Staphylococcus aureus Bacteremia at 5 US Academic Medical Centers, 2008–2011: Significant Geographic Variation in Community-Onset Infections
Staphylococcus aureus Bacteremia at 5 US Academic Medical Centers, 2008-2011: Significant Geographic Variation in Community- Onset Infections
0 David Geffen School of Medicine, University of California , Los Angeles, UCLA
1 Pediatrics, University of Chicago , Illinois
2 Medicine and Health Studies
3 Departments of
4 Montefiore Medical Center/Albert Einstein College of Medicine , New York
5 Department of Medicine, Duke University Medical Center , Raleigh-Durham, North Carolina
6 Department of Medicine, University of California , San Francisco
7 Division of Infectious Diseases, Harbor-UCLA Medical Center and the Los Angeles Biomedical Research Institute , Torrance, California
8 Division of General Internal Medicine, Harbor-UCLA Medical Center and the Los Angeles Biomedical Research Institute , Torrance, California
9 Infection Control Program, University of Chicago Medicine , Illinois
Background. The incidence of community-onset (CO) methicillin-resistant Staphylococcus aureus (MRSA) bacteremia rose from the late 1990s through the 2000s. However, hospital-onset (HO) MRSA rates have recently declined in the United States and Europe. Methods. Data were abstracted from infection prevention databases between 1 January 2008 and 31 December 2011 at 5 US academic medical centers to determine the number of single-patient blood cultures positive for MRSA and methicillin-susceptible S. aureus (MSSA) per calendar year, stratified into CO and HO infections. Results. Across the 5 centers, 4171 episodes of bacteremia were identified. Center A (Los Angeles, California) experienced a significant decline in CO-MRSA bacteremia rates (from a peak in 2009 of 0.42 to 0.18 per 1000 patient-days in 2011 [P = .005]), whereas CO-MSSA rates remained stable. Centers B (San Francisco, California), D (Chicago, Illinois), and E (Raleigh-Durham, North Carolina) experienced a stable incidence of CO-MRSA and CO-MSSA bacteremia. In contrast, at center C (New York, New York), the incidence of CO-MRSA increased >3fold (from 0.11 to 0.34 cases per 1000 patient-days [P < .001]). At most of the sites, HO-MRSA decreased and HOMSSA rates were stable. USA300 accounted for 52% (104/202) of genotyped MRSA isolates overall, but this varied by center, ranging from 35% to 80%. Conclusions. CO-MRSA rates and the contribution of USA300 MRSA varied dramatically across diverse geographical areas in the United States. Enhanced infection control efforts are unlikely to account for such variation in CO infection rates. Bioecological and clinical explanations for geographical differences in CO-MRSA bacteremia rates merit further study.
bacteremia; epidemiology; genotyping; MRSA; Staphylococcus aureus
Methicillin-resistant Staphylococcus aureus (MRSA)
first appeared clinically in the early 1960s [
isolates became endemic in most US hospitals by the
late 1980s [
]. In the 1990s, a new wave of MRSA
infections occurred in community settings [
In many centers, MRSA isolates predominate as a
cause of S. aureus community-onset (CO) infections,
including bacteremia [
]. Rather than being “escaped”
hospital-based MRSA clones, the community MRSA strains were
derived from methicillin-susceptible S. aureus (MSSA) strains that
acquired a novel resistance element, SCCmec type IV [
2, 4, 10–
]. These strain types, especially the USA300 genetic
background, were highly virulent, often susceptible to multiple non–
β-lactam antibiotics, and carried signature toxin genes (most
commonly Panton-Valentine leukocidin [PVL]) rarely found in
the older, hospital-acquired strain types [
]. USA300 also had
a constitutive upregulation of several key virulence genes [
Driven by emergence of USA300, the incidence of MRSA
infections rose dramatically in the early 2000s. In 2000, the
Centers for Disease Control and Prevention (CDC) estimated
that there were >30 000 hospitalizations for MRSA bacteremia
]; however, more recent published reports have described
declines in invasive MRSA infection rates, particularly in
healthcare settings [
]. In the United States and the United
Kingdom, the decline in MRSA bacteremia rates appears to
have preceded enhanced infection prevention efforts in
]. In a study of US military personnel and their
dependents, rates of MRSA bacteremia declined between 2008 and
]. However, to date, few reports have documented the
incidence of MSSA bacteremia in this time period, or assessed
potential geographic variations in CO infection rates, as
opposed to hospital-onset (HO) rates.
The current study was conducted to define trends in the
annual incidence of MRSA and MSSA bacteremia at 5 large,
geographically dispersed US academic medical centers to assess the
incidence of HO and CO MSSA and MRSA bacteremia during a
Aim and Data Abstraction
We conducted a retrospective clinical review and microbial
sample study using existing databases and stored bacterial samples.
The hypothesis tested was that rates of CO-MRSA bacteremia
had declined between 2008 and 2011, whereas MSSA rates
were stable or rising. Community-onset bacteremia was defined
as a case in which a blood culture yielded S. aureus in a specimen
collected ≤48 hours after hospital admission. Resistance to
methicillin was determined in the clinical microbiology laboratory at
each institution. The 5 participating tertiary care academic
medical centers included center A, a 470-bed public hospital in Los
Angeles County, California; center B, a 460-bed public hospital
in San Francisco, California; center C, a private 1491-bed hospital
in New York, New York; center D, a 558-bed private university
hospital in Chicago, Illinois; and center E, a private 813-bed
university hospital in Raleigh-Durham, North Carolina.
Data were abstracted from existing comprehensive infection
prevention databases at each institution to determine how
many MRSA and MSSA blood cultures from distinct patients
were recorded per calendar year; whether the positive culture
was drawn at ≤48 hours or >48 hours from hospital admission;
and whether the case was in a child (defined as <18 years of age)
or an adult (≥18 years of age) patient. In addition, to calculate
standardized annual incidences of infection, the number of
patient-days for all admitted patients (all services) was
abstracted from electronic records. The incidence of infections was
reported as the number of patients divided by 1000 patient-days
per calendar year. The study was approved by the institutional
review board at each institution.
A sample of isolates, stratified proportionally by the total
number of MRSA or MSSA bacteremia episodes at each center and
by 2 time periods (2008–2009 and 2010–2011), was retrieved
from frozen repositories maintained by the infection prevention
services at the 5 institutions. The sample size calculation for the
total number of MRSA bloodstream isolates to be typed was
determined by the hypothesis that there was an increase in the
percentage of MRSA isolates that were USA300, from 20% in
2008–2009 to 35% in 2010–2011. To achieve 80% power with
a 2-tailed test for a P value <.05 to assess this hypothesis, we
required 122 isolates from each time period. Given anecdotal
evidence that PVL-bearing (PVL+) MSSA isolates were
becoming more common among bloodstream infection isolates in the
United States, a secondary hypothesis was that the percentage
of MSSA isolates that were PVL+ would increase from 5% in
2008–2009 to 18% in 2010–2011. To achieve 80% power with
a 2-tailed test for a P value <.05 to assess this hypothesis, we
needed 114 MSSA isolates from each of the 2-year periods.
Isolates were randomly selected from those available at each center;
in the event that a center did not have the requested number of
isolates available, all available isolates for a given time period
and MSSA/MRSA group were included.
In total, 472 bloodstream isolates (244 MRSA and 228 MSSA
isolates) were requested from the 5 centers, and 363 (76.9%)
were available for genotyping. Thus, 109 requested isolates
were not available for genotyping. Center C had no stored
isolates from 2008–2009 (n = 53 MSSA and n = 29 MRSA were
requested for this period), accounting for the majority of the 109
unavailable isolates (82/109 [75%]).
For each isolate, multilocus sequence typing (MLST) was
performed as described [
]. Detection of the PVL genes was
], as was polymerase chain reaction (PCR) for arcA
]. The SCCmec type in mecA-bearing isolates was
determined by a panel of PCR assays defining the mec and ccr
]. Isolates that were sequence type (ST) 8, PVL+, and
carried SCCmec type IV were considered to be USA300, with a
Abbreviations: CO, community onset; HO, hospital onset; MRSA, methicillin-resistant Staphylococcus aureus; MSSA, methicillin-susceptible Staphylococcus
99% positive predictive value . Isolates that were ST5, lacked
the PVL genes (PVL–), and carried SCCmec type II were
considered to be USA100.
Endpoints and Statistical Analysis
The primary endpoint was the rate per 1000 patient-days of
CO-MRSA and CO-MSSA infections from 2008 to 2011,
combining pediatric and adult cases, but separately analyzing the 5
centers. Secondary analyses included the incidence of blood
cultures positive >48 hours after admission (ie, HO-MSSA and
HO-MRSA), as well as analyses of the cumulative data across
all 5 centers. Changes in incidence over time were assessed by
Poisson regression, and changes in ratios over time were
compared by the χ2 test (SPSS). Changes over time in molecular
types were assessed by the χ2 test comparing 2008–2009 with
2010–2011. For all analyses, a P value ≤.05 was considered
significant (Stata, version 12, StataCorp, College Station, Texas).
At the 5 centers combined in 2008–2011, there were 4171
patients with S. aureus bacteremia, of which 42.9% (1791/4171)
were MRSA infections and 57.1% (2380/4171) were MSSA.
Among MRSA patients, 76.3% (1367/1791) were CO- and
23.6% (424/1791) were HO-MRSA. Among MSSA patients,
74.9% (1783/2380) were CO- and 25.1% (597/2380) were
HO-MSSA. Among all S. aureus bacteremia patients, 92.6%
(3864/4171) were adults and 7.4% (307/4171) were children.
Among adults, 43.9% (1697/3864) had MRSA infections,
whereas among children, a much smaller percentage (30.6%
[94/307]) had MRSA. In adults, 77.1% (1309/1697) of MRSA
patients had CO-MRSA, and among children, 62% (58/94)
had CO-MRSA. Among adult patients, more than
threefourths, 77.4% (1678/2167), of MSSA patients had CO-MSSA;
among children, only approximately half (49.3% [105/213]) had
CO-MSSA (Table 1).
Incidence of Bacteremia at All Sites
At the 5 centers combined, there was not a significant change in
the incidence of all S. aureus bacteremia (P = .4) from 2008 to
2011. However, MSSA bacteremia rates significantly declined
from 0.73 to 0.59 per 1000 patient-days (P < .001), whereas
MRSA bacteremia rates rose from 0.41 to 0.50 per 1000
patientdays (P < .001) (Figure 1). Both changes were driven by CO
bacteremia. Specifically, there was a 33% increase in the incidence of
CO-MRSA bacteremia, from 0.30 per 1000 patient-days in 2008
to 0.40 in 2011 (P < .001; Figure 1). At the same time, CO-MSSA
bacteremia incidence decreased significantly, from 0.52 to 0.45 per
1000 patient-days (P < .001) in 2008 to 2011.
The incidence of HO-MRSA bacteremia was consistently
considerably lower than the CO-MRSA rate (Figure 1). At the
5 centers combined, there was a small but significant decrease in
the incidence of HO-MRSA bacteremia from 2008 to 2011,
from 0.12 to 0.10 per 1000 patient-days (P < .001). There was
also a more dramatic decrease in incidence of HO-MSSA
bacteremia, from 0.21 to 0.14 per 1000 patient days (P < .001).
Significant variation was seen in bacteremia rates across the
centers studied, particularly in the incidence of CO-MRSA. At
center A (Los Angeles), the incidence of CO-MRSA bacteremia
declined by 57%, from a peak of 0.42 per 1000 patient-days in
2009 to 0.18 per 1000 patient days in 2011 (P = .005; Figure 2).
In contrast, at center C (New York), CO-MRSA bacteremia
incidence more than tripled, from 0.11 to 0.34 cases per 1000
patient days (P < .001; Figure 2). At centers B (San Francisco), D
(Chicago), and E (Raleigh-Durham), the CO-MRSA rate was
stable (Figure 2). CO-MSSA bacteremia rates were stable at
each of the medical centers except for center C, where the
rates declined significantly from 0.53 to 0.39 per 1000
patientdays (P < .001).
At center B (San Francisco), the HO-MRSA rate declined by
69% (P = .03), and a similar drop was observed at center E
(Raleigh-Durham) (P = .009), whereas at center D (Chicago)
(P = .6) and center A (Los Angeles) (P = .6), the HO-MRSA
bacteremia rate did not change significantly. In contrast, at
center C (New York), the HO-MRSA rate increased significantly
from 0.057 to 0.097 per 1000 patient-days from 2008 to 2011
(P = .046).
HO-MSSA rates did not change significantly at centers A (Los
Angeles), B (San Francisco), D (Chicago), or E (Raleigh-Durham).
Abbreviations: NA, not applicable; NT, new type; slv, single locus variant.
a Shown as multilocus sequence type/presence (+) or absence (−) of Panton-Valentine leukocidin.
b Includes the following single isolates from center A in 2008–2009: 332/−; center A in 2010–2011: 12/−, 45slv/−; center B in 2008–2009: 9/−, 121/−, 121/+, 582/−;
center B in 2010–2011: 97slv/−, 333/−, 731/−; center C in 2010–2011: 7/−, 8/+ (carries arcA), 47/−, 72slv/−, 182slv/−, 231/−; center D in 2008–2009: 74/−, 868/+,
322/−, 109slv/−, 45slv/+; center D in 2010–2011: 5slv/+ (carries arcA), 508slv/−, 580/−, 714/−, 2229/+ (carries arcA); center E in 2008–2009: 54/−, 59/−, 536/+;
center E in 2010–2011: 398/−.
However, at center C (New York), the HO-MSSA decrease
was significant, from 0.25 to 0.14 per 1000 patient-days
(P < .001).
Isolate genotyping results are shown in Table 2 for 161 MSSA
and in Table 3 for 202 MRSA isolates. Tested MSSA isolates
were polyclonal, with 45 STs represented. The 3 most common
STs were ST5 (n = 25 [15.5%]), ST8 (n = 21 [12.4%]), and ST30
(n = 13 [8.1%]). Among the 161 MSSA isolates tested, 18
(11.1%) were PVL+, and the most common background
among these was ST8 (8/18 [44%]). Because center C did not
contribute isolates for genotyping in 2008–2009, the 2010–
2011 center C isolates were eliminated from consideration;
without center C isolates included in the analysis, there was a
significant decrease in PVL+ MSSA isolates at the remaining 4
centers, from 17% (10/58) to 6.0% (4/67) (P = .046).
Tested MRSA isolates were less polyclonal; as shown in Table 3,
for the period 2008–2011, there were only 17 STs among the 202
isolates typed. ST8 (n = 110 [55%]) and ST5 (n = 57 [28%])
accounted for 83% of the MRSA isolates. Of the 202 typed
isolates, 126 (62%) were PVL+. The arcA gene, which is often carried
by USA300 isolates and a proxy marker for the ACME element,
was identified by PCR in 52% (104/202) of the MRSA isolates of
several different genetic backgrounds, including ST5, ST8, ST30,
ST632, and a number of STs not previously defined in the MLST
database. The USA300 phenotype (ie, ST8, PVL+, SCCmec type
IV) accounted for 52% (104/202) of isolates, whereas
USA100like isolates (ie, ST5, PVL–, SCCmec type II) accounted for
16.8% (34/202). Excluding the 2010–2011 isolates from center
C, the percentage of USA300 isolates increased in 2010–2011
(40/71 [56%]) compared with 2008–2009 (42/94 [45%]), but
the difference was not statistically significant (P = .1). Among
typed CO-MRSA, 55% (84/154) were USA300 while among
HO-MRSA, 42% (20/48) were USA300.
The percentage of MRSA isolates that were USA300 obtained
from each center differed in 2008–2011: They accounted for
35% (28/80) at center E, 68% (25/37) at center C, 70%
(14/20) at center A, 80% (20/25) at center B, and 43% (17/40)
at center D.
Abbreviation: NA, not applicable.
a Shown as multilocus sequence type/SCCmec type/presence (+) or absence (−) of Panton-Valentine leukocidin genes. The 8/IV/+ isolates carrying arcA were
considered to be USA300; the 5/II/− isolates were considered to be USA100, as indicated.
b Carries ACME arcA.
c Includes the following single isolates from center A 2008–2009: 840/IV/−; center A in 2010–2011: 1/III/−; center B in 2008–2009: 59/IV/−; center B in 2010–2011:
267/IV/−, 576slv/IV/+ (carries arcA); center C in 2010–2011: 105/C2,5/−; center D in 2008–2009: 8/C2,4/+, 30/II/−, 72/C2,4/+, 231/II/+, 1771/C2, 4/−, 2512/II/−;
center E in 2008–2009: 1/IV/−, 59/II/−, NT/IV/+ (carries arcA); center E in 2010–2011: 5slv/IV/−, 30/II/+ (carries arcA), 231/II/−, 632/II/+, 632/II/+ (carries arcA).
At 5 large, geographically distant US medical centers between
2008 and 2011, we examined >4100 patients with S. aureus
bacteremia. The rates of and trends in HO-MRSA,
COMRSA, HO-MSSA, and CO-MSSA bloodstream infections
varied greatly from center to center. This suggests that there
is not a uniform “national epidemic” of MRSA infections,
but that the epidemiologic and ecologic dynamics of MRSA
vary geographically. At each center, CO bacteremia rates
were higher than HO bacteremia rates, underscoring the
importance of CO infections as contributors to the infectious
burden in subsequently hospitalized patients. This suggests
that efforts to prevent MRSA infections should be focused on
outpatient settings as well as inpatient settings, as noted by
previous studies [
The rise in CO-MRSA bacteremia rates, and the inverse
decline in CO-MSSA bacteremia rates, suggest ongoing,
fundamental changes in bacterial ecology. In contrast, the HO rates
of both MRSA and MSSA bacteremia declined, possibly
indicative of successful horizontal infection control practices [
However, evidence suggesting that this was not universally
true is indicated by the fact that at some centers, change in
HO-MSSA and HO-MRSA infection incidence was not
There was substantial geographical variation in both MRSA
and MSSA bloodstream infection rates and in the molecular
epidemiology of S. aureus bacteremia at the 5 studied centers.
At 3 centers, CO-MRSA bacteremia rates did not change
significantly during the study period, in contrast to recent US
reports on the decreasing trend in HO-MRSA infections [
]. However, at one center (Los Angeles), CO-MRSA
bacteremia rates markedly declined, whereas at another center
(New York) rates significantly rose 3-fold over the same time
period. Of note, Los Angeles and Chicago were 2 of the earliest
geographical areas in the United States to be affected by the
emergence of CO-MRSA infections in the mid-1990s [
]. In contrast, the New England/New York areas were spared
initially and only more recently substantially affected by this
“epidemic” . Thus, it is possible that the relatively stable
incidence of CO-MRSA in Chicago and the decline seen in Los
Angeles will begin to be seen in cities on the East Coast in the
near future. Geographic differences in the incidence of
invasive CA-MRSA infections were noted in CDC surveillance as
We found that most hospitals experienced declines in the rate
of HO-MRSA bacteremia, consistent with the trends identified
in CDC surveillance reports [
]. Our results, however, contrast
with a recent publication that described a decline in rates of
both CO- and HO-MRSA bacteremia in US military personnel
and their dependents [
]. It may be that different bacterial
ecology or infection control practices predominate in military
populations, or that geographic variation is masked in
country-wide studies that provide summary statistics of average
trends from across the nation.
Surprisingly, in contrast to our expectations, there was a
significant decrease in the percentage of PVL+ MSSA isolates
causing bloodstream infections in 2008–2011. PVL positivity in
MSSA and MRSA strain types has been associated with isolates
causing skin and pulmonary infections, although there has been
controversy about the importance of PVL as a virulence factor.
USA300 MSSA carrying PVL have been identified in a number
of studies [
]; we did not find them to be common causes of
bacteremia in our study.
We found that there was a nonsignificant increase in the
percentage of USA300 as a cause of MRSA bacteremia. Genotyping
revealed important center-specific differences in the prevalence
of USA300 among MRSA bacteremia isolates, with a higher
prevalence at the 2 California centers (70% and 80%) than
the others. This may reflect a difference in the patient
populations of the various centers, and perhaps whether hospitals are
public or private, sociodemographic variation in patient
populations, or a difference in the proportion of patients who have
chronic comorbid conditions and therefore more healthcare
exposure. Alternatively, this difference in the genotypic spectrum,
also noted in CDC surveillance [
], may reflect geographic
differences in the spread of USA300 or that each individual center
may represent different “time-points on the curve” of the
epidemic of USA300 in the United States. These data suggest
that a shift in S. aureus strain types is involved in changing
CO-MRSA bacteremia rates, although the relationship is not
Sequence types of MRSA isolates from invasive infections
common in other parts of the world such as ST59 [
ST30, and ST239 [
] were rarely identified, whereas others,
such as ST93, common in Australia [
], and ST228, ST241,
ST22 or ST80, common in Europe [
35, 36, 39–41
], were entirely
absent from our typed isolate sample. Why there are marked
differences in genotypes of MRSA in different continents is
A strength of our study was inclusion of data from patients in
diverse settings across the United States in a large collective
catchment area including millions of residents. Furthermore,
use of infection prevention databases designed to capture all
patients with a diagnosis of bacteremia enabled comprehensive
and unbiased detection of infections.
A limitation of our study was the inclusion of only large,
urban academic medical centers. Because a principal finding
was the diversity of changing MRSA and MSSA bacteremia
rates stratified by geographic areas, it would be anticipated
that the diversity would further increase when evaluating
smaller, community-based, or rural medical centers. The criteria
used for CO and HO infection did not differentiate
communityfrom healthcare-associated infection; we did not abstract
medical records to distinguish CO from healthcare-associated
CO–S. aureus infections. Nevertheless, clear differences in
trends over time in CO as compared with HO bacteremia
rates were seen, indicating that our classification of “CO” was
not merely a surrogate for healthcare-associated, CO infections.
We did not have a complete sample of isolates for genotyping
from all 5 centers, decreasing our power to compare stratified
samples over time. However, our sample size was large, and
most missing isolates were from a single center, allowing us to
analyze data from just the other 4 centers when assessing
change over time in genotypic characteristics of S. aureus isolates
In conclusion, our study demonstrates that both CO– and
HO–S. aureus bacteremia remain common at large US
academic medical centers in different regions of the country and that
the molecular epidemiology of MRSA and MSSA causing
bloodstream infections is not homogeneous. MRSA bacteremia
was not uniformly decreasing at such centers between 2008 and
2011. Future surveillance studies should be particularly
sensitive to the substantial geographical variation in CO-MRSA
incidence and ecology.
Acknowledgments. The authors thank Dr Younju Pak for statistical
support, and support from the National Center for Advancing Translational
Sciences (NCATS) of the National Institutes of Health (NIH) through
UCLA CTSI grant UL1TR000124.
Financial support. A. S. B. was supported by the National Institute of
Allergy and Infectious Diseases (NIAID) of the NIH (AI-039108). R. S. D.,
H. F. C., V. G. F., and B. S. were supported by the NIAID (UM1AI104681).
M. Z. D. was supported by the NIAID (K23 AI095361). R. S. D. and
B. S. were supported by the NIAID (R01 AI103342). V. G. F. was supported
by the NIH (K24-AI093969). S. B.-V. was supported by NCATS (grant UL1
TR000430), Grant Healthcare Foundation, and the NIAID (R21AI111760).
Potential conflicts of interest. A. S. B. has received research grants from
Theravance Pharmaceuticals, Cubist Pharmaceuticals, ContraFect
Corporation, and University of Wurzberg. R. S. D. has served as a paid consultant to
Epoch Biosciences and Clorox, and board membership at Affinium
Pharmaceuticals. M. Z. D. has received an investigator-initiated research
grant from Pfizer. V. G. F. served as chair of the V710 Scientific Advisory
Committee (Merck); has received grant support from Cerexa, Pfizer,
Advanced Liquid Logic, and MedImmune; has been a paid consultant for
Merck, Astellas, Affinium, Theravance, Cubist, Cerexa, Durata, Pfizer,
NovaDigm, Novartis, Medicines Company, Biosynexus, MedImmune, and
Inimex; and has received honoraria from Merck, Astellas, Cubist, Pfizer,
Theravance, and Novartis. S. G.-H. has served as a consultant for Joint
Commission Resources and has received payment for serving on a speaker’s
bureau for Covidian and for Joint Commission Resources. B. S.’s institution
has received consulting fees on his behalf from GSK, Adenium, Spero
Therapeutics, Anacor, Synthetic Biologics, and Novartis. All other authors report
no potential conflicts.
All authors have submitted the ICMJE Form for Disclosure of Potential
Conflicts of Interest. Conflicts that the editors consider relevant to the
content of the manuscript have been disclosed.
1. Jevons MP . “ Celbenin ”-resistant Staphylococci . British Med J 1961 ; 1 : 113 - 4 .
2. Chambers HF , Deleo FR . Waves of resistance: Staphylococcus aureus in the antibiotic era . Nat Rev Microbiol 2009 ; 7: 629 - 41 .
3. Chambers HF . The changing epidemiology of Staphylococcus aureus? Emerg Infect Dis 2001 ; 7 : 178 - 82 .
4. David MZ , Daum RS . Community-associated methicillin-resistant Staphylococcus aureus: epidemiology and clinical consequences of an emerging epidemic . Clin Microbiol Rev 2010 ; 23 : 616 - 87 .
5. Moran GJ , Amii RN , Abrahamian FM , Talan DA . Methicillin-resistant Staphylococcus aureus in community-acquired skin infections . Emerg Infect Dis 2005 ; 11 : 928 - 30 .
6. Moran GJ , Krishnadasan A , Gorwitz RJ , et al. Methicillin-resistant S. aureus infections among patients in the emergency department . N Engl J Med 2006 ; 355 : 666 - 74 .
7. Fergie JE , Purcell K. Community-acquired methicillin-resistant Staphylococcus aureus infections in south Texas children . Pediatr Infect Dis J 2001 ; 20 : 860 - 3 .
8. Herold BC , Immergluck LC , Maranan MC , et al. Community-acquired methicillin-resistant Staphylococcus aureus in children with no identified predisposing risk . JAMA 1998 ; 279 : 593 - 8 .
9. Fridkin SK , Hageman JC , Morrison M , et al. Methicillin-resistant Staphylococcus aureus disease in three communities . N Engl J Med 2005 ; 352 : 1436 - 44 .
10. Enright MC , Robinson DA , Randle G , Feil EJ , Grundmann H , Spratt BG . The evolutionary history of methicillin-resistant Staphylococcus aureus (MRSA) . Proc Natl Acad Sci U S A 2002 ; 99 : 7687 - 92 .
11. Diep BA , Gill SR , Chang RF , et al. Complete genome sequence of USA300, an epidemic clone of community-acquired meticillin-resistant Staphylococcus aureus . Lancet 2006 ; 367 : 731 - 9 .
12. Ma XX , Ito T , Tiensasitorn C , et al. Novel type of staphylococcal cassette chromosome mec identified in community-acquired methicillinresistant Staphylococcus aureus strains . Antimicrob Agents Chemother 2002 ; 46 : 1147 - 52 .
13. Naimi TS , LeDell KH , Como-Sabetti K , et al. Comparison of community- and health care-associated methicillin-resistant Staphylococcus aureus infection . JAMA 2003 ; 290 : 2976 - 84 .
14. Montgomery CP , Boyle-Vavra S , Daum RS . Importance of the global regulators Agr and SaeRS in the pathogenesis of CA-MRSA USA300 infection . PLoS One 2010 ; 5 : e15177 .
15. Kuehnert MJ , Hill HA , Kupronis BA , Tokars JI , Solomon SL , Jernigan DB . Methicillin-resistant Staphylococcus aureus hospitalizations, United States . Emerg Infect Dis 2005 ; 11 : 868 - 72 .
16. Johnson AP . Methicillin-resistant Staphylococcus aureus: the European landscape . J Antimicrob Chemother 2011 ; 66 ( suppl 4 ): iv43 - iv8 .
17. Rao GG , Kearns AM , Edwards GF . Decline and fall of epidemic meticillin-resistant Staphylococcus aureus-16 . J Hosp Infect 2011 ; 79 : 269 - 70 .
18. Wilson J , Guy R , Elgohari S , et al. Trends in sources of meticillinresistant Staphylococcus aureus (MRSA) bacteraemia: data from the national mandatory surveillance of MRSA bacteraemia in England, 2006 - 2009 . J Hosp Infect 2011 ; 79 : 211 - 7 .
19. Kallen AJ , Mu Y , Bulens S , et al. Health care-associated invasive MRSA infections, 2005 - 2008 . JAMA 2010 ; 304 : 641 - 8 .
20. Dantes R , Mu Y , Belflower R , et al. National burden of invasive methicillin-resistant Staphylococcus aureus infections , United States , 2011 . JAMA Intern Med 2013 ; 173 : 1970 - 8 .
21. Wyllie D , Paul J , Crook D. Waves of trouble: MRSA strain dynamics and assessment of the impact of infection control . J Antimicrob Chemother 2011 ; 66 : 2685 - 8 .
22. Landrum ML , Neumann C , Cook C , et al. Epidemiology of Staphylococcus aureus blood and skin and soft tissue infections in the US military health system, 2005 - 2010 . JAMA 2012 ; 308 : 50 - 9 .
23. Enright MC , Day NP , Davies CE , Peacock SJ , Spratt BG . Multilocus sequence typing for characterization of methicillin-resistant and methicillin-susceptible clones of Staphylococcus aureus . J Clin Microbiol 2000 ; 38 : 1008 - 15 .
24. Lina G , Piémont Y , Godail-Gamot F , Bes M , Peter MO , et al. Involvement of Panton-Valentine leukocidin-producing Staphylococcus aureus in primary skin infections and pneumonia . Clin Infect Dis 1999 ; 29 : 1128 - 32 .
25. Diep BA , Stone GG , Basuino L , et al. The arginine catabolic mobile element and staphylococcal chromosomal cassette mec linkage: convergence of virulence and resistance in the USA300 clone of methicillin-resistant Staphylococcus aureus . J Infect Dis 2008 ; 197 : 1523 - 30 .
26. Boyle-Vavra S , Ereshefsky B , Wang CC , Daum RS . Successful multiresistant community-associated methicillin-resistant Staphylococcus aureus lineage from Taipei, Taiwan, that carries either the novel staphylococcal chromosome cassette mec (SCCmec) type VT or SCCmec type IV . J Clin Microbiol 2005 ; 43 : 4719 - 30 .
27. International Working Group on the Classification of Staphylococcal Cassette Chromosome Elements (IWG-SCC) . Classification of staphylococcal cassette chromosome mec (SCCmec): guidelines for reporting novel SCCmec elements . Antimicrob Agents Chemother 2009 ; 53 : 4961 - 7 .
28. David MZ , Taylor A , Lynfield R , et al. Comparing pulsed-field gel electrophoresis with multilocus sequence typing, spa typing, staphylococcal cassette chromosome mec (SCCmec) typing, and PCR for PantonValentine leukocidin, arcA, and opp3 in methicillin-resistant Staphylococcus aureus isolates at a U.S. medical center . J Clin Microbiol 2013 ; 51 : 814 - 9 .
29. Edmond MB , Wenzel RP. Screening inpatients for MRSA-case closed . N Engl J Med 2013 ; 368 : 2314 - 5 .
30. Klevens RM , Morrison MA , Nadle J , et al. Invasive methicillin-resistant Staphylococcus aureus infections in the United States . JAMA 2007 ; 298 : 1763 - 71 .
31. Hultén KG , Kaplan SL , Gonzalez BE , et al. Three-year surveillance of community onset health care-associated Staphylococcus aureus infections in children . Pediatr Infect Dis J 2006 ; 25 : 349 - 53 .
32. Orscheln RC , Hunstad DA , Fritz SA , et al. contribution of genetically restricted, methicillin-susceptible strains to the ongoing epidemic of community-acquired Staphylococcus aureus infections . Clin Infect Dis 2009 ; 49 : 536 - 42 .
33. Wang WY , Chiueh TS , Sun JR , Tsao SM , Lu JJ . Molecular typing and phenotype characterization of methicillin-resistant Staphylococcus aureus isolates from blood in Taiwan . PLoS One 2012 ; 7 : e30394 .
34. Song JH , Hsueh PR , Chung DR , et al. Spread of methicillin-resistant Staphylococcus aureus between the community and the hospitals in Asian countries: an ANSORP study . J Antimicrob Chemother 2011 ; 66 : 1061 - 9 .
35. He W , Chen H , Zhao C , et al. Population structure and characterisation of Staphylococcus aureus from bacteraemia at multiple hospitals in China: association between antimicrobial resistance, toxin genes and genotypes . Int J Antimicrob Agents 2013 ; 42 : 211 - 9 .
36. Grundmann H , Aanensen DM , van den Wijngaard CC , Spratt BG , Harmsen D , Friedrich AW ; European Staphylococcal Reference Laboratory Working Group. Geographic distribution of Staphylococcus aureus causing invasive infections in Europe: a molecular-epidemiological analysis . PLoS Med 2010 ; 7 : e1000215 .
37. Peleg AY , Munckhof WJ , Kleinschmidt SL , Stephens AJ , Huygens F . Life-threatening community-acquired methicillin-resistant Staphylococcus aureus infection in Australia . Eur J Clin Microbiol Infect Dis 2005 ; 24 : 384 - 7 .
38. Townell NJ , Munckhof WJ , Nimmo G , et al. Community-associated methicillin-resistant Staphylococcus aureus endocarditis 'down under': case series and literature review . Scand J Infect Dis 2012 ; 44 : 536 - 40 .
39. Baldan R , Testa F , Lorè NI , et al. Factors contributing to epidemic MRSA clones replacement in a hospital setting . PLoS One 2012 ; 7 : e43153 .
40. Denis O , Deplano A , De Beenhouwer H , et al. Polyclonal emergence and importation of community-acquired methicillin-resistant Staphylococcus aureus strains harbouring Panton-Valentine leucocidin genes in Belgium . J Antimicrob Chemother 2005 ; 56 : 1103 - 6 .
41. Knight GM , Budd EL , Whitney L , et al. Shift in dominant hospitalassociated methicillin-resistant Staphylococcus aureus (HA-MRSA) clones over time . J Antimicrob Chemother 2012 ; 67 : 2514 - 22 .