Panton-Valentine Leukocidin–Positive MRSA, Shanghai

To the Editor: The development of methicillin resistance in community strains of Staphylococcus aureus is a notable step in the evolution of this pathogen. Unlike their equivalents in the hospital environment, community-associated methicillin-resistant S. aureus (CA-MRSA) strains tend to cause infections in children and young adults who have few known healthcare risks (1). CA-MRSA strains usually possess the Panton-Valentine leukocidin (PVL) genes and staphylococcal cassette chromosome (SCC) mec type IV or V (1,2). 
 
We studied 72 S. aureus isolates (49 MRSA and 23 methicillin-susceptible [MSSA]) by pulsed-field gel electrophoresis and by SCCmec, staphylococcal protein A (spa), and multilocus sequence typing (1,3). These isolates were recovered from clinical specimens (52 respiratory specimens, 9 wound, 4 urine, 2 blood, and 5 other body fluids) from 72 patients treated in 5 district hospitals in Shanghai, People’s Republic of China, during October 2005 through January 2007. The isolates were randomly chosen. In the hospitals, ≈1,000 S. aureus isolates were recovered annually during the time period of our study. The 5 hospitals are estimated to serve a population of 3.4 million, equivalent to one fourth of the total population in Shanghai. Hospital D is a children’s hospital. The other 4 hospitals (A, B, C, and E) have all the major clinical specialties, emergency departments, and outpatient clinics. 
 
The isolates were identified as S. aureus by Gram stain, latex agglutination (Slide StaphPlus; bioMerieux, Marcy l’Etoile, France), and tube coagulase, mannitol, ornithine, and deoxyribonuclease reactions (1,4). Methicillin resistance in the isolates was detected by cefoxitin disc screening and confirmed by mecA PCR (1,4). For patients with PVL-positive MRSA, the computerized discharge records in the hospitals were retrospectively reviewed to ascertain demographic and clinical information. A MRSA case was considered to be community associated if it was isolated from an outpatient or within 2 days of a patient’s hospitalization. Exclusion criteria included a history of hospitalization for illness (except birth), surgery, or dialysis in the previous year or the presence of indwelling catheters or other medical devices (1). Conversely, healthcare-associated MRSA was defined by isolation >2 days after hospitalization or presence of any of the aforementioned healthcare risks. 
 
PVL genes were detected in 9 (18.4%) of the 49 MRSA isolates (Table) and 4 (17.3%) of the 23 MSSA isolates. The 9 MRSA case-patients included 8 infants with pneumonia and 1 adult with prostatitis. Pulsed-field gel electrophoresis clustered 8 of the 9 PVL-positive MRSA isolates into 2 groups: 6 isolates as SH100 and 2 isolates as SH200. Strains of SH100 were spa type/MLST-SCCmec type t318/ST30-IV or t318/ST1114-V, and SH200 strains had t1376/ST88-V. Similar to the PVL-positive MRSA isolates, a limited number of spa types were found among the 40 PVL-negative MRSA isolates. These were t037/ST239-III (n = 19), t002/ST5-II (n = 14), t030/ST239-III (n = 5), t459/ST239-III (n = 1), and t1764/ST88-IV (n = 1). 
 
 
 
Table 
 
Epidemiologic and microbiologic characteristics for Panton-Valentine leukocidin–positive MRSA infections in 9 case-patients, Shanghai, People’s Republic of China, 2006* 
 
 
 
In contrast, spa and sequence types (STs) among the 23 MSSA isolates were highly diverse. There were 20 spa types and 14 STs, giving a total of 20 distinct patterns. Three patterns (t091/ST7, t3388/ST630, t3389/ST15) had 2 isolates, and 17 patterns (t002/ST5, t1077/ST121, t127/ST1, t1376/ST88, t189/ST188, t2024/ST30, t2092/ST121, t2207/ST1206, t2471/ST25, t258/ST25, t3383/ST20, t3386/ST630, t377/ST630, t437/ST1205, t548/ST5, t701/ST6, t796/ST7) had 1 isolate only. The 4 PVL-positive MSSA isolates were t1376/ST88, t2471/ST25, t258/ST25, and t3383/ST20. 
 
Mupirocin resistance rates among the PVL-positive and PVL-negative MRSA isolates were 33.3% (3/9) and 7.5% (3/40), respectively (p = 0.07). All MSSA isolates were susceptible to mupirocin. 
 
It is notable that of the 9 PVL-positive MRSA isolates, 8 of them came from hospital D and all were from children 1–4 months of age. Others have noted that the epidemiology of MRSA differs for children and adults (1,2,7,8). Molecular typing showed that the PVL-positive CA-MRSA isolates were attributed to 2 clones with genotypes t318/ST30-IV (or t318/ST1114-V) and t1376/ST88-V. Detection of t318/ST30 strains in 4 patients with healthcare-associated infections suggested hospital transmission of this CA-MRSA clone, corroborating reports elsewhere (9). Worldwide, ST30 is a common CA-MRSA genetic lineage (1,2). Besides t318, strains related to the ST30 clone have been reported to be spa types t019, t021, and t1273 (2). ST88 PVL-positive MRSA is relatively less common but has been found in Wenzhou (People’s Republic of China), Bangladesh, Belgium, and Nigeria (2,7,8,10). 
 
Because the number of isolates tested in this study is relatively small, no firm conclusion could be drawn on the prevalence of PVL-positive CA-MRSA among S. aureus isolates. Nonetheless, our findings agree with previous reports that the genotypes of MSSA isolates are more diverse than are those for PVL-positive and -negative MRSA isolates and that genotypes for some CA-MRSA strains are shared by a few of the MSSA strains (1).

To the Editor: The development of methicillin resistance in community strains of Staphylococcus aureus is a notable step in the evolution of this pathogen. Unlike their equivalents in the hospital environment, communityassociated methicillin-resistant S. aureus (CA-MRSA) strains tend to cause infections in children and young adults who have few known healthcare risks (1). CA-MRSA strains usually possess the Panton-Valentine leukocidin (PVL) genes and staphylococcal cassette chromosome (SCC) mec type IV or V (1,2).
We studied 72 S. aureus isolates (49 MRSA and 23 methicillinsusceptible [MSSA]) by pulsed-fi eld gel electrophoresis and by SCCmec, staphylococcal protein A (spa), and multilocus sequence typing (1,3). These isolates were recovered from clinical specimens (52 respiratory specimens, 9 wound, 4 urine, 2 blood, and 5 other body fl uids) from 72 patients treated in 5 district hospitals in Shanghai, People's Republic of China, during October 2005 through January 2007. The isolates were randomly chosen. In the hospitals, ≈1,000 S. aureus isolates were recovered annually during the time period of our study. The 5 hospitals are estimated to serve a population of 3.4 million, equivalent to one fourth of the total population in Shanghai. Hospital D is a children's hospital. The other 4 hospitals (A, B, C, and E) have all the major clinical specialties, emergency departments, and outpatient clinics.
The isolates were identifi ed as S. aureus by Gram stain, latex agglutination (Slide StaphPlus; bioMérieux, Marcy l'Etoile, France), and tube coagulase, mannitol, ornithine, and deoxyribonuclease reactions (1,4). Methicillin resistance in the isolates was detected by cefoxitin disc screening and confi rmed by mecA PCR (1,4). For patients with PVL-positive MRSA, the computerized discharge records in the hospitals were retrospectively reviewed to ascertain demographic and clinical information. A MRSA case was considered to be community associated if it was isolated from an outpatient or within 2 days of a patient's hospitalization. Exclusion criteria included a history of hospitalization for illness (except birth), surgery, or dialysis in the previous year or the presence of indwelling catheters or other medical devices (1). Conversely, healthcare-associated MRSA was defi ned by isolation >2 days after hospitalization or presence of any of the aforementioned healthcare risks.
It is notable that of the 9 PVL-positive MRSA isolates, 8 of them came from hospital D and all were from children 1-4 months of age. Others have noted that the epidemiology of MRSA differs for children and adults (1,2,7,8). Molecular typing showed that the PVL-positive CA-MRSA isolates were attributed to 2 clones with genotypes t318/ST30-IV (or t318/ ST1114-V) and t1376/ST88-V. Detection of t318/ST30 strains in 4 patients with healthcare-associated infections suggested hospital transmission of this CA-MRSA clone, corroborating reports elsewhere (9). Worldwide, ST30 is a common CA-MRSA genetic lineage (1,2). Besides t318, strains related to the ST30 clone have been reported to be spa types t019, t021, and t1273 (2). ST88 PVL-positive MRSA is relatively less common but has been found in Wenzhou (People's Republic of China), Bangladesh, Belgium, and Nigeria (2,7,8,10).
Because the number of isolates tested in this study is relatively small, no fi rm conclusion could be drawn on the prevalence of PVL-positive CA-MRSA among S. aureus isolates. Nonetheless, our fi ndings agree with previous reports that the genotypes of MSSA isolates are more diverse than are those for PVL-positive and -nega-  †According to the diagnosis given in the computerized record. In the 8 children, MRSA was isolated from sputum specimens. One child also had MRSA isolated from pleural fluid. Because limited information was provided in the computerized records, some of the isolations may represent respiratory colonization. The outcomes of the 8 children were unknown.
tive MRSA isolates and that genotypes for some CA-MRSA strains are shared by a few of the MSSA strains (1).
The experimental work was conducted in the Department of Microbiology, University of Hong Kong, and supported by grants from the University Development Fund Project-Research Centre of Emerging Infectious Diseases of the University of Hong Kong.