某院2015-2019年肺炎克雷伯菌的耐药性及耐药基因分析

摘要：目的 分析某院临床分离的肺炎克雷伯菌(Klebsiella pneumoniae，KP)耐药情况及耐药基因，指导临床规范使用抗菌药物。方法 收集该院2015-2019年临床分离的2416株肺炎克雷伯菌，使用VITEK-2 Compact系统对临床分离株进行细菌鉴定，药敏试验采用MIC法和K-B纸片扩散法，采用WHONET 5.6软件对菌株的标本来源、科室分布、检出率及耐药率进行统计，采用SPSS 26.0软件对5年期间菌株的检出率、耐药率进行比较及趋势分析，选取2019年内科重症监护病房(Intensive care unit，ICU)的19株耐碳青霉烯肺炎克雷伯菌(carbapenem-resistant Klebsiella pneumoniae，CR-KP)，通过二代测序平台对其进行全基因组测序(whole genome sequencing, WGS)，运用丹麦技术大学(DTU)基因组流行病学中心(CGE)(https://cge.cbs.dtu.dk/services/ResFinder/)的ResFinder 4.0工具进行耐药基因筛查，运用MORPHEUS(https://software.broadinstitute.org/morpheus/)在线制作耐药基因热图。结果 2015-2019年临床标本共分离出2416株KP，主要来自痰液、尿液，分别占51.7%(1249/2416)、11.3%(273/2416)，科室主要分布于内科ICU、外科ICU，分别占23.2%(561/2416)、19.1%(461/2416)。每年KP的分离率呈上升趋势(χ2趋势=18.190，P＜0.001)，其对常用抗菌药物的耐药率呈逐年升高趋势，除对替加环素的耐药率无法统计外，对其他各类抗菌药物的耐药率差异均有统计学意义(P＜0.001)，对亚胺培南、美罗培南耐药率分别从2015年的2.9%、3.0%上升至2019年的16.0%、13.7%。5年间共检出354株CR-KP，主要来自痰液、尿液，分别占43.8%(155/354)、10.7%(38/354)，分布较高的科室为外科ICU、内科ICU，分别占14.4%(51/354)、10.2%(36/354)，每年的检出率呈上升趋势(χ2趋势=46.176，P＜0.001)，其对各类抗菌药物的耐药率呈逐年升高趋势，除对庆大霉素的耐药率差异无统计学意义(P=0.105)、对替加环素的耐药率差异无法统计外，对其他各类抗菌药物的耐药率差异均有统计学意义(P＜0.05)。采用WGS技术对19株CR-KP进行分析，共筛查出15种耐药基因，主要包括β-内酰胺酶耐药基因(blaKPC-2)、喹诺酮类耐药基因(qnrS1)、磷霉素耐药基因(fosA)、四环素耐药基因(tetA)、氯霉素耐药基因(catA2)以及氨基糖苷类耐药基因(aadA2b、rmtB)等。结论 该院KP、CR-KP的耐药情况日趋严重，CR-KP同时携带多种耐药基因，耐药表型多样化。运用WGS结合生物信息学分析技术研究细菌耐药机制，有助于耐药菌防控措施的精准实施，未来将成为耐药菌研究的主流技术之一。

Analysis of antibiotics resistance and resistance genes of Klebsiella pneumoniae in a hospital from 2015 to 2019

Abstract Objective To analyze the antibiotic resistance and drug resistance genes of Klebsiella pneumoniae (KP) isolated from clinical departments in the second hospital of Anhui Medical University, so as to guide the rationaluse of antibacterial drugs in clinical departments. Methods 2,416 strians of Klebsiella pneumoniae isolated from clinical in the hospital from 2015 to 2019 were collected, and the VITEK-2 compact automated microorganism system was used for bacterial identification. The MIC and K-B methods were used for drug susceptibility testing. The WHONET 5.6 software was used to perform statistical analysis on the source of strains, the distribution of departments, the detection rate, and the drug resistance rate. The SPSS 26.0 software was used to compare the detection rate and drug resistance rate of strains during the five-year period and analyze the trend. The 19 strains of carbapenem-type Klebsiella pneumoniae (CR-KP) in the internal medicine Intensive Care Unit (ICU) in 2019 were selected to perform whole genome sequencing (WGS) on the next-generation sequencing platform. The ResFinder 4.0 of the Danish Technical University (DTU) Center for Genomic Epidemiology (CGE) (https://cge.cbs.dtu.dk/services/ResFinder/) was used to select resistance genes, and the MORPHEUS (https: //software.broadinstitute.org/morpheus/) was used to make a heat map of resistance genes online. Results A total of 2,416 KP strains were isolated from clinical specimens from 2015 to 2019, mainly from sputum and urine, accounting for 51.7% (1,249/2,416) and 11.3% (273/2,416) respectively. The departments were mainly distributed in internal medicine ICU and surgical ICU, and respectively accounted for 23.2% (561/2,416) and 19.1% (461/2,416). The annual separation rate of KP strains was increasing (χ2 trend=18.190, P＜0.001), and the resistance rate to commonly used antibiotics was increasing year by year. Except that the resistance rate to tigecycline cannot be counted, other differences in the resistance rates of various antibiotics were statistically significant (P<0.001). The resistance rates to imipenem and meropenem increased from 2.9% and 3.0% in 2015 to 16.0% and 13.7% in 2019 respectively. A total of 354 CR-KP strains were detected in five years, mainly from sputum and urine, which accounted for 43.8% (155/354) and 10.7% (38/354) respectively. The departments with higher distribution were surgical ICU and internal medicine ICU, and respectively accounted for 14.4% (51/354) and 10.2% (36/354). The annual detection rate was increasing (χ2 trend=46.176, P＜0.001), and the resistance rate to various drugs was increasing year by year. Except for the difference in the resistance rate to gentamicin which was not statistically significant (P=0.105), and the difference in the resistance rate to tigecycline cannot be counted, other differences in the resistance rates of various antibiotics were statistically significant (P<0.001). WGS technology was used to analyze 19 CR-KP strains. A total of 15 kinds of resistance genes were detected, including β-lactamase resistance gene (blaKPC-2), quinolone resistance gene (qnrS1), fosfomycin resistance gene (fosA), tetracycline resistance gene (tetA), chloramphenicol resistance gene (catA2), and aminoglycoside resistance gene (aadA2b and rmtB). Conclusion The drug resistance of KP and CR-KP strains in this hospital was becoming more and more serious. CR-KP strains carried multiple drug resistance genes, and their drug resistance phenotype were diverse. The use of WGS combined with bioinformatics analysis technology to research the mechanism of bacterial resistance will help the precise implementation of prevention and control measures for resistant bacteria, and will become one of the mainstream technologies for the research of resistant bacteria in the