbapenem（χ~2=39.703,P<0.05）,exposure history of β-lactamase inhibitors（χ~2=58.810,P<0.05）,non carbapenem combination therapy（χ~2=39.938,P<0.05）,exposure history of third and fourth generation cephalosporins（χ~2=5.098,P<0.05）,Acute Physiology and Chronic Health evaluation Scoring System Ⅱ（APACHE Ⅱ）at admis sion(U =4 367.000,P<0.05),APACHE Ⅱ≥11（χ~2=4.210,P<0.05）,neutrophil percentage(t=12.648,P<0.05),lymphocyte percentage(U =4 205.000,P<0.05),granulocyte to lymphocyte ratio(U=3 241.000,P<0.05),prothrombintime(t=7.802,P<0.05),activated partial thromboplastin time(U=3 662.000,P<0.05),D-dimer(U=2 616.000,P<0.05)were related to CRE infection in hospitalized patients. Age(U =625.500,P < 0.05),age ≥78 （χ~2=17.824,P <0.05）,cardiovascular and cerebrovascular diseases（χ~2=9.737,P <0.05）,indwelling venous catheter（χ~2=5.379,P<0.05）,urea(U=606.000,P<0.05),creatinine(U=788.000,P<0.05),uricacid(U=714.500,P<0.05)were related to the death of CRE-infected patients. Multivariate analysis showed that urinary system diseases(OR=14.222,95%CI:1.474-137.270),indwelling venous catheter(OR=8.735,95%CI:1.620-47.083),exposure history of carbapenem(OR=11.520,95%CI:2.781-47.716),exposure history of β-lactamase inhibitors(OR=5.763,95%CI:1.368-24.280),non carbapenem combination therapy(OR=8.705,95%CI:2.114-35.841),neutrophil percentage(OR=1.083,95%CI:1.026-1.142),prothrombin time(OR=1.714,95%CI:1.109-2.648)were independent risk factors for CRE infection in hospitalized patients. Age ≥78(OR=26.831,95%CI:2.745-262.287)and cardiovascular and cerebrovascular diseases(OR=28.427,95%CI:2.379-339.668)were independent risk factors for death in CREinfected patients. The ROC curve analysis of the continuous variable neutrophil percentage and prothrombin time among the independent risk factors of CRE infection showed that the area under the curve(AUC)of neutrophil percentage was 0.645,higher than the AUC of prothrombin time(0.598). The optimal cut-off value was 69.45,and the corresponding sensitivity and specificity were 75.5% and 52.2%,respectively.Conclusion:Independent risk factors of CRE infection in hospitalized patients include urinary system disease,indwelling venous catheter,carbapenem drug exposure history,exposure history of β-lactamase inhibitors,noncarbapenem combination therapy,neutrophil percentage and prothrombin time. The independent risk factors for death of infected patients with CRE are age≥78 and suffering from cardiovascular and cerebrovascular diseases. Neutrophil percentage has a certain predictive value for the infection of CRE in hospitalized patients.更多还原