MRSA in Austria—an overview
Abstract
The aim of this study was to provide an overview of predominant and sporadic methicillin-resistant Staphylococcus aureus (MRSA) strains in large regions of Austria, and to compare the results with those from other European countries. In total, 1439 MRSA isolates, collected routinely between January 1996 and June 2006 from five Austrian federal provinces, were investigated. The isolates were confirmed as MRSA using mecA/femA multiplex PCR assays. Genes encoding Panton-Valentine leukocidin (PVL), which are characteristic of community-acquired MRSA, were also detected by PCR. Subtyping was performed using SmaI macrorestriction digestion of genomic DNA, followed by pulsed-field gel electrophoresis (PFGE) and cluster analysis. Isolates that could not be assigned to clusters were further analysed by spa typing and/or multilocus sequence typing. The predominant clones detected in Austria were ST228 (southern German epidemic clone), ST5 (Rhine-Hessen MRSA), the ST8 Austrian clone and CC8/ST8. Whereas the frequencies of lineages corresponding to ST247, ST45 and ST22 remained comparably low, an increase in the frequency of lineages corresponding to ST5 and to ST228 was recorded. Overall, 20 different MRSA types and 321 subtypes were recognised according to PFGE analysis. The prevalence of different strains varied considerably in the different Austrian regions. When compared to other European countries, the situation in Austria was most similar to that found in Germany.
Introduction
Since their discovery in 1961, infections with methicillin-resistant Staphylococcus aureus (MRSA) have been a growing problem in hospitals worldwide, as well as in nursing homes and other medical care facilities [1,2]. These Gram-positive pathogens are responsible for a wide range of diseases, e.g., bacteraemia, septic arthritis, endocarditis, pneumonia, osteomyelitis, infections of wounds, bones and joints, and deep abscess formation [3,4]. Most MRSA strains are multiresistant and show susceptibility only to glycopeptide antibiotics (e.g., vancomycin), linezolid and some newly licensed drugs. Insusceptibility to glycopeptides has been reported from many parts of the world [5,6], but is so far infrequent. There are also a few reports concerning linezolid resistance in MRSA [7].
The prevalence of MRSA varies widely among regions. In Europe, the prevalence of MRSA ranges from 0% in Iceland to 1–5% in most northern European countries, 10–25% in central and eastern Europe, and >50% in Romania, Malta and Cyprus [8]. In western Pacific countries, the MRSA prevalence ranges from 24% in Australia to c. 70% in Japan and Hong Kong [9–11]. In the USA, the prevalence is reported to be c. 50% [12]. Increasing frequencies of MRSA are associated mainly with the emergence and spread of epidemic hospital-associated MRSA (haMRSA); however the emergence and spread of MRSA strains that are independent of healthcare facilities has also been recorded in the community worldwide (community-associated MRSA; caMRSA).
In Austria, staphylococci play a major role in hospital-acquired infections, and are the most common cause of bacterial infection among inpatients. The incidence of MRSA infection ranged between 0.85 and 2.42/1000 inpatients during 1994–1998 [13], and the proportion of MRSA among all S. aureus isolates increased from 8.8% in 1999-2002 [1] to 13.7% (mean value) in 2002–2005 [14]. To date, no data concerning the occurrence and distribution of the epidemic haMRSA genotypes in Austria have been published. The aim of the present study was to provide an overview of the genotypes present throughout Austria, with a special focus on the federal provinces of Carinthia, Salzburg, Upper Austria (UA), Lower Austria (LA) and Vienna. The situation in Austria was also compared to that in other European countries.
Materials and methods
Bacterial isolates
Methicillin-resistant S. aureus isolates received at the National Reference Centre (Linz, Austra) between January 1996 and June 2006 were collected. In total, 1790 isolates that the senders considered to be MRSA were received, but the isolates were not collected systematically. The isolates were sent from 45 hospitals, 11 physicians and two private laboratories in Carinthia, Salzburg, UA, LA, Styria and Vienna. The isolates were grouped into six regions for the analyses, based on the regional geographical conditions and socio-economic characteristics: i.e., Carinthia, Salzburg, Salzkammergut (which is mainly the southern part of UA), central UA, eastern UA/western LA and central/eastern LA/Vienna (Viennese region). The different areas started sending isolates at different times (see Results). Isolates from Styria and from northwestern UA were not included in the geographical analysis as the numbers were too low. Isolates were derived from patients with either MRSA infection or colonisation. For the present study, only primary isolates were investigated.
Table 1 lists the reference strains used; these belonged to ten clonal lineages and either formed part of the Harmony collection of European epidemic MRSA [15] or were kindly provided by H. de Lencastre (Instituto de Tecnologia Quimica e Biologica, Laboratory of Molecular Genetics, Oeiras, Portugal) or W. Witte (Robert-Koch-Institute, Wernigerode, Germany).
| MLST | Straina | Isolate no. | Reference/source |
|---|---|---|---|
| ST228 | Southern German EMRSA | 131/98 | HC |
| ST228 | Southern German EMRSA | 1155-2/98 | HC |
| ST247 | Northern German EMRSA | 134/93 | HC |
| ST247 | Northern German EMRSA/Belgium | EC-1 97S96 | HC |
| ST5 | Rhine-Hessen EMRSA | 2387/00 | RKI |
| ST5 | Rhine-Hessen EMRSA/Belgium | EC-3 97S101 | HC |
| ST45 | Berlin EMRSA | 1150/93 | RKI |
| ST45 | Berlin EMRSA/Belgium | EC-2 97S99 | HC |
| ST22 | Barnim MRSA | 1678/96 | RKI |
| ST22 | Barnim MRSA/UK-EMRSA-15 | 90/10685 | HC |
| ST239 | Vienna EMRSA | 635/93 | RKI |
| ST239/241 | Greece 1 | 3680 | HC |
| ST239/241 | UK-EMRSA-1 | NCTC 11939 | HC |
| ST239 | HUSA304—Hungary | (ATCC BAA-39) | [33,44,54], ITB |
| ST239 | HU25—Brazil | [38,55], ITB | |
| ST239 | 2HK—Czech Republic | [56], ITB | |
| ST239 | GRE18—Greece | [40], ITB | |
| ST239 | CPS22—Portugal | [57], ITB | |
| ST241 | TAW9—Taiwan | [36], ITB | |
| ST36 | Finland E5 | 75916 | HC |
| ST36 | UK-EMRSA-16 | 96/32010 | HC |
| ST8 | France A | 162 | HC |
| ST8 | France B | 97121 | HC |
| ST254 | Hanover area EMRSA | 1000/93 | RKI |
- MLST, multilocus sequence typing; HC, HARMONY collection of European epidemic MRSA; ITB, Instituto de Tecnologia Quimica e Biologica Laboratory of Molecular Genetics, Oeiras, Portugal; RKI, Robert-Koch-Institute Wernigerode Branch, Germany.
- aIsolates used as reference strains in Fig. 1 are shown in bold type
Molecular typing strategy
All isolates were subjected to cluster analysis of SmaI macrorestriction patterns (see below) and compared with major epidemic clonal lineages of haMRSA (Table 1). For comparisons, the clusters were named according to multilocus sequence types (STs). Isolates that could not be attributed to clusters were further typed by means of spa sequence typing and BURP (based upon repeat patterns) analysis, as described below, and/or by means of multilocus sequence typing (MLST) at the German National Reference Centre for Staphylococci (Robert-Koch-Institute, Wernigerode Branch, Germany) or the UK Staphylococcus Reference Laboratory (Centre for Infections, Health Protection Agency, London, UK). Previous investigations have revealed that SmaI macrorestriction/cluster analysis, spa typing and MLST analysis yield highly concordant results [16,17] that allow isolates to be attributed to clonal lineages and clonal complexes.
DNA isolation
Pure cultures were grown on TSS (trypticase soy agar containing sheep blood 5% v/v) plates (bioMérieux, Marcy l’Etoile, France) overnight at 37°C in an atmosphere containing CO2 5% v/v. A pipette-tip of bacterial culture was resuspended in phosphate-buffered saline, centrifuged at 12 000 g for 1 min and then resuspended in lysis buffer (8 mM Tris-HCl, pH 7.5, containing lysostaphin 20 mg/L, and proteinase K 20 mg/L) for 1 h at 37°C. After boiling for 5 min, DNA was extracted from 200 μL of bacterial suspension, using a QIAamp DNA Mini kit according to the manufacturer’s instructions (Qiagen, Hilden, Germany), and eluted in 200 μL of distilled water.
PCR and restriction digestion
Identification as MRSA was confirmed using a multiplex PCR for the femA and mecA genes [18,19]. The primers used [20,21] yielded fragments of 533 bp (mecA) and 467 bp (femA). In brief, the multiplex PCR comprised 200 μM dNTPs, 1.5 mM MgCl2, 1× reaction buffer (Applied Biosystems, Foster City, CA, USA), 1.25 U of AmpliTaq DNA Polymerase (Applied Biosystems), and 1 μM of each primer, with 30 cycles of 95°C for 1 min, 57°C for 1 min and 72°C for 1 min.
Coagulase gene polymorphism was detected following digestion with AluI, as described by Schwarzkopf [22], using the primers described by Goh et al. [23]. Fragment lengths depend on the polymorphism, and vary from 654 to 1059 bp. PCR conditions comprised 95°C for 3 min, followed by 30 cycles of 95°C for 30 s, 57°C for 1 min and 72°C for 2 min, with a final extension at 72°C for 5 min.
A third PCR was performed as described by Lina et al. [24] to detect genes encoding Panton–Valentine leukocidin (PVL), which is characteristic of caMRSA strains.
SmaI macrorestriction digestion, pulsed-field gel electrophoresis (PFGE) and cluster analyses
All isolates were investigated by SmaI macrorestriction digestion and subsequent PFGE according to Bannerman et al. [25] and the recommendations of the EU Harmony project [15]. Analysis of PFGE band patterns was according to the guidelines of Tenover et al. [26]. Cluster analyses of SmaI macrorestriction patterns were performed using the UPGMA (unweighted pair-group method with arithmetic mean) clustering algorithm and the Dice correlation coefficient (Fingerprinting II Software; BioRad, Hercules, CA, USA). Clusters were defined using a similarity cut-off value of 70%.
spa typing
Methicillin-resistant S. aureus isolates that could not be attributed to clusters according to their PFGE patterns were further analysed by spa typing. Bacterial DNA was prepared as described on the Ridom Bioinformatics website (http://www.ridom.de/doc/Ridom_spa_sequencing.pdf). Sequencing was performed using an ABI Prism 310 Genetic Analyzer (Applied Biosystems). Ridom StaphType software (Ridom GmbH, Würzburg, Germany) was used for subsequent spa sequence analysis.
Results
Distribution and occurrence of MRSA types
The number of regions and hospitals submitting isolates, as well as the overall number of isolates, increased steadily from the year 2000 onwards. A mean of 30 isolates/year was received in 1996–2001, and this increased to a mean of 416 isolates/year in the last 2.5 years of the study. Table 2 shows that the year in which the MRSA strains with a higher incidence were detected was, in most cases, earlier than that in which sporadically emerging strains were detected. Macrorestriction digestion of the genomic DNA and subsequent PFGE revealed 321 different band patterns (MRSA subtypes), which were assigned to 20 MRSA types as described in Materials and methods.
| MRSA type | n | % | No. subtypes | No. of PVL-positive MRSA | Year first detected |
|---|---|---|---|---|---|
| CC5/ST228 (southern German) | 709 | 49 | 131 | 0 | 1996 |
| CC5/ST5 (Rhine-Hessen) | 155 | 11 | 30 | 7 | 1999 |
| CC8/ST8 Austrian clone | 143 | 10 | 34 | 0 | 1996 |
| CC8/ST8 | 130 | 9 | 23 | 22 | 2000 |
| CC8/ST247 (northern German) | 73 | 5 | 11 | 0 | 1996 |
| CC22/ST22 (Barnim) | 56 | 4 | 13 | 1 | 2002 |
| CC30/ST36 (Finland E5) | 50 | 4 | 13 | 0 | 1998 |
| CC8/ST239 (Vienna) | 21 | 2 | 7 | 0 | 1996 |
| CC8/ST239/241 | 17 | 1 | 17 | 0 | 1996 |
| CC45/ST45 (Berlin) | 12 | 0.8 | 7 | 0 | 1996 |
| CC30/ST30 | 12 | 0.8 | 11 | 10 | 2003 |
| ST80 | 12 | 0.8 | 9 | 11 | 2001 |
| ST152 | 11 | 0.8 | 4 | 11 | 2002 |
| ST778 | 10 | 0.7 | 1 | 0 | 2003 |
| CC8/ST254 (Hanover) | 8 | 0.6 | 4 | 0 | 1996 |
| Not known | 6 | 0.4 | – | 0 | 2006 |
| Not typeable | 5 | 0.3 | – | 0 | 2006 |
| CC8/ST72 | 3 | 0.2 | 2 | 0 | 2004 |
| ST12 | 2 | 0.1 | 1 | 0 | 2002 |
| ST777 | 2 | 0.1 | 1 | 2 | 2004 |
| ST97 | 1 | 0.1 | 1 | 0 | 2005 |
| ST250 | 1 | 0.1 | 1 | 0 | 2004 |
| In total | 1439 | 321 | 64 |
- CC, clonal complex; ST, sequence type; PVL, Panton–Valentine leukocidin.
The present study analysed 1439 primary MRSA isolates from among the 1790 isolates received; the remaining 351 isolates comprised methicillin-susceptible S. aureus (n = 182), repeat isolates (n = 35), specimens other than S. aureus (n = 59), untyped isolates from 1996–2000 (n = 69), or others (n = 6). Almost half (49%) of the isolates investigated belonged to ST228 (southern German epidemic strain), which was found throughout the entire study period in all Austrian regions (Table 2). The diversity of subtypes of this strain was extraordinarily high (131 subtypes; Table 2). The remaining 51% of isolates included several MRSA strains that also occur in other countries, e.g., ST5 (Rhine-Hessen MRSA), CC8/ST8, ST247 (northern German epidemic strain), ST22 (Barnim-MRSA), ST36 (Finland E5), ST239 (Vienna epidemic strain) and CC8/ST239/241.
One particular pattern was related to isolates attributed to ST8 and ST247, but was clearly separated (Fig. 1). As isolates with this pattern had a wide dissemination, this pattern was designated the ‘Austrian clone’. These isolates belonged to spa types t190 and t594, and BURP analysis attributed this pattern to ST8. This was confirmed for 12 representative isolates by spa typing and/or MLST.
Dendrogram illustrating the similarity of SmaI macrorestriction profiles of the Austrian clone and other methicillin-resistant Staphylococcus aureus strains. • Indicates reference strains, with the isolate numbers shown in Table 1. The spa types of the reference strains were as follows: northern German, t051; CC8/ST8, t008; Vienna epidemic strain, t037; southern German, t001; Rhine-Hessen, t002; and Barnim, t032.
The other MRSA types identified (Table 2) had a prevalence of <1% and occurred sporadically. A further 0.4% of all isolates (n = 6) could not be subtyped by coagulase polymorphism and PFGE because they yielded previously unknown band patterns. In addition, it was not possible to (sub)type 0.3% of all isolates (n = 5), because no bands were detectable after SmaI macrorestriction digestion (in three independent experiments).
PVL-positive MRSA
The presence of the genes encoding PVL was investigated in 672 randomly chosen MRSA isolates. Of these, 64 were PVL-positive, and most of these belonged to ST8, followed by ST80, ST152, and ST30 (Table 2).
Predominant MRSA strains in Austria
The data in Tables 2 and 3 show that the most predominant MRSA types in Austria were ST228, ST5, ST8 Austrian clone, CC8/ST8, ST247, ST22 and ST36. Fig. 2 summarises the occurrence of these strains between January 1996 and June 2006. ST228 was predominant throughout this period, although its proportion seemed to decrease slightly. ST247 emerged between 2000 and 2003. ST36 accounted for a significant proportion of isolates between 2000 and 2004, after which it nearly disappeared. ST5 steadily increased in frequency, beginning in 2003 and continuing to the end of the study. The frequency of occurrence of ST22 also increased after 2004, albeit to a lesser extent. CC8/ST8 remained at a constant frequency of 9–13% after 2003. The incidence of the ST8 Austrian clone fluctuated between 3% and 40%; from 2002, its mean incidence was 11%.
| Carinthia | 2000 (%) | 2001 (%) | 2002 (%) | 2003 (%) | 2004 (%) | 2005 (%) | 2006 (%) | Total (%) |
|---|---|---|---|---|---|---|---|---|
| CC5/ST228 | 0 | 15 (44) | 24 (42) | 41 (66) | 120 (78) | 189 (80) | 92 (87) | 481 (73) |
| CC8/ST247 | 6 (100) | 18 (53) | 29 (51) | 11 (18) | 4 (3) | 1 (0.4) | 0 | 69 (11) |
| Others | 0 | 1 (3) | 4 | 10 (16) | 29 (19) | 46 (19) | 14 (13) | 104 (16) |
| Total | 6 (100) | 34 (100) | 57 (100) | 62 (100) | 153 (100) | 236 (100) | 106 (100) | 654 (100) |
| Salzburg | 2002 (%) | 2003 (%) | 2004 (%) | 2005 (%) | 2006 (%) | Total (%) |
|---|---|---|---|---|---|---|
| CC5/ST228 | 11 (69) | 0 | 2 (100) | 18 (35) | 6 (29) | 37 (41) |
| CC5/ST5 | 2 (13) | 0 | 0 | 5 (10) | 2 (10) | 9 (10) |
| CC8/ST8 | 0 | 0 | 0 | 10 (19) | 3 (14) | 13 (14) |
| CC22/ST22 | 0 | 0 | 0 | 7 (14) | 5 (24) | 12 (13) |
| Others | 3 (18) | 0 | 0 | 12 (23) | 5 (24) | 20 (22) |
| Total | 16 (100) | 0 | 2 (100) | 52 (100) | 21 (100) | 91 (100) |
| Salzkammergut | 2002 (%) | 2003 (%) | 2004 (%) | 2005 (%) | 2006 (%) | Total (%) |
|---|---|---|---|---|---|---|
| CC5/ST228 | 0 | 0 | 9 (33) | 4 (10) | 1 (6) | 14 (15) |
| CC5/ST5 | 0 | 0 | 7 (26) | 26 (62) | 16 (89) | 49 (53) |
| CC8/ST8 | 0 | 0 | 4 (15) | 6 (14) | 0 | 10 (11) |
| CC8/ST239 | 5 (100) | 0 | 6 (22) | 3 (7) | 0 | 14 (15) |
| Others | 0 | 0 | 1 (4) | 3 (7) | 1 (5) | 5 (5) |
| Total | 5 (100) | 0 | 27 (100) | 42 (100) | 18 (100) | 92 (100) |
| Central UA | 1997 (%) | 1998 (%) | 1999 (%) | 2000 (%) | 2001 (%) | 2002 (%) | 2003 (%) | 2004 (%) | 2005 (%) | 2006 (%) | Total (%) |
|---|---|---|---|---|---|---|---|---|---|---|---|
| CC5/ST228 | 3 (60) | 28 (71) | 10 (71) | 14 (67) | 12 (44) | 8 (40) | 11 (37) | 5 (12) | 8 (12) | 12 (11) | 111 (30) |
| CC5/ST5 | 0 | 0 | 1 (7) | 0 | 0 | 1 (5) | 0 | 3 (7) | 11 (16) | 34 (30) | 50 (13) |
| Austrian clone | 2 (40) | 2 (5) | 0 | 4 (19) | 1 (4) | 4 (20) | 4 (13) | 4 (9) | 9 (13) | 5 (5) | 35 (9) |
| CC8/ST8 | 0 | 0 | 0 | 1 (5) | 1 (4) | 2 (10) | 8 (27) | 16 (37) | 21 (30) | 26 (23) | 75 (20) |
| CC22/ST22 | 0 | 0 | 0 | 0 | 0 | 1 (5) | 2 (7) | 5 (12) | 11 (16) | 13 (12) | 32 (9) |
| CC30/ST36 | 0 | 3 (8) | 0 | 2 (10) | 11 (41) | 3 (15) | 3 (10) | 5 (12) | 1 (1) | 5 (5) | 33 (9) |
| Others | 0 | 5 (13) | 3 (21) | 0 | 2 (7) | 1 (5) | 2 (6) | 5 (12) | 8 (12) | 15 (14) | 41 (11) |
| Total | 5 (100) | 38 (100) | 14 (100) | 21 (100) | 27 (100) | 20 (100) | 30 (100) | 43 (100) | 69 (100) | 110 (100) | 377 (100) |
| Eastern UA/western LA | 2001 (%) | 2002 (%) | 2003 (%) | 2004 (%) | 2005 (%) | 2006 (%) | Total (%) |
|---|---|---|---|---|---|---|---|
| CC5/ST228 | 2 (100) | 0 | 2 (13) | 10 (21) | 1 (17) | 3 (17) | 18 (19) |
| Austrian clone | 0 | 6 (86) | 9 (60) | 26 (55) | 5 (83) | 12 (67) | 58 (61) |
| CC30/ST36 | 0 | 1 (14) | 2 (13) | 8 (17) | 0 | 0 | 11 (12) |
| Others | 0 | 0 | 2 (13) | 3 (6) | 0 | 3 (17) | 8 (8) |
| Total | 2 (100) | 7 (100) | 15 (100) | 47 (100) | 6 (100) | 18 (100) | 95 (100) |
| Viennese region | 1996 (%) | 2000 (%) | 2001 (%) | 2003 (%) | 2004 (%) | 2005 (%) | 2006 (%) | Total (%) |
|---|---|---|---|---|---|---|---|---|
| CC5/ST228 | 15 (56) | 5 (100) | 0 | 3 (100) | 0 | 15 (43) | 2 (11) | 40 (38) |
| CC5/ST5 | 0 | 0 | 1 (50) | 0 | 5 (36) | 1 (3) | 5 (26) | 12 (11) |
| Austrian clone | 4 (15) | 0 | 0 | 0 | 2 (14) | 6 (17) | 5 (26) | 17 (16) |
| CC8/ST8 | 0 | 0 | 0 | 0 | 3 (21) | 2 (6) | 1 (5) | 6 (6) |
| CC8/ST239/241 | 2 (7) | 0 | 0 | 0 | 0 | 4 (11) | 0 | 6 (6) |
| CC8/ST239 | 1 (4) | 0 | 1 (50) | 0 | 0 | 0 | 0 | 2 (2) |
| Others | 5 (18) | 0 | 0 | 0 | 4 (29) | 7 (20) | 6 (32) | 22 (21) |
| Total | 27 (100) | 5 (100) | 2 (100) | 3 (100) | 14 (100) | 35 (100) | 19 (100) | 105 (100) |
- aYear not shown indicates that no samples were received.
- CC, clonal complex; ST, sequence type; UA, Upper Austria; LA, Lower Austria.
Predominant methicillin-resistant Staphylococcus aureus strains in Austria between January 1996 and June 2006.
Predominant MRSA strains at the regional level
One particular MRSA strain was predominant in each of Carinthia, Salzkammergut and eastern UA/western LA (Table 3). In Carinthia, 73% (mean value) of all isolates belonged to ST228. From 2001, the frequency of this strain increased from 44% to 87%. ST247 accounted for 100% of isolates in the year 2000, probably as a result of an outbreak.
In the Salzkammergut, the proportion of ST5 isolates increased from 26% in 2004 to 89% in 2006, and this strain accounted for 53% of all isolates in the Salzkammergut region during the study period (Table 3). Other strains that occurred relatively frequently in this region were ST228 (which decreased from 33% in 2004 to 6% in 2006), ST239 and CC8/ST8 (Table 3).
In eastern UA/western LA, the ST8 Austrian clone was clearly predominant, accounting for 61% (mean value) of all isolates from this region from 2002 (Table 3). Two other MRSA strains showed a somewhat higher incidence in this region, namely ST228 (19%) and ST36 (12%).
The Salzburg, central UA and Viennese regions showed a higher diversity of strains, and no single MRSA strain was predominant (Table 3). However, some strains seemed to occur preferentially in these regions (e.g., ST22 in Salzburg, and ST239 and ST239/241 in the Viennese region).
Discussion
The data presented above give an overview of the MRSA types currently circulating in Austria, and allow a comparison with other European countries. Relatively few isolates were received and typed between 1996 and 2000, but the number of isolates steadily increased from 2001 onwards, and each MRSA isolate was routinely typed. Thus, the recent data are more comprehensive than those collected earlier.
The increasing number of isolates reflects an increasing awareness of the importance of MRSA epidemiology, and the need to increase efforts to detect and resolve outbreaks of MRSA infection as early as possible. However, under-reporting is still widespread, and the large differences in the numbers of isolates from the different regions do not necessarily represent differences in MRSA prevalence among these regions. Nevertheless, the large number of isolates received from Carinthia and central UA over a long period of time probably provides a fairly accurate view of the epidemiological situation in at least these two regions.
As reported in other European countries, c. 80% of all isolates belonged to only four MRSA types. ST228 was the largest group and represented 50% of all isolates; ST5, ST8 Austrian clone and CC8/ST8 each accounted for c. 10% of all isolates. If ST247, ST22 and ST36 are also included, these seven lineages account for 92% of all Austrian isolates. A similar range of MRSA types can be found in certain countries, e.g., Germany [27], Switzerland [28], Poland [29] and France [30].
ST228 has been predominant in Carinthia since 2003, increasing in proportion during every subsequent year. Interestingly, ST228 isolates have also been reported from Italy [31] and Slovenia (http://www.harmony-microbe.net). However, some differences in the spa types suggest that the ST228 lineages in Austria, Italy and Slovenia have developed independently of each other, perhaps because of the natural boundary of the Alps. Indeed, in Austria, several subtypes of ST228 were found only, or almost only, in Carinthia. An explanation for this might again be the natural boundary of the Alps, which separates Carinthia from all flanking Austrian federal provinces and countries. These subtypes indicate that genomic rearrangements in MRSA belonging to ST228 are quite frequent. However, some subtypes remained rare and appear to have a limited capacity for spread (93 of 131 subtypes were isolated only once or twice).
Also of interest was the occurrence of MRSA type ST247 (northern German epidemic strain) in Carinthia during 2000–2003. Of the 73 isolates of that type analysed, 69 were from Carinthia, and 66 were from an outbreak in a single hospital. A single subtype predominated, with 86% of the ST247 isolates in the hospital belonging to this subtype, which is rare in other Carinthian hospitals. This suggests that the outbreak caused by this strain must have been associated with intra-hospital transfer. ST247 has been reported from many European countries [32,33], being frequent during the 1990s, but becoming more rare subsequently [34].
ST5 was found in many regions of Austria, except eastern UA/western LA, with a remarkable increase in frequency during the past 3 years (Fig. 2). A comparable rise has been reported in Germany [35]. The situation in the Salzkammergut region was unusual in that it was dominated by ST5, with only four MRSA types (ST5, ST239, ST228, CC8/ST8) representing 96% of all isolates. As in Carinthia, this could be explained by the geographical location (partly bordered by the Alps), and also by the fact that it is a less important business location/industrial area than central UA or Vienna (see below), and thus has a low migration rate.
In eastern UA/western LA, 93% of all isolates belonged to only three MRSA strains, i.e., the ST8 Austrian clone, ST228 and ST36. The predominant ST8 Austrian clone was similar, but not identical, to isolates attributed to ST8 and ST247. Whereas the Austrian clone mainly occurred in central UA, eastern UA/western LA and the Viennese region, the CC8/ST8 clone was predominant in central UA (as was the Austrian clone), Salzburg and the Salzkammergut. ST247 was found mainly in Carinthia.
In contrast to Carinthia, the Salzkammergut and eastern UA/western LA, the central UA and Viennese regions showed a high diversity of MRSA types. The latter two regions attract many workers and business employees, which might result in a high migration rate, a high population density and, especially in Vienna, citizens of very diverse origins.
Notable MRSA strains in the Viennese region were ST239, which also occurred in the Salzkammergut, and ST239/241, which occurred only in the Viennese region at a significant frequency (6%). Both strains are very similar, since ST241 (Taiwan clone) is a single-locus variant of ST239 (Hungarian clone) [36]. MRSA isolates of lineage ST239 were reported as ‘Vienna’ MRSA in the mid-1990s as part of an outbreak of nosocomial infections in Vienna [37]. This clone has been recorded worldwide, particularly in Brazil [38], Portugal [39], Greece [40] and Mongolia [41], and also in countries neighbouring Austria, i.e., Germany [42], the Czech Republic [43] and Hungary [44]. The low prevalence of the Vienna MRSA clearly distinguishes Austria, as well as Germany and Switzerland, from southern European countries, e.g., Italy, Portugal and Greece, and middle European countries, e.g., the Czech Republic and Hungary, all of which have a high prevalence of this strain.
Eastern UA/western LA and central UA were the only regions with a noteworthy frequency (9–12%) of ST36. However, in eastern UA/western LA, this strain was detected only between 2002 and 2004, and its prevalence in central UA seems to have been decreasing since 2001. If this trend continues, ST36 might disappear from Austria. However, this clone is very similar to the epidemic MRSA clone UK-EMRSA-16, which is the second most frequent epidemic MRSA clone in the UK, with both clones being assigned to ST36 in the MLST database (http://www.mlst.net). Comparison of the PFGE patterns reveals no differences between these two strains, and the present study found that they had identical coagulase patterns.
Compared to the other Austrian regions, ST22 occurred most frequently in Salzburg, perhaps indicating a possible epidemiological link between Salzburg and Germany, where ST22 is the most frequent epidemic MRSA clone [27,35]. During the 2 years in which this strain was detected (2005 and 2006), its percentage rose from 14% to 24%, but this period was too short to confirm a genuine upward tendency.
With respect to MRSA, Switzerland shows little similarity to Austria, with 29.5% of all MRSA isolates belonging to lineage ST45 (Belgium EC-2 strain, Berlin epidemic strain), 17.7% to ST247 (Belgium EC-1 strain; related to the northern German epidemic strain and the Iberian clone [28]), 9% to a Switzerland-specific strain, and 35% to 62 sporadically occurring PFGE types [28].
It has been reported that MRSA strains expressing PVL are emerging community-acquired pathogens, such that PVL-positive MRSA strains are also referred to as caMRSA [45]. However, other characteristics, e.g., resistance patterns and the presence of type IV and V SCCmec elements, as well as the patient’s history, also define caMRSA [46]. The present study revealed PVL-positive MRSA isolates belonging to seven different STs, with the most frequent corresponding to ST80, ST152, ST30 and ST8. ST80 seems to be one of the most widespread caMRSA strains in Europe [46–52]. ST30 is also spread widely throughout European countries [47,50–52] and in the Pacific region [47,48]. In contrast, ST152 seems to be restricted to only a few countries [48,50,53].
In conclusion, the high number of MRSA isolates in the present study means that the data can serve as a basis for infection control studies in the separate Austrian regions. The pattern of clonal lineages of haMRSA observed was similar to that observed in Germany, although the frequency distributions differed. The lineages of PVL-positive MRSA were also largely in accord. The reasons for the differences in MRSA strains among particular countries, despite worldwide dissemination of epidemic haMRSA, are still unexplored.
Acknowledgements
We thank H. de Lencastre for providing isolates of the MRSA strain ST239/241 from different countries. K. Krziwanek and C. Luger contributed equally to this work. No information has been provided by the authors concerning the existence or absence of conflicting or dual interests.
