Background: Acinetobacter baumannii is a significant nosocomial infectious pathogen worldwide. The aim of this study is to characterize the molecular epidemiology of Acinetobacter baumannii isolated from the clinical infection, providing the epidemiology data for prevention and control. Four patients hospitalized in EICU on January 31st, 2014, and then Acinetobacter baumannii infection was observed. Antimicrobial resistance and resistance genes were analyzed by antimicrobial susceptibility testing and PCR sequencing. Pulse field gel electrophoresis (PFGE) and multilocus sequence typing (MLST) were used to analyze these strains’ clonal relatedness.
Results: Sixteen strains were recovered, of which 4 strains were isolated from 4 patients, and others were from environment in EICU, such as air, phone and ventilator. All strains belonged to clonal pulsotype A and ST369. Sixteen antibiotics were used to perform the susceptibility testing, and all strains were extensively drug resistant (XDR) Acinetobacter baumannii, they were only susceptible to tigecycline and polymyxin B, but resistant to others, including carbapenems and aminoglycoside antibiotics. Furthermore, all strains carried blaOXA-23-like carbapenemases gene with ISAba1 insertion sequence in the upstream, aminoglycoside resistance genes ant(3″)-I, 16S rRNA methylase gene armA and disinfectant resistant gene qacE△1, which were mainly responsible for the spread of antimicrobial resistance. Fortunately, enhanced control measures were immediately implemented after this infection, and new strains were no longer detected for consecutive three months.
Conclusions: molecular epidemiology of blaOXA-23-like carbapenemase-producing Acinetobacter baumannii ST369 in EICU of a hospital was characterized. Routine monitoring should be strengthen to prevent outbreaks of this disease.
Acinetobacter baumannii is an important opportunistic pathogen that causes life-threatening nosocomial infection and outbreak in intensive care unit (ICU) . Acinetobacter baumannii infections can increase the mortality rates in neonatal ICUs , moreover, the mortality rate associated with neonatal outbreaks of Acinetobacter baumannii infection could be reaching up to 83% .
Although carbapenems were effective antibiotics for the treatment of Acinetobacter baumannii infection, the efficacy of carbapenems has been reducing with the emergence of multidrug resistant (MDR) strains, especially carbapenem-resistant strains in recent years [2, 4]. The blaOXA−type resistance genes, including blaOXA−23−like and blaOXA−72 genes, are mainly responsible for this event due to their ability of hydrolyzing carbapenems, which further facilitates the spread of Acinetobacter baumannii worldwide and increases the challenge for infection control [5–7]. To date, outbreaks of blaOXA−type-producing MDR Acinetobacter baumannii have been reported throughout the world, and different sequence type (ST) have been observed in various outbreaks throughout Asia [8–10].
According to the results of China Antimicrobial Resistance Surveillance System, the amount of Acinetobacter baumannii clinical isolates have been increasing rapidly from 2012 to 2016 in China with the carbapenem resistance also increasing from 45.8–59.2% (http://www.carss.cn/). Therefore, the aim of the present study was to analyze the epidemiological characterization of Acinetobacter baumannii isolates caused this infection in our hospital.
The susceptibility to 16 antibiotics were perform, all strains were extensively drug resistant (XDR) Acinetobacter baumannii and minimum inhibitory concentrations (MICs) were recorded. Specifically, all 16 strains were resistant to imipenem (≥ 16 µg/ml), meropenem (≥ 16 µg/ml), cefepime (≥ 32 µg/ml), ceftazidime (≥ 32 µg/ml), ceftriaxon (≥ 64 µg/ml), piperacillin (≥ 64 µg/ml), piperacillin/tazobactam (≥ 128/4 µg/ml), ampicillin/sulbactam (≥ 32/16 µg/ml), amikacin (≥ 64 µg/ml), gentamycin (≥ 16 µg/ml), tobramycin (≥ 16 µg/ml), levofloxacin (≥ 16 µg/ml) and ciprofloxacine (≥ 4 µg/ml), but susceptible to tigecycline (≤ 2 µg/ml) and polymyxin B (≤ 0.5 µg/ml). The results of cefoperazone/sulbactam was determined with disk diffusion method, and the inhibition zone diameters were within the range of 10 mm to 14 mm.
The results of PCR and sequencing showed that all strains encoded the acquired blaOXA−23−like carbapenemase-resistant gene except the intrinsic blaOXA−51−like gene, moreover, ISAba1 insertion sequence was found in the upstream of all blaOXA−23−like genes but not blaOXA−51−like genes. Additionally, 16 strains had aminoglycoside resistance gene ant(3’’)-I, 16S rRNA methylase gene armA and disinfectant resistant gene qacE△1. However, other resistance genes, such as blaOXA−type genes, 16S rRNA methylase gene and disinfectant resistant genes were not detected in these Acinetobacter baumannii strains.
As shown in Fig. 1, 16 XDR Acinetobacter baumannii strains were clustered into one single clonal type (A) by PFGE analysis, and they can be further divided into three subtypes (A1, A2, and A3). Among them, 8 strains belonged to subtype A1 (50.0%), 3 strains for subtype A2 (18.8%), and 5 strains for subtype A3 (31.2%). Only ST369 was detected by the MLST.
Acinetobacter baumannii has the capability of long-term survival and can persistently inhabit most inanimate surfaces in hospital, including the air, medical equipment system and sickrooms occupied by patients with Acinetobacter baumannii infection or colonization. It has been reported that sputum aspirator, respirator, air conditioning, mechanical ventilation equipment could be contaminated by Acinetobacter baumannii and function as risk factors for facilitating dissemination [11–13]. In our study, 16 strains of XDR Acinetobacter baumannii ST369 were isolated and characterized from the clinical samples in the tertiary care teaching hospital. Among them, 4 strains were isolated from the patients hospitalized in EICU, and others were isolated from the environment, such as air, computer mouse, ambulance and mechanical ventilator. Recently, some studies demonstrated that tapwater contaminated by Acinetobacter baumannii was the important transmission route . Therefore, these results suggested that we should thoroughly disinfect environment contaminated by Acinetobacter baumannii once this pathogen was observed in hospital.
Acinetobacter baumannii is an important causal agent for nosocomial infection . In the present study, infection caused by blaOXA−23−like harboring Acinetobacter baumannii also occurred in EICU in our hospital. The therapeutic effect of carbapenems is getting worse with the emergence and increasing spread of blaOXA−type-producing Acinetobacter baumannii. The majority of hospital outbreaks reported in the recent years were caused by carbapenem-resistant Acinetobacter baumannii [15–17]. In this study, 16 antibiotics were used to perform the susceptibility testing, but all strains were only susceptible to tigecycline and polymyxin B, which indicated that maybe the early fair use of tigecycline and polymyxin B therapy was a good option. Further study showed that all strains had an intrinsic blaOXA−51−like gene and the acquired blaOXA−23−like carbapenemase gene. The blaOXA−23−like gene was widespread, and, it is the most common carbapenemase gene in Acinetobacter baumannii in China . Although hydrolytic efficiencies of enzymes produced by Acinetobacter baumannii carrying blaOXA−23−like gene are relatively low and the resistant to carbapenem antibiotics are not 100%, the various insertion sequences (ISs) located in the upstream of the blaOXA carbapenemase genes, including ISAba1, ISAba2 and ISAba4, provide promoters for blaOXA carbapenemase genes and facilitate their expressions [19, 20]. In our study, ISAba1 gene was found upstream of blaOXA−23−like gene, which govern the expression of the blaOXA−23−like gene and mediate resistance to carbapenems. Yokoyama et al reported 16S rRNA methylase could result in the resistance to aminoglycosides , only ant(3’’)-I aminoglycoside resistance gene and armA 16S rRNA methylase gene were seen in all strains of the present study, which were mainly responsible for the aminoglycoside resistance.
To prevent and control infections, disinfectants are frequently used in hospitals. But the excessive use of disinfectants can lead to a wide distribution of disinfectant resistance genes. Many disinfectant resistance genes have been confirmed in MDR bacteria, such as qacA/B, qacC, qacD, qacE, qacE△1, qacH, and qacJ genes [22, 23]. Among them, qacE△1 evolved from qacE, and it is the deletion form of qacE gene. It was found that qacE△1 existed in all strains in our study, but not other disinfectant resistance genes. More recently, a total of 51 carbapenem-resistant strains collected between 2014 and 2015 were used to detect the frequency of qac-resistant genes, and result showed that the qacE△1 gene was detected in 96.07% with the highest detection rate . These data indicated that the reduced susceptibility to compound disinfectant of quaternary ammonium salt, such as benzalkonium chloride and chlorhexidine, in carbapenem-resistant Acinetobacter baumannii. Therefore, scientific and effective disinfectants should be considered. The hospital in our study is a referral hospital, a large number of patients are transferred from outpatient clinics to our hospital for care. Therefore, we were unable to identify the index case for this infection. However, infection control measures were immediately implemented, mainly including the isolation of infected patients, environmental disinfection, antimicrobial stewardship, and the improvement of the hand hygiene compliance of medical staffs. Since then, we consecutively monitored the clinical samples from patients, medical personnel and equipment in EICU for two months without detection of the Acinetobacter baumannii isolate.
Currently, many STs were involved in the outbreak of Acinetobacter baumannii, such as Acinetobacter baumannii ST208 was a predominant prevalence clone in Sichuan and Zhejiang provinces of China . ST219 infection was observed in a hospital in Japan . The clonal spread of ST417 occurred in one hospital in Mexico, and some novel STs, such as ST758, ST762 and ST777 were also reported . More recently, some researches revealed that Acinetobacter baumannii ST369 was implicated in the outbreak occurred in hospitals of China and Korean, but it was not the dominant ST [5, 25, 26]. However, ST369 was identified and responsible for this infection in our hospital according to MLST analysis. To our knowledge, this is the first report about infection of Acinetobacter baumannii only caused by ST369. It was unclear whether the virulence of Acinetobacter baumannii ST369 was increasing, but given that it can cause small-scale infection or outbreak in the hospital, we should raise awareness to strengthen epidemiological surveillance and infection control measures to prevent spread.
We characterized the molecular epidemiology of blaOXA−23−like carbapenemase-producing Acinetobacter baumannii ST369 in EICU of a hospital in Shandong. Routine monitoring of Acinetobacter baumannii and strict sanitation and disinfection management should be implemented to prevent outbreaks of this disease.
This study was conducted in Taian City Central Hospital, a regional tertiary teaching hospital with 1906 general ward beds, 12 emergency intensive care unit (EICU) beds and 28 beds ICU. Our hospital provides medical service to residents around the area, with admissions of approximately 80–100 and 70–90 patients in EICU and ICU per month, respectively. Four patients involved in this infection were all in EICU with severe illness, and their sputum specimens were collected for bacterial identification and further study.
Four Acinetobacter baumannii strains were isolated from sputum of 4 patients hospitalized in EICU on January 31st, 2014. Then on February 1st, the environment of EICU contaminated by patients, including the phone, computer mouse, ambulance, defibrillator, mechanical ventilator, stethoscope, electrocardiogram monitor, injection pump, and oxygen valve were detected, and 1 strain was isolated from each site. Other 3 strains were from the air samples collected by the sedimentation method. A total of 16 Acinetobacter baumannii strains were isolated from the patients and environment in EICU.
Bacterial identification and susceptibility testing were performed using WalkAway 96 Plus System (Siemens, Erlangen, Germany). But the sensitivity tests of cefoperazone/sulbactam and tigecycline, polymyxin B were detected by the disk diffusion and E test method, respectively. The susceptibility testing result of tigecycline was interpreted according to Food and Drug Administration (FDA) guideline (http://www.fda.org.uk/sitemap.aspx), and other antimicrobial sensitivity tests results (imipenem, meropenem, ceftazidime, cefepime, ceftriaxone, piperacillin, piperacillin/tazobactam, ampicillin/sulbactam, amikacin, gentamicin, tobramycin, levofloxacin, and ciprofloxacine) were referred to Clinical and Laboratory Standards Institute (CLSI) guidelines (http://clsi.org/standards/). Cefoperazone/sulbactam susceptibility test was explained by cefoperazone criteria made by CLSI (http://clsi.org/standards/).
Bacterial DNA was extracted according to the instructions of the TIANamp Bacteria DNA Kit (TIANGEN, Beijing, China), and carbapenemase genes, including blaOXA−23−like, blaOXA−51−like, blaOXA−58−like , blaOXA−24 , blaOXA−48, blaOXA−50, blaOXA−55, blaOXA−60  and blaOXA−143 , aminoglycoside resistance genes aac(3)-Ⅰ, aac(3)-Ⅱ, aac(3)-Ⅲ, aac(3)-Ⅳ, aac(6′)-Ⅰ, aac(6′)-Ⅱ, aph(3′)-Ⅵ, ant(3″)-Ⅰ, ant(2″)-Ⅰ, and 16S rRNA methylase gene (armA and rmtB) , disinfectant resistant gene quaternary ammonium compound (qac)E△1, qacA/B, qacC, qacG and qacJ were detected by PCR method as previously described . The insertion sequence (IS) ISAba1 and ISAba125 linkages to the blaOXA−type genes were determined according to the Zhou et al  and Lopes et al , respectively. The positive PCR products were sequenced and the sequences were analyzed using the BLAST program (https://blast.ncbi.nlm.nih.gov/Blast.cgi).
Clonal relatedness of 16 Acinetobacter baumannii strains isolated in this study was determined by PFGE method. ApaI restriction enzyme (TaKaRa, Dalian, China) was used for the bacterial pulsotype using the DRII PFGE system (Bio-Rad, Marnes-la-Coquette, France) as previously described , and the PFGE data was analyzed according to the criteria of Tenover et al . The genetic relatedness of the Acinetobacter baumannii isolates was evaluated by the unweighted pair group method using Bionumerics version 6.01 (AppliedMaths, Sint-Martens-Latem, Belgium). The cutoff level of Hierarchial clustering was set to 85% similarity to identify the pulsotype. The alleles and STs of the isolates were performed by the MLST method . The data were upload to the Acinetobacter baumannii MLST database to determine the STs (http://pubmlst.org/abaumannii/).
Meijie Jiang performed the main experiment, analyzed data and wrote the manuscript. Lin Li and Shuang Liu performed the experiment, Zhijun Zhang analyzed data. Ning Li analyzed data and reviewed the manuscript. Shuping Zhao conceived the experiment and reviewed the manuscript.
This work was supported by Natural Science Foundation of Shandong Province (ZR2016HL44).
Consent for publication
Ethics approval and consent to participate
The study was performed in accordance with the approved guidelines of the Ethics Committee of Taian City Central Hospital with written informed consent from all subjects, and it was also approved by the Ethics Committee of Taian City Central Hospital. The informed consents were obtained from the patients for the purpose of publication.
The authors declare that there is no conflict of interest regarding the publication of this article.