Phenotypic and molecular detection of the blaKPC gene in clinical isolates from inpatients at hospitals in São Luis, MA, Brazil

Background Bacteria that produce Klebsiella pneumoniae carbapenemases (KPCs) are resistant to broad-spectrum β-lactam antibiotics. The objective of this study was to phenotypically and genotypically characterize the antibiotic susceptibility to carbapenems of 297 isolates recovered from clinical samples obtained from inpatients at 16 hospitals in São Luis (Maranhão, Brazil). Methods The study was conducted using phenotypic tests and molecular methods, including polymerase chain reaction (PCR), sequencing and enterobacterial repetitive intergenic consensus (ERIC)-PCR. The nonparametric chi-square test of independence was used to evaluate the associations between the bacterial bla KPC gene and the modified Hodge test, and the chi-square adherence test was used to assess the frequency of carbapenemases and their association with the bla KPC gene. Results The most frequently isolated species were Acinetobacter baumannii (n = 128; 43.0%), K. pneumoniae (n = 75; 25.2%), and Pseudomonas aeruginosa (n = 42; 14.1%). Susceptibility assays showed that polymixin B was active against 89.3% of the bacterial isolates. The Acinetobacter spp. and K. pneumoniae strains were susceptible to amikacin and tigecycline, and Pseudomonas spp. were sensitive to gentamicin and amikacin. Among the 297 isolates, 100 (33.7%) were positive for the bla KPC gene, including non-fermentative bacteria (A. baumannii) and Enterobacteriaceae species. Among the isolates positive for the bla KPC gene, K. pneumoniae isolates had the highest positivity rate of 60.0%. The bla KPC gene variants detected included KPC-2, which was found in all isolates belonging to species of the Enterobacteriaceae family. KPC-2 and KPC-3 were observed in A. baumannii isolates. Importantly, the bla KPC gene was also detected in three Raoultella isolates and one isolate of the Pantoea genus. ERIC-PCR patterns showed a high level of genetic diversity among the bacterial isolates; it was capable of distinguishing 34 clones among 100 strains that were positive for bla KPC and were circulating in 11 of the surveyed hospitals. Conclusions The high frequency of the bla KPC gene and the high degree of clonal diversity among microorganisms isolated from patients from different hospitals in São Luis suggest the need to improve the quality of health care to reduce the incidence of infections and the emergence of carbapenem resistance in these bacteria as well as other Gram-negative pathogens.


Results:
The most frequently isolated species were Acinetobacter baumannii (n = 128; 43.0%), K. pneumoniae (n = 75; 25. 2%), and Pseudomonas aeruginosa (n = 42; 14.1%). Susceptibility assays showed that polymixin B was active against 89. 3% of the bacterial isolates. The Acinetobacter spp. and K. pneumoniae strains were susceptible to amikacin and tigecycline, and Pseudomonas spp. were sensitive to gentamicin and amikacin. Among the 297 isolates, 100 (33. 7%) were positive for the bla KPC gene, including non-fermentative bacteria (A. baumannii) and Enterobacteriaceae species. Among the isolates positive for the bla KPC gene, K. pneumoniae isolates had the highest positivity rate of 60.0%. The bla KPC gene variants detected included KPC-2, which was found in all isolates belonging to species of the Enterobacteriaceae family. KPC-2 and KPC-3 were observed in A. baumannii isolates. Importantly, the bla KPC gene was also detected in three Raoultella isolates and one isolate of the Pantoea genus. ERIC-PCR patterns showed a high level of genetic diversity among the bacterial isolates; it was capable of distinguishing 34 clones among 100 strains that were positive for bla KPC and were circulating in 11 of the surveyed hospitals.
(Continued on next page)

Isolation and identification of bacterial strains
Biological specimens were processed in the Clinical Microbiology Section from Laboratory Cedro, where bacterial strains were isolated using MacConkey agar (Difco, Detroit, MI, USA), blood agar (bioMérieux), brain heart infusion (BHI) broth (Difco, Detroit, MI, USA) and CPS chromogenic agar (bioMérieux). Subsequently, the isolates were identified with Gram-negative (GN) cards, and their susceptibility profiles were determined using antimicrobial susceptibility (AST) test cards (N209 cards) by using the Vitek® 2 Compact system (bio-Mérieux, Marcy l'Etoile, France), according to the manufacturer's instructions.
Of the isolates evaluated, 654 showed low sensitivity to carbapenems imipenem, meropenem or ertapenem and were frozen at −80°C in BHI with 20% glycerol. We randomly selected 297 (45.4%) strains for this study. To ensure that there were no repeated bacterial isolates from the same patient, each selected bacterial strain had a code related to a patient's medical record. At the Molecular Biology Laboratory of Microorganisms from University Ceuma, the bacterial strains were checked for purity. Confirmation of biochemical identification of all bacterial isolates was carried out by conventional tests.    [16].

Modified Hodge testing
The confirmation of carbapenemase activity was performed using a MHT as previously described [16]. The K. pneumoniae strain ATCC BAA-1706® was used as a negative control and K. pneumoniae ATCC BAA-1705® was used as a positive control. The presence of a distorted or clover leaf-shaped inhibition zone was interpreted as positive for carbapenemase-producing isolates, as recommended by the document M100-S24 of the Clinical and Laboratory Standards Institute methods [16].

Molecular identification of isolated bacterial strains resistant to carbapenem
Total genomic DNA from the clinical isolates was obtained using a Wizard® Genomic DNA Purification Kit (Promega Corporation, Madison, USA), according to the manufacturer's instructions. The concentration and purity of the extracted DNA were verified using a Nanodrop-ND1000 (Thermo Fisher Scientific, Waltham, MA, USA). PCR assays were performed with primers for the bla KPC family, Uni-KPC-F (5'-ATGTCACTGTATCGCCGTCT-3') and Uni-KPC-R (5'-TTACTGCCCGTTGACGCCC-3'), as previously described [17]. This pair of primers amplifies the complete sequence of the bla KPC gene (882 nucleotides). The PCR was performed in a Mastercycler thermocycler (Eppendorf, Foster City, California, USA) with AccuPrime DNA polymerase mix (1X buffer BII; 2 mM each dNTPs; 1.5 mM MgCl2; 200 mM Tris-HCl pH 8.4 and 1.5 U AccuPrime Taq DNA polymerase) (Invitrogen/Stratagene, La Jolla, CA, USA), 20 pmol of each primer, and 50 ng of genomic DNA to a final volume of 50 μl. The amplification reaction was performed using the following conditions: 94°C for 3 min (initial denaturation) followed by 30 cycles of 94°C for 1 min, 55°C for 1 min and 72°C for 1 min, and a final extension step at 72°C for 5 min.
To verify the efficiency of the PCR and determine the sizes of the amplified DNA fragments, 8 μl of each PCR product was analyzed by electrophoresis on a 1.2% (wt/ vol) agarose gel in Tris-acetate-EDTA buffer (TAE: 40 mM Tris-acetate and 1 mM EDTA). A 100 bp DNA ladder (Promega Corporation, Madison, USA) was included in each run. After electrophoresis, the agarose gels were stained with ethidium bromide (0.5 μg/ml) and photographed under ultraviolet light (UV) at 260 nm.

Purification and sequencing of PCR products
To determine which bla KPC variants were circulating in patients from hospitals in São Luis, Maranhão, the 882-bp amplified PCR products were purified using a commercial Wizard® SV Gel and PCR Clean-Up System Kit (both from Promega Corporation, Madison, USA), according to the manufacturer's instructions. PCR purified products were sequenced by Myleus Biotechnology (Belo Horizonte, Minas Gerais, Brazil) using an ABI 3730XL DNA analyzer (Applied Biosystems, Carlsbad, CA, USA). Amplicons were bidirectionally sequenced at least three times; thus, each PCR product was sequenced six times.
The quality of the sequence electropherograms obtained during the sequencing process was analyzed with ChromasPro (http://www.technelysium.com.au/ chromas.html) software. At least two sequences of each type of KPC were chosen from the database for alignments using MEGA 4.0 [18]. Similarities between the nucleotide sequences obtained were verified using BLASTN (https://blast.ncbi.nlm.nih.gov/Blast.cgi).
To identify the bla KPC gene variants, all sequences were translated into amino acids using the software ExPASy [translate tool (http://web.expasy.org/translate/) in the six reading frames]. The correct translation was chosen based on the available data in GenBank. Deduced amino acid sequences were compared with KPC protein sequences from GenBank using BLASTX. Similarity values for amino acid sequences ranged from 99 to 100%, indicating a highly conserved region.
Identification of Acinetobacter baumannii by detection of the bla OXA-51-like carbapenemase gene PCR assays for bla OXA-51 -like carbapenemase, a gene intrinsic to this species, were performed for all isolates identified by biochemical tests as Acinetobacter. As a positive control, we used a strain of A. baumannii ATCC19606. Amplifications were carried out with specific primers previously described [19], and the PCR products were analysed by electrophoresis on a 2.0% (wt/vol) agarose gel containing 0.5 μg/ml ethidium bromide.

ERIC-PCR amplifications and profile analysis
ERIC-PCR amplifications were used for molecular typing of all isolates analyzed in this study. PCR assays were performed with 100 ng of genomic bacterial DNA, 10 pmol of each ERIC-1R (5'-ATGTAAGCTCCTGGGGATTCAC-3') and ERIC-2 (5'-AAGTAAGTGACTGGGGTGAGCG-3') primer (Invitrogen, Carlsbad, CA, USA), as previously described [20], plus 12.5 μl of Master Mix (Promega Corporation, Madison, USA) and ultra-pure water to bring the final reaction volume to 25 μl. The PCR conditions included an initial denaturation at 94°C for 3 min followed by 35 cycles of 94°C for 30 s, 52°C for 1 min, and 72°C for 2 min, with a final extension at 72°C for 6 min. PCR amplicons were resolved by electrophoresis on a 2% (w/v) agarose gel in acetate-EDTA buffer. A 100 bp DNA ladder (Promega Corporation, Madison, USA) was included in each run. After electrophoresis, the gels were stained with ethidium bromide (0.5 μg/ml), and photographed under UV at 260 nm.
ERIC-PCR profiles were scored by fragment size with the LabImage-1D gel analysis software, Version 3.2 (1D: V6.2. Available in: http://www.kapelanbio.com/). Amplified fragments were scored as absent (0) or present (1) to construct a dendrogram. ERIC-PCR genotype dendrograms were constructed based on the average similarity of the matrix using the unweighted pair group method with arithmetic mean and the DICE similarity coefficient using NTSYS-pc version 2.1, Exeter Software (New York, NY, USA). The nearest neighbor-joining clustering method was used to show relationships between similar groups.

Statistical analysis
To evaluate the associations between the bla KPC gene variants and the modified Hodge test, the nonparametric chi-square test of independence was used. To assess the frequency of carbapenemases and their association with the presence of the bla KPC gene, we used the chi-square adherence test. The data were considered statistically significant at p < 0.05.
The susceptibility testing and minimum inhibitory concentrations of bacterial species indicated that members of the Enterobacteriaceae family had elevated MICs for most antimicrobial agents tested, except for Polymixin. For meropenem and imipenem, a MIC of ≥ 32 μg/ml was detected for all strains of K. pneumoniae. Simultaneous drug resistance to fluoroquinolones, aminoglycosides and other β-lactam antibiotics was observed for 66 bacterial isolates ( Table 2). In addition, low percentages of resistance to amikacin were observed for most strains. K. pneumoniae isolates were susceptible to amikacin and tigecycline; Raoultella planticola isolates showed sensitivity only to amikacin; and R. ornithinolytica showed sensitivity to amikacin, gentamicin, and tigecycline. S. marcescens strains, which are intrinsically resistant to Polymixin B, were highly sensitive to fluoroquinolone (ciprofloxacin) ( Table 2).
The multidrug-resistant (MDR) non-fermentative isolates were susceptible to Polymixin B, with an MIC of ≤ 0.5 μg/ml for all isolates. A. baumannii and P. aeruginosa had MICs of ≥16 μg/ml for meropenem and imipenem. In addition, MICs of ≥ 64 μg/ml for amikacin and 4 μg/ml for tigecycline were found in Acinetobacter spp. isolates, and MICs of ≥ 64 μg/ml for amikacin and ≥ 16 μg/ml for gentamicin were found in Pseudomonas spp. isolates; the antimicrobial resistance profiles of the other isolates showed values above 50% (Table 3).

Evaluation of the modified Hodge test (MHT)
The results of the MHT indicated a statistically significant association (p = 0.0001) with the assessed bacterial species. We observed that the microorganism with the highest positive test results was K. pneumoniae [60 (80%) of 75 isolates], followed by Enterobacter cloacae [7 (36.8%)]. (Table 4). There is no standardization of this test for non-fermenting bacteria (Acinetobacter spp. and Pseudomonas spp.) in the document M100-S24 of the Clinical and Laboratory Standards Institute methods [16]. Thus, for these microorganisms, the MHT was not performed.
Regarding the association between the modified Hodge test and the presence of the bla KPC gene in species of the family Enterobacteriaceae, we observed positive results in both tests for 64 isolates (Table 4). Of importance, the concordance rate between the two tests was of 81% (64 out of 79 that were positive for the bla KPC gene. Of the non-glucose fermenting bacteria, only A. baumannii isolates carried the bla KPC gene, and the largest number of isolates was recovered from three public hospitals in São Luis, MA: H02, 50.0% (5/10); H01, 19.0% (7/37) and H10, 18.1% (8/44).
The phenotypic profiles obtained by different methods (Vitek® 2, E-Test®, and the modified Hodge test) for all isolates that showed resistance to carbapenem antibiotics are shown in Table 5. Importantly, a high percentage of antimicrobial-resistant microorganisms harbored the bla KPC gene.
Statistical analyses using chi-square adherence tests revealed that the presence of the bla KPC gene was significantly correlated with the distribution of the isolates at the site of infection (p < 0.0001). The highest frequency of this gene in the clinical isolates was 46.9% (30/64) in isolates from urine, followed by 39.5% (17/43) from blood, 39.4% (13/33) in nasal swabs, 33.3% (5/15) in rectal swabs and 21.4% in tracheal secretions (24/112). Specimens obtained from other anatomical sites showed low detection frequencies, which were probably due to the lack of quantitative uniformity among the materials collected during the study. Additionally, chi-square adherence tests indicated a correlation between the frequency of hospital accommodation and the presence of the bla KPC gene (p < 0.0001).
The distribution of KPC-positive microorganisms by hospital sector revealed that 56 (32.7%) isolates were positive for the bla KPC gene in 171 samples derived from the intensive care units (ICUs). By contrast, KPCproducing bacteria were found in several other units, except the pediatric ICU and the ward and labor rooms. The bla KPC gene was also detected in bacterial isolates from patients admitted to sectors, such as the intermediate care unit, with four (20.0%) of the 20 cases from H10 and three (60.0%) of the five cases from H05, a hospital emergency room.

Sequencing of the bla KPC gene and amino acid sequence analysis
The bla KPC gene was detected in 100 (33.7%) resistant bacterial isolates, and similarities among the sequences were assessed by performing alignments with nucleotide sequences from different bacteria. All sequences were deposited in GenBank with accession numbers KU695917 to KU696016. The similarity indices ranged from 99 to 100%, indicating that the sequenced region was highly conserved.
Nucleotide and amino acid differences between KPC enzymes (bla KPC variants) were determined based on a study [6]. Amino acid analysis showed that type 2 (KPC-2) was the predominant variant for species from the Enterobacteriaceae family, including R. planticola, R. ornithinolytica and Pantoea spp. In addition, the KPC-2 and KPC-3 variants were predominant in A. baumannii strains.

Detection of genes encoding OXA carbapenemase by Polymerase chain reaction
The primers specific for amplifying of the fragment of 353 bp for bla OXA-51-like gene intrinsic to A. baumannii species showed that of the 129 isolates 128 harboured this gene. Only the isolate number 159, identified as A.
ursingii not amplified with this pair of primer. Some representative results of amplification are shown in Fig. 1.
Among isolates positive and negative for the bla KPC gene, ERIC-PCR amplicons ranged between 80 and 900 bp for nine isolates identified as Enterobacter aerogenes. Two different profiles were observed for three strains negative for this gene. For the SL147 and SL193 isolates, bands of 80 and 160 bp were observed, and for SL80, the profile was 100, 280, 310, 450, 550, (Fig. 3).
For five isolates identified as Serratia marcescens, ERIC-PCR fragments ranged between 80 and 310 bp. Isolates SL55, SL56 and SL209, which were negative for the bla KPC gene, exhibited a single profile of 80 and 160 bp. The isolates SL04 and SL111, positive for bla KPC , showed bands of 80, 160, and 310 bp. These strains were isolated from patients treated at H05, a public hospital emergency room (Fig. 3).
Two strains identified as E. coli showed distinct ERIC-PCR profiles; bla KPC -positive SL141 was recovered from a patient at the H09 hospital, and showed fragments of 80, 160, 280, 350, and 550 bp, and the bla KPC -negative SL322 isolate had fragments of 160 and 250 bp (Fig. 3).
ERIC-PCR patterns of 80 and 160 bp were observed for three different isolates as follows: for those identified as Raoultella planticola, which included SL36 obtained from H10 and SL196 from H01, and for SL110, which was identified as Raoultella ornithinolytica and recovered from H10. The bla KPC gene was detected in all these isolates.
For three Proteus mirabilis isolates, SL26, SL310 and SL313, a unique profile of 160 bp was observed. For the Aeromonas salmonicida strain SL210, fragments of 80 and 160 bp were found. All these isolates were negative for the bla KPC gene (Fig. 3).
The distribution of different bla KPC -positive species in hospitals was as follows: for H01, K. pneumoniae, Enterobacter cloacae and Raoultella planticola; for H02, only Enterobacter cloacae; for H05, K. pneumoniae, Enterobacter aerogenes and Serratia marcescens; for H07, only Enterobacter aerogenes; for H08, only K. pneumoniae; for H09, K. pneumoniae, Enterobacter cloacae, Enterobacter aerogenes and E. coli; for H10, K. pneumoniae, Raoultella planticola and Raoultella ornithinolytica; for H11, only Enterobacter cloacae; for H12, K. pneumoniae and Enterobacter cloacae; and for H15, only Pantoea spp. It is important to note that H09, a public emergency room, had the highest number of different bacterial species carrying the bla KPC gene.
In summary, the results of ERIC-PCR for 120 isolates belonging to the Enterobacteriaceae family showed that 79 (65.8%) bacterial strains carried the bla KPC gene. Among these isolates, 26 (33.0%) clones were genetically distinct and were circulating in most [10/16 (62.6%)] of the hospitals surveyed.
Among the 297 bacterial isolates, 177 (59.6%) were non-fermenting Gram-negative bacilli from the genera Acinetobacter (n = 129) and Pseudomonas (n = 48). The profiles produced by ERIC-PCR with DNA of these microorganisms were quite heterogeneous, with fragments ranging from 70 to 1500 bp. Of the 177 isolates, 128 (72.3%) were identified as A. baumannii and one A. ursingii (SL159). Of the 128 isolates of A. baumannii, the majority (107, 83.5%) was negative for the bla KPC gene, and this gene was found in only 21 (16.5%) strains.
ERIC-PCR profiles for Pseudomonas spp.

Discussion
This study showed the spread of multi-drug-resistant (MDR) bacterial strains isolated from clinical samples from patients in different healthcare facilities (public and private hospitals in northeastern Brazil) who showed resistance or decreased carbapenem antimicrobial susceptibility to imipenem, meropenem or ertapenem. The increasing prevalence of the clinical MDR-KPC phenotype has been associated with higher mortality rates, thereby posing a considerable threat to public health [21].
Our analysis indicated that A. baumannii, K. pneumoniae and P. aeruginosa species were the most frequently observed in clinical samples, with carbapenem resistance Fig. 4 Dendrogram showing the genetic relatedness of 129 Acinetobacter spp. isolates. Cluster analysis of enterobacterial repetitive intergenic consensus PCR (ERIC-PCR) fingerprinting patterns using the DICE similarity coefficient and the UPGMA cluster method. The scale indicates the percentage of genetic similarity frequencies of 100, 84 and 100%, respectively. The high percentages of bacterial resistance to carbapenems and the cross-resistance observed to various antimicrobials are of concern in clinical medicine, especially in intensive care units [22]. It is possible that the genes encoding carbapenemases are located on genetic elements, such as integrons and transposons, in association with conjugative plasmids typically carrying genes for resistance to other antimicrobials [23]. Recently, the bla KPC-3 , bla VIM-1 , bla SHV-12, bla OXA-9 and bla CMY-2 genes were detected in a unique clinical multidrug-resistant E. coli isolate (clone ST448) carrying a Tn4401 transposon associated with an IncFII plasmid [24]. Moreover, bla KPC variant genes are typically located in Tn4401 transposable elements and their isoforms, supporting their dispersion [25].
We found that several isolates of the Enterobacteriaceae family were resistant to most of the antimicrobials tested, except for Polymixin and amikacin. We observed that 10.7% of the K. pneumoniae strains were resistant to Polymixin B, with an MIC of 64 μg/ml. The other species were sensitive to Polymixin B. This antimicrobial drug is a good treatment option for A. baumannii, E. aerogenes, E. cloacae and E. coli, followed by amikacin, even in the presence of varying percentages of resistance [26].
This study has thus far been categorical in stating that the antimicrobial agent amikacin is a therapeutic option for the treatment of infections. Nevertheless, it has emerged, alone or in combination, as a major drug of interest for treating isolates that possess intrinsic resistance to Polymixin [26]. The combination of Polymixin with tigecycline has recently been described as a suitable option for the treatment of infections with MDR Gramnegative pathogens extended spectrum beta lactamase (ESBL-and carbapenemase-producing strains) [15]. However, increased expression of the efflux system in A. baumannii isolates has been correlated with a low level of susceptibility to tigecycline [27].
Molecular approaches aimed at detecting strains harboring carbapenemase genes are highly sensitive and efficient for confirmation of cases [28,29]. Conversely, there are no methodologies to routinely conduct clinical and laboratory diagnoses [30]. In this respect, the presence of the bla KPC gene must be confirmed by molecular biology techniques to define the production of KPCs [28]. The production of KPCs determined by the phenotypic modified Hodge test (MHT) showed a strong association with the presence of the bla KPC gene by PCR (p < 0.0001). However, analyses showed that MHT positivity does not confirm the presence of KPCs, only the involvement of some KPCs that may or may not indicate a resistant phenotype. It is important to note that false positive results can occur when the MHT is used to detect carbapenemase in ESBL-producing isolates [7]. However, only one isolate was bla KPC gene-positive and MHT-negative. Similar to our findings, patterns of antimicrobial resistance and prevalence of the bla KPC gene were determined in Enterobacteriaceae species isolated from a hospital in India. Among 46 strains resistant to carbapenem antimicrobials, 38 (82.6%) tested positive by MHT, but the bla KPC gene was detected in only 31 (67.4%) [31].
The emergence of antibiotic-resistant organisms is a major public health concern, particularly in hospitals and other health care settings, and represents a serious challenge for surveillance systems [32,33]. Previous studies have shown the spread of KPC-type 2 and KPCtype 3 strains and have also identified new variants, such as bla   [7,34,35]. Mortality rates of 18-72% have been reported in patients infected with bacteria carrying the bla KPC gene in recent years [33,36,37], including those treated with combined therapy (tigecycline-gentamicin or tigecycline-colistin) or monotherapy using colistin or tigecycline for Enterobacteriaceae infections [15].
In Brazil, KPC-2 has been endemic since 2006 [10], and our surveillance systems have failed to detect KPC, its variants and their incidence rates. Moreover, we do not know the true prevalence of these carbapenemases and their impacts on hospital mortality rates. However, the few existing studies have suggested that the mortality rate associated with KPC infection in intensive care units in Brazil can reach 18% [38]. Recently, the first description of the K. pneumoniae clone ST258 associated with an outbreak in Ribeirão Preto city, Brazil was reported. This clone carried the transposon Tn4401a, which likely contributed to its spread [39].
Previous studies have indicated that the prevalence of the bla KPC gene in A. baumannii is variable and not always indicative of β-lactam antimicrobial resistance with this type of KPC [40]. Other reports have indicated that high levels of resistance to carbapenems are strongly related to the presence of the bla OXA-23 and bla OXA-24 genes [41]. Furthermore, this study showed that 21 (16.4%) Acinetobacter strains harbored the bla KPC gene among the 128 bacterial strains evaluated. A more recent study found that despite the apparent spread of the bla KPC gene in A. baumannii isolates causing nosocomial outbreaks, additional types of carbapenemases are involved in the activity of this bacterial species and are associated (or not) with other non-enzymatic mechanisms, including changes in outer membrane protein (OMPs) efflux pumps and penicillin-binding proteins (PBP) [42].
Moreover, it is important to note that all isolates previously identified by biochemical tests as Acinetobacter spp were confirmed by PCR for identification at the species-level. Among the various genotypic methods used to identify isolates of the genus Acinetobacter, the PCR of the bla oxa51 gene has been proposed as a method to identify A. baumannii [19].
In Enterobacteriaceae carrying the bla KPC gene, a high level of sensitivity to Polymixin was observed; however, it is worrisome that some strains showing resistance to this antimicrobial agent have emerged, including pathogens with intrinsic resistance. Previous reports have described the presence of KPC-3-producing K. pneumoniae that are resistant to colistin (Polymixin E) in hospitals in Sicily, Italy [21,43]. In addition, high percentages (83.1%) of ciprofloxacin resistance have been observed in K. pneumoniae strains carrying the bla KPC gene, indicating that few antimicrobial options may be available because these microorganisms accumulate different mechanisms of resistance.
Molecular genotyping using ERIC-PCR fingerprinting showed heterogeneous profiles that could differentiate bacterial species that contained these repetitive elements. ERIC-PCR patterns showed 26 distinct clones for eight bacterial species of the family Enterobacteriaceae and eight different clones for strains of Acinetobacter baumannii that were resistant to carbapenems and that were circulating in 11/16 (68.7%) of the hospitals surveyed.