Journal of Global Antimicrobial Resistance 2 (2014) 43–47

Short communication

Antibiotic susceptibility and molecular epidemiology of Panton–Valentineleukocidin-positive meticillin-resistant Staphylococcus aureus:An international survey

Marina Macedo-Vinas a,*, John Conly a,b, Patrice Francois a, Richard Aschbacher c,Dominique S. Blanc d, Geoffrey Coombs e, George Daikos f, Benu Dhawan g, Joanna Empel h,Jerome Etienne i, Agnes Marie Sa Figueiredo j, George Golding & CNISP k, Lizhong Han l,Hong Bin Kim m, Robin Kock n, Anders Larsen o, Franziska Layer p, Janice Lo q,Tadashi Maeda r, Michael Mulvey k, Annalisa Pantosti s, Tomoo Saga r,Jacques Schrenzel a, Andrew Simor t, Robert Skov o, Miranda Van Rijen u,Hui Wang v, Zunita Zakaria w, Stephan Harbarth a

a Hopitaux Universitaires de Geneve, Geneva, Switzerlandb University of Calgary, Calgary, Canadac Laboratorio Aziendale de Microbiologia e Virologia, Bolzano, Italyd Centre Hospitalier Universitaire Vaudois, Lausanne, Switzerlande Curtin University, Perth, Australiaf Laiko General Hospital, Athens, Greeceg All India Institute of Medical Sciences, New Delhi, Indiah National Medicine Institute (NMI), Warsaw, Polandi National Reference Center for Staphylococci, Lyon, Francej Universidade Federal do Rio de Janeiro, Rio de Janeiro, Brazilk National Microbiology Laboratory and Canadian Nosocomial Infection Surveillance Program (CNISP), Winnipeg, Canadal Shanghai Ruijin Hospital, Shanghai, Chinam Seoul National University, Bundang Hospital, Seongnam, South Korean University Hospital Munster, Munster, Germanyo Statens Serum Institut, Copenhagen, Denmarkp National Reference Centre for Staphylococci and Enterococci, Robert Koch Institute, Wernigerode, Germanyq Public Health Laboratory Services Branch Centre for Health Protection, Hong Kong Special Administrative Regionr Toho University Omori Hospital, Tokyo, Japans Instituto Superiore di Sanita, Rome, Italyt Sunnybrook Health Sciences Centre, Toronto, Canadau Laboratory for Microbiology and Infection Control, Breda, The Netherlandsv Peking Union Medical College Hospital, Beijing, Chinaw University Veterinary Hospital, Universiti Putra, Serdang, Malaysia

A R T I C L E I N F O

Article history:

Received 21 May 2013

Accepted 8 August 2013

Keywords:

Meticillin-resistant Staphylococcus aureus

Panton–Valentine leukocidin

Antibiotic multiresistance

Molecular epidemiology

A B S T R A C T

The antibiotic susceptibility and molecular epidemiology of Panton–Valentine leukocidin (PVL)-positive

meticillin-resistant Staphylococcus aureus (MRSA) isolates reported from 17 countries in the Americas,

Europe and, Australia-Asia were analysed. Among a total of 3236 non-duplicate isolates, the lowest

susceptibility was observed to erythromycin in all regions. Susceptibility to ciprofloxacin showed large

variation (25%, 75% and 84% in the Americas, Europe and Australia-Asia, respectively). Two vancomycin-

intermediate PVL-positive MRSA isolates were reported, one from Hong Kong and the other from The

Netherlands. Resistance to trimethoprim/sulfamethoxazole and linezolid was <1%. Among 1798 MRSA

isolates from 13 countries that were tested for the requested 10 non-b-lactam antibiotics, 49.4% were

multisusceptible. However, multiresistant isolates (resistant to at least three classes of non-b-lactam

antibiotics) were reported from all regions. Sequence type 30 (ST30) was reported worldwide, whereas

ST80 and ST93 were exclusive to Europe and Australia, respectively. USA300 and related clones (ST8) are

progressively replacing the ST80 clone in several European countries. Eight major clusters were

Contents lists available at ScienceDirect

Journal of Global Antimicrobial Resistance

jo u rn al h om ep age: w ww.els evier .c o m/lo c ate / jg ar

* Corresponding author. Present address: Centro Nacional de Quemados CENAQUE, Hospital de Clınicas, Facultad de Medicina, Avenida Italia s/n. Piso 13, CP 11600

Montevideo, Uruguay. Tel.: +598 2 487 2020x859; fax: +598 2 487 4521.

E-mail addresses: mmacedo@cenaque.org.uy, mmacedov@gmail.com (M. Macedo-Vinas).

2213-7165/$ – see front matter � 2013 International Society for Chemotherapy of Infection and Cancer. Published by Elsevier Ltd. All rights reserved.

http://dx.doi.org/10.1016/j.jgar.2013.08.003

M. Macedo-Vinas et al. / Journal of Global Antimicrobial Resistance 2 (2014) 43–4744

discriminated by multilocus variable-number tandem repeat assay (MLVA), showing a certain

geographic specificity. PVL-positive MRSA isolates frequently remain multisusceptible to non-b-lactam

agents, but multiresistance is already prevalent in all regions. Surveillance of MRSA susceptibility

patterns should be monitored to provide clinicians with the most current information regarding changes

in resistance patterns.

� 2013 International Society for Chemotherapy of Infection and Cancer. Published by Elsevier Ltd. All

rights reserved.

1. Introduction

Meticillin-resistant Staphylococcus aureus (MRSA) is a majorhealthcare-associated (HA) pathogen and an emerging cause ofcommunity-associated (CA) infections in numerous countries [1].Historically, and in contrast to HA-MRSA, CA-MRSA has beendistinguished by being susceptible to most non-b-lactam anti-biotics [2], carrying new staphylococcal chromosomal cassettesmec elements (SCCmec IV or SCCmec V) [3,4] and often harbouringparticular exotoxins such as the Panton–Valentine leukocidin(PVL) [5,6]. Nevertheless, the presence of the PVL-encoding genes(lukF-PV and lukS-PV) is neither required [6] nor exclusive of CA-MRSA [7]. In addition, two recent epidemiological trends suggestthat the conventional classification of CA-MRSA may need to bereconsidered [6]. First, CA-MRSA-related clones are emerging asnosocomial pathogens [8]. Second, resistance to non-b-lactamantibiotics is increasingly reported, including intermediate-levelresistance to vancomycin as well as resistance to numerous otherantimicrobial classes [9,10].

We aimed to evaluate the current global situation of PVL-positive MRSA with regard to resistance patterns and molecularepidemiology. Thus, a multicentre survey examining a large seriesof PVL-positive MRSA isolates reported from various countriesaround the world was performed.

2. Materials and methods

2.1. Antibiotic resistance

A retrospective laboratory-based survey was performed usingdata sent from study sites in different countries around the worldthat agreed to participate. A convenience sample of 22 centres,groups or institutions (hereafter referred as ‘sites’) from 17countries in three regions was analysed. Sites were asked toprovide laboratory data between 2009 and 2010 according to theiravailability and to report on: the number of PVL-positive MRSAisolates; the number of these isolates that were susceptible toerythromycin (ERY), clindamycin (CLI), gentamicin (GEN), cipro-floxacin (CIP), tetracycline (TET), fusidic acid (FA), rifampicin (RIF),trimethoprim/sulfamethoxazole (SXT), vancomycin (VAN) andlinezolid (LZD); and the three most frequent antibiotic resistancepatterns. Resistance pattern analysis only included isolates fromsites that tested all antibiotics except for VAN, LZD and FA.Nevertheless, FA was included in the analysis whenever available.

Sites reported susceptibility testing results according to theirlocal methods and interpretation standards. For the purpose ofthis analysis, ‘intermediate’ was considered as ‘resistant’. If adifferent but equivalent agent within a given class of antibioticswas tested according to Clinical and Laboratory StandardsInstitute (CLSI) guidelines [11] (e.g. levofloxacin instead of CIP),the result was considered equivalent; otherwise the antibioticwas removed from the analysis. Multiresistance was defined asacquired resistance to at least three classes of non-b-lactamantibiotics. The most resistant pattern, defined as the patternshowing resistance to a greater number of classes of antibiotics,was described by region.

2.2. Molecular epidemiology

Sites were asked to report on the three most frequentlyidentified multilocus sequence typing (MLST) types and/or S.

aureus protein A (spa) types among the PVL-positive MRSA isolatesthat they reported.

In addition, a nested laboratory study was conducted at theGenomics Research Laboratory at Geneva University Hospitals(Geneva, Switzerland) to characterise a sample of PVL-positiveMRSA strains provided by some of the participating sites bymultilocus variable-number tandem repeat assay (MLVA). Thestrains were selected by each site to be representative of their mostfrequent isolates. The MLVA consisted of a multiplex PCR using 10primer pairs [12].

3. Results

3.1. Antibiotic resistance

The 22 participating sites were located in 17 countries in theAmericas (Brazil and Canada), Australia-Asia (Australia, China andHong Kong, India, Japan, South Korea and Malaysia) and Europe(Denmark, France, Germany, Greece, Italy, The Netherlands, Polandand Switzerland). Overall, 3236 non-duplicate PVL-positive MRSAisolates were included. The number of reported isolates by siteranged from 3 (South Korea and Malaysia) to 1222 (China–HongKong). Of the 22 sites, 19 reported data from 2009 and/or 2010,whereas 3 sites also reported data from 2008.

Reported susceptibility data are shown in Fig. 1, where thepercentage of susceptibility of isolates to each antibiotic wascalculated as percentage by region (number of susceptible isolateswithin a given region/number of reported isolates from thatregion). The lowest susceptibility was observed to ERY in allregions, varying between 17% (Americas) and ca. 60% (Europe andAustralia-Asia). A large difference was observed for CIP, showingsusceptibilities of 25%, 75% and 84% in the Americas, Europe andAustralia-Asia, respectively. Europe reported the lowest suscepti-bility to GEN (72%), TET (65%) and FA (71%). Australia-Asia reportedthe lowest susceptibility to CLI (70%). Two VAN-intermediate PVL-positive MRSA isolates were reported, one from Hong Kong andone from The Netherlands.

Fifteen sites located in 13 countries tested the available PVL-positive MRSA isolates for all of the requested 10 non-b-lactamantibiotics. Only the three most frequent resistance patterns fromeach institution are represented. A total of 1798 isolates weretherefore included in the resistance patterns analysis. Thedistribution of resistance patterns is shown in Table 1. Of the1798 isolates, 49.4% were susceptible to all non-b-lactamantibiotics and were thus defined as multisusceptible. Multi-susceptible was the most frequent pattern in Australia-Asia,whereas this pattern represented <10% of isolates both in theAmericas and in Europe. The most frequent resistance pattern inthe Americas was ERY-CIP, whereas TET with or without associatedresistance to FA was unique to Europe.

Multiresistant isolates were reported from 12 countries,except for Brazil where the most resistant pattern corresponded

Fig. 1. Antibiotic susceptibility (in % of tested isolates) by region. ERY, erythromycin; CLI, clindamycin; GEN, gentamicin; CIP, ciprofloxacin; TET, tetracycline; FA, fusidic acid;

RIF, rifampicin; SXT, trimethoprim/sulfamethoxazole; VAN, vancomycin; LZD, linezolid.

M. Macedo-Vinas et al. / Journal of Global Antimicrobial Resistance 2 (2014) 43–47 45

to ERY-CLI resistance, whilst in Canada resistance to ERY-CLI-GEN-CIP-FA was the most resistant. The most resistant pattern from thewhole series was reported from Europe (Greece) and includedresistance to ERY-CLI-GEN-CIP-FA-TET-SXT. The most resistantpatterns from Australia-Asia were ERY-CLI-TET (Australia) andERY-CLI-GEN-TET-VAN (Hong Kong).

3.2. Molecular epidemiology

Sixteen sites from 13 countries reported their three mostfrequent MLST or spa types. Isolates belonged to eight clonalcomplexes (CC1, CC5, CC8, CC22, CC30, CC59, CC80 and ST93).Sequence type 30 (ST30) was reported worldwide, being the mostfrequent type in Brazil and Asia. ST8 was the most frequent inCanada. In Europe, ST80 and ST8 were the most common types.ST80 was not reported from other regions. ST93 was exclusive toand the most frequent type reported from Australia. Otherfrequently reported types were ST1 from Canada and Australia,ST59 from Asia and ST338 from Asia and Poland.

Forty-nine representative isolates were sent from 10 countries tobe analysed by MLVA. Eight major clusters were discriminated bythis method, showing a certain geographic specificity and agree-ment with MLST (Fig. 2). Cluster one corresponded to ST1, clustertwo to ST8, cluster 3 to ST22, cluster 4 to ST722, cluster 5 to ST80,cluster 6 to ST5, cluster 7 to ST22 (SCCmec IV) and cluster 8 to ST59

Table 1Distribution of resistance patterns of 1798 isolates from 15 sites in 13 countries.

Resistance pattern Americas [n (%)] Australia

Multisusceptiblea 19 (2.1) 803 (90

ERY-CLI-TET-(FA)b 0 (0) 296 (94

ERY-CIP 127 (69.8) 1 (0.5

TET-(FA)b 0 (0) 0 (0)

ERY-CLI 4 (2.9) 118 (86

ERY 19 (35.2) 10 (18

Other 30 (40.5) 16 (21

Total 199 (11.1) 1244 (69

ERY, erythromycin; CLI, clindamycin; TET, tetracycline; FA, fusidic acid; CIP, ciprofloxaa Susceptible to all non-b-lactam antibiotics.b FA included when tested.

and ST388. Switzerland was the only country with isolates belongingto at least five different clusters. Those clusters corresponded to ST8,ST80, ST5, ST772 and ST338/59, which is in accordance withprevious reports [13]. The Netherlands, Greece and Italy wererepresented in three different clusters, whereas Australian isolatesonly clustered in two distinct groups. In contrast, Poland and Canadashowed strictly clonal populations of ST338 and ST8, respectively.

4. Discussion

To our knowledge, this is one of the largest studies attemptingto study the global epidemiology of PVL-positive MRSA strains.Despite the convenience sampling scheme and lack of representa-tiveness (i.e. predominance of European sites, absence of data fromthe USA, different proportion of strains reported by sites), thesedata provide an illustrative picture of the current distribution andmolecular epidemiology of PVL-positive MRSA in different regionsof the world. The distribution of genotypes in Europe is congruentwith other recent reports showing a high genetic heterogeneityamong PVL-positive MRSA [14,15] in that continent. Theseobservations as well as our results also suggest that the USA300and related clones (ST8, CC8) are progressively replacing theEuropean ST80 clone [15]. Poland was the only European countrythat did not report ST8 isolates, whilst ST338 (CC59), common inAsia, was the most frequent type in this country.

-Asia [n (%)] Europe [n (%)] Total [n (%)]

.4) 66 (7.4) 888 (49.4)

.9) 16 (5.1) 312 (17.4)

) 54 (29.7) 182 (10.1)

152 (100) 152 (8.5)

.8) 14 (10.3) 136 (7.6)

.5) 25 (46.3) 54 (3.0)

.6) 28 (37.8) 74 (4.1)

.2) 355 (19.7) 1798 (100)

cin.

Fig. 2. Multilocus variable-number tandem repeat analysis (MLVA) of 49 representative isolates from ten countries. The main clusters belong to major sequence types; ST8

(second cluster from the top), ST722 and ST80 (4th and 5th cluster from the top) containing isolates from divers countries. SWITZERL, Switzerland; NETHERL, The

Netherlands; AUSTRAL, Australia.

M. Macedo-Vinas et al. / Journal of Global Antimicrobial Resistance 2 (2014) 43–4746

M. Macedo-Vinas et al. / Journal of Global Antimicrobial Resistance 2 (2014) 43–47 47

The variability in the number of reported PVL-positive MRSAisolates relates to different data availability at each site. Further-more, not all sites investigated on a routine basis all MRSA isolatesfor the presence of the PVL toxin. Thus, the numbers presentedabove do not represent the true prevalence of PVL-positive MRSAat each site.

PVL-positive MRSA frequently remains multisusceptible tonon-b-lactam agents but it is worrisome that multiresistance wasreported from all regions, with the most resistant pattern fromGreece. Of note, VAN, RIF, SXT and LZD retain activity against thesePVL-positive MRSA isolates. By contrast, increasing resistance tosecond-line antibiotics potentially active against CA-MRSA causingskin infections (CIP, CLI and TET) may compromise treatment.

It is expected that the epidemiology of MRSA will continue toevolve as medical practices changes and new susceptibilitypatterns emerge. It is imperative that surveillance of MRSAsusceptibility patterns be monitored to provide clinicians with themost up-to-date information regarding changes in resistancepatterns.

Funding

This work was supported by an unrestricted educational grantfrom the competitive Europe ASPIRE Research Program 2010,provided by Pfizer International Operations. The work performedat the National Medicine Institute (Warsaw, Poland) was partiallysupported by project CONCORD-HEALTH-F3-2008-2227 18 fromthe European Commission, grant n8 1216/7, PR UE/2009/7 and theMikrobank 2 Programme from the Ministry of Science and HigherEducation, Poland. MMV was partially supported by a grant of theCentre de Recherche Clinique of the University of Geneva Hospitalsand Faculty of Medicine.

Competing interests

JC has received honoraria from the Canadian Agency for Drugsand Technologies in Health for work as an expert reviewer andclinical expert, has received speaker’s honoraria related to newantibacterial agents from Janssen-Ortho and Pfizer during the past3 years, and has received financial support for MRSA researchactivities from the Alberta Heritage Foundation for MedicalResearch, the Canadian Institutes for Health Research and Pfizer;SH is a member of the speakers’ bureau for bioMerieux and Pfizer,is a member of the scientific advisory board of Destiny Pharma, DaVolterra and bioMerieux, and has received financial support forMRSA research activities from B. Braun, Pfizer and the EuropeanCommission [MOSAR network contract LSHP-CT-2007-037941];AS has received funding for investigator-initiated research fromPfizer Canada, and in the past 2 years has served on advisory boards

and as a paid speaker for Pfizer Canada, Novartis and Sunovion; JSis Chief Medical Advisor for bioMerieux. All other authors declareno competing interests.

Ethical approval

Not required.

References

[1] DeLeo FR, Otto M, Kreiswirth BN, Chambers HF. Community-associated meti-cillin-resistant Staphylococcus aureus. Lancet 2010;375:1557–68.

[2] Herold BC, Immergluck LC, Maranan MC, Lauderdale DS, Gaskin RE, Boyle-Vavra S, et al. Community-acquired methicillin-resistant Staphylococcus aure-us in children with no identified predisposing risk. JAMA 1998;279:593–8.

[3] Ito T, Ma XX, Takeuchi F, Okuma K, Yuzawa H, Hiramatsu K. Novel type Vstaphylococcal cassette chromosome mec driven by a novel cassette chromo-some recombinase, ccrC. Antimicrob Agents Chemother 2004;48:2637–51.

[4] Daum RS, Ito T, Hiramatsu K, Hussain F, Mongkolrattanothai K, Jamklang M,et al. A novel methicillin-resistance cassette in community-acquired methi-cillin-resistant Staphylococcus aureus isolates of diverse genetic backgrounds. JInfect Dis 2002;186:1344–7.

[5] Dufour P, Gillet Y, Bes M, Lina G, Vandenesch F, Floret D, et al. Community-acquired methicillin-resistant Staphylococcus aureus infections in France:emergence of a single clone that produces Panton–Valentine leukocidin. ClinInfect Dis 2002;35:819–24.

[6] Otter JA, French GL. Community-associated meticillin-resistant Staphylococcusaureus: the case for a genotypic definition. J Hosp Infect 2012;81:143–8.

[7] Otokunefor K, Sloan T, Kearns AM, James R. Molecular characterization andPanton–Valentine leucocidin typing of community-acquired methicillin-sensitive Staphylococcus aureus clinical isolates. J Clin Microbiol 2012;50:3069–72.

[8] Otter JA, French GL. Community-associated meticillin-resistant Staphylococcusaureus strains as a cause of healthcare-associated infection. J Hosp Infect2011;79:189–93.

[9] Graber CJ, Wong MK, Carleton HA, Perdreau-Remington F, Haller BL, ChambersHF. Intermediate vancomycin susceptibility in a community-associated MRSAclone. Emerg Infect Dis 2007;13:491–3.

[10] Ellington MJ, Ganner M, Warner M, Cookson BD, Kearns AM. Polyclonalmultiply antibiotic-resistant methicillin-resistant Staphylococcus aureus withPanton–Valentine leucocidin in England. J Antimicrob Chemother 2010;65:46–50.

[11] Clinical and Laboratory Standards Institute. Performance standards for anti-microbial susceptibility testing: twenty-second informational supplement.Document M100-S22. Wayne, PA: CLSI; 2012.

[12] Francois P, Huyghe A, Charbonnier Y, Bento M, Herzig S, Topolski I, et al. Use ofan automated multiple-locus, variable-number tandem repeat-based methodfor rapid and high-throughput genotyping of Staphylococcus aureus isolates. JClin Microbiol 2005;43:3346–55.

[13] Francois P, Harbarth S, Huyghe A, Renzi G, Bento M, Gervaix A, et al. Methicil-lin-resistant Staphylococcus aureus, Geneva, Switzerland, 1993–2005. EmergInfect Dis 2008;14:304–7.

[14] Aschbacher R, Pichon B, Spoladore G, Pagani E, Innocenti P, Moroder L, et al.High clonal heterogeneity of Panton–Valentine leukocidin-positive meticillin-resistant Staphylococcus aureus strains from skin and soft-tissue infections inthe Province of Bolzano, Northern Italy. Int J Antimicrob Agents 2012;39:522–5.

[15] Rolo J, Miragaia M, Turlej-Rogacka A, Empel J, Bouchami O, Faria NA, et al. Highgenetic diversity among community-associated Staphylococcus aureus in Eur-ope: results from a multicenter study. PLoS ONE 2012;7:e34768.