Abstract
The epidemiology of community-acquired methicillin-resistant Staphylococcus aureus (CA-MRSA) differs from country to country. We assess the features of the ST80 European clone, which is the most prevalent PVL-positive CA-MRSA clone in Europe, and the TSST-1 ST5 clone that was recently described in France. In 2008, all MRSA strains susceptible to fluoroquinolones and gentamicin and resistant to fusidic acid that were isolated in 104 French laboratories were characterized using agr alleles, spa typing, and the staphylococcal cassette chromosome mec element and PCR profiling of 21 toxin genes. Three phenotypes were defined: (i) kanamycin resistant, associated with the ST80 clone; (ii) kanamycin and tobramycin resistant, associated with the ST5 clone; and (iii) aminoglycoside susceptible, which was less frequently associated with the ST5 clone. Among the 7,253 MRSA strains isolated, 91 (1.3%) were ST80 CA-MRSA (89 phenotype 1) and 190 (2.6%) were ST5 CA-MRSA (146 phenotype 2, 42 phenotype 3). Compared to the latter, ST80 CA-MRSAs were more likely to be community acquired (80% versus 46%) and found in young patients (median age, 26.0 years versus 49.5 years) with deep cutaneous infections (48% versus 6%). They were less likely to be tetracycline susceptible (22% versus 85%) and to be isolated from respiratory infections (6% versus 27%). The TSST-1 ST5 clone has rapidly emerged in France and has become even more prevalent than the ST80 European clone, whose prevalence has remained stable. The epidemiological and clinical patterns of the two clones differ drastically. Given the low prevalence of both among all staphylococcal infections, no modification of antibiotic recommendations is required yet.
Community-acquired methicillin-resistant Staphylococcus aureus (CA-MRSA) infection was first recognized about a decade ago. Initially causing outbreaks in various well-defined populations, such as children, incarcerated persons, or sports teams, it now has spread worldwide (15). Several CA-MRSA clones have been identified throughout the world, mainly on the basis of molecular typing tools (15). They differ from the classical health care-associated (HA) MRSA in several important ways, including the following: (i) the size of the staphylococcal cassette chromosome mec element (SCCmec) containing the mecA gene, which is usually smaller and of type IV or V (7); (ii) some virulence factors that are frequently detected in these various clones, mainly Panton-Valentine leukocidin (PVL) and toxic shock syndrome toxin 1 (TSST-1) (20, 23); and (iii) specific susceptibility patterns with fewer resistances to antimicrobial drugs that could ease their recognition (16). Of interest, some authors have shown that phenotypic rules based on these specific antimicrobial resistance patterns are useful in routine tests for CA-MRSA strains and even for clone assignment (8, 10, 12, 16, 22).
The high prevalence of CA-MRSA in the United States is due mainly to the spread of the PVL-positive USA300 clone. Conversely, in Europe the prevalence remains low, except in Greece, and epidemics are due mainly to the European PVL-positive ST80 clone (15). In France, this clone emerged 10 years ago, but a new CA-MRSA clone containing the tst gene (encoding TSST-1) was discovered 6 years ago. It contains an SCCmec type I variant that is truncated for the pls gene, and it was named the ST5 Geraldine clone. The antimicrobial resistance patterns of these two clones are relatively stable. Both are susceptible to fluoroquinolones (FQ) (9, 19), whereas most nosocomial MRSA strains isolated in French hospitals are FQ resistant (2, 21). The European ST80 clone is resistant to kanamycin, tetracycline, and fusidic acid, and the ST5 Geraldine clone is resistant to fusidic acid and possibly to kanamycin and tobramycin (9, 23).
In 2004, an initial prospective study based on antimicrobial susceptibility patterns demonstrated that the European ST80 clone represented only 1.4% of clinical MRSA strains in France, confirming findings of a previous retrospective study (18, 19). The prevalence of the ST5 Geraldine clone currently is not known. To evaluate the recent epidemiology of these two CA-MRSA clones in France, we performed a large prospective study in 2008 in which the ST80 and ST5 clones were searched for by microbiological laboratories throughout France on the basis of their specific antibiotic susceptibility patterns. Corresponding strains then were sent to the French National Reference Center for Staphylococci (FNRCS) for molecular characterization using agr alleles, spa typing, and SCCmec and PCR profiling of 21 toxin genes.
MATERIALS AND METHODS
The study was promoted by ONERBA (National Observatory for Epidemiology of Bacterial Resistance to Antibiotics; http://www.onerba.org), a nonprofit organization established in 1997 and composed of 15 French surveillance networks for the surveillance of bacterial resistance to antibiotics in collaboration with the FNRCS. During a 6-month period in 2008, all laboratories participating in one of ONERBA's networks were asked to include all patients harboring a MRSA strain that was susceptible to fluoroquinolones and gentamicin and resistant to fusidic acid. In addition, MRSA strains had to match one of the following antibiotic susceptibility patterns: (i) resistant to kanamycin and susceptible to tobramycin (phenotype 1; considered typical for ST80 CA-MRSA in Europe); (ii) resistant to kanamycin and tobramycin (phenotype 2; considered to be frequently associated with the ST5 Geraldine clone); and (iii) susceptible to kanamycin and tobramycin (phenotype 3; less frequently associated with the ST5 Geraldine clone).
Only clinical specimens were considered, and duplicates were excluded. Susceptibility testing was performed in each participating laboratory by following guidelines from the French Society for Microbiology for antibiotic susceptibility testing (CA-SFM) (4). For each patient harboring one of the three defined MRSA patterns, basic demographic and clinical information were collected. In addition, each laboratory recorded the total number of patients harboring S. aureus that were susceptible or resistant to methicillin during the study period.
When available, the strains were sent to the FNRCS to be further characterized. The strains were screened for genes encoding methicillin resistance (mecA), staphylococcal enterotoxins A, B, C, D, H, K, L, M, O, P, Q, and R (sea to sed, seh, sek to sem, and seo to ser, respectively), toxic shock syndrome toxin 1 (tst), exfoliative toxins A, B, and D (eta, etb, and etd, respectively), Panton-Valentine leukocidin (PVL; luk-PV), class F LukM leukocidin (lukM), beta-hemolysin (hlb), and epidermal cell differentiation inhibitor (edinA, edinB, and edinC), as previously described (13, 22). Multiplex PCR amplification was used to determine the agr allelic group (13). spa typing was performed with the Ridom Staph Type standard protocol and using the Ridom SpaServer, which automatically analyzes spa repeats and assigns spa types (http://spa.ridom.de/index.shtml); the multilocus sequence type was inferred from the spa type according to the Ridom SpaServer (11). SCCmec types were determined by PCR with a simplified version of Kondo's typing system, including M-PCR-1 and M-PCR-2, without determining differences in the junkyard region (14).
MRSA strains with pvl, etd, and edin genes, as well as agr allele type 3, and that belonged to spa-CC t044 were classified as the European CA-MRSA ST80 clone. MRSA strains with tst, sec, sel, sem, seo toxin genes, agr allele type 2, and SCCmec type I and that belonged to spa-CC t002 were classified as the ST5 Geraldine clone.
Data were computerized using EpiData software and were analyzed using Stata 10 (Stata Corp, College Station, TX). Categorical variables were compared using the χ2 test or the Fischer's exact test when appropriate, and the Mann-Whitney test was used for continuous variables. P values are two tailed, and P < 0.05 was considered statistically significant.
RESULTS
A total of 104 laboratories distributed throughout the French territory (including La Réunion Island) and Monaco participated in the 2008 study. There were 60 general hospital laboratories, 32 university hospitals, 7 army hospitals, and 5 private laboratories working for ambulatory patients and private hospitals.
A total of 34,970 patients harboring S. aureus in a clinical specimen were diagnosed, including 7,253 with MRSA (20.7%). Among the latter, 391 (5.4%) strains displayed one of the three predefined phenotypes, and 354 were sent to the FNRCS (37 strains were lost). Among those sent to the NRC, 14 belong to a species other than S. aureus and 7 were mecA-negative S. aureus. Finally, 333 MRSA strains were analyzed, including 93 (1.3% of all MRSA strains) displaying phenotype 1, 176 (2.4%) with phenotype 2, and 64 (0.9%) having phenotype 3.
Phenotypic analysis.
A total of 90 of the 93 phenotype 1 MRSA (97%) strains had agr allele type 3, including 89 pvl+, etd+, and edin+ strains (Table 1). One of the pvl-positive strains also was hlb positive. The most frequent spa type among the 89 pvl-positive agr allele type 3 MRSA strains was spa t044 (n = 77; 87%). Among the 12 remaining strains, 11 presented spa types belonging to the same spa-CC as spa t044 but differed from the main spa type by one (spa t042, spa t131, spa t376, and spa t4725) or two repeats (spa t639 and spa t5088). The last strain had an indeterminate spa type, although it shared all other microbiological characteristics of the ST80 European clone, and MLST confirmed (data not shown) that it belonged to ST80. Therefore, a total of 89 of the 93 phenotype 1 MRSA belonged definitively to the ST80 CA-MRSA European clone. Moreover, one strain shared all of the other features of the ST80 clone (agr allele type 3, etd+, edin+, and spa t376) except for the absence of pvl, which likely was due to the excision of the phage harboring pvl. Because the latter strain did not strictly match the ST80 European clone definition, it was not further considered to belong to this clone. The three remaining phenotype 1 MRSA strains were tst positive, harbored sec, sel, sem, and seo genes, had agr allele type 2, and belonged to spa t002. In addition, two of the latter were SCCmec type I and therefore were considered to belong to the ST5 Geraldine clone. The third remaining strain was SCCmec type IV.
TABLE 1.
Microbiological characteristics of MRSA strains according to the three predefined phenotypes
| Microbiological characteristic | Phenotype |
||
|---|---|---|---|
| 1 (n = 93) (%) | 2 (n = 176) (%) | 3 (n = 64) (%) | |
| Antibiotic susceptibility | |||
| Fusidic acid | 0 | 0 | 0 |
| Fluoroquinolones | 93 (100) | 176 (100) | 64 (100) |
| Gentamicin | 93 (100) | 176 (100) | 64 (100) |
| Kanamycin | 0 | 0 | 64 (100) |
| Tobramycin | 93 (100) | 0 | 64 (100) |
| agr allele type | |||
| 1 | 0 | 11 (6) | 5 (8) |
| 2 | 3 (3) | 163 (93) | 56 (88) |
| 3 | 90 (97) | 1 (1) | 3 (5) |
| Protein encoded | |||
| PVL | 89 (96) | 0 | 3 (5) |
| EDIN | 90 (97) | 0 | 4 (6) |
| ETD | 90 (97) | 0 | 2 (3) |
| TST | 3 (3) | 152 (86) | 47 (73) |
| SEA | 0 | 13 (7) | 7 (11) |
| SEC | 3 (3) | 152 (86) | 47 (73) |
| SED | 3 (3) | 143 (81) | 44 (69) |
| SEL | 3 (3) | 152 (86) | 47 (73) |
| SEM | 3 (3) | 167 (89) | 56 (88) |
| SEO | 3 (3) | 167 (89) | 56 (88) |
| SEP | 0 | 4 (2) | 2 (3) |
| SEQ | 0 | 0 | 3 (5) |
| SER | 3 (3) | 141 (80) | 44 (69) |
| ETA, ETB, LUKM | 0 | 0 | 0 |
| HLB | 1 (1) | 14 (8) | 8 (13) |
| spa type | |||
| Clonal complex 2 | 3 (3) | 153 (87) | 50 (78) |
| Clonal complex 44 | 89 (96) | 2 (1) | 2 (3) |
| Other spa types | 0 | 20 (11) | 12 (19) |
| Indeterminate | 1 (1) | 1 (1) | 0 |
Of the 176 phenotype 2 MRSA strains, 163 (92.6%) had agr allele type 2, including 151 tst-positive strains harboring the sec, sel, sem, and seo toxin genes. Among these 151 tst-positive strains, 145 had SCCmec type I, and one strain had a variant of SCCmec type I that included an additional ccrC recombinase gene. All 146 SCCmec type I MRSA strains had spa types belonging to the same spa-CC: 113 (77.4%) spa type 2, 19 (13.0%) differing from spa t002 by one repeat (spa t010, spa t067, spa t088, spa t105, spa t242, spa t269, spa t306, spa t447, spa t856, spa t3230, and spa t5046), 6 (4.1%) differing by two repeats (spa t509, spa t1300, spa t1683, and spa t1815), and 8 (5.5%) differing by more than two repeats (spa t045, spa t062, spa t688, spa t1340, spa t3152, spa t4971, and spa t5094). Therefore, 146 of the 176 (83.0%) phenotype 2 MRSA strains matched the characteristics of the ST5 Geraldine clone. Among the five remaining tst-positive strains, two had SCCmec type II, two SCCmec type IV, and one had a nontypeable SCCmec element. Of the 12 agr allele type 2 tst-negative strains, 11 harbored sed, sem, seo, and ser toxin genes and one harbored only sem and seo toxin genes. Finally, 11 of the 176 phenotype 2 MRSA strains had agr allele type 1 and were tst and pvl negative, one had agr allele type 3 and was tst and pvl negative, and one had a nontypeable agr allele and was tst positive and pvl negative.
Among the 64 phenotype 3 MRSA strains, 56 (87.5%) had agr allele type 2, including 47 tst-positive strains harboring the sec, sel, sem, and seo toxin genes. Of the 47 tst-positive strains, 41 had SCCmec type I and 2 had a variant of SCCmec type I, including an additional ccrC recombinase gene. All 42 SCCmec type I MRSA strains had spa types belonging to the same spa-CC: 31 isolates (73.8%) with spa t002, 7 (16.7%) with spa types differing from spa t002 by one repeat (spa t010, spa t105, spa t242, spa t447, spa t2487, and spa t5037), 1 with a spa type differing by two repeats (spa t1815), and 3 with spa types differing by more than two repeats (spa t045 and spa t4909). Therefore, 42 of the 64 (65.6%) phenotype 3 MRSA strains matched the features of the ST5 Geraldine clone. The five remaining tst-positive strains had SCCmec type IV. Among the nine agr allele type 2 tst-negative strains, six harbored sem and seo toxin genes and three had sem, seo, sed, and ser toxin genes. Finally, five of the strains were agr allele type 1 and three were agr allele type 3. Among the latter, two were pvl positive and had spa t044, therefore belonging to the ST80 clone.
ST80 European clone.
The 91 patients with an ST80 strain (phenotype 1, n = 89; phenotype 3, n = 2) had a median age of 27 (range, 0.1 to 83) years and were male in 53% of the cases. The ST80 clone represented 0.26% of all S. aureus and 1.25% of all MRSA strains found. ST80 MRSA-positive samples were drawn in the outpatient clinic or within 2 days following admission in 82 (91%) cases. However, in 7 of these 82 cases, infection was considered hospital acquired by the physician in charge due to associated risk factors; in 6 additional cases, the origin of acquisition was doubtful, resulting in at least 73 (80%) of the strains being community acquired. Only six (7%) patients had recognized links with other patients harboring similar strains. Most patients had either deep (48%) or superficial (34%) cutaneous infections. Less than half (45%) of the patients were treated with antibiotics, and 40% had surgery. Details of resistance profiles and antibiotic treatment are presented in Table 2.
TABLE 2.
Characteristics of patients harboring defined clones of toxin-producing MRSA
| Characteristic | ST80 European clone (n = 91) | ST5 Geraldine clone |
Pb | ||
|---|---|---|---|---|---|
| Phenotype 2 (n = 146) | Phenotype 3 (n = 42) | Total (n = 190c) | |||
| Proportion (%) | |||||
| Among S. aureus strains (n = 34,970) | 0.26 | 0.42 | 0.12 | 0.54 | <.001 |
| Among MRSA strains (n = 7,253) | 1.25 | 2.01 | 0.58 | 2.61 | <.001 |
| Days in hospital before sampling, median (10p-90pa) | 0 (0-2) | 0 (0-21) | 1 (0-12) | 0 (0-19) | 0.001 |
| Age, median (10p-90p) | 26.0 (2-64) | 48.5 (3-94) | 53.5 (1-84) | 49.5 (2-84) | <.001 |
| Male gender | 48 (53%) | 68 (47%) | 24 (57%) | 94 (49%) | 0.61 |
| Type of sample | |||||
| Deep cutaneous | 44 (48%) | 8 (5%) | 3 (7%) | 11 (6%) | <.001 |
| Superficial | 31 (34%) | 54 (37%) | 17 (41%) | 71 (37%) | 0.69 |
| Bloodstream infection | 9 (6%) | 3 (7%) | 12 (6%) | 0.01 | |
| Respiratory | 7 (8%) | 39 (27%) | 10 (24%) | 51 (27%) | <.001 |
| Urine | 11 (8%) | 1 (2%) | 12 (6%) | 0.01 | |
| Other | 9 (10%) | 25 (17%) | 8 (19%) | 33 (17%) | 0.11 |
| Toxic shock syndromed | 1 (1%) | 13 (9%) | 6 (14%) | 19 (10%) | 0.007 |
| Community acquired | 73 (80%) | 65 (45%) | 21 (50%) | 87 (46%) | <.001 |
| Linked to another case | 6 (7%) | 2 (1%) | 0 | 2 (1%) | 0.01 |
| Antibiotic treatment | 41 (45%) | 85 (62%) | 20 (48%) | 107 (56%) | 0.06 |
| Surgical treatment | 36 (40%) | 16 (11%) | 5 (12%) | 21 (11%) | <.001 |
| Associated susceptibility of strains | |||||
| Erythromycin | 64 (70%) | 105 (72%) | 33 (79%) | 140 (74%) | 0.57 |
| Tetracycline | 18 (22%) | 121 (83%) | 38 (90%) | 161 (85%) | <.001 |
| Rifampin | 91 (100%) | 136 (93%) | 41 (98%) | 179 (94%) | 0.02 |
10p-90p, 10th and 90th percentiles.
P values after comparison between ST80 and total ST5 patients.
Two strains harbored phenotype 1 (see Results).
As reported by physicians, even if the classification was incomplete according to CDC criteria (http://www.cdc.gov/ncphi/disss/nndss/casedef/toxicsscurrent.htm).
ST5 Geraldine clone.
Strains belonging to the ST5 Geraldine clone were distributed into phenotype 2 (n = 146), phenotype 3 (n = 42), and phenotype 1 (n = 2). No statistical difference was observed between microbiological features of strains or between patient characteristics according to the phenotype (Table 2). Therefore, the three populations were merged for further analysis (Table 2).
The 190 patients had a median age of 49.5 (range, 0.1 to 96) years and were male in 49% of the cases. This clone represented 0.54% of all S. aureus and 2.61% of all MRSA strains isolated. A total of 135 (71%) of the strains were drawn within the first 48 h following admission; the strain was considered hospital acquired by the physician in charge in 45 of these 135 cases, and the origin of the strain was doubtful for 12 cases. Therefore, only 87 (46%) of the cases were considered community acquired. A majority of patients (37%) had superficial cutaneous infections, while only 6% had deep cutaneous infections. Among other patients, 51 (27%) had respiratory infections, 12 (6%) had urinary tract infections, and 33 (17%) displayed other types of infections. Toxic shock syndrome was reported for 19 (10%) patients, but only 3 of the 19 patients had clinical signs matching the CDC definition of TSS (http://www.cdc.gov/ncphi/disss/nndss/casedef/toxicsscurrent.htm). Details of resistance profiles and antibiotic treatment are presented in Table 2.
Comparison of patient populations.
Compared to patients harboring a strain belonging to the ST5 Geraldine clone, those harboring a strain belonging to the ST80 European clone were younger (27.0 versus 49.5; P < 0.001). They also were more likely to have deep cutaneous infections (48% versus 6%; P < 0.001), a community-acquired infection (80% versus 46%; P < 0.001), and a recognized contact case (7% versus 1%; P = 0.02), to have received a beta-lactam-containing regimen (16% versus 11%; P = 0.03), and to have benefited from surgical treatment (42% versus 11%; P < 0.001) (Table 2). On the other hand, they were less likely to have bloodstream (0% versus 6%; P = 0.01), respiratory (8% versus 27%; P < 0.001), or urinary tract infections (0% versus 6%; P = 0.01) and to have a strain susceptible to tetracycline (22% versus 85%; P < 0.001).
DISCUSSION
The principal epidemiological characteristics of two CA-MRSA clones, the ST80 European clone and the ST5 Geraldine clone, are reported for France using a screening method based on typical antimicrobial susceptibility patterns that subsequently was complemented by molecular analysis. The participation of 104 laboratories throughout France during a 6-month period, with the screening of 34,970 S. aureus strains, including 7,253 MRSA strains, gives a national perspective to the results. We highlight that (i) the use of specific antimicrobial resistance patterns allowed for the efficient screening of the two clones; (ii) the ST5 Geraldine clone (0.54% of S. aureus, 2.61% of MRSA) is more prevalent than the ST80 European clone (0.26% of S. aureus, 1.26% of MRSA), with the presence of the latter clone being stable since 2004; and (iii) the two CA-MRSA clones have highly different epidemiological features.
As suggested by others (10, 16, 22), phenotypic rules based on antimicrobial resistance patterns are useful tools for suspecting pvl- and tst-positive MRSA strains in France: phenotype 1 was highly predictive of the ST80 European clone (97%), whereas phenotype 2 and phenotype 3 were highly predictive of the ST5 Geraldine clone (83 and 66%, respectively). This is of major importance, because no method, except for specific PCR, currently is available to easily obtain information about the toxins present in S. aureus strains. Nevertheless, CA-MRSA clones have revealed worrisome and gradually increasing multidrug resistance (1, 8), indicating that genetic drift is associated with a wide dissemination of these clones observed in the community and/or in hospitals. Such evolution demonstrates that the susceptibility pattern, although currently very helpful, must be regarded with caution when used as a single tool for CA-MRSA surveillance over time. In any case, phenotypic rules must be periodically reassessed to be accurate.
The large scale of the present study gave us the opportunity to clearly define and compare microbiological and clinical features of these two CA-MRSA clones that differ drastically. Surprisingly, it appears that, among CA-MRSA, the circulation of the Geraldine clone in France is higher than that of the ST80 European clone. This clone was reported for the first time in France in 2003 and was involved in neonatal toxic shock-like exanthematous disease (NTED) (9). According to the database of the FNRCS, this clone has been present in France since at least 2000 (one strain; unpublished data). Since then, it has been described as an emerging CA-MRSA clone in France (5, 9, 17). Two susceptibility patterns, probably related by acquisition or the loss of a mobile genetic element harboring the 4′-4″-aminoglycoside nucleotidyltransferase gene (ant4′) responsible for kanamycin and tobramycin resistance, are spreading in France. No statistical difference regarding epidemiological, clinical, or microbiological characteristics was observed between strains and patients harboring the two phenotypes, suggesting that they represent only resistance phenotype variants. The second CA-MRSA clone, namely, the ST80 European clone, initially considered to be the European counterpart of the USA300 clone, has not had the same epidemiological success. First described in France in 1999, its prevalence remains low (1.26% of all MRSA strains) and stable (on the basis of data from a study performed in 2004 with the same design as the present one; 1.4% of all MRSA strains [18]). Conversely, the PVL-positive USA300 clone has spread rapidly in the U.S. community since 2000. The genetic background of the two clones is different and could explain the epidemiological differences observed between the two PVL-positive clones. In fact, the reason for this favorable but unexpected stagnation of the ST80 European clone in France is unclear. Indeed, the ST80 clone has spread widely throughout Greece, Tunisia, and Algeria, and there is a high flow of travel and migrants toward France from the latter two countries. In these countries, more than 50% of staphylococcal infections in the community and the hospital are due to pvl-positive ST80 MRSA (1) (F. Laurent, personal data).
The present study highlighted many differences between the two CA-MRSA clones. Infections due to the ST5 Geraldine clone (i) were acquired equally in the community and in the hospital, whereas those due to the ST80 European clone were mainly community acquired (80%), and(ii) involved an older population than that affected by the ST80 European clone. The ST80 European clone was associated with a limited range of cutaneous infections and, to a lesser extent, with pulmonary infections. The sites involved are in agreement with the reported literature and match the polynuclear cytolytic and/or apoptotic activity of PVL (3, 6). Conversely, the ST5 Geraldine clone was associated with a wider variety of clinical manifestations, including a few cases of toxic shock syndrome. This large panel of infections and the low proportion of toxic shock syndrome cannot be directly related to the superantigenic activity of TSST-1. However, it suggests that TSST-1 is not a major virulence determinant but is mainly a highly linked epidemiological marker for the ST5 Geraldine clone. The diversity of infection sites may be linked to the fact that a large proportion of infections due to the ST5 Geraldine clone were hospital acquired. According to the low prevalence of ST80 and the Geraldine MRSA clone, no modification of antibiotic recommendations for staphylococcal disease is required yet.
The present study demonstrated the stability of the prevalence of the ST80 European clone in France and the rapid spread of the ST5 Geraldine clone. On the basis of the number of strains in the present study and the respective prevalence of community-acquired and hospital-acquired strains, one can estimate that the number of strains of the two CA-MRSA clones circulating in the community is roughly the same. These results contrast with those from some other European countries, where the ST80 European clone is more prevalent and the ST5 Geraldine clone has not been reported. The increase in reports of local and large-scale outbreaks over the past few years stresses the need for the surveillance of CA-MRSA throughout the world. Additional studies with different methodological approaches may be of interest to confirm our data. For instance, community-based surveys of CA-MRSA infections with active case findings may be of interest to confirm the level of circulation of these CA-MRSA clones in France.
Acknowledgments
We thank our colleagues who sent French MRSA isolates. We thank C. Courtier, C. Gardon, C. Spinelli, Caroline Bouveyron, Annie Martra, and Martine Rougier for technical help.
The following laboratories (microbiologists) participated in the study: Aix en Provence (O. Bellon, H. Chardon), Amboise (B. Cattier), Angers (M. Eveillard), Argenteuil (A. Scanvic), Arras (M.-N. Noulard), Aubagne (P. Stolidi), Aulnay sous Bois (H. Porcheret), Aurillac (M. Villemain), Bagnols sur Ceze (G. Khatib), Bayeux (E. Heusse), Bergerac (C. Fabe), Besançon (X. Bertrand), Bethune (D. Descamps), Blois (C. Hombrouck-Alet), Bondy (I. Poilane), Bordeaux HIA (B. Soullié), Boulogne-Billancourt (B. Heym), Bron (M. Bes), Cahors (A. LeCoustumier), Calais (J.-M. Pannecoulk), Cambrai (B. Dumoulard, C. Tiry), Chambéry (M. Levast), Chartres (R. Vergez-Pascal), Chartres (O. Zamfir), Cherbourg (F. Bessis), Clamart (J.-W. Decousser), Clamart HIA (V. Hervé, C. Mac-Nab), Clermont-Ferrand (F. Robin), Clichy (F. Bert), Colombes (C. Branger), Créteil (A. Goubart), Dunkerque (A. Verhaeghe), Elbeuf (R. Fabre), Elbeuf (G. Grise), Fougères (J. Rome), Garches (E. Ronco), Grasse (S. Léotard), La Roche sur Yon (G. Chambreuil), Lagny sur Marne (Y. Costa), Landerneau (M.-P. Pelletier), Laval (D. Jan), Le Havre (F. Evreux), Le Kremlin Bicetre (N. Fortineau), Le Mans (C. Varache), Lille (M. Roussel-Delvallez), Limeil-Brevannes (C.-A. Akpabie), Limoges (O. Barraud), Lomme (A. Decoster), Lyon (Y. Boucaud-Maitre), Mantes La Jolie (F. Richardin), Marseille HIA (E. Garnotel), Maubeuge (M. Vasseur), Meaux (F. Faibis), Metz (M.-C. Moulhade), Metz HIA (J. Puyhardy), Monaco (S. Gabriel), Montbrisson (F. Crepet), Montceau les Mines (O. Menouni), Montfermeil (N. Luizy), Mulhouse (J.-M. Delarbre), Nancy (N. Aissa), Nantes (M.-E. Juvin, J. Caillon), Nice (F. Girard-Pipau), Niort (P. Lureau, P. Ochocki), Orleans (L. Bret), Orsay (M.-N. Adam), Paris-HIA (E. Nicand), Paris Bichat (L. Armand-Lefèvre), Paris Cochin (A. Doloy), Paris L'Institut Mutualiste Montsouris (Y. Péan), Paris Lariboisiere (L. Raskine), Paris Necker (C. Guillet), Paris Pitié-Salpêtrierè (J. Robert), Paris St-Anne (S. Brovedani), Paris St-Antoine (V. Lalande), Paris Trousseau (H. Vu Thien), Perigueux (R. Sanchez), Perpignan (E. Lecaillon-Thibon), Provins (F. Pateyron), Rang du Fliers (M. Menouar), Reims (C. de Champs), Rennes (P.-Y. Donnio), Rodez (B. Dubourdieu), Roscoff (M.-P. Pelletier), Roubaix (A. Vachée), Saint-Amand-Montrond (P. Harriau), Saint-Brieuc (J. Vaucel), Saint-Etienne (A.-C. Vautrin), Saint-Mandé HIA (A. Mérens), Saint-Denis de la Réunion (C. Denoix), Selestat (C. Lemble), Sevran (I. Poilane), Toulon (D. Sansot), Toulon Naval (P. Brisou), Toulouse (J.-L. Galinier), Toulouse (L. Cavalié), Tourcoing (M. Caillaux), Tours (P. Laudat), Tours (V. Morange, N. van der Mee-Marquet), Troyes (C. Eloy), Valenciennes (C. Cattoen), Vannes (P. Pouedras), Vesoul (V. Cailleaux), Villejuif (N. Khassis-Chikhani).
Footnotes
Published ahead of print on 10 January 2011.
REFERENCES
- 1.Antri, K., et al. 2010. High prevalence of methicillin-resistant Staphylococcus aureus clone ST80-IV in hospital and community settings in Algiers. Clin. Microbiol. Infect. doi: 10.1111/j.1469-0691.2010.03273.x. [DOI] [PubMed]
- 2.Bertrand, X., et al. 2004. Regional surveillance of the evolution of methicillin resistant Staphylococcus aureus (MRSA): what results for what teaching? Med. Mal. Infect. 34:264-269. [DOI] [PubMed] [Google Scholar]
- 3.Brown, E. L., et al. 2009. The Panton-Valentine leukocidin vaccine protects mice against lung and skin infections caused by Staphylococcus aureus USA300. Clin. Microbiol. Infect. 15:156-164. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Comité de l'Antibiogramme de la Société Française de Microbiologie. 2008. Recommendations 2008. www.sfm=microbiologie.org/UserFiles/file/CASFM/casfm_2008.pdf.
- 5.Dauwalder, O., et al. 2008. Epidemiology of invasive methicillin-resistant Staphylococcus aureus clones collected in France in 2006 and 2007. J. Clin. Microbiol. 46:3454-3458. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.del Giudice, P., et al. 2009. Primary skin abscesses are mainly caused by Panton-Valentine leukocidin-positive Staphylococcus aureus strains. Dermatology 219:299-302. [DOI] [PubMed] [Google Scholar]
- 7.Deurenberg, R. H., and E. E. Stobberingh. 2008. The evolution of Staphylococcus aureus. Infect. Genet. Evol. 8:747-763. [DOI] [PubMed] [Google Scholar]
- 8.Diep, B. A., et al. 2008. Emergence of multidrug-resistant, community-associated, methicillin-resistant Staphylococcus aureus clone USA300 in men who have sex with men. Ann. Intern. Med. 148:249-257. [DOI] [PubMed] [Google Scholar]
- 9.Durand, G., et al. 2006. Detection of new methicillin-resistant Staphylococcus aureus clones containing the toxic shock syndrome toxin 1 gene responsible for hospital- and community-acquired infections in France. J. Clin. Microbiol. 44:847-853. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Gbaguidi-Haore, H., et al. 2009. Usefulness of antimicrobial resistance pattern for detecting PVL- or TSST-1-producing meticillin-resistant Staphylococcus aureus in a French university hospital. J. Med. Microbiol. 58:1337-1342. [DOI] [PubMed] [Google Scholar]
- 11.Harmsen, D., et al. 2003. Typing of methicillin-resistant Staphylococcus aureus in a university hospital setting by using novel software for spa repeat determination and database management. J. Clin. Microbiol. 41:5442-5448. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Jappe, U., et al. 2008. Staphylococcus aureus in dermatology outpatients with special emphasis on community-associated methicillin-resistant strains. J. Investig. Dermatol. 128:2655-2664. [DOI] [PubMed] [Google Scholar]
- 13.Jarraud, S., et al. 2002. Relationships between Staphylococcus aureus genetic background, virulence factors, agr groups (alleles), and human disease. Infect. Immun. 70:631-641. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Kondo, Y., et al. 2007. Combination of multiplex PCRs for staphylococcal cassette chromosome mec type assignment: rapid identification system for mec, ccr, and major differences in junkyard regions. Antimicrob. Agents Chemother. 51:264-274. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Otter, J. A., and G. L. French. 2010. Molecular epidemiology of community-associated meticillin-resistant Staphylococcus aureus in Europe. Lancet Infect. Dis. 10:227-239. [DOI] [PubMed] [Google Scholar]
- 16.Popovich, K., B. Hota, T. Rice, A. Aroutcheva, and R. A. Weinstein. 2007. Phenotypic prediction rule for community-associated methicillin-resistant Staphylococcus aureus. J. Clin. Microbiol. 45:2293-2295. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Raulin, O., et al. 2010. Toxin profiling of Staphylococcus aureus strains involved in varicella superinfection. J. Clin. Microbiol. 48:1696-1700. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Robert, J. 2004. Panton-Valentine leukocidin-producing methicillin-resistant S. aureus in France, 2001-2004. 6th Eur. Cong. Chemother. Infect. 24th Interdisc. Meet. Anti-Infect. Chemother., Paris, France, 1 to 3 December 2004. www.onerba.org.
- 19.Robert, J., J. Etienne, and X. Bertrand. 2005. Methicillin-resistant Staphylococcus aureus producing Panton-Valentine leukocidin in a retrospective case series from 12 French hospital laboratories, 2000-2003. Clin. Microbiol. Infect. 11:585-587. [DOI] [PubMed] [Google Scholar]
- 20.Saïd-Salim, B., B. Mathema, and B. N. Kreiswirth. 2003. Community-acquired methicillin-resistant Staphylococcus aureus: an emerging pathogen. Infect. Control Hosp. Epidemiol. 24:451-455. [DOI] [PubMed] [Google Scholar]
- 21.Thouverez, M., A. Muller, D. Hocquet, D. Talon, and X. Bertrand. 2003. Relationship between molecular epidemiology and antibiotic susceptibility of methicillin-resistant Staphylococcus aureus (MRSA) in a French teaching hospital. J. Med. Microbiol. 52:801-806. [DOI] [PubMed] [Google Scholar]
- 22.Tristan, A., et al. 2007. Virulence determinants in community and hospital meticillin-resistant Staphylococcus aureus. J. Hosp. Infect. 65(Suppl. 2):105-109. [DOI] [PubMed] [Google Scholar]
- 23.Vandenesch, F., et al. 2003. Community-acquired methicillin-resistant Staphylococcus aureus carrying Panton-Valentine leukocidin genes: worldwide emergence. Emerg. Infect. Dis. 9:978-984. [DOI] [PMC free article] [PubMed] [Google Scholar]