成人耐碳青霉烯革兰阴性菌血流感染临床特征及死亡危险因素分析

The clinical characteristics and risk factors of mortality in adult patients with carbapenem resistant Gram-negative bacteria blood infections

目的探讨碳青霉烯类耐药革兰阴性菌(CR-GNB)血流感染的临床特征及死亡危险因素,为临床耐药菌感染合理治疗提供依据。方法回顾性调查2014年7月—2016年6月四川大学华西医院明确诊断的CR-GNB血流感染患者157例,根据致病菌是否为广泛耐药,分为XDR组及非XDR组,分析其临床特征及治疗方案,并对CR-GNB血流感染者临床转归进行多因素Logistic回归分析。结果 CR-GNB血流感染者XDR组68例,非XDR组89例。患者主要分布在ICU病房及血液科,基础疾病以重症胰腺炎及肿瘤多见。CR-GNB血流感染全因病死率55.41%,XDR组高于非XDR组(P<0.05)。CR-GNB以鲍曼不动杆菌、铜绿假单胞菌和大肠埃希菌多见。根据药敏结果调整治疗方案似能改善CR-GNB血流感染者的临床转归,但差异无统计学意义(P>0.05)。多因素Logistic回归分析显示,年龄>65岁(OR=3.059,95%CI 1.117~8.380)及入住重症监护病房(ICU)(OR=2.541,95%CI 1.179~5.477)是CR-GNB血流感染患者死亡的可能危险因素(P<0.05)。结论目前CR-GNB感染形势严峻,入住ICU的高龄患者预后差,需要加强各种感控措施,尽量延缓或避免CR-GNB的产生与感染,需积极探索新的治疗方法,提高疗效。

Objective To analyze the clinical characteristics and risk factors of mortality in adult patients with carbapenem resistant Gram-negative bacteria （CR-GNB） blood infections. Methods A retrospective study was carried out in 157 patients with confirmed diagnosis of CR-GNB blood infections from July 2014 to June 2016 in West China hospital, Sichuan University. The patients were divided into XDR group and non-XDR group according to whether the isolates were extensively drug-resistant bacteria. The clinical features and antibacterial drug therapies were reviewed. The risk factors of mortality associated with CR-GNB blood infections were examined with multivariate logistic regression analysis. Results There were 68 cases in XDR group and 89 cases in non-XDR group. The patients with CR-GNB blood infections mainly distributed in intensive care unit （ICU） and hematology department. The common underlying diseases were severe acute pancreatitis and tumor. All-cause mortality of CRGNB blood infections was 55.41%, and all-cause mortality in XDR group higher than that in non-XDR group （P〈0.05）. The most common CR-GNB were Acinetobacter baumannii, Pseudomonas aeruginosa and Escherichia coli. Adjusting the antibacterial drug therapy according to the antimicrobial susceptibility results could improve the clinical outcome of CR-GNB blood infections, however, this difference was not statistically significant （P〉0.05）. The multivariate logistic regression analysis showed that the risk factors of mortality associated with CR-GNB blood infections were patients over 65 （OR=3.059, 95% CI 1.117-8.380, P=0.030） and admission to ICU （OR=2.541, 95% CI 1.179-5.477, P=0.017） （P〈0.05）. Conclusion The situation of CR-GNB infection is fairly severe. The elder patients in ICU were with poor prognosis. We should adopt effective measure to prevent the emerging of carbapenem resistant strains and explore the new treatment methods to improving outcomes for these patients.