JAC Advance Access published online on November 13, 2007
Journal of Antimicrobial Chemotherapy, doi:10.1093/jac/dkm443
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Ongoing epidemic of blaVIM-1-positive Klebsiella pneumoniae in Athens, Greece: a prospective survey
1 First Department of Propaedeutic Medicine, Medical School, University of Athens, Athens, Greece 2 Department of Microbiology, Medical School, University of Athens, Athens, Greece 3 Department of Microbiology, Laikon General Hospital, Athens, Greece 4 Department of Microbiology, Evangelismos General Hospital, Athens, Greece 5 Department of Microbiology, Hippokration General Hospital, Athens, Greece
* Corresponding author. Tel: +30-210-7462152; Fax: +30-210-7462010; E-mail: ltzouvel{at}cc.uoa.gr
Received 25 July 2007; returned 29 August 2007; revised 15 October 2007; accepted 19 October 2007
| Abstract |
|---|
|
|
|---|
Objectives: To determine the current frequency and study the characteristics of VIM-1-producing Klebsiella pneumoniae isolates from bloodstream infections in Greek hospitals.
Methods: All blood isolates of K. pneumoniae were prospectively collected during 2004–06 in three teaching hospitals located in Athens. MICs of antibiotics were determined by the Etest. Extended-spectrum- (ESBL) and metallo-ß-lactamase (MBL) production was examined by clavulanate- and EDTA-based techniques, respectively. Isolates were typed by PFGE of XbaI-digested genomic DNA. Detection of blaVIM-1 and mapping of the VIM-1-encoding integrons were performed by PCR and sequencing.ß-Lactamase activities were analysed by IEF and imipenem hydrolysis was assessed by spectrophotometry. VIM-1-encoding plasmids were transferred to Escherichia coli by conjugation and transformation and characterized by Inc/rep typing and RFLP.
Results: Sixty-seven (37.6%) of 178 K. pneumoniae blood isolates were blaVIM-1-positive (VPKP); 77.8% of these were from ICUs. All VPKP isolates were multidrug-resistant. The MICs of carbapenems for VPKP varied from the susceptible range to high-level resistance overlapping with those of MBL-negative isolates. The EDTA-imipenem synergy methods had reduced sensitivity in detecting VPKP isolates when the MICs were in the susceptible range. ESBL production was common among VPKP isolates (n = 45, 67.2%) as indicated by resistance to aztreonam and confirmed by a clavulanate-based double-disc synergy test. The responsible ESBL was always an SHV-5-type enzyme as indicated by IEF. PFGE identified eight clusters (A–H) of VPKP isolates with related (>80%) patterns, as well as four unique types. Both inter-hospital spread of several clones and genotypic similarities among susceptible, ESBL-positive and VPKP isolates were also observed. Location of blaVIM-1 and expression of VIM-1 were studied in 12 isolates representing the eight PFGE clusters. In all isolates, blaVIM-1 was part of a class 1 integron that also carried aacA4, dhfrI, aadA and sulI. In eight isolates (clusters C, D, G and H), the blaVIM-1 integron was located in transferable IncN plasmids. A cluster F isolate carried a VIM-1-encoding, self-transferable plasmid that was not typeable by Inc/rep typing. VIM-1-encodingreplicons were not identified in three isolates (PFGE clusters A, B and E). VPKP isolates exhibited differences in imipenem-hydrolysing activities which, however, were not correlated with the respective carbapenem MICs.
Conclusions: A multiclonal epidemic of blaVIM-1-carrying K. pneumoniae is under way in the majorhospitals in Greece. Microorganisms producing both VIM-1 and SHV-5 constitute the prevalent multidrug-resistant population of K. pneumoniae in this setting.
Key Words: K. pneumoniae , carbapenem resistance , metallo-ß-lactamases
| Introduction |
|---|
|
|
|---|
Carbapenems are considered as a reliable choice for the treatment of infections caused by Klebsiella pneumoniae producing various types of extended-spectrum ß-lactamases (ESBLs) and cephalosporinases. A recent development, however, is the emergence of strains producing metallo-ß-lactamases (MBLs) hydrolysing carbapenems. The main foci of MBL-producing Gram-negative microorganisms, including K. pneumoniae, seem to be in Southern Europe and the Far East.1 blaVIM-1-positive K. pneumoniae (VPKP) strains have been isolated with increasing frequency in Greek hospitals since 20002–5 (see also www.rivm.nl/earss/database/). Yet, the extent of their spread is unclear. Previous studies covered limited time periods and were based on phenotypic criteria, mainly decreased susceptibility to imipenem and/or results of various imipenem-EDTA synergy methods. We present here the results of a 1 year prospective study aiming at (i) the determination of the actual prevalence of VPKP in three of the major hospitals in Athens, (ii) the genotypic and phenotypic characterization of the VPKP strains, and (iii) the evaluation of MBL detection methods as applied to this bacterial population.
| Materials and methods |
|---|
|
|
|---|
Bacterial isolates
A total of 178 consecutive blood isolates of K. pneumoniae were collected from an equal number of patients in three hospitals in Athens, I (500 beds), II (1000 beds), and III (400 beds), during February 2004–March 2006 (hospital I) and from February 2005 to March 2006 in hospitals II and III. Species identification was confirmed with the API 20E (bioMérieux, Marcy lÉtoile, France).
Susceptibility and imipenem-EDTA synergy testing
MICs of ß-lactams were determined by the Etest (AB Biodisk, Solna, Sweden). Susceptibility to other antibiotics was assessed by a disc diffusion method. All isolates were examined blind by two MBL screening tests based on imipenem-EDTA synergy: (i) the double disc test (MBL-DDT);6 and (ii) the combined disc test (MBL-CDT).7 In the MBL-DDT, a clear and reproducible synergy image was interpreted as a positive result. Isolates exhibiting an increase of
5 mm in the inhibition zone of imipenem in the MBL-CDT were characterized as MBL-positive. Selected isolates with varying imipenem MICs were also examined by the MBL-Etest (AB Biodisk). Isolates were screened for ESBL production by the double-disc synergy test (DDST) using amoxicillin/clavulanate combined with ceftazidime, cefotaxime, aztreonam and cefepime discs.
ß-Lactamase-containing extracts were prepared by ultrasonic treatment of cell suspensions. IEF of ß-lactamases was performed in polyacrylamide gels containing ampholines (pH range 3.5–9.5) using nitrocefin as a chromogenic substrate. To facilitate the detection of MBLs, a ZnCl2 solution (1 mM) was used to overlay the gels before application of nitrocefin. Imipenem hydrolytic activity of the ß-lactamase extracts was determined by spectrophotometry as described previously8 and expressed as units (U) (1 U was the amount of enzyme hydrolysing 1 nmol of substrate per min per mg of protein).
Transfer of resistance and plasmid characterization
Transfer of MBL-encoding plasmids to susceptible Escherichia coli recipient cells was carried out by either conjugation in mixed broth cultures2 or transformation. Transferable plasmids were purified using a Nucleobond BAC100 kit (Macherey-Nagel, Duren, Germany). Replicon typing was performed by a PCR-based method.9 For comparison of RFLPs, plasmids were digested with PstI and fragments were separated by electrophoresis in 0.8% agarose gels.
Detection of blaVIM-1 and characterization of integrons
Total bacterial DNA was extracted with a QIAamp DNA mini kit (Qiagen GmbH, Hilden, Germany) and used as template in PCR assays for detection of blaVIM, IMP and SPM MBL genes with consensus primers.3 PCR mapping of blaVIM-carrying class 1 integrons was carried out as described previously.10 Nucleotide sequences of selected PCR products were determined on both strands using an automated sequencer (Avant 3100; Applied Biosystems).
K. pneumoniae isolates were typed by PFGE.2 Agarose-embedded genomic DNA was digested with XbaI and the resulting fragments were separated by PFGE using a CHEF DR-III apparatus (Bio-Rad, Athens, Greece). Electrophoresis was at 6.0 V/cm for 22 h with a pulsing time linearly ramped from 5 to 40 s. Un-weighted pair group method with arithmetic means (UPGMA) dendrograms were constructed with the GelCompar I software (Applied-Maths, Sint-Martens-Latem, Belgium), using the Dice similarity coefficient, with optimization and position tolerance settings of 1.0% and 0.5%, respectively.
Data were analysed by the
2 test using the SPSS v. 12.0 software.
| Results and discussion |
|---|
|
|
|---|
VPKP isolation frequency
Hospital I contributed 90, hospital II 62 and hospital III 26 K. pneumoniae isolates. Ninety-six isolates (53.9%) were from medical ward patients, 54 (30.3%) from ICUs and 28 (15.7%) from surgical wards. All isolates were studied for carriage of MBL genes irrespective of their resistance phenotype. Sixty-seven (37.6%) isolates were positive for blaVIM-1 by PCR and DNA sequencing. Other types of MBL genes were not detected. Isolation rates of VPKP were similar in hospitals II and III (53.2% and 53.8%, respectively), whereas in hospital I the rate was lower (22.2%). The highest frequency of VPKP was observed in ICUs (77.8%), whereas it was significantly lower in the surgical and medical wards (28.5% and 17.7%, respectively; P < 0.001).
Resistance phenotypes and MBL detection
Imipenem MICs for the 67 VPKP isolates ranged from 0.125 to 32 mg/L. MIC50 and MIC90 were 1 and 16 mg/L, respectively. Consequently, 48 (71.6%) VPKP isolates would have been classified as phenotypically fully susceptible to imipenem according to the current breakpoints. For 31 (46.3%) VPKP isolates, imipenem MICs were
0.5 mg/L. For 35 (52.2%) of the VPKP isolates, meropenem MICs (range 0.03–32, MIC50 = 1 and MIC90 = 32 mg/L) were lower than those of imipenem by one to three doubling dilutions. All VPKP isolates were resistant to amoxicillin/clavulanate, ticarcillin/clavulanate, piperacillin, cefoxitin, cefotaxime, ceftriaxone and ceftazidime. One and two isolates were susceptible to piperacillin/tazobactam and cefepime, respectively. Forty-five (67.2%) of the VPKP isolates exhibited either resistance or decreased susceptibility to aztreonam, a ß-lactam that is not inactivated by VIM-1. These isolates were all ESBL-positive as indicated by the synergy between aztreonam and clavulanate in the respective DDST. High resistance rates were also observed for tobramycin (100%), gentamicin (47.8%), amikacin (86.6%), co-trimoxazole (100%), tetracycline (85.1%) and ciprofloxacin (86.6%) (Table 1). Thus, the VPKP isolates were invariably multidrug-resistant. None of the 111 blaVIM-1-negative isolates exhibited either resistance or decreased susceptibility to carbapenems (imipenem MICs ranged from 0.064 to 0.5 mg/L). Nevertheless, 18 (16.2%) of the latter isolates were ESBL-positive and displayed a multidrug resistance phenotype including newer ß-lactams (third-generation cephalosporins and aztreonam) and aminoglycosides. Fluoroquinolone resistance was also common in this group and the respective rate (83.3%) was comparable with that of the VIM producers (P > 0.05). The remaining 93 isolates were susceptible to newer ß-lactams. Also, resistance rates to aminoglycosides, such as gentamicin, as well as co-trimoxazole and ciprofloxacin in the latter group (5.4%, 27% and 20.6%, respectively) were significantly lower than in the isolates possessing VIM and/or ESBL (P < 0.001).
|
High sensitivity scores were observed for both the EDTA-based techniques for MBL detection. The MBL-DDT correctly identified 63 of the VPKP isolates (94% sensitivity). The MBL-CDT using imipenem discs was positive for 62 VPKP isolates (92.5% sensitivity). False positives were not observed by either method. Four VPKP isolates with imipenem MICs ranging from 0.125 to 0.5 mg/L were misidentified by both these methods. Preliminary experiments with the MBL-Etest showed that this method performed poorly even with VPKP isolates with high-level resistance to imipenem (data not shown). This was probably due to the relatively high drug content of the strips.
A total of 94 K. pneumoniae isolates were studied by PFGE. These included 61 VPKP and 33 MBL-negative isolates (18 ESBL producers and 15 randomly selected isolates negative for ESBL). Fifty-seven of the VPKP isolates were distributed into 8 PFGE clusters (A–H), each grouping together 2–14 isolates with chromosomal fingerprints displaying >80% similarity. The remaining four VPKP isolates displayed unique PFGE patterns (Figure 1 and Table 1). Three of the less populous clusters, B, G and H, were each associated with a single hospital. The remaining five clusters were composed of isolates from either two (clusters A and F) or all three hospitals (clusters C, D and E) (Figure 1 and Table 1). There were certain associations between PFGE types and antibiotic resistance phenotypes of the VPKP isolates. Clusters A, D and H included isolates with increased resistance to carbapenems, whereas most isolates belonging to clusters B, E, F and G exhibited low carbapenem MICs (Table 1). In contrast, isolates in PFGE cluster C exhibited a wide variety of carbapenem MICs, though they were uniformly resistant to the other ß-lactams tested. VPKP isolates that were also positive for ESBL production were distributed in all clusters except G. Interestingly, 16 of the 18 ESBL-positive isolates lacking blaVIM-1 clustered together with VPKP isolates, mainly in clusters F (9 isolates) and E (6 isolates). Also, three of the MBL- and ESBL-negative isolates belonged to clusters E (two isolates) and C (Figure 1).
|
Expression of VIM-1 and location of blaVIM-1
Twelve VPKP isolates representing the main PFGE clusters (one from each of the clusters A, B, E, F and G, two from each of the clusters D and H and three from cluster C) were studied. IEF of ß-lactamase-containing extracts confirmed production of VIM-1 in all 12 VPKP isolates. These experiments also showed that the eight ESBL-positive and aztreonam-resistant isolates included in this group produced a ß-lactamase with an isoelectric point of 8.2, consistent with an SHV-5-type enzyme. Differences were observed among imipenem hydrolytic activities ranging from 30 to 50 U. However, these values did not correlate with the respective carbapenem MICs.
PCR mapping and sequencing showed that the variable region of the blaVIM-1-carrying integrons from all 12 VPKP isolates contained blaVIM-1, aacA4, dhfrI, aadA and sulI similarly to the In-e541 integron (AY339625
[GenBank]
in GenBank) commonly found among Enterobacteriaceae in this setting.4,8,10 VIM-1-encoding plasmids were transferred to E. coli by conjugation from eight VPKP isolates belonging to PFGE clusters C, D, F and H. Transformation experiments revealed an additional VIM-1-encoding plasmid from an isolate of cluster G. blaVIM-1-carrying replicons were not identified from the remaining three isolates (clusters A, B and E). Eight of the plasmids transferring VIM-1, although varied in size (50–70 kb) and exhibiting RFLP differences (data not shown), belonged to IncN. The self-transferable plasmid from the cluster F isolate was
150 kb in size and not typeable by Inc/rep typing. None of the blaVIM-1-carrying plasmids identified here encoded production of SHV-5.
In contrast to previous studies, describing either clonal outbreaks in single hospitals or occurrence of a limited number of VPKP strains, this work documents the inter- and intra-hospital dissemination of several distinct clones. It also shows that VPKP strains are currently prevalent in the major hospitals in Athens. They constitute the majority of multidrug-resistant K. pneumoniae, whereas isolates with only an ESBL phenotype, which were predominant in this setting up to 2001, are observed less frequently nowadays. Results from previous studies on acquired bla genes in enterobacteria in Athens hospitals11 as well as the data available in the National Surveillance System for Antibiotic Resistance database (www.mednet.gr/whonet/top.htm) suggest that this development must have taken place during a relatively short time period, probably between 2002 and 2003. Apart from the intensive use of carbapenems, the dissemination of VPKP isolates is probably associated with ineffective infection control practices. Evidence in support of this in the present study includes the persistence for at least one year, as well as the inter-hospital spread of several strains. Additionally, identification of multiple VPKP clones underscores the transfer potential of blaVIM-1 that is disseminated mostly via self-transmissible IncN plasmids as also observed in previous studies.4,8–10 A strong association of VIM-1 and SHV-5 production was noticeable; this has also been described previously2,4,12 and is due to acquisition of distinct plasmids. It is possible that some of the VPKP strains may have been derived from endemic SHV-5-producing progenitors. However, the presence of clonally related isolates producing none, one or both VIM-1 and SHV-5 indicates that frequent events of acquisition and loss of bla genes probably operate alongside the more forbidding gradual accumulation of resistance.
Imipenem MICs for the previously described VPKP in Greece ranged from 1 to 64 mg/L.2–5 The present findings not only extend the lower limit to 0.125 mg/L, but also show that an imipenem MIC <1 mg/L is common. The wide range of carbapenem MICs among VPKP, including genotypically related isolates, may arise from differences in the outer membrane permeability,8 whereas the diversity in the levels of VIM-1 production probably plays a lesser role. The low carbapenem MICs of a significant proportion of the VPKP isolates makes their direct phenotypic identification difficult. Based on the present findings, the use of either the MBL-DDT or the MBL-CDT is proposed for all newer ß-lactam-resistant K. pneumoniae isolated in this setting.
| Funding |
|---|
|
|
|---|
This study was supported in part by a research grant from the Kapodistrias program of the University of Athens.
| Transparency declarations |
|---|
|
|
|---|
None to declare.
| Acknowledgements |
|---|
We thank Georgia Diamantopoulou for expert assistance with PFGE.
| References |
|---|
|
|
|---|
1 . Walsh TR, Toleman MA, Poirel L, et al. Metallo-ß-lactamases: the quiet before the storm? Clin Microbiol Rev (2005) 18:306–25.
2
.
Giakkoupi P, Xanthaki A, Kanelopoulou M, et al. VIM-1 metallo-ß-lactamase-producing Klebsiella pneumoniae strains in Greek hospitals. J Clin Microbiol (2003) 41:3893–6.
3
.
Ikonomidis A, Tokatlidou D, Kristo I, et al. Outbreaks in distinct regions due to a single Klebsiella pneumoniae clone carrying a blaVIM-1 metallo-ß-lactamase gene. J Clin Microbiol (2005) 43:5344–7.
4
.
Miriagou V, Tzelepi E, Daikos GL, et al. Panresistance in VIM-1-producing Klebsiella pneumoniae. J Antimicrob Chemother (2005) 55:810–1.
5 . Deshpande LM, Jones RN, Fritsche TR, et al. Occurrence and characterization of carbapenemase-producing Enterobacteriaceae: report from the SENTRY Antimicrobial Surveillance Program 2000–2004. Microb Drug Resist (2006) 12:223–30.[CrossRef][Web of Science][Medline]
6
.
Lee K, Lim YS, Yong D, et al. Evaluation of the Hodge test and the imipenem-EDTA double-disk synergy test for differentiating metallo-ß-lactamase-producing isolates of Pseudomonas spp. and Acinetobacter spp. J Clin Microbiol (2003) 41:4623–9.
7
.
Franklin C, Liolios L, Peleg AY. Phenotypic detection of carbapenem-susceptible metallo-ß-lactamase-producing Gram-negative bacilli in the clinical laboratory. J Clin Microbiol (2006) 44:3139–44.
8
.
Loli A, Tzouvelekis LS, Tzelepi E, et al. Sources of diversity of carbapenem resistance levels in Klebsiella pneumoniae carrying blaVIM-1. J Antimicrob Chemother (2006) 58:669–72.
9 . Carattoli A, Miriagou V, Bertini A, et al. Replicon typing of plasmids encoding resistance to newer ß-lactams. Emerg Infect Dis (2006) 12:1145–8.[Web of Science][Medline]
10
.
Miriagou V, Tzelepi E, Gianneli D, et al. Escherichia coli with a self-transferable, multiresistant plasmid coding for the metallo-ß-lactamase VIM-1. Antimicrob Agents Chemother (2003) 47:395–7.
11 . Tzelepi E, Magana C, Platsouka E, et al. Extended-spectrum ß-lactamase types in Klebsiella pneumoniae and Escherichia coli in two Greek hospitals. Int J Antimicrob Agents (2003) 21:285–8.[CrossRef][Web of Science][Medline]
12
.
Kassis-Chikhani N, Decre D, Gautier V, et al. First outbreak of multidrug-resistant Klebsiella pneumoniae carrying blaVIM-1 and blaSHV-5 in a French university hospital. J Antimicrob Chemother (2006) 57:142–5.
![]()
CiteULike
Connotea
Del.icio.us What's this?
This article has been cited by other articles:
![]() |
M. Falcone, M. L. Mezzatesta, M. Perilli, C. Forcella, A. Giordano, V. Cafiso, G. Amicosante, S. Stefani, and M. Venditti Infections with VIM-1 Metallo-{beta}-Lactamase-Producing Enterobacter cloacae and Their Correlation with Clinical Outcome J. Clin. Microbiol., November 1, 2009; 47(11): 3514 - 3519. [Abstract] [Full Text] [PDF] |
||||
![]() |
L. F. Mataseje, N. Neumann, B. Crago, P. Baudry, G. G. Zhanel, M. Louie, M. R. Mulvey, and and the ARO Water Study Group Characterization of Cefoxitin-Resistant Escherichia coli Isolates from Recreational Beaches and Private Drinking Water in Canada between 2004 and 2006 Antimicrob. Agents Chemother., July 1, 2009; 53(7): 3126 - 3130. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. Carattoli Resistance Plasmid Families in Enterobacteriaceae Antimicrob. Agents Chemother., June 1, 2009; 53(6): 2227 - 2238. [Full Text] [PDF] |
||||
![]() |
G. L. Daikos, P. Petrikkos, M. Psichogiou, C. Kosmidis, E. Vryonis, A. Skoutelis, K. Georgousi, L. S. Tzouvelekis, P. T. Tassios, C. Bamia, et al. Prospective Observational Study of the Impact of VIM-1 Metallo-{beta}-Lactamase on the Outcome of Patients with Klebsiella pneumoniae Bloodstream Infections Antimicrob. Agents Chemother., May 1, 2009; 53(5): 1868 - 1873. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. Tsakris, I. Kristo, A. Poulou, K. Themeli-Digalaki, A. Ikonomidis, D. Petropoulou, S. Pournaras, and D. Sofianou Evaluation of Boronic Acid Disk Tests for Differentiating KPC-Possessing Klebsiella pneumoniae Isolates in the Clinical Laboratory J. Clin. Microbiol., February 1, 2009; 47(2): 362 - 367. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||


