Journal of Antimicrobial Chemotherapy Advance Access published September 1, 2014
J Antimicrob Chemother doi:10.1093/jac/dku339
Dominance of international ‘high-risk clones’ among metallob-lactamase-producing Pseudomonas aeruginosa in the UK Laura L. Wright1,2*, Jane F. Turton1, David M. Livermore1,2, Katie L. Hopkins1 and Neil Woodford1 1
Antimicrobial Resistance and Healthcare Associated Infections Reference Unit, Public Health England, Colindale, London NW9 5EQ, UK; 2 Norwich Medical School, University of East Anglia, Norwich, Norfolk NR4 7TJ, UK
Received 9 June 2014; returned 10 July 2014; revised 31 July 2014; accepted 4 August 2014 Objectives: Carbapenem-resistant isolates of Pseudomonas aeruginosa producing metallo-b-lactamases (MBLs) are increasingly reported worldwide and often belong to particular ‘high-risk clones’. This study aimed to characterize a comprehensive collection of MBL-producing P. aeruginosa isolates referred to the UK national reference laboratory from multiple UK laboratories over a 10 year period. Methods: Isolates were referred to the UK national reference laboratory between 2003 and 2012 for investigation of resistance mechanisms and/or outbreaks. MBL genes were detected by PCR. Typing was carried out by ninelocus variable-number tandem repeat (VNTR) analysis and MLST. Results: MBL-producing P. aeruginosa isolates were referred from 267 source patients and 89 UK laboratories. The most common isolation sites were urine (24%), respiratory (18%), wounds (17%) and blood (13%). VIM-type MBLs predominated (91% of all MBLs found), but a few IMP- and NDM-type enzymes were also identified. Diverse VNTR types were seen, but 86% of isolates belonged to six major complexes. MLST of representative isolates from each complex showed that they corresponded to STs 111, 233, 235, 357, 654 and 773, respectively. Isolates belonging to these complexes were received from between 9 and 25 UK referring laboratories each. Conclusions: The incidence of MBL-producing P. aeruginosa is increasing in the UK. The majority of these isolates belong to several ‘high-risk clones’, which have been previously reported internationally as host clones of MBLs. Keywords: VIM, ST111, ST235, carbapenemases, MBLs
Introduction Pseudomonas aeruginosa is a common opportunistic pathogen responsible for many hospital-acquired infections. Carbapenems are important antibiotics for treatment of these infections, but resistance to these agents is increasing worldwide. Although most carbapenem resistance in P. aeruginosa arises by mutations that lead to the loss of the porin OprD or up-regulation of efflux pumps (e.g. MexAB-OprM),1 carbapenemases are also increasingly reported. Serine carbapenemases of the KPC, GES and OXA families have occasionally been reported in this species, with limited geographical scatter,2 – 4 but metallo-b-lactamases (MBLs), particularly VIM and IMP types, are the most widespread and have been reported globally.5 MBLs hydrolyse almost all b-lactam antibiotics, including carbapenems, and most MBL-producing P. aeruginosa strains are resistant to other antibiotic classes, including fluoroquinolones and aminoglycosides, often leaving polymyxins as the sole therapeutic options. The MBL genes often reside on mobile genetic elements able to be transmitted between strains,5 but are
commonly associated with multiresistant ‘high-risk clones’. These have been identified in several bacterial species, and in the case of P. aeruginosa the most commonly reported ‘high-risk clones’ belong to STs 111, 235 and 175.6 Recent studies report major dissemination of an ST235 lineage with the VIM-2 MBL in Russia, Belarus and Kazakhstan,7 and of ST277 with the SPM-1 MBL in Brazil.8 In recent years, the number of MBL-producing P. aeruginosa referred to the UK national reference laboratory has steadily risen. We sought to determine the contribution of internationally recognized ‘high-risk clones’ to this increase.
Materials and methods Bacterial isolates Three-hundred and thirty-four MBL-positive isolates were identified amongst P. aeruginosa isolates referred from UK hospital laboratories to PHE’s Antimicrobial Resistance and Healthcare Associated Infections Reference Unit (formerly the HPA’s ARMRL and LHCAI laboratories)
# Crown copyright 2014.
1 of 8
Downloaded from http://jac.oxfordjournals.org/ at University of Tasmania Library on September 3, 2014
*Corresponding author. Antimicrobial Resistance and Healthcare Associated Infections Reference Unit, Public Health England, 61 Colindale Avenue, London NW9 5EQ, UK. Tel: +44(0)2083276764; E-mail: [email protected]
Wright et al.
between 2003 and 2012 for susceptibility testing, investigation of resistance mechanisms and/or strain typing. Isolates were identified as MBL producers by PCR, either at the time of referral (304 isolates) or by retrospective screening of isolates submitted from 2009 – 12 that shared similar variable-number tandem repeat (VNTR) types with other MBLproducing isolates (30 isolates). Demographic and isolation site data were obtained from the PHE’s laboratory information management system. Referring laboratories were assigned codes to indicate the UK region and given a unique number within the region, in the format ‘region_number’ (e.g. North West_1). As a comparator group, all 209 P. aeruginosa isolates collected in 2011 as part of the BSAC’s Bacteraemia Surveillance Programme from hospitals in the UK and Irish Republic were also studied (http://www.bsacsurv.org/).
Genes encoding VIM-, IMP-, SPM-, GIM- and SIM-type MBLs were sought by multiplex PCR, as detailed by Ellington et al.9 Genes for NDM-type MBLs were sought by a single PCR, as previously described.10 Sequencing of blaVIM and blaIMP MBL genes was carried out using previously described primers specific to MBL genes and class 1 integrons;11 sequencing of blaNDM genes was with primers NDM-orfF (5′ -ATGGAATTGCCCAATATTATG-3′ ) and NDM-orfR (5′ -TCAGCGCAGCTTGTCGGCCA-3′ ).
Typing PFGE was the routine typing and outbreak investigation method used in the reference laboratory for P. aeruginosa between 2003 and 2009, and was carried out using SpeI-digested genomic DNA.12 Nine-locus VNTR analysis became the routine method from 2009 onwards and was performed as previously described.12 Minimum spanning trees were produced using Bionumerics software v6.1 (Applied Maths, Sint-Martens-Latem, Belgium). MLST analysis was undertaken as described by Curran et al.;13 results were analysed using Bionumerics software and STs were assigned using the P. aeruginosa MLST database (http://pubmlst.org/paeruginosa/).
Patient demographics The 334 MBL-positive isolates were from a total of 267 patients, with VNTR profiles remaining consistent when multiple isolates were received from the same patient. Accordingly, one isolate per patient was selected for further study, leaving 267 nonduplicate-patient isolates from 89 UK laboratories. Ages of the source patients ranged from 0 to 94, mean 54 years; proportions in age bands were as follows: 0 – 2 years, 4%; 3 – 15 years, 2%; 16 – 29 years, 6%; 30 – 44 years, 11%; 45 – 59 years, 22%; 60 – 74 years, 32%; and .74 years, 12% (age was unknown for the remaining 11%). Predominant isolation sites were urine (24%), respiratory (18%), wounds (17%), blood (13%) and indwelling devices (7%), whereas few were from skin (3%) and faecal (3%) specimens; the remaining isolates were from other (8%) or unknown (7%) isolation sites. Sixty percent of patients were male, 35% female and for 5% gender was not stated. blaVIM genes alone were detected in 243 isolates, blaIMP in 22 isolates and blaNDM in 1 isolate; 1 further isolate had both blaVIM and blaNDM. Of the 89 referring laboratories, 83 submitted isolates from fewer than 10 patients over the 10 year period, with 71 submitting only one or two isolates each. The remaining six laboratories, which submitted between 13 and 31 isolates each, collectively accounted for 43% of all patients (116/267).
Typing Among the 267 non-duplicate-patient isolates, 232 had VNTR data available or generated in this study, and a minimum spanning tree based on these data is shown in Figure 1. The remaining 35 isolates were not available in the archive for retrospective VNTR typing, but 19 had previously determined PFGE profiles (not shown) identical to those of isolates with known VNTR types
D B F A
E C
Figure 1. Minimum spanning tree, based on clustering at the first eight VNTR loci for MBL-positive P. aeruginosa, with one isolate included per patient (n ¼232). The six main complexes, A–F, are labelled. Coloured segments of the circles indicate laboratories that submitted three or more isolates and white segments represent laboratories submitting one or two isolates. The diameter of the circle represents the number of isolates with that VNTR profile. Shading indicates complexes. Thick solid lines represent single-locus variants while thin solid lines and dotted lines represent multilocus variants.
2 of 8
Downloaded from http://jac.oxfordjournals.org/ at University of Tasmania Library on September 3, 2014
Analysis of MBL genes
Results
JAC
MBL-producing P. aeruginosa in the UK
Table 1. Typing data for the six main VNTR complexes identified (n ¼251) No. of different VNTR profiles
A
11,3,4,3,2,2,x,4,x
6
B
13,3,6,4,5,1,x,2,x
C D
E F Others
MLST type(s) (no. of isolates tested)
No. of isolatesb
No. of submitting laboratories
MBLs detected (no. of isolates)
ST111 (11)
75
25
16
ST235 (18)
52
25
12,3,4,5,3,1,x,2,x 11,3,2,15,3,1,x,3,x
11 6
ST233 (10) ST654 (10), ST964 (1)
26 19
16 11
13,2,1,5,2,3,6,x,x 12,4,6,5,3,1,10,x,x diverse
7 3 26
ST357 (9) ST773 (5) not done
30 13 36
9 11 25
VIM (70) IMP (5) VIM (46) IMP (6) VIM (26) VIM (17) IMP (1) NDM (1) VIM (30) VIM (13) VIM (25) IMP (10) VIM and NDM (1)
a
x represents loci where the repeat number varies between isolates within a complex. One isolate per patient was included; these numbers include 4 isolates (complex B), 14 isolates (complex E) and 1 isolate (complex F) where the MBL-positive organisms were no longer available in the archive for VNTR analysis, but which were previously found to share a PFGE profile, and are from the same hospital outbreak as other isolates in the respective complex. Isolates were also received from an additional 39 patients at London_17 with a PFGE profile corresponding to complex A. These are not included here as they had not been screened for MBL genes and were no longer available in our archives.
b
from a suspected outbreak at the same hospital. These isolates were assumed to belong to the corresponding VNTR complex and are included in Table 1. The remaining 16/35 unavailable isolates had no typing data, or belonged to PFGE types unique to their respective hospital; these latter isolates, from 12 laboratories, are excluded from Table 1; all carried blaVIM. Six VNTR complexes (designated A – F) accounted for 86% of the 251 isolates with VNTR data available or inferred from PFGE data. Isolates belonging to these six complexes were referred from between 9 and 25 laboratories and, owing to their predominance, became the focus of further study. Sixty-four representatives, covering the variation in VNTR profile within each complex, were selected for MLST analysis. Isolates belonging to complexes A, B, C, D, E and F were found to belong to ST111, ST235, ST233, ST654/ST964, ST357 and ST773, respectively (Table 1). Ten of 11 MLST-typed complex D isolates belonged to ST654, but one belonged to ST964; this is a single-locus variant differing only in the acsA allele, where ST964 has allele 145 with a single CT substitution compared with allele 17 in ST654. Table 2 shows the distribution of each of the major complexes among referring laboratories. The largest group of isolates was complex A (corresponding to ST111; VNTR type 11,3,4,3,2,2,x,4,x, where x is variable), with 75 representatives. It included isolates from 25 laboratories spread across the UK, but with more than half of the isolates coming from outbreaks at London_17 and Wales_1, referring 29 and 13 isolates, respectively, all with blaVIM. Most other complex A isolates (28/33) also had blaVIM, but five, from two laboratories, had blaIMP. VNTR profiles were highly conserved amongst all complex A isolates, with most only differing by repeat numbers between six and eight at locus 61. Complex B (ST235; 13,3,6,4,5,1,x,2,x) was the second largest group, comprising 52 isolates referred from 25 different laboratories. There were potential outbreaks at sites London_13 and
Scotland_2, referring eight and six isolates, respectively, all with blaVIM; these groups were received over 7 months and 2 years, respectively. Another laboratory (London_7) also submitted isolates carrying blaVIM from eight patients over an 8 year period. The remaining 30 complex B isolates were from 22 laboratories, each submitting 1 to 3 isolates; 24 isolates had blaVIM and 6 had blaIMP. Twenty-six isolates belonged to complex C (ST233; 12,3,4,5, 3,1,x,2,x). These all carried blaVIM and were from 16 laboratories, which submitted one to four representatives each. Nineteen isolates from 11 laboratories belonged to complex D (ST654; 11,3,2,15,3,1,x,3,x); 7 of these were submitted from South East_6 and all had blaVIM and shared an identical VNTR profile, whilst 1 or 2 isolates were referred from each of the remaining 10 laboratories; 10 isolates had blaVIM, 1 isolate had blaIMP and 1 isolate had blaNDM. Most of the 30 complex E isolates (ST357; 13,2,1,5,2,3,6,x,x), all with blaVIM, were from an outbreak at North West_15, with 22 isolates referred over a 9 year period; 12 of these 22 were submitted in 2007. Each of the remaining eight complex E isolates was referred from a different laboratory. Finally, 13 isolates, all with blaVIM, belonging to complex F (ST773; 12,4,6,5,3,1,10,x,x) were received from 11 laboratories in diverse areas of the UK. Representative isolates were selected to cover the VNTR variation seen in blaVIM-positive isolates from the six major complexes. In complexes A, C, D, E and F (STs 111, 233, 654, 357 and 773) all representatives had blaVIM-2 genes, whereas eight isolates selected from complex B (ST235) variously had blaVIM-1 (two isolates), blaVIM-2 (four isolates), blaVIM-4 (one isolate) or blaVIM-6 (one isolate). Amongst seven representatives of the 22 blaIMP-positive isolates, blaIMP-1 (three isolates), blaIMP-7 (one isolate), blaIMP-10 (one isolate) and blaIMP-13 (two isolates) alleles were seen. Both blaNDM-positive isolates had the blaNDM-1 allele.
3 of 8
Downloaded from http://jac.oxfordjournals.org/ at University of Tasmania Library on September 3, 2014
VNTR typea
VNTR complex
Wright et al.
Table 2. Geographical and temporal distribution of MBL-producing P. aeruginosa belonging to the six major complexes (215 isolates) among referring laboratories Major contributors (≥5 referred isolates)
VNTR Complex
referring laboratorya
MBL genes detected
Minor contributors 2 laboratories with 4 representatives each 2 laboratories with 3 representatives each 19 laboratories with 1 representative each 1 laboratory with 3 representatives 6 laboratories with 2 representatives each 15 laboratories with 1 representative each 3 laboratories with 2 representatives each 11 laboratories with 1 representative each 2 laboratories with 2 representatives each 8 laboratories with 1 representative each 8 laboratories with 1 representative each 2 laboratories with 2 representatives each 9 laboratories with 1 representative each
A (n¼75)
London_17c Wales_1
29 (39%) 13 (17%)
80 months 18 months
all with blaVIM all with blaVIM
B (n¼52)
London_7 London_13 Scotland_2 London_7 London_13 South East_6
8 (15%) 8 (15%) 6 (12%) 5 (19%) 4 (15%) 7 (37%)
87 months 7 months 27 months 27 months 18 months 36 months
all with blaVIM all with blaVIM all with blaVIM all with blaVIM all with blaVIM all with blaVIM
22 (73%) 103 months no major contributors
all with blaVIM
C (n¼26) D (n¼19) E (n¼30) F (n¼13)
North West_15
a
Referring laboratories are coded in the format ‘UK region_number’. One isolate per patient is included; these numbers include 4 isolates (complex B), 14 isolates (complex E) and 1 isolate (complex F) that were not available in the archive for VNTR analysis, but which shared a PFGE profile, and are from the same hospital outbreak as other isolates in the complex. c Isolates were also received from an additional 39 patients at London_17 with a PFGE type corresponding to complex A. These are not included here as they had not been screened for MBL production and were no longer available in our archives. b
The 14% of isolates (n¼ 36) that did not belong to any of these six major complexes were diverse by VNTR, with 26 different VNTR profiles represented. They were referred from 24 laboratories across the UK, each submitting one to three isolates; 26 had blaVIM, 10 had blaIMP and one had both blaVIM and blaNDM genes.
Regional distribution of MBL-positive isolates The UK distribution of the 267 non-duplicate-patient isolates is shown in Figure 2. London accounted for 47% of these organisms, with MBL-positive isolates referred from 28 London region laboratories; all six major complexes (A – F) were represented and nine other VNTR types seen. Four London laboratories were amongst the six sites that submitted .10 MBL-positive isolates. London_17 submitted 29 complex A (ST111) isolates over 6 years, with source patients on multiple wards. London_13 had separate outbreaks of MBL-positive isolates belonging to complexes B and C (STs 235 and 233), with eight and four representatives, respectively. Each ‘outbreak’ lasted 1 –2 years separated by a 6 year gap. London_7 and London_12 referred 19 and 15 isolates, respectively, over 7 – 8 years, with isolates belonging to diverse VNTR complexes. Greater Manchester accounted for 12% of isolates (31/267). Isolates belonging to complex E (ST357) predominated and included all 22 from an outbreak at North West_13, although isolates belonging to complexes A, B and D (STs 111, 235 and 654) were also seen. In Sussex, Surrey and Kent half of the 18 isolates received belonged to complex D (ST654); 7 of these 9 isolates were from South East_6; the remaining 9 representatives from this region belonged to complexes B and C (STs 235 and 233) or
4 of 8
had other VNTR profiles. Wales accounted for 6% of isolates, mostly (13/15) from an outbreak of ST111 P. aeruginosa at Wales_1, although one ST111 isolate was from another laboratory and one ST773 isolate also was referred from the region. Finally, the 12 (4%) isolates submitted from Scotland belonged to complexes A, B, C and F (STs 111, 235, 233 and 773). Fewer than 10 isolates were referred from each of the remaining UK regions.
Typing of comparator P. aeruginosa isolates The comparator set, comprising P. aeruginosa isolates collected as part of the BSAC Bacteraemia Resistance Surveillance Programme in 2011, showed far greater diversity than the MBL producers, with 136 different VNTR profiles represented amongst the 209 isolates (Figure 3). Although 4.8% (10 isolates) were imipenem resistant, none produced MBLs. Two isolates shared VNTR profiles with complex B (ST235) and one shared a profile with complex C (ST233); the remaining 206 isolates did not share VNTR profiles with any of the major complexes identified amongst the MBL-producing isolates.
Discussion Most (86%) MBL-producing P. aeruginosa isolates referred to the UK reference laboratory between 2003 and 2012 belonged to six VNTR complexes, corresponding to internationally recognized ‘high-risk clones’, STs 111, 235, 233, 357, 654 and 773, and most had VIM-type MBLs. These findings are in striking contrast to the general population structure of P. aeruginosa, for which a
Downloaded from http://jac.oxfordjournals.org/ at University of Tasmania Library on September 3, 2014
number of isolatesb
time period over which isolates were referred
JAC
MBL-producing P. aeruginosa in the UK
London
126
28
Greater Manchester
6
Sussex, Surrey and Kent
5
Wales
2
Scotland
5
31 18 15 12
9 5
West Midlands
9 7
Norfolk, Suffolk, Cambridgeshire and Essex
7 3
Cumbria and Lancashire
7 3
Cheshire and Merseyside
7 6
Thames Valley
6 3
Lincolnshire, Leicestershire, Nottinghamshire and Derbyshire
5 2
Hampshire, Isle of Wight and Dorset
5 4 3 3
Bedfordshire, Hertfordshire and Northamptonshire Northern Ireland
2 2
North East
2 2
Avon, Gloucestershire and Wiltshire
2 2
Devon, Cornwall and Somerset
1 1 0
Isolates
A (ST111) D (ST654) Other types
Referring hospitals
50
100
B (ST235) E (ST357) No typing data
C (ST233) F (ST773)
150
Figure 2. Geographical sources of isolates in the UK and distribution of the six main complexes (A –F) in each of the five regions referring .10 isolates. This figure appears in colour in the online version of JAC and in black and white in the print version of JAC.
recent UK study14 shows considerable diversity, with overlap between environmental isolates and those causing clinical infection, although with a few prevalent clusters in diverse locations. The major complexes seen here were not prevalent amongst MBL-negative isolates in this previous study, nor were they prevalent in our comparator collection of (largely susceptible) P. aeruginosa isolates from the BSAC Bacteraemia Resistance Surveillance Programme. Rather, these ‘high-risk clones’ seem to represent a distinct subset of P. aeruginosa lineages, which may be successful precisely due to a particularly strong ability to acquire and/or maintain resistance genes compared with the general P. aeruginosa population. Notably, the reference laboratory has also received MBL-negative representatives corresponding to these six major complexes, including a few ST235 isolates carrying blaGES-5 and a single ST773 isolate with blaOXA-181, both genes encoding non-metallo-carbapenemases (J. F. Turton, K. L. Hopkins and N. Woodford, unpublished results). Moreover, in three of these complexes we encountered isolates that variously had either blaVIM or blaIMP genes and, in one complex (B; ST235), various different blaVIM alleles. Each of the six main VNTR complexes has been reported internationally as a host for MBLs. Complex A (corresponding to ST111)
was the largest group in our collection, with most isolates carrying blaVIM, generally blaVIM-2. VNTR profiles within this complex were highly similar, most differing only at locus 61, which is the most variable of the nine VNTR loci.12 ST111 is frequently recorded amongst MBL-producing P. aeruginosa, with ST111 isolates reported to be of serotype O12, which has been recognized as a common host of multiresistance in Europe since the 1980s.15,16 ST111 was the predominant ST type seen in a nationwide study of MBL-producing P. aeruginosa in the Netherlands in 2010 – 11,16 and was responsible for outbreaks affecting hospitals in central Greece17 and Italy.18 In all these settings, as in the UK, the ST111 isolates generally carried blaVIM-2 . Smaller numbers of ST111 isolates with blaVIM-2 have been reported in other European countries including Sweden,11 Croatia,19 Spain20 and Germany,21 whilst isolates with this ST have been associated with blaVIM-4 in Hungary,22 blaVIM-1 in Greece23 and blaIMP-13 in France.24 Outside Europe, an isolate of ST111 from a patient in Colombia was recently found to carry both blaVIM-2 and blaKPC-2.25 The second most prevalent complex (B) in our collection corresponded to ST235. Isolates in this complex had more diverse VNTR profiles than those in complex A and were more widely disseminated among referring laboratories, with no more than eight
5 of 8
Downloaded from http://jac.oxfordjournals.org/ at University of Tasmania Library on September 3, 2014
Yorkshire and Humber
Wright et al.
Novel cluster 2 Complex B (ST235)
Clone C (ST17) Novel cluster 1
Cluster A (ST27)
Cluster D (ST395)
Cluster E
Cluster H PA14 (ST253) Figure 3. Minimum spanning tree based on clustering at the first eight VNTR loci for P. aeruginosa isolates from the BSAC Bacteraemia Resistance Surveillance Programme (n¼209). The diameter of the circle represents the number of isolates with that VNTR profile. Shading indicates complexes. Thick solid lines represent single-locus variants while thin solid lines and dotted lines represent multilocus variants. Isolates corresponding to VNTR complexes B and C (STs 235 and 233, respectively) are shown in black circles. Isolates corresponding to previously reported clones as described by Martin et al.14 are indicated and had the following VNTR profiles: cluster A, 8,3,4,5,2,3,5,2,x; cluster D, 10,3,5,5,4,1,3,x,x; cluster E, 11,4,5,2,2,1,x,2,x; cluster H, 12,5,1,5,2,2,x,x,x; clone C, 11,4,5,2,2,1,x,2,x; and PA14, 12,2,1,5,5,2,x,5,x. Two novel clusters are indicated with VNTR profiles 12,5,5,5,4,3,7,6,x and 12,8,2,2,4,3,5,1,x, respectively.
from any single site. Again, most isolates had blaVIM, but with diverse blaVIM alleles identified. Together with the VNTR diversity, this carbapenemase diversity suggests multiple imports and/or acquisitions of blaVIM by ST235 P. aeruginosa. This variation is in contrast to the ST235 clone with blaVIM-2 that is widespread across Russia, Belarus and Kazakhstan.7 ST235 was the most commonly identified MBL-producing P. aeruginosa type in a study of five Mediterranean countries,26 and ST235 isolates with blaVIM-2 genes have also been reported in Spain,20 Croatia,19 Germany21 and Greece,23 blaVIM-4 in Hungary,22 Norway11 and Belgium27 and blaVIM-13 in Spain.28 In Asia, outbreaks of ST235 clones with blaIMP-6 genes have occurred in Japan29 and South Korea,30 with blaVIM-positive isolates with this ST also seen in Thailand, Malaysia, Sri Lanka and Korea.31 There are single reports of ST235 isolates carrying blaNDM-1 and blaSPM-1 genes in France32 and Brazil,8 respectively. The other four major complexes (C – F) also correspond to ‘high-risk clones’, previously reported as hosts for MBLs. ST233 (complex C) isolates with blaVIM-2 have been found in Norway (imported from Ghana),11 Japan,33 South Africa34 and, carrying a blaIMP-1 variant, in Singapore.35 We likewise consistently saw blaVIM-2 in complex C isolates. ST654 (complex D) has been reported in Sweden (imported from Tunisia) carrying blaVIM-2 genes,11 in Singapore carrying blaIMP-1 and blaIMP-2635 and in Argentina carrying a KPC carbapenemase.4 We predominantly saw blaVIM-2 in ST654 isolates, but one isolate had blaNDM-1. The sole ST964 isolate (also complex D) had blaIMP-1, as also reported in this ST in Singapore.35 We saw only blaVIM-2 in representatives of
6 of 8
complex E (ST357), but this ST has previously been reported with blaVIM-2 or blaIMP-7 genes in a hospital in the Czech Republic,36 whilst a few ST357 isolates with blaIMP-7 were reported in Poland.37 ST773 (complex F) has been recently reported among Indian isolates carrying blaVIM-2.31 Interestingly, all complex F isolates tested here also had blaVIM-2 and four isolates were from patients who had recently travelled to India; travel history for the other nine patients was not available. The remaining VNTR profiles, accounting for 14% of MBL producers, were diverse and probably represent separate acquisitions of blaVIM and blaIMP genes. Ongoing research indicates a variety of VIM-containing class 1 integron structures amongst these UK MBL producers. Each of complexes A, C, D and F (STs 111, 233, 654 and 773) typically have different predominant blaVIM-2-carrying integrons, with more diverse types seen amongst isolates belonging to complexes B and E (STs 235 and 357). This included the blaVIM-2carrying integron seen in isolates of complex C (ST233) and also seen in some representatives of complex B (ST235) (L. L. Wright, unpublished results). Although comprehensive epidemiological data are lacking, VIM-type MBLs generally are the predominant carbapenemases seen in P. aeruginosa in Europe, with only sporadic isolations and/or local spread of strains producing IMP or NDM types. The 22 isolates harbouring blaIMP genes belonged to diverse VNTR types, with diverse blaIMP alleles and no known epidemiological links between most isolates. Just two blaNDM-1-positive isolates were found, both with different VNTR types. These IMP and NDM
Downloaded from http://jac.oxfordjournals.org/ at University of Tasmania Library on September 3, 2014
Complex C (ST233)
JAC
MBL-producing P. aeruginosa in the UK
Acknowledgements Part of this work was presented at the Fifty-third Interscience Conference on Antimicrobial Agents and Chemotherapy, Denver, CO, USA, 2013 (Abstract C2-1597). We thank staff at the Antimicrobial Resistance and Healthcare Associated Infections Reference Unit for carrying out PFGE and some of the VNTR typing and MBL detection, performed as part of the reference service. We are grateful to the UK hospital laboratories for submitting isolates to us and to the BSAC for allowing us to access isolates from their Bacteraemia Surveillance Programme.
Funding This work was supported by Public Health England through a competitively awarded PhD studentship.
Transparency declarations D. M. L. and N. W. have received research grants and/or fees from various pharmaceutical companies. D. M. L. also holds shares in several
pharmaceutical companies. None of these poses a conflict of interest with this work. All other authors: none to declare.
References 1 Breidenstein EBM, de la Fuente-Nunez C, Hancock REW. Pseudomonas aeruginosa: all roads lead to resistance. Trends Microbiol 2011; 19: 419–26. 2 Sevillano E, Gallego L, Garcı´a-Lobo JM. First detection of the OXA-40 carbapenemase in P. aeruginosa isolates, located on a plasmid also found in A. baumannii. Pathol Biol (Paris) 2009; 57: 493–5. 3 Viedma E, Juan C, Acosta J et al. Nosocomial spread of colistin-onlysensitive sequence type 235 Pseudomonas aeruginosa isolates producing the extended-spectrum b-lactamases GES-1 and GES-5 in Spain. Antimicrob Agents Chemother 2009; 53: 4930– 3. 4 Pasteran F, Faccone D, Gomez S et al. Detection of an international multiresistant clone belonging to sequence type 654 involved in the dissemination of KPC-producing Pseudomonas aeruginosa in Argentina. J Antimicrob Chemother 2012; 67: 1291 –3. 5 Cornaglia G, Giamarellou H, Rossolini GM. Metallo-b-lactamases: a last frontier for b-lactams? Lancet Infect Dis 2011; 11: 381– 93. 6 Woodford N, Turton JF, Livermore DM. Multiresistant Gram-negative bacteria: the role of high-risk clones in the dissemination of antibiotic resistance. FEMS Microbiol Rev 2011; 35: 736–55. 7 Edelstein MV, Skleenova EN, Shevchenko OV et al. Spread of extensively resistant VIM-2-positive ST235 Pseudomonas aeruginosa in Belarus, Kazakhstan, and Russia: a longitudinal epidemiological and clinical study. Lancet Infect Dis 2013; 13: 1 –10. 8 Silva FM, Carmo MS, Silbert S et al. SPM-1-producing Pseudomonas aeruginosa: analysis of the ancestor relationship using multilocus sequence typing, pulsed-field gel electrophoresis, and automated ribotyping. Microb Drug Resist 2011; 17: 215–20. 9 Ellington MJ, Kistler J, Livermore DM et al. Multiplex PCR for rapid detection of genes encoding acquired metallo-b-lactamases. J Antimicrob Chemother 2007; 59: 321–2. 10 Mushtaq S, Irfan S, Sarma JB et al. Phylogenetic diversity of Escherichia coli strains producing NDM-type carbapenemases. J Antimicrob Chemother 2011; 66: 2002– 5. 11 Samuelsen O, Toleman MA, Sundsfjord A et al. Molecular epidemiology of metallo-b-lactamase-producing Pseudomonas aeruginosa isolates from Norway and Sweden shows import of international clones and local clonal expansion. Antimicrob Agents Chemother 2010; 54: 346–52. 12 Turton JF, Turton SE, Yearwood L et al. Evaluation of a nine-locus variable-number tandem-repeat scheme for typing of Pseudomonas aeruginosa. Clin Microbiol Infect 2010; 16: 1111 –6. 13 Curran B, Jonas D, Grundmann H et al. Development of a multilocus sequence typing scheme for the opportunistic pathogen Pseudomonas aeruginosa. J Clin Microbiol 2004; 42: 5644 –9. 14 Martin K, Baddal B, Mustafa N et al. Clusters of genetically similar isolates of Pseudomonas aeruginosa from multiple hospitals in the UK. J Med Microbiol 2013; 62: 988–1000. 15 Pitt TL, Livermore DM, Pitcher D et al. Multiresistant serotype O12 Pseudomonas aeruginosa: evidence for a common strain in Europe. Epidemiol Infect 1989; 103: 565– 76. 16 Van der Bij AK, Van der Zwan D, Peirano G et al. Metallo-b-lactamaseproducing Pseudomonas aeruginosa in the Netherlands: the nationwide emergence of a single sequence type. Clin Microbiol Infect 2012; 18: E369–72.
7 of 8
Downloaded from http://jac.oxfordjournals.org/ at University of Tasmania Library on September 3, 2014
MBL-producing isolates may have been imported from outside the UK, or acquired locally, with local spread at a small number of sites. Unfortunately, data on patient travel were not available for most isolates, but single isolates carrying blaNDM and blaIMP were from patients who had travelled to India and Pakistan, respectively. MBL-positive P. aeruginosa were referred from around half the hospital laboratories in the UK, with all six major complexes found in multiple UK regions. Referral of suspect isolates is not mandatory and some likely MBL producers were no longer viable in our archives, so the numbers reported here underestimate the true incidence of MBL-producing P. aeruginosa. They were rare at most referring sites, but a few sites did have persistent problems with single clones. These include an outbreak of ST111 isolates at London_17 and of ST654 isolates at South East_6, both associated with contamination of the waste-water networks,38 and an outbreak of ST357 P. aeruginosa at North West_15, where the strain, which also produced a VEB-1a ESBL, was thought to have been imported via a patient transferred from an Indian hospital but to have acquired the VIM MBL locally in the UK.39 In contrast, two laboratories (London_7 and London_12) referred MBL-positive isolates of diverse types over 7 and 8 year periods, implying that MBL-producing P. aeruginosa had been introduced repeatedly. Since these ‘high-risk clones’ are reported amongst MBLproducing P. aeruginosa worldwide, it is important not to assume that UK isolates with the same VNTR or ST profile are directly related unless this view is supported by evidence of epidemiological links between affected patients; it is just as likely that cases could represent repeated imports of the same clone from different sources. Although MBL-producing P. aeruginosa have so far caused outbreaks at few UK hospitals, those that have occurred have been linked to environmental reservoirs within the hospitals, including waste-water networks, as well as to patient-to-patient cross-infection.
Wright et al.
17 Liakopoulos A, Mavroidi A, Katsifas EA et al. Carbapenemase-producing Pseudomonas aeruginosa from central Greece: molecular epidemiology and genetic analysis of class I integrons. BMC Infect Dis 2013; 13: 505. 18 Edalucci E, Spinelli R, Dolzani L et al. Acquisition of different carbapenem resistance mechanisms by an epidemic clonal lineage of Pseudomonas aeruginosa. Clin Microbiol Infect 2008; 14: 88– 90. 19 Sardelic S, Bedenic B, Colinon-Dupuich C et al. Infrequent finding of metallo-b-lactamase VIM-2 in carbapenem-resistant Pseudomonas aeruginosa strains from Croatia. Antimicrob Agents Chemother 2012; 56: 2746– 9.
21 Hentschke M, Goritzka V, Campos CB et al. Emergence of carbapenemases in Gram-negative bacteria in Hamburg, Germany. Diagn Microbiol Infect Dis 2011; 71: 312–5. 22 Libisch B, Watine J, Balogh B et al. Molecular typing indicates an important role for two international clonal complexes in dissemination of VIM-producing Pseudomonas aeruginosa clinical isolates in Hungary. Res Microbiol 2008; 159: 162–8. 23 Koutsogiannou M, Drougka E, Liakopoulos A et al. Spread of multidrug-resistant Pseudomonas aeruginosa clones in a university hospital. J Clin Microbiol 2013; 51: 665–8. 24 Fournier D, Jeannot K, Robert-Nicoud M et al. Spread of the blaIMP-13 gene in French Pseudomonas aeruginosa through sequence types ST621, ST308 and ST111. Int J Antimicrob Agents 2012; 40: 562– 3. 25 Correa A, Montealegre MC, Mojica MF et al. First report of a Pseudomonas aeruginosa isolate coharboring KPC and VIM carbapenemases. Antimicrob Agents Chemother 2012; 56: 5422– 3. 26 Maatallah M, Cheriaa J, Backhrouf A et al. Population structure of Pseudomonas aeruginosa from five Mediterranean countries: evidence for frequent recombination and epidemic occurrence of CC235. PLoS One 2011; 6: e25617. 27 Glupczynski Y, Bogaerts P, Deplano A et al. Detection and characterization of class A extended-spectrum-b-lactamase-producing Pseudomonas aeruginosa isolates in Belgian hospitals. J Antimicrob Chemother 2010; 65: 866–71. 28 Juan C, Beceiro A, Gutie´rrez O et al. Characterization of the new metallo-b-lactamase VIM-13 and its integron-borne gene from a
8 of 8
29 Kitao T, Tada T, Tanaka M et al. Emergence of a novel multidrug-resistant Pseudomonas aeruginosa strain producing IMP-type metallo-b-lactamases and AAC(6′ )-Iae in Japan. Int J Antimicrob Agents 2012; 39: 518–21. 30 Seok Y, Bae IK, Jeong SH et al. Dissemination of IMP-6 metallo-b-lactamase-producing Pseudomonas aeruginosa sequence type 235 in Korea. J Antimicrob Chemother 2011; 66: 2791 –6. 31 Kim MJ, Bae IK, Jeong SH et al. Dissemination of metallo-b-lactamaseproducing Pseudomonas aeruginosa of sequence type 235 in Asian countries. J Antimicrob Chemother 2013; 68: 2820 –4. 32 Janvier F, Jeannot K, Tesse´ S et al. Molecular characterization of blaNDM-1 in a sequence type 235 Pseudomonas aeruginosa isolate from France. Antimicrob Agents Chemother 2013; 57: 3408– 11. 33 Tsutsui A, Suzuki S, Yamane K et al. Genotypes and infection sites in an outbreak of multidrug-resistant Pseudomonas aeruginosa. J Hosp Infect 2011; 78: 317–22. 34 Jacobson RK, Minenza N, Nicol M et al. VIM-2 metallo-b-lactamaseproducing Pseudomonas aeruginosa causing an outbreak in South Africa. J Antimicrob Chemother 2012; 67: 1797 –8. 35 Koh TH, Khoo CT, Tan TT et al. Multilocus sequence types of carbapenem-resistant Pseudomonas aeruginosa in Singapore carrying metallo-b-lactamase genes, including the novel blaIMP-26 gene. J Clin Microbiol 2010; 48: 2563– 4. 36 Papagiannitsis CC, Studentova V, Ruzicka F et al. Molecular characterization of metallo-b-lactamase-producing Pseudomonas aeruginosa in a Czech hospital (2009 – 2011). J Med Microbiol 2013; 62: 945– 7. 37 Hraba´k J, Cervena´ D, Izdebski R et al. Regional spread of Pseudomonas aeruginosa ST357 producing IMP-7 metallo-b-lactamase in Central Europe. J Clin Microbiol 2011; 49: 474– 5. 38 Breathnach AS, Cubbon MD, Karunaharan RN et al. Multidrug-resistant Pseudomonas aeruginosa outbreaks in two hospitals: association with contaminated hospital waste-water systems. J Hosp Infect 2012; 82: 19 –24. 39 Woodford N, Zhang J, Kaufmann ME et al. Detection of Pseudomonas aeruginosa isolates producing VEB-type extended-spectrum b-lactamases in the United Kingdom. J Antimicrob Chemother 2008; 62: 1265– 8.
Downloaded from http://jac.oxfordjournals.org/ at University of Tasmania Library on September 3, 2014
20 Viedma E, Estepa V, Juan C et al. Comparison of local features from two Spanish hospitals reveals common and specific traits at multiple levels of the molecular epidemiology of metallo-b-lactamase-producing Pseudomonas spp. Antimicrob Agents Chemother 2014; 58: 4992.
Pseudomonas aeruginosa clinical isolate in Spain. Antimicrob Agents Chemother 2008; 52: 3589– 96.
J Antimicrob Chemother doi:10.1093/jac/dku339
Dominance of international ‘high-risk clones’ among metallob-lactamase-producing Pseudomonas aeruginosa in the UK Laura L. Wright1,2*, Jane F. Turton1, David M. Livermore1,2, Katie L. Hopkins1 and Neil Woodford1 1
Antimicrobial Resistance and Healthcare Associated Infections Reference Unit, Public Health England, Colindale, London NW9 5EQ, UK; 2 Norwich Medical School, University of East Anglia, Norwich, Norfolk NR4 7TJ, UK
Received 9 June 2014; returned 10 July 2014; revised 31 July 2014; accepted 4 August 2014 Objectives: Carbapenem-resistant isolates of Pseudomonas aeruginosa producing metallo-b-lactamases (MBLs) are increasingly reported worldwide and often belong to particular ‘high-risk clones’. This study aimed to characterize a comprehensive collection of MBL-producing P. aeruginosa isolates referred to the UK national reference laboratory from multiple UK laboratories over a 10 year period. Methods: Isolates were referred to the UK national reference laboratory between 2003 and 2012 for investigation of resistance mechanisms and/or outbreaks. MBL genes were detected by PCR. Typing was carried out by ninelocus variable-number tandem repeat (VNTR) analysis and MLST. Results: MBL-producing P. aeruginosa isolates were referred from 267 source patients and 89 UK laboratories. The most common isolation sites were urine (24%), respiratory (18%), wounds (17%) and blood (13%). VIM-type MBLs predominated (91% of all MBLs found), but a few IMP- and NDM-type enzymes were also identified. Diverse VNTR types were seen, but 86% of isolates belonged to six major complexes. MLST of representative isolates from each complex showed that they corresponded to STs 111, 233, 235, 357, 654 and 773, respectively. Isolates belonging to these complexes were received from between 9 and 25 UK referring laboratories each. Conclusions: The incidence of MBL-producing P. aeruginosa is increasing in the UK. The majority of these isolates belong to several ‘high-risk clones’, which have been previously reported internationally as host clones of MBLs. Keywords: VIM, ST111, ST235, carbapenemases, MBLs
Introduction Pseudomonas aeruginosa is a common opportunistic pathogen responsible for many hospital-acquired infections. Carbapenems are important antibiotics for treatment of these infections, but resistance to these agents is increasing worldwide. Although most carbapenem resistance in P. aeruginosa arises by mutations that lead to the loss of the porin OprD or up-regulation of efflux pumps (e.g. MexAB-OprM),1 carbapenemases are also increasingly reported. Serine carbapenemases of the KPC, GES and OXA families have occasionally been reported in this species, with limited geographical scatter,2 – 4 but metallo-b-lactamases (MBLs), particularly VIM and IMP types, are the most widespread and have been reported globally.5 MBLs hydrolyse almost all b-lactam antibiotics, including carbapenems, and most MBL-producing P. aeruginosa strains are resistant to other antibiotic classes, including fluoroquinolones and aminoglycosides, often leaving polymyxins as the sole therapeutic options. The MBL genes often reside on mobile genetic elements able to be transmitted between strains,5 but are
commonly associated with multiresistant ‘high-risk clones’. These have been identified in several bacterial species, and in the case of P. aeruginosa the most commonly reported ‘high-risk clones’ belong to STs 111, 235 and 175.6 Recent studies report major dissemination of an ST235 lineage with the VIM-2 MBL in Russia, Belarus and Kazakhstan,7 and of ST277 with the SPM-1 MBL in Brazil.8 In recent years, the number of MBL-producing P. aeruginosa referred to the UK national reference laboratory has steadily risen. We sought to determine the contribution of internationally recognized ‘high-risk clones’ to this increase.
Materials and methods Bacterial isolates Three-hundred and thirty-four MBL-positive isolates were identified amongst P. aeruginosa isolates referred from UK hospital laboratories to PHE’s Antimicrobial Resistance and Healthcare Associated Infections Reference Unit (formerly the HPA’s ARMRL and LHCAI laboratories)
# Crown copyright 2014.
1 of 8
Downloaded from http://jac.oxfordjournals.org/ at University of Tasmania Library on September 3, 2014
*Corresponding author. Antimicrobial Resistance and Healthcare Associated Infections Reference Unit, Public Health England, 61 Colindale Avenue, London NW9 5EQ, UK. Tel: +44(0)2083276764; E-mail: [email protected]
Wright et al.
between 2003 and 2012 for susceptibility testing, investigation of resistance mechanisms and/or strain typing. Isolates were identified as MBL producers by PCR, either at the time of referral (304 isolates) or by retrospective screening of isolates submitted from 2009 – 12 that shared similar variable-number tandem repeat (VNTR) types with other MBLproducing isolates (30 isolates). Demographic and isolation site data were obtained from the PHE’s laboratory information management system. Referring laboratories were assigned codes to indicate the UK region and given a unique number within the region, in the format ‘region_number’ (e.g. North West_1). As a comparator group, all 209 P. aeruginosa isolates collected in 2011 as part of the BSAC’s Bacteraemia Surveillance Programme from hospitals in the UK and Irish Republic were also studied (http://www.bsacsurv.org/).
Genes encoding VIM-, IMP-, SPM-, GIM- and SIM-type MBLs were sought by multiplex PCR, as detailed by Ellington et al.9 Genes for NDM-type MBLs were sought by a single PCR, as previously described.10 Sequencing of blaVIM and blaIMP MBL genes was carried out using previously described primers specific to MBL genes and class 1 integrons;11 sequencing of blaNDM genes was with primers NDM-orfF (5′ -ATGGAATTGCCCAATATTATG-3′ ) and NDM-orfR (5′ -TCAGCGCAGCTTGTCGGCCA-3′ ).
Typing PFGE was the routine typing and outbreak investigation method used in the reference laboratory for P. aeruginosa between 2003 and 2009, and was carried out using SpeI-digested genomic DNA.12 Nine-locus VNTR analysis became the routine method from 2009 onwards and was performed as previously described.12 Minimum spanning trees were produced using Bionumerics software v6.1 (Applied Maths, Sint-Martens-Latem, Belgium). MLST analysis was undertaken as described by Curran et al.;13 results were analysed using Bionumerics software and STs were assigned using the P. aeruginosa MLST database (http://pubmlst.org/paeruginosa/).
Patient demographics The 334 MBL-positive isolates were from a total of 267 patients, with VNTR profiles remaining consistent when multiple isolates were received from the same patient. Accordingly, one isolate per patient was selected for further study, leaving 267 nonduplicate-patient isolates from 89 UK laboratories. Ages of the source patients ranged from 0 to 94, mean 54 years; proportions in age bands were as follows: 0 – 2 years, 4%; 3 – 15 years, 2%; 16 – 29 years, 6%; 30 – 44 years, 11%; 45 – 59 years, 22%; 60 – 74 years, 32%; and .74 years, 12% (age was unknown for the remaining 11%). Predominant isolation sites were urine (24%), respiratory (18%), wounds (17%), blood (13%) and indwelling devices (7%), whereas few were from skin (3%) and faecal (3%) specimens; the remaining isolates were from other (8%) or unknown (7%) isolation sites. Sixty percent of patients were male, 35% female and for 5% gender was not stated. blaVIM genes alone were detected in 243 isolates, blaIMP in 22 isolates and blaNDM in 1 isolate; 1 further isolate had both blaVIM and blaNDM. Of the 89 referring laboratories, 83 submitted isolates from fewer than 10 patients over the 10 year period, with 71 submitting only one or two isolates each. The remaining six laboratories, which submitted between 13 and 31 isolates each, collectively accounted for 43% of all patients (116/267).
Typing Among the 267 non-duplicate-patient isolates, 232 had VNTR data available or generated in this study, and a minimum spanning tree based on these data is shown in Figure 1. The remaining 35 isolates were not available in the archive for retrospective VNTR typing, but 19 had previously determined PFGE profiles (not shown) identical to those of isolates with known VNTR types
D B F A
E C
Figure 1. Minimum spanning tree, based on clustering at the first eight VNTR loci for MBL-positive P. aeruginosa, with one isolate included per patient (n ¼232). The six main complexes, A–F, are labelled. Coloured segments of the circles indicate laboratories that submitted three or more isolates and white segments represent laboratories submitting one or two isolates. The diameter of the circle represents the number of isolates with that VNTR profile. Shading indicates complexes. Thick solid lines represent single-locus variants while thin solid lines and dotted lines represent multilocus variants.
2 of 8
Downloaded from http://jac.oxfordjournals.org/ at University of Tasmania Library on September 3, 2014
Analysis of MBL genes
Results
JAC
MBL-producing P. aeruginosa in the UK
Table 1. Typing data for the six main VNTR complexes identified (n ¼251) No. of different VNTR profiles
A
11,3,4,3,2,2,x,4,x
6
B
13,3,6,4,5,1,x,2,x
C D
E F Others
MLST type(s) (no. of isolates tested)
No. of isolatesb
No. of submitting laboratories
MBLs detected (no. of isolates)
ST111 (11)
75
25
16
ST235 (18)
52
25
12,3,4,5,3,1,x,2,x 11,3,2,15,3,1,x,3,x
11 6
ST233 (10) ST654 (10), ST964 (1)
26 19
16 11
13,2,1,5,2,3,6,x,x 12,4,6,5,3,1,10,x,x diverse
7 3 26
ST357 (9) ST773 (5) not done
30 13 36
9 11 25
VIM (70) IMP (5) VIM (46) IMP (6) VIM (26) VIM (17) IMP (1) NDM (1) VIM (30) VIM (13) VIM (25) IMP (10) VIM and NDM (1)
a
x represents loci where the repeat number varies between isolates within a complex. One isolate per patient was included; these numbers include 4 isolates (complex B), 14 isolates (complex E) and 1 isolate (complex F) where the MBL-positive organisms were no longer available in the archive for VNTR analysis, but which were previously found to share a PFGE profile, and are from the same hospital outbreak as other isolates in the respective complex. Isolates were also received from an additional 39 patients at London_17 with a PFGE profile corresponding to complex A. These are not included here as they had not been screened for MBL genes and were no longer available in our archives.
b
from a suspected outbreak at the same hospital. These isolates were assumed to belong to the corresponding VNTR complex and are included in Table 1. The remaining 16/35 unavailable isolates had no typing data, or belonged to PFGE types unique to their respective hospital; these latter isolates, from 12 laboratories, are excluded from Table 1; all carried blaVIM. Six VNTR complexes (designated A – F) accounted for 86% of the 251 isolates with VNTR data available or inferred from PFGE data. Isolates belonging to these six complexes were referred from between 9 and 25 laboratories and, owing to their predominance, became the focus of further study. Sixty-four representatives, covering the variation in VNTR profile within each complex, were selected for MLST analysis. Isolates belonging to complexes A, B, C, D, E and F were found to belong to ST111, ST235, ST233, ST654/ST964, ST357 and ST773, respectively (Table 1). Ten of 11 MLST-typed complex D isolates belonged to ST654, but one belonged to ST964; this is a single-locus variant differing only in the acsA allele, where ST964 has allele 145 with a single CT substitution compared with allele 17 in ST654. Table 2 shows the distribution of each of the major complexes among referring laboratories. The largest group of isolates was complex A (corresponding to ST111; VNTR type 11,3,4,3,2,2,x,4,x, where x is variable), with 75 representatives. It included isolates from 25 laboratories spread across the UK, but with more than half of the isolates coming from outbreaks at London_17 and Wales_1, referring 29 and 13 isolates, respectively, all with blaVIM. Most other complex A isolates (28/33) also had blaVIM, but five, from two laboratories, had blaIMP. VNTR profiles were highly conserved amongst all complex A isolates, with most only differing by repeat numbers between six and eight at locus 61. Complex B (ST235; 13,3,6,4,5,1,x,2,x) was the second largest group, comprising 52 isolates referred from 25 different laboratories. There were potential outbreaks at sites London_13 and
Scotland_2, referring eight and six isolates, respectively, all with blaVIM; these groups were received over 7 months and 2 years, respectively. Another laboratory (London_7) also submitted isolates carrying blaVIM from eight patients over an 8 year period. The remaining 30 complex B isolates were from 22 laboratories, each submitting 1 to 3 isolates; 24 isolates had blaVIM and 6 had blaIMP. Twenty-six isolates belonged to complex C (ST233; 12,3,4,5, 3,1,x,2,x). These all carried blaVIM and were from 16 laboratories, which submitted one to four representatives each. Nineteen isolates from 11 laboratories belonged to complex D (ST654; 11,3,2,15,3,1,x,3,x); 7 of these were submitted from South East_6 and all had blaVIM and shared an identical VNTR profile, whilst 1 or 2 isolates were referred from each of the remaining 10 laboratories; 10 isolates had blaVIM, 1 isolate had blaIMP and 1 isolate had blaNDM. Most of the 30 complex E isolates (ST357; 13,2,1,5,2,3,6,x,x), all with blaVIM, were from an outbreak at North West_15, with 22 isolates referred over a 9 year period; 12 of these 22 were submitted in 2007. Each of the remaining eight complex E isolates was referred from a different laboratory. Finally, 13 isolates, all with blaVIM, belonging to complex F (ST773; 12,4,6,5,3,1,10,x,x) were received from 11 laboratories in diverse areas of the UK. Representative isolates were selected to cover the VNTR variation seen in blaVIM-positive isolates from the six major complexes. In complexes A, C, D, E and F (STs 111, 233, 654, 357 and 773) all representatives had blaVIM-2 genes, whereas eight isolates selected from complex B (ST235) variously had blaVIM-1 (two isolates), blaVIM-2 (four isolates), blaVIM-4 (one isolate) or blaVIM-6 (one isolate). Amongst seven representatives of the 22 blaIMP-positive isolates, blaIMP-1 (three isolates), blaIMP-7 (one isolate), blaIMP-10 (one isolate) and blaIMP-13 (two isolates) alleles were seen. Both blaNDM-positive isolates had the blaNDM-1 allele.
3 of 8
Downloaded from http://jac.oxfordjournals.org/ at University of Tasmania Library on September 3, 2014
VNTR typea
VNTR complex
Wright et al.
Table 2. Geographical and temporal distribution of MBL-producing P. aeruginosa belonging to the six major complexes (215 isolates) among referring laboratories Major contributors (≥5 referred isolates)
VNTR Complex
referring laboratorya
MBL genes detected
Minor contributors 2 laboratories with 4 representatives each 2 laboratories with 3 representatives each 19 laboratories with 1 representative each 1 laboratory with 3 representatives 6 laboratories with 2 representatives each 15 laboratories with 1 representative each 3 laboratories with 2 representatives each 11 laboratories with 1 representative each 2 laboratories with 2 representatives each 8 laboratories with 1 representative each 8 laboratories with 1 representative each 2 laboratories with 2 representatives each 9 laboratories with 1 representative each
A (n¼75)
London_17c Wales_1
29 (39%) 13 (17%)
80 months 18 months
all with blaVIM all with blaVIM
B (n¼52)
London_7 London_13 Scotland_2 London_7 London_13 South East_6
8 (15%) 8 (15%) 6 (12%) 5 (19%) 4 (15%) 7 (37%)
87 months 7 months 27 months 27 months 18 months 36 months
all with blaVIM all with blaVIM all with blaVIM all with blaVIM all with blaVIM all with blaVIM
22 (73%) 103 months no major contributors
all with blaVIM
C (n¼26) D (n¼19) E (n¼30) F (n¼13)
North West_15
a
Referring laboratories are coded in the format ‘UK region_number’. One isolate per patient is included; these numbers include 4 isolates (complex B), 14 isolates (complex E) and 1 isolate (complex F) that were not available in the archive for VNTR analysis, but which shared a PFGE profile, and are from the same hospital outbreak as other isolates in the complex. c Isolates were also received from an additional 39 patients at London_17 with a PFGE type corresponding to complex A. These are not included here as they had not been screened for MBL production and were no longer available in our archives. b
The 14% of isolates (n¼ 36) that did not belong to any of these six major complexes were diverse by VNTR, with 26 different VNTR profiles represented. They were referred from 24 laboratories across the UK, each submitting one to three isolates; 26 had blaVIM, 10 had blaIMP and one had both blaVIM and blaNDM genes.
Regional distribution of MBL-positive isolates The UK distribution of the 267 non-duplicate-patient isolates is shown in Figure 2. London accounted for 47% of these organisms, with MBL-positive isolates referred from 28 London region laboratories; all six major complexes (A – F) were represented and nine other VNTR types seen. Four London laboratories were amongst the six sites that submitted .10 MBL-positive isolates. London_17 submitted 29 complex A (ST111) isolates over 6 years, with source patients on multiple wards. London_13 had separate outbreaks of MBL-positive isolates belonging to complexes B and C (STs 235 and 233), with eight and four representatives, respectively. Each ‘outbreak’ lasted 1 –2 years separated by a 6 year gap. London_7 and London_12 referred 19 and 15 isolates, respectively, over 7 – 8 years, with isolates belonging to diverse VNTR complexes. Greater Manchester accounted for 12% of isolates (31/267). Isolates belonging to complex E (ST357) predominated and included all 22 from an outbreak at North West_13, although isolates belonging to complexes A, B and D (STs 111, 235 and 654) were also seen. In Sussex, Surrey and Kent half of the 18 isolates received belonged to complex D (ST654); 7 of these 9 isolates were from South East_6; the remaining 9 representatives from this region belonged to complexes B and C (STs 235 and 233) or
4 of 8
had other VNTR profiles. Wales accounted for 6% of isolates, mostly (13/15) from an outbreak of ST111 P. aeruginosa at Wales_1, although one ST111 isolate was from another laboratory and one ST773 isolate also was referred from the region. Finally, the 12 (4%) isolates submitted from Scotland belonged to complexes A, B, C and F (STs 111, 235, 233 and 773). Fewer than 10 isolates were referred from each of the remaining UK regions.
Typing of comparator P. aeruginosa isolates The comparator set, comprising P. aeruginosa isolates collected as part of the BSAC Bacteraemia Resistance Surveillance Programme in 2011, showed far greater diversity than the MBL producers, with 136 different VNTR profiles represented amongst the 209 isolates (Figure 3). Although 4.8% (10 isolates) were imipenem resistant, none produced MBLs. Two isolates shared VNTR profiles with complex B (ST235) and one shared a profile with complex C (ST233); the remaining 206 isolates did not share VNTR profiles with any of the major complexes identified amongst the MBL-producing isolates.
Discussion Most (86%) MBL-producing P. aeruginosa isolates referred to the UK reference laboratory between 2003 and 2012 belonged to six VNTR complexes, corresponding to internationally recognized ‘high-risk clones’, STs 111, 235, 233, 357, 654 and 773, and most had VIM-type MBLs. These findings are in striking contrast to the general population structure of P. aeruginosa, for which a
Downloaded from http://jac.oxfordjournals.org/ at University of Tasmania Library on September 3, 2014
number of isolatesb
time period over which isolates were referred
JAC
MBL-producing P. aeruginosa in the UK
London
126
28
Greater Manchester
6
Sussex, Surrey and Kent
5
Wales
2
Scotland
5
31 18 15 12
9 5
West Midlands
9 7
Norfolk, Suffolk, Cambridgeshire and Essex
7 3
Cumbria and Lancashire
7 3
Cheshire and Merseyside
7 6
Thames Valley
6 3
Lincolnshire, Leicestershire, Nottinghamshire and Derbyshire
5 2
Hampshire, Isle of Wight and Dorset
5 4 3 3
Bedfordshire, Hertfordshire and Northamptonshire Northern Ireland
2 2
North East
2 2
Avon, Gloucestershire and Wiltshire
2 2
Devon, Cornwall and Somerset
1 1 0
Isolates
A (ST111) D (ST654) Other types
Referring hospitals
50
100
B (ST235) E (ST357) No typing data
C (ST233) F (ST773)
150
Figure 2. Geographical sources of isolates in the UK and distribution of the six main complexes (A –F) in each of the five regions referring .10 isolates. This figure appears in colour in the online version of JAC and in black and white in the print version of JAC.
recent UK study14 shows considerable diversity, with overlap between environmental isolates and those causing clinical infection, although with a few prevalent clusters in diverse locations. The major complexes seen here were not prevalent amongst MBL-negative isolates in this previous study, nor were they prevalent in our comparator collection of (largely susceptible) P. aeruginosa isolates from the BSAC Bacteraemia Resistance Surveillance Programme. Rather, these ‘high-risk clones’ seem to represent a distinct subset of P. aeruginosa lineages, which may be successful precisely due to a particularly strong ability to acquire and/or maintain resistance genes compared with the general P. aeruginosa population. Notably, the reference laboratory has also received MBL-negative representatives corresponding to these six major complexes, including a few ST235 isolates carrying blaGES-5 and a single ST773 isolate with blaOXA-181, both genes encoding non-metallo-carbapenemases (J. F. Turton, K. L. Hopkins and N. Woodford, unpublished results). Moreover, in three of these complexes we encountered isolates that variously had either blaVIM or blaIMP genes and, in one complex (B; ST235), various different blaVIM alleles. Each of the six main VNTR complexes has been reported internationally as a host for MBLs. Complex A (corresponding to ST111)
was the largest group in our collection, with most isolates carrying blaVIM, generally blaVIM-2. VNTR profiles within this complex were highly similar, most differing only at locus 61, which is the most variable of the nine VNTR loci.12 ST111 is frequently recorded amongst MBL-producing P. aeruginosa, with ST111 isolates reported to be of serotype O12, which has been recognized as a common host of multiresistance in Europe since the 1980s.15,16 ST111 was the predominant ST type seen in a nationwide study of MBL-producing P. aeruginosa in the Netherlands in 2010 – 11,16 and was responsible for outbreaks affecting hospitals in central Greece17 and Italy.18 In all these settings, as in the UK, the ST111 isolates generally carried blaVIM-2 . Smaller numbers of ST111 isolates with blaVIM-2 have been reported in other European countries including Sweden,11 Croatia,19 Spain20 and Germany,21 whilst isolates with this ST have been associated with blaVIM-4 in Hungary,22 blaVIM-1 in Greece23 and blaIMP-13 in France.24 Outside Europe, an isolate of ST111 from a patient in Colombia was recently found to carry both blaVIM-2 and blaKPC-2.25 The second most prevalent complex (B) in our collection corresponded to ST235. Isolates in this complex had more diverse VNTR profiles than those in complex A and were more widely disseminated among referring laboratories, with no more than eight
5 of 8
Downloaded from http://jac.oxfordjournals.org/ at University of Tasmania Library on September 3, 2014
Yorkshire and Humber
Wright et al.
Novel cluster 2 Complex B (ST235)
Clone C (ST17) Novel cluster 1
Cluster A (ST27)
Cluster D (ST395)
Cluster E
Cluster H PA14 (ST253) Figure 3. Minimum spanning tree based on clustering at the first eight VNTR loci for P. aeruginosa isolates from the BSAC Bacteraemia Resistance Surveillance Programme (n¼209). The diameter of the circle represents the number of isolates with that VNTR profile. Shading indicates complexes. Thick solid lines represent single-locus variants while thin solid lines and dotted lines represent multilocus variants. Isolates corresponding to VNTR complexes B and C (STs 235 and 233, respectively) are shown in black circles. Isolates corresponding to previously reported clones as described by Martin et al.14 are indicated and had the following VNTR profiles: cluster A, 8,3,4,5,2,3,5,2,x; cluster D, 10,3,5,5,4,1,3,x,x; cluster E, 11,4,5,2,2,1,x,2,x; cluster H, 12,5,1,5,2,2,x,x,x; clone C, 11,4,5,2,2,1,x,2,x; and PA14, 12,2,1,5,5,2,x,5,x. Two novel clusters are indicated with VNTR profiles 12,5,5,5,4,3,7,6,x and 12,8,2,2,4,3,5,1,x, respectively.
from any single site. Again, most isolates had blaVIM, but with diverse blaVIM alleles identified. Together with the VNTR diversity, this carbapenemase diversity suggests multiple imports and/or acquisitions of blaVIM by ST235 P. aeruginosa. This variation is in contrast to the ST235 clone with blaVIM-2 that is widespread across Russia, Belarus and Kazakhstan.7 ST235 was the most commonly identified MBL-producing P. aeruginosa type in a study of five Mediterranean countries,26 and ST235 isolates with blaVIM-2 genes have also been reported in Spain,20 Croatia,19 Germany21 and Greece,23 blaVIM-4 in Hungary,22 Norway11 and Belgium27 and blaVIM-13 in Spain.28 In Asia, outbreaks of ST235 clones with blaIMP-6 genes have occurred in Japan29 and South Korea,30 with blaVIM-positive isolates with this ST also seen in Thailand, Malaysia, Sri Lanka and Korea.31 There are single reports of ST235 isolates carrying blaNDM-1 and blaSPM-1 genes in France32 and Brazil,8 respectively. The other four major complexes (C – F) also correspond to ‘high-risk clones’, previously reported as hosts for MBLs. ST233 (complex C) isolates with blaVIM-2 have been found in Norway (imported from Ghana),11 Japan,33 South Africa34 and, carrying a blaIMP-1 variant, in Singapore.35 We likewise consistently saw blaVIM-2 in complex C isolates. ST654 (complex D) has been reported in Sweden (imported from Tunisia) carrying blaVIM-2 genes,11 in Singapore carrying blaIMP-1 and blaIMP-2635 and in Argentina carrying a KPC carbapenemase.4 We predominantly saw blaVIM-2 in ST654 isolates, but one isolate had blaNDM-1. The sole ST964 isolate (also complex D) had blaIMP-1, as also reported in this ST in Singapore.35 We saw only blaVIM-2 in representatives of
6 of 8
complex E (ST357), but this ST has previously been reported with blaVIM-2 or blaIMP-7 genes in a hospital in the Czech Republic,36 whilst a few ST357 isolates with blaIMP-7 were reported in Poland.37 ST773 (complex F) has been recently reported among Indian isolates carrying blaVIM-2.31 Interestingly, all complex F isolates tested here also had blaVIM-2 and four isolates were from patients who had recently travelled to India; travel history for the other nine patients was not available. The remaining VNTR profiles, accounting for 14% of MBL producers, were diverse and probably represent separate acquisitions of blaVIM and blaIMP genes. Ongoing research indicates a variety of VIM-containing class 1 integron structures amongst these UK MBL producers. Each of complexes A, C, D and F (STs 111, 233, 654 and 773) typically have different predominant blaVIM-2-carrying integrons, with more diverse types seen amongst isolates belonging to complexes B and E (STs 235 and 357). This included the blaVIM-2carrying integron seen in isolates of complex C (ST233) and also seen in some representatives of complex B (ST235) (L. L. Wright, unpublished results). Although comprehensive epidemiological data are lacking, VIM-type MBLs generally are the predominant carbapenemases seen in P. aeruginosa in Europe, with only sporadic isolations and/or local spread of strains producing IMP or NDM types. The 22 isolates harbouring blaIMP genes belonged to diverse VNTR types, with diverse blaIMP alleles and no known epidemiological links between most isolates. Just two blaNDM-1-positive isolates were found, both with different VNTR types. These IMP and NDM
Downloaded from http://jac.oxfordjournals.org/ at University of Tasmania Library on September 3, 2014
Complex C (ST233)
JAC
MBL-producing P. aeruginosa in the UK
Acknowledgements Part of this work was presented at the Fifty-third Interscience Conference on Antimicrobial Agents and Chemotherapy, Denver, CO, USA, 2013 (Abstract C2-1597). We thank staff at the Antimicrobial Resistance and Healthcare Associated Infections Reference Unit for carrying out PFGE and some of the VNTR typing and MBL detection, performed as part of the reference service. We are grateful to the UK hospital laboratories for submitting isolates to us and to the BSAC for allowing us to access isolates from their Bacteraemia Surveillance Programme.
Funding This work was supported by Public Health England through a competitively awarded PhD studentship.
Transparency declarations D. M. L. and N. W. have received research grants and/or fees from various pharmaceutical companies. D. M. L. also holds shares in several
pharmaceutical companies. None of these poses a conflict of interest with this work. All other authors: none to declare.
References 1 Breidenstein EBM, de la Fuente-Nunez C, Hancock REW. Pseudomonas aeruginosa: all roads lead to resistance. Trends Microbiol 2011; 19: 419–26. 2 Sevillano E, Gallego L, Garcı´a-Lobo JM. First detection of the OXA-40 carbapenemase in P. aeruginosa isolates, located on a plasmid also found in A. baumannii. Pathol Biol (Paris) 2009; 57: 493–5. 3 Viedma E, Juan C, Acosta J et al. Nosocomial spread of colistin-onlysensitive sequence type 235 Pseudomonas aeruginosa isolates producing the extended-spectrum b-lactamases GES-1 and GES-5 in Spain. Antimicrob Agents Chemother 2009; 53: 4930– 3. 4 Pasteran F, Faccone D, Gomez S et al. Detection of an international multiresistant clone belonging to sequence type 654 involved in the dissemination of KPC-producing Pseudomonas aeruginosa in Argentina. J Antimicrob Chemother 2012; 67: 1291 –3. 5 Cornaglia G, Giamarellou H, Rossolini GM. Metallo-b-lactamases: a last frontier for b-lactams? Lancet Infect Dis 2011; 11: 381– 93. 6 Woodford N, Turton JF, Livermore DM. Multiresistant Gram-negative bacteria: the role of high-risk clones in the dissemination of antibiotic resistance. FEMS Microbiol Rev 2011; 35: 736–55. 7 Edelstein MV, Skleenova EN, Shevchenko OV et al. Spread of extensively resistant VIM-2-positive ST235 Pseudomonas aeruginosa in Belarus, Kazakhstan, and Russia: a longitudinal epidemiological and clinical study. Lancet Infect Dis 2013; 13: 1 –10. 8 Silva FM, Carmo MS, Silbert S et al. SPM-1-producing Pseudomonas aeruginosa: analysis of the ancestor relationship using multilocus sequence typing, pulsed-field gel electrophoresis, and automated ribotyping. Microb Drug Resist 2011; 17: 215–20. 9 Ellington MJ, Kistler J, Livermore DM et al. Multiplex PCR for rapid detection of genes encoding acquired metallo-b-lactamases. J Antimicrob Chemother 2007; 59: 321–2. 10 Mushtaq S, Irfan S, Sarma JB et al. Phylogenetic diversity of Escherichia coli strains producing NDM-type carbapenemases. J Antimicrob Chemother 2011; 66: 2002– 5. 11 Samuelsen O, Toleman MA, Sundsfjord A et al. Molecular epidemiology of metallo-b-lactamase-producing Pseudomonas aeruginosa isolates from Norway and Sweden shows import of international clones and local clonal expansion. Antimicrob Agents Chemother 2010; 54: 346–52. 12 Turton JF, Turton SE, Yearwood L et al. Evaluation of a nine-locus variable-number tandem-repeat scheme for typing of Pseudomonas aeruginosa. Clin Microbiol Infect 2010; 16: 1111 –6. 13 Curran B, Jonas D, Grundmann H et al. Development of a multilocus sequence typing scheme for the opportunistic pathogen Pseudomonas aeruginosa. J Clin Microbiol 2004; 42: 5644 –9. 14 Martin K, Baddal B, Mustafa N et al. Clusters of genetically similar isolates of Pseudomonas aeruginosa from multiple hospitals in the UK. J Med Microbiol 2013; 62: 988–1000. 15 Pitt TL, Livermore DM, Pitcher D et al. Multiresistant serotype O12 Pseudomonas aeruginosa: evidence for a common strain in Europe. Epidemiol Infect 1989; 103: 565– 76. 16 Van der Bij AK, Van der Zwan D, Peirano G et al. Metallo-b-lactamaseproducing Pseudomonas aeruginosa in the Netherlands: the nationwide emergence of a single sequence type. Clin Microbiol Infect 2012; 18: E369–72.
7 of 8
Downloaded from http://jac.oxfordjournals.org/ at University of Tasmania Library on September 3, 2014
MBL-producing isolates may have been imported from outside the UK, or acquired locally, with local spread at a small number of sites. Unfortunately, data on patient travel were not available for most isolates, but single isolates carrying blaNDM and blaIMP were from patients who had travelled to India and Pakistan, respectively. MBL-positive P. aeruginosa were referred from around half the hospital laboratories in the UK, with all six major complexes found in multiple UK regions. Referral of suspect isolates is not mandatory and some likely MBL producers were no longer viable in our archives, so the numbers reported here underestimate the true incidence of MBL-producing P. aeruginosa. They were rare at most referring sites, but a few sites did have persistent problems with single clones. These include an outbreak of ST111 isolates at London_17 and of ST654 isolates at South East_6, both associated with contamination of the waste-water networks,38 and an outbreak of ST357 P. aeruginosa at North West_15, where the strain, which also produced a VEB-1a ESBL, was thought to have been imported via a patient transferred from an Indian hospital but to have acquired the VIM MBL locally in the UK.39 In contrast, two laboratories (London_7 and London_12) referred MBL-positive isolates of diverse types over 7 and 8 year periods, implying that MBL-producing P. aeruginosa had been introduced repeatedly. Since these ‘high-risk clones’ are reported amongst MBLproducing P. aeruginosa worldwide, it is important not to assume that UK isolates with the same VNTR or ST profile are directly related unless this view is supported by evidence of epidemiological links between affected patients; it is just as likely that cases could represent repeated imports of the same clone from different sources. Although MBL-producing P. aeruginosa have so far caused outbreaks at few UK hospitals, those that have occurred have been linked to environmental reservoirs within the hospitals, including waste-water networks, as well as to patient-to-patient cross-infection.
Wright et al.
17 Liakopoulos A, Mavroidi A, Katsifas EA et al. Carbapenemase-producing Pseudomonas aeruginosa from central Greece: molecular epidemiology and genetic analysis of class I integrons. BMC Infect Dis 2013; 13: 505. 18 Edalucci E, Spinelli R, Dolzani L et al. Acquisition of different carbapenem resistance mechanisms by an epidemic clonal lineage of Pseudomonas aeruginosa. Clin Microbiol Infect 2008; 14: 88– 90. 19 Sardelic S, Bedenic B, Colinon-Dupuich C et al. Infrequent finding of metallo-b-lactamase VIM-2 in carbapenem-resistant Pseudomonas aeruginosa strains from Croatia. Antimicrob Agents Chemother 2012; 56: 2746– 9.
21 Hentschke M, Goritzka V, Campos CB et al. Emergence of carbapenemases in Gram-negative bacteria in Hamburg, Germany. Diagn Microbiol Infect Dis 2011; 71: 312–5. 22 Libisch B, Watine J, Balogh B et al. Molecular typing indicates an important role for two international clonal complexes in dissemination of VIM-producing Pseudomonas aeruginosa clinical isolates in Hungary. Res Microbiol 2008; 159: 162–8. 23 Koutsogiannou M, Drougka E, Liakopoulos A et al. Spread of multidrug-resistant Pseudomonas aeruginosa clones in a university hospital. J Clin Microbiol 2013; 51: 665–8. 24 Fournier D, Jeannot K, Robert-Nicoud M et al. Spread of the blaIMP-13 gene in French Pseudomonas aeruginosa through sequence types ST621, ST308 and ST111. Int J Antimicrob Agents 2012; 40: 562– 3. 25 Correa A, Montealegre MC, Mojica MF et al. First report of a Pseudomonas aeruginosa isolate coharboring KPC and VIM carbapenemases. Antimicrob Agents Chemother 2012; 56: 5422– 3. 26 Maatallah M, Cheriaa J, Backhrouf A et al. Population structure of Pseudomonas aeruginosa from five Mediterranean countries: evidence for frequent recombination and epidemic occurrence of CC235. PLoS One 2011; 6: e25617. 27 Glupczynski Y, Bogaerts P, Deplano A et al. Detection and characterization of class A extended-spectrum-b-lactamase-producing Pseudomonas aeruginosa isolates in Belgian hospitals. J Antimicrob Chemother 2010; 65: 866–71. 28 Juan C, Beceiro A, Gutie´rrez O et al. Characterization of the new metallo-b-lactamase VIM-13 and its integron-borne gene from a
8 of 8
29 Kitao T, Tada T, Tanaka M et al. Emergence of a novel multidrug-resistant Pseudomonas aeruginosa strain producing IMP-type metallo-b-lactamases and AAC(6′ )-Iae in Japan. Int J Antimicrob Agents 2012; 39: 518–21. 30 Seok Y, Bae IK, Jeong SH et al. Dissemination of IMP-6 metallo-b-lactamase-producing Pseudomonas aeruginosa sequence type 235 in Korea. J Antimicrob Chemother 2011; 66: 2791 –6. 31 Kim MJ, Bae IK, Jeong SH et al. Dissemination of metallo-b-lactamaseproducing Pseudomonas aeruginosa of sequence type 235 in Asian countries. J Antimicrob Chemother 2013; 68: 2820 –4. 32 Janvier F, Jeannot K, Tesse´ S et al. Molecular characterization of blaNDM-1 in a sequence type 235 Pseudomonas aeruginosa isolate from France. Antimicrob Agents Chemother 2013; 57: 3408– 11. 33 Tsutsui A, Suzuki S, Yamane K et al. Genotypes and infection sites in an outbreak of multidrug-resistant Pseudomonas aeruginosa. J Hosp Infect 2011; 78: 317–22. 34 Jacobson RK, Minenza N, Nicol M et al. VIM-2 metallo-b-lactamaseproducing Pseudomonas aeruginosa causing an outbreak in South Africa. J Antimicrob Chemother 2012; 67: 1797 –8. 35 Koh TH, Khoo CT, Tan TT et al. Multilocus sequence types of carbapenem-resistant Pseudomonas aeruginosa in Singapore carrying metallo-b-lactamase genes, including the novel blaIMP-26 gene. J Clin Microbiol 2010; 48: 2563– 4. 36 Papagiannitsis CC, Studentova V, Ruzicka F et al. Molecular characterization of metallo-b-lactamase-producing Pseudomonas aeruginosa in a Czech hospital (2009 – 2011). J Med Microbiol 2013; 62: 945– 7. 37 Hraba´k J, Cervena´ D, Izdebski R et al. Regional spread of Pseudomonas aeruginosa ST357 producing IMP-7 metallo-b-lactamase in Central Europe. J Clin Microbiol 2011; 49: 474– 5. 38 Breathnach AS, Cubbon MD, Karunaharan RN et al. Multidrug-resistant Pseudomonas aeruginosa outbreaks in two hospitals: association with contaminated hospital waste-water systems. J Hosp Infect 2012; 82: 19 –24. 39 Woodford N, Zhang J, Kaufmann ME et al. Detection of Pseudomonas aeruginosa isolates producing VEB-type extended-spectrum b-lactamases in the United Kingdom. J Antimicrob Chemother 2008; 62: 1265– 8.
Downloaded from http://jac.oxfordjournals.org/ at University of Tasmania Library on September 3, 2014
20 Viedma E, Estepa V, Juan C et al. Comparison of local features from two Spanish hospitals reveals common and specific traits at multiple levels of the molecular epidemiology of metallo-b-lactamase-producing Pseudomonas spp. Antimicrob Agents Chemother 2014; 58: 4992.
Pseudomonas aeruginosa clinical isolate in Spain. Antimicrob Agents Chemother 2008; 52: 3589– 96.
