Identification of multi-drug resistant genes in P. aeruginosa isolates from patients under mechanical ventilation and respiratory support in an intensive care unit

Purpose: To determine multi-drug resistant (MDR) and metallo β-lactamase (MBL)-resistant genes from Pseudomonas aeruginosa isolated from intensive care unit (ICU) patients under mechanical ventilation and respiratory support. Methods: P. aeruginosa was isolated from 387 purulent tracheobronchial secretions collected from ICU patients who were intubated and mechanically ventilated for at least 48 h. Antibiotic resistance was determined by minimum inhibitory concentration (MIC) assay while MDR genes, viz, blaTEM, blaOXA, blaVIM, blaCTX-M-15 were determined by polymerase chain reaction (PCR). Results: A total of 144 (37.2 %) P. aeruginosa were isolated from the purulent tracheobronchial secretions. A majority of the isolates (51.4 %) were resistant to gentamicin. Meropenem-gentamicin was the predominant (35.4 %) resistant combination. Out of the 144 isolates, 102 (70.8 %) were positive for blaTEM gene, 51 (35.4 %) for were positive for blaOXA gene, 22 (15.3 %) were positive for blaVIM gene, while 19 (13.2 %) were positive for blaCTX-M gene. Conclusion: The high prevalence of MDR P. aeruginosa indicates the need for continued monitoring of MDR P. aeruginosa especially in ICU patients who are under mechanical respiratory support.


INTRODUCTION
Mechanical ventilation is one of the fundamental elements of therapy for patients admitted in ICUs [1]. The ability of Pseudomonas aeruginosa to colonize in-patients is highly critical within ICUs. Mechanical ventilators used in the ICUs are associated with a higher risk of respiratory tract infections leading to ventilator-associated respiratory infections (VARI) and ventilatorassociated pneumonia (VAP) [1]. Pseudomonas aeruginosa is one of the most common organisms associated with nosocomial pneumonia [2]. According to the United States National Healthcare Safety Network (NHSN) data, P. aeruginosa was the second most common pathogen isolated from patients with VAP [2]. The National Nosocomial Infection Surveillance System (NNISS) in China reported that P. aeruginosa is the predominant pathogen isolated from the lower respiratory tract, accounting for 12. 8 [3,4].
A combination of β-lactams either with antipseudomonal quinolone or an aminoglycoside is the primary treatment choice for P. aeruginosa infections. However, increasing resistance towards various antibiotics has led to severe lifethreatening conditions which pose a challenge in the treatment of P. aeruginosa infections. Pseudomonas aeruginosa exhibits resistance towards multiple antibiotics leading to the development of multidrug-resistant (MDR) strains. The increase in MDR strains is a global problem. Multi-drug resistant P. aeruginosa strains are associated with increased morbidity and mortality, prolonged hospital stay, and higher costs of treatment [5]. Infections caused by resistant P. aeruginosa are often associated with excessive use of antibiotics, and invasive procedures including hemodialysis, tracheostomy and mechanical ventilation catheter [6].

Carbapenems
were considered effective antibiotics against P. aeruginosa infections. However, due to extensive use of these antibiotics, the resistance mechanism spread across hospitals. Multi-drug resistance especially the MBL-producing strains has been commonly reported in all regions of the globe. The most common MBL genes are imipenem (IMP), verona integron-encoded metallo-β-lactamase (VIM), Sao Paulo MBL (SPM), German imepenemase (GIM), and the recently reported Seoul imepenemase (SIM) families.
The true prevalence of MDR P. aeruginosa has not been well established mainly due to the ambiguity existing in the definition of MDR [5]. Different definitions of MDR have been used in the literature [7]. Majority of the published literature define MDR as strains which possess resistance to a minimum of three different antibiotic classes, mainly the anti-pseudomonal penicillins, aminoglycosides, fluoroquinolones, cephalosporins and carbapenems [5]. However, a group of experts from European Centre for Disease Prevention and Control and the Centers for Disease Control and Prevention, defined MDR as acquired non-susceptibility to at least one agent in three or more antimicrobial categories [7].
Strains are categorized as MDR strains based on the definition published by Magiorakos et al [7]. In addition, due to geographical variations and participating centers, the SENTRY antimicrobial surveillance program was developed to track antimicrobial resistance trends nationally and internationally, fail to report the true prevalence of MDR P. aeruginosa [8]. With the lack of safe therapeutic antibiotics against MDR and pandrug resistant strains, continued presence of such strains would pose serious challenges in infection control management [9]. Hence, it is extremely vital to understand the distribution of pathogens and antibiotic resistance patterns of pneumonia so as to achieve optimal antibiotic therapy. Although several studies have reported MDR P. aeruginosa, [5] limited data are available from China. The present study determined MDR resistant and MBL resistant genes from P. aeruginosa isolated from ICU patients who were under mechanical ventilation and respiratory support.

Sample collection
A total of 387 purulent tracheobronchial secretions were collected from 387 non-repetitive patients who were admitted in ICU for various disabilities, and who were intubated and mechanically ventilated for at least 48 h between July 2015 and June 2017. Written informed consents were obtained from all the patients or their legal representatives after duly explaining the nature of the study. The study was approved by the institutional review board of The First Affiliated Hospital of Xi'an Jiaotong University, Xi'an, China (approval no. TKG155661) and was carried out as per WHO guidelines [10]. Samples were collected in a sterile container and sent immediately for microbial analysis. After collection, the samples were liquefied by the addition of 1 % N-acetyl cysteine (equal volume v/v) and homogenized by vortexing (3000 rpm for one min). Then, 0.1 ml of the sample was diluted (1:100 with 9.9 ml of sterile physiological saline) and it was subjected to microbial culture [11].

Isolation and identification of P. aeruginosa
The samples were sub-cultured onto MacConkey agar at 37 ⁰C for 18 -24 h. The non-fermenting, irregular, green-to-brown, catalase-and oxidasepositive colonies showing typical colony morphology of P. aeruginosa were subjected to identification using Vitek® 2 microbial identification system (Vitek® 2 software, version R02.03, Advanced Expert System software, version R02.00N). The isolates were stored in nutrient broth (with 20 % glycerol) at -40 ⁰C until used for further analysis.

Minimum inhibitory concentration (MIC) assay
The MICs of various antibiotics i.e. gentamicin, amikacin, piperacillin-tazobactam, ciprofloxacin, cefepime, ceftazidime, polymyxin B, imipenem and meropenem (Sigma-Aldrich, USA) against P. aeruginosa were determined. The inoculum was prepared by direct suspension of colonies grown overnight on nutrient agar in 0.85 % saline, with turbidity adjusted to 0.5 McFarland's standard for inoculation. The MIC assay (micro broth dilution method) was performed at concentrations of each antibiotic ranging from 0.03 to 128 µg/mL; using Muller-Hinton broth (MHB) as described in Clinical Laboratory Standard Institute (CLSI) guidelines [12]. In essence, 10 µL of culture was inoculated into the various concentrations of MHB and incubated at 37 ⁰C for 24 h. After incubation, MIC was determined visually by the highest concentration showing the absence of growth.

Multiplex PCR
DNA was extracted from pure cultures by alkali lysis method and stored at -20 ⁰C until used for PCR. Multiplex PCR was used to determine extended-spectrum β-lactamases (ESBL)-MBL resistant genes such as bla TEM , bla OXA , bla CTX-M- 15 , bla VIM [9]. The PCR was performed using a 50-μL master mix containing 5 μL of template DNA, 5 pmoL of each primer (Table 1), dNTPs (2 mM), 3 units of Taq polymerase enzyme, 5 μL of 10x reaction buffer, and molecular grade PCR water in a total volume of 50 μL. PCR was performed using the thermocycler (Applied Biosystems, Verti Thermal Cycler, Thermo Fisher Scientific). The PCR cycling conditions were: initial denaturation at 94°C for 5 min followed by 35 cycles at 94 °C for 30 s; 56 °C for 30 s, extension at 72 °C for 1.5 min, and final extension at 72 °C for 7 min. After PCR, amplicons were resolved in 1.2 % agarose gel electrophoresis.

Statistical analysis
Continuous and categorical variables are presented as mean/ ranges and numbers/ percentages, respectively. ANOVA and chisquare tests were performed to determine the statistical significance using MINITAB statistical software (Minitab version 13.1; Minitab Inc, PA, USA). Values of p < 0.05 were considered statistically significant.  Only one (0.7 %) isolate was positive for all four genes tested (Table 4). A comparison of antibiotic resistance and the presence of genes are shown in Table 5. Out of the 58 MDR isolates, 39 (67.2 %) were positive for bla TEM , 32 (55.2 %) were positive for bla OXA , 16 (27.6 %) were positive for bla VIM , while 11 (19 %) were positive for bla CTX-M genes. All the resistant isolates tested positive for any of the four tested genes. A total of 28 isolates which showed intermediate resistance towards the tested antibiotics were positive for bla TEM (12), bla OXA (8), bla CTX-M (5) and bla VIM (3) genes.

DISCUSSION
Rapid and accurate identification of infectious agents is crucial for the initiation of appropriate therapy that has important consequences in patient's clinical outcome. It has been reported that rapid identification of infectious agents leads to a substantial reduction in the time taken to initiate effective antimicrobial therapy, and also decreases hospital cost and mortality [13].
Several studies have reported that Vitek® 2 microbial identification system correctly identifies bacterial strains, with accuracy ranging from 85.3 -100 % [14 -16]. In this study, rapid Vitek® 2 microbial identification system was used successfully to identify the isolates up to the species level.
The prevalence of P. aeruginosa was 37.2 %. A meta-analysis by Ding et al. [4] on 28 studies in China reported an overall 19.4 % prevalence of P. aeruginosa in VAP, which is much lower than that obtained in the present study (37.2 %). Similarly, another systematic review from Mainland China reported that 20.6 % of isolates from ICU patients with VAP were P. aeruginosa [17]. In other Asian countries, the prevalence of P. aeruginosa was much lower than that reported in this study, as indicated by these results: Thailand (18 %), Malaysia (18 %) and the Philippines (19 %) [18].
However, a study in Brazil reported a higher percentage (51.9 %) of P. aeruginosa isolates from ICU patients, when compared with the prevalence of 37.2 % obtained in the present study. Studies have also reported that mechanical ventilator is a risk factor for P. aeruginosa infections [19]. A declining trend in the prevalence of P. aeruginosa isolates associated with VAP has been reported [4]. However, the present study obtained much higher percentage of P. aeruginosa than those reported in other Asian countries. This could possibly be due to the differences in periods of the studies. The Asian studies were conducted between 1999 and 2012. Thus, there could be fluctuations in the prevalence of P. aeruginosa over time, which may account for the differences in the prevalence figures. This variation in the prevalence implies that it is risky to adopt an attitude of complacence by continuing to refer to declining trend in the prevalence of P. aeruginosa in the Asian region based on previous findings. There should be a continuous monitoring system focused on P. aeruginosa infections in the clinical departments of hospitals, especially in the ICUs.
The inherent resistance of P. aeruginosa to a broad range of antibiotics, and its ability to develop MDR and acquired resistance through chromosomal mutations pose serious challenges during treatment [8,16]. Given the increasing resistance towards various antibiotics, MDR is expected to become more prevalent in several hospitals. In this study, 51.4 % (MIC range: 0.5 -64 µg/mL) of the isolates were found to be resistant to gentamicin, while only 24.3 % (MIC range: 0.06 ≥ 128 µg/mL) of isolates were resistant to amikacin. Similarly, a meta-analysis in China reported gentamicin resistance in 51.1 % and amikacin resistance in 22.5% of the isolates in VAP cases [4].
A comparable resistance to imipenem (41.1 %) and meropenem (38.9 %) was reported in China [4]. The study also reported comparable resistance values for ceftazidime (40.3 %), cefepime (38.5%) and piperacillin/tazobactam (38.9 %), while a study in India reported much lower resistance (25 %) for meropenem and comparable resistance (25 %) for piperacillin/tazobactam [9]. In contrast, a study in Brazil reported much higher degrees of resistance (64.8 %) to meropenem and imipenem in P. aeruginosa isolated from ICU patients [19].Although the majority of the isolates were resistant to gentamicin, antibiotic resistance did not differ significantly among the isolates (p = 1.00). In the present study, 40.3 % of isolates were MDR strains, which is higher than that reported in Brazil (37 %) [19]. All the identified MDR isolates were resistant to gentamicin.
The ESBL enzymes encoded by TEM, CTX-M, GES, PER, SHV, VEB and IBC family of genes were prevalent among P. aeruginosa. In addition, OXA-type ESBL enzymes have been reported. The bla TEM gene was predominantly carried by P. aeruginosa, being narrow spectrum βlactamases derived from mutation of single nucleotide from the TEM-1 and TEM-2, leading to TEM-3 and other variants which confer resistance to penicillin group of antibiotics. In this study significantly higher number of isolates (70.8 %) was positive for bla TEM gene. This is similar to current findings in a study in India which reported that bla TEM was the predominant gene present among the P. aeruginosa isolates (72.5 %) [9]. In that study [9], the bla OXA (33.5 %) and bla VIM (11.5%) genes were comparable. However, the bla CTX-M gene (5 %) was lower than that the value obtained in the present study. In another study, 93 % of the isolates were positive for bla OXA gene which was much higher than that reported in the present study [16]. In P. aeruginosa, the bla OXA gene is considered a naturally occurring gene.
The high prevalence of this gene raises an alarm because this may lead to potential horizontal gene transfer where other co-inhabiting bacteria species may possess Class D β-lactamases. In one study [25], 19.6 % of P. aeruginosa isolates were positive for bla CTX-M gene, which is higher than that seen in the present this study. Although the presence of bla TEM gene was significantly higher, the presence of resistance genes did not differ significantly among the isolates which were resistant to various antibiotics.
It has been reported that if an isolate was positive for bla CTX-M gene but did not possess any of the other beta-lactamase genes, then the patient from whom the isolate was obtained can be treated using aminoglycosides, quinolones, carbapenems, and fourth and fifth generation cephalosporins, to the exclusion of cefotaxime and other third-generation cephalosporins [9]. In this study, a total of 7 (4.9 %) isolates which were positive for bla CTX-M gene, did not amplify any of the other tested genes. The study [9] also reported that if an isolate showed presence of bla VIM/DIM , bla TEM and bla CTX-M genes, the patient should be treated with colistin, polymyxin, aztreonam, or a combination of drugs [9]. In the present study, 2 (1.4 %) isolates were found to possess this combination of genes, which suggests that these patients should be treated with polymyxin, aztreonam, or a combination of drugs for better patient management.