MAJOR ARTICLE
Clinical and Epidemiologic Characteristics Cannot Distinguish Community-Associated MethicillinResistant Staphylococcus aureus Infection from Methicillin-Susceptible S. aureus Infection: A Prospective Investigation Loren G. Miller,1,2,3 Franciose Perdreau-Remington,4 Arnold S. Bayer,1,2,3 Binh Diep,4 Nelly Tan,3,5 Kiran Bharadwa,6 Jennifer Tsui,7 Joshua Perlroth,1,2,3 Anthony Shay,1,2,3 Grace Tagudar,1,2 Uzoma Ibebuogu,8 and Brad Spellberg1,2,3 1 Division of Infectious Diseases and 2Los Angeles Biomedical Research Institute, Harbor–University of California–Los Angeles (UCLA) Medical Center, Torrance, 3David Geffen School of Medicine, UCLA, Los Angeles, 4Division of Infectious Diseases, University of California, San Francisco, and 5University of California, Riverside, California; 6School of Medicine and Biomedical Sciences, State University of New York, Buffalo, and 7 Mailman School of Public Health, Columbia University, New York, New York; and 8Medical College of Georgia, Augusta, Georgia
Background. Community-associated (CA) methicillin-resistant Staphylococcus aureus (MRSA) infection has become common worldwide. Some researchers have argued that empirical therapy for MRSA should be given only to patients with suspected CA S. aureus infections who have risk factors for acquisition of MRSA. However, there are no prospective data examining this approach. Methods. We prospectively enrolled consecutive patients who were hospitalized with S. aureus infection, administered a detailed questionnaire, and collected clinical and microbiological information. Results. Of the 280 consenting patients, 180 were adults with CA S. aureus infection. Among these subjects, 108 (60%) had MRSA infection, and 78 (40%) had methicillin-susceptible S. aureus (MSSA) infection. MRSA infection was associated with younger age (P ! .0001 ); skin/soft-tissue infection (P p .015 ); snorting/smoking illegal drugs (P p .01); recent incarceration (P p .03 ); lower comorbidity index (P p .01 ); more frequent visits to bars, raves, and/or clubs (P p .03 ); and higher frequency of laundering clothes in hot water (P p .05). However, the sensitivity, specificity, and predictive values for these factors for discriminating CA-MRSA infection from CAMSSA infection were relatively poor. Post-hoc modeling revealed that, even in a 10% (i.e., low) MRSA prevalence population, patients lacking the 3 strongest MRSA risk factors would still have a 7% posttest probability of MRSA. Most MRSA strains belonged to the ST-8/USA300 genotype, contained SCCmec type IV, and shared virulence factors commonly found in the ST1:USA400 clone. MSSA strains were genotypically heterogeneous. Conclusions. We found that clinical and epidemiological risk factors in persons hospitalized for CA S. aureus infection cannot reliably distinguish between MRSA and MSSA. Our findings have important implications for the choice of empirical antibiotic therapy for suspected S. aureus infections and for infection control. Staphylococcus aureus is a common cause of community-associated (CA) and health care–associated (HCA) infections [1, 2]—especially infection of skin and soft tissues [2–4]. Until recently, CA strains of S. aureus
Received 4 August 2006; accepted 6 November 2006; electronically published 19 January 2007. Reprints or correspondence: Dr. Loren G. Miller, Div. of Infectious Diseases, Harbor-UCLA Medical Center, 1000 W. Carson St., Box 466, Torrance, CA 90509 (
[email protected]). Clinical Infectious Diseases 2007; 44:471–82 2007 by the Infectious Diseases Society of America. All rights reserved. 1058-4838/2007/4404-0001$15.00
were nearly uniformly susceptible to penicillinase-resistant b-lactams (e.g., methicillin, oxacillin, and cephalosporins). However, during this decade, outbreaks of CA methicillin-resistant S. aureus (MRSA) infection have been increasingly reported worldwide [5–13]. In many places, CA-MRSA infection has become endemic and more prevalent than CA methicillin-susceptible S. aureus (MSSA) infection [9, 14–17]. Traditionally, risk factors for MRSA infection among
Presented in part: 42nd Annual Meeting of the Infectious Diseases Society of America, Boston, Massachusetts, October 2004 (abstract LB-7).
CA MRSA and MSSA Are Indistinguishable • CID 2007:44 (15 February) • 471
Table 1. Characteristics of adult patients with community-associated (CA) Staphylococcus aureus infection and clinical and epidemiologic risk factors for methicillin-resistant S. aureus (MRSA).
Characteristic
All patients (n p 180)
Patients with CA- MRSA infection (n p 108)
Patients with CA-MSSA infection (n p 72)
OR (95% CI)
P
44.5 13.1
41.4 12.2
49.1 13.1
0.952 (0.96–0.98)
!.0001
Demographic characteristic Age, years Mean SD Median
45
42
49
Male sex
78
79
76
0.88 (0.43–1.8)
.71 .062
Race White
26
30
19
Referent
African American
18
20
14
0.96 (0.36–2.6)
Hispanic
51
47
57
0.54 (0.26–1.2)
6
3
10
0.19 (0.04–0.83)
Hospitalizations within past 12 months
24
24
24
Non–skin-related infection
10
6
Other/mixed Clinical characteristic
a
b
17
1.0 (0.51–2.1)
0.94
0.29 (0.10–0.82)
0.015 0.85
APACHE II score Mean SD Median
7.7 3.3
7.8 3.3
7.7 3.4
0.99 (0.91–1.1)
8
8
7
…
Charlson comorbidity score 0.91 1.4
0.68 1.3
1.3 1.6
0.76 (0.61–0.94)
Median
Mean SD
1
0
1
…
HIV infection
4
5
4
1.1 (0.26–4.8)
.01 1.0
Antibiotic use Recent b-lactam use
c
Any b-lactam use in the prior 12 months Recent antibiotic use
d
c
Any antibiotic use in the prior 12 months
d
7
8
6
1.5 (0.46–5.2)
.57
10
11
7
1.7 (0.57–5.0)
.34
18
18
19
0.88 (0.41–1.9)
.75
34
36
30
1.3 (0.69–2.5)
.40
.21
Exposure Duration of symptoms prior to admission, days Mean SD
40.4 321.8
8.5 13.7
88.5 507.9
1.005 (0.996–1.014)
5.0
4.5
5.0
…
2.4 8.9
2.8 11.4
1.7 0.96
0.99 (0.92–1.1)
1.5
1.5
1.5
…
6.0 1.9
6.0 2.0
6.0 1.9
1.0 (0.86–1.2)
6
6
6
…
1.7 6.6
1.1 1.9
2.6 10.4
1.046 (0.96–1.1)
1
1
0
…
2.1 0.85
2.1 0.84
2.2 0.86
1.2 (0.83–1.7)
2
2
2
…
2.8 1.3
2.8 1.1
2.6 1.5
0.88 (0.70–1.1)
3.0
3.0
2.5
…
Recent contact with known skin infection
12
16
7
2.5 (0.87–7.2)
Snorted/smoked illicit drugs
21
28
12
2.9 (1.2–6.8)
.01
Incarcerated in prior 12 months
16
20
8
2.8 (1.1–7.3)
.03
Used public spa, pool, shower, and/or gym
19
23
13
2.1 (0.92–4.8)
.07
Homelessness in prior 12 months
26
27
24
1.2 (0.60–2.4)
.63
Close contact with health care worker
11
10
13
0.79 (0.31–2.0)
.63
High-risk sexual behavior
22
25
17
1.8 (0.73–3.4)
.24
Injection drug use
20
25
14
2.0 (0.90–4.5)
Median No. of people per room Mean SD Median
.38
e
Hygiene score
Mean SD Median
1.0
No. of visits to other health care providers for illness leading to infection Mean SD Median
.29
f
Bathing score
Mean SD Median
.30
f
Education score Mean SD Median
.19 .08
.083 (continued)
Table 1. (Continued.)
All patients (n p 180)
Characteristic Close contact with people requiring frequent hospitalizations or antibiotic use
11
Patients with CA- MRSA infection (n p 108)
14
Patients with CA-MSSA infection (n p 72)
OR (95% CI)
P
7
2.0 (0.69–5.9)
.19
.62
No. of household contacts with frequent hospitalizations and/or antibiotic use Mean SD
0.04 0.19
0.03 0.17
0.04 0.21
1.5 (0.30–7.7)
0
0
0
…
Close contact with injection drug user
20
24
15
1.9 (0.81–4.4)
Lives with children
28
26
32
0.75 (0.39–1.4)
.38
Lives with children who go to out-of-home day care
16
17
14
1.2 (0.52–2.8)
.65
Visits bars, raves, or clubs frequented by men who have sex with men
8
10
4
2.6 (0.70–9.7)
.17 .03
Median
0.14
f
Visits bars, raves, or clubs Mean SD Median
0.44 0.69
0.52 0.72
0.32 0.65
0.64 (0.40–1.0)
0
0
0
…
3.3 0.98
3.4 0.94
3.1 1.0
0.76 (0.56–1.0)
4
4
4
…
1.6 1.6
1.8 1.7
1.4 1.6
0.85 (0.70–1.0)
2
2
0
…
7
6
7
0.93 (0.28–3.0)
f
Launders clothes in hot water Mean SD Median
.05
f
Antibacterial soap use Mean SD Median Plays contact sports
.10 .90
NOTE. Data are percentage of patients, unless otherwise indicated. Variables associated with MRSA with a P value of ⭐.05 are in boldface font. MSSA, methicillin-susceptible S. aureus. a
Included bacteremia (n p 2 ), bacteremia with endopthalmitis (n p 1 ), bacteremia with osteomyelitis (n p 2 ), and bacteremia and urinary tract infection (n p 1). b Included urinary tract infection (n p 2 ), bacteremia and prostatitis (n p 1 ), bacteremia (n p 1 ), bacteremia and urinary tract infection (n p 3), bacteremia and pneumonia (n p 1), pneumonia (n p 2), endocarditis (n p 2), and bronchitis (n p 1). c Use of antibiotics for the infection that eventually led to hospitalization. d Excludes antibiotics used for the infection that eventually led to hospitalization. e A composite score based on 4 survey items (related to rewearing clothes, sharing unwashed clothes, sharing towels, and sharing razors). Each of the 4 individual components was measured on a 4-point Likert type scale. A higher number means lower hygiene. f Scores measured on 5-point Likert-type scales. Higher value means lower hygiene; more education; more visits to bars, raves, or clubs; clothes laundered more often; less frequent bathing; and more antibacterial soap use.
patients with S. aureus infections have been exposures to health care settings, such as hospitals, nursing homes, and dialysis centers [11, 18]; among community dwellers, the risk factors have been injection drug use and diabetes [18–21], Recent outbreaks of CA-MRSA infection have occurred in populations with few or none of these risk factors, including athletes [6, 8, 22], prisoners [6, 23], and healthy children [9, 12, 24]. These outbreaks suggest that, among patients with CA S. aureus infection, exposure to specific non–health care environments may increase the likelihood that the infection is caused by MRSA. The ability of traditional [11, 18–21] and more recently recognized [6, 8, 9, 12, 22–24] risk factors for CA-MRSA infection to distinguish between CA-MRSA infection and CA-MSSA infection is not well defined. Studies of the risk factors for acquisition of CA-MRSA infection have been gathered almost uniformly via retrospective case series, typically in the context of an outbreak among a well-defined population [7, 15, 17, 22, 25, 26]. Retrospective studies often lack comprehensive assess-
ments of risk factors or may be subject to bias [27]. To our knowledge, only 1 investigation of pediatric patients has prospectively evaluated risk factors for CA-MRSA [9]. The lack of prospective data, especially among adults, hampers the ability of clinicians to properly determine which patients require empirical therapy for MRSA infection and which do not. Los Angeles County, California, has been affected by numerous, well-described outbreaks of CA-MRSA infection among prisoners, athletes, and men who have sex with men [6, 25, 28]. Because populations at higher risk for acquiring CA-MRSA infection may exist in the county, it may be an ideal setting to investigate the clinical utility of CA-MRSA risk factors when evaluating patients with CA S. aureus infection. To this end, we performed a prospective investigation of patients with CA S. aureus infection at Harbor–University of California–Los Angeles (UCLA) Medical Center, a public tertiary care hospital in Los Angeles County, which has seen a rapid increase in the number of CA-MRSA infections.
CA MRSA and MSSA Are Indistinguishable • CID 2007:44 (15 February) • 473
Figure 1. Algorithm of patient recruitment and final subject categorization. In this investigation, 409 consecutive hospitalized patients from whom Staphylococcus aureus had been isolated from a wound, blood, urine, or sputum culture within 72 h after hospital admission were eligible. Of these, 326 were approached; the remaining 83 had been discharged, had died, or had transferred. Of the 326 patients approached, 280 consented, and 23 refused; in 23 cases, the informed consent process could not be completed, largely because of the patient’s altered mental status. Once patients were enrolled and surveyed and clinical data were collected, the infections were categorized as either community associated or health care associated, and S. aureus isolates were determined to be methicillin resistant or methicillin susceptible (see Methods for details).
METHODS
From 23 February through 15 October 2004, we screened the clinical microbiology laboratory at Harbor-UCLA Medical Center daily for new cultures of wound, blood, urine, or sputum specimens that were positive for S. aureus. Patients were considered eligible if the culture was positive for S. aureus, as determined using a rapid S. aureus–specific latex agglutination test (Staphaurex; Remel); the patient was still hospitalized; and the culture sample was obtained ⭐72 h after hospital admission. Patients were excluded from the study if they had previously been enrolled in the protocol or refused to participate. Study staff approached all eligible patients and attempted to complete the informed consent process. The study design was approved by the Institutional Review Board at Harbor-UCLA Medical Center.
vestigation of previous outbreaks of CA-MRSA infection [25, 33], if available, and standard measures of severity of illness [34] and comorbidities [35]. We also obtained data on hygiene, bathing frequency, antibacterial soap use, and other habits or exposures (table 1), which were measured using Likert-type scales (e.g., bathing/showering more than once daily to bathing/ showering less than once per week). The in vitro susceptibility of S. aureus isolates was determined using the Vitek system (bioMe´rieux), according to Clinical and Laboratory Standards Institute (formerly the NCCLS) protocols. Enrollment was targeted to allow an 80% power to detect a 10% difference in prevalence for a given risk factor, assuming a risk factor prevalence of 20% in the MRSA group and 1:1 ratio of CA-MRSA infections to CA-MSSA infections. Interim analysis of enrollment was performed monthly, and recruitment ceased when the enrollment target was met.
Data Collection
Case Definitions
A trained research assistant administered a standardized survey to all consenting patients. This detailed questionnaire explored a wide variety of exposures, demographic characteristics, and clinical characteristics that either have been classically associated with HCA-MRSA infection [18, 29] or are hypothesized to have an association with CA-MRSA infection on the basis of recent outbreaks (table 1) [5–7, 18–20, 25, 28, 30–32]. Measures incorporated survey items that were developed for in-
Antimicrobial susceptibility was determined using the Vitek system (bioMe´rieux) in accordance with the protocols of the Clinical and Laboratory Standards Institute. Isolates that were resistant to oxacillin were considered to be MRSA; isolates that were susceptible to oxacillin were considered to be MSSA. Subjects were considered to have CA infection if the culture specimen was not from a surgical site and if, in the past 12 months, the subject (1) had not resided in a long-term care
Surveillance Population
474 • CID 2007:44 (15 February) • Miller et al.
facility, such as a nursing home or rehabilitation facility; (2) had no indwelling devices, such as intravenous catheters or venous access ports; (3) had not visited an infusion clinic; and (4) had not received peritoneal or hemodialysis. Any subjects who did not fulfill the aforementioned criteria were classified as having HCA infection. This definition of CA infection versus HCA infection is consistent with the Centers for Disease Control and Prevention’s ABC criteria [33]. However, we considered patients who were hospitalized in the past 12 months to have CA infection, as pilot data suggested that prior hospitalization was typically unrelated to infection acquisition, and exclusion of this population would have limited the generalizability of our findings. Therefore, the criteria above allowed the categorization of all cases as CA-MRSA, CA-MSSA, HCAMRSA, or HCA-MSSA (figure 1). Strain Characterization
Genotyping. Molecular typing was performed on isolates from the first 123 patients enrolled at an independent site by investigators who were blinded to the clinical details (F.P.R. and B.D.); the number of patients was chosen on the basis of available resources. PFGE with SmaI [36] and multilocus restriction fragment typing genome [37–39] were used to characterize all S. aureus isolates. A PFGE group consisted of isolates with a 0–6-band difference from at least 1 other isolate in the group [40], and we used the genotype nomenclature of McDougal et al. [41]. Sequence types (STs) were assigned with reference to the MLST database (http://www.mlst.net). Strains were designated by ST and PFGE genotype to provide a simplified, uniform nomenclature for describing MRSA (e.g., ST8/USA300). PCR assays for virulence genes and SCCmec allotypes. PCR-based assays for virulence factors (sea, seb, sec, sed, see, seg, seh, sei, sej, sen, seo, sem, tst, and lukM [42] and bsaA, lukE, sek, sel, sep, and seq [43]), adhesins (cna, fnbA, and icaA) [44], Panton-Valentine leukocidin (lukS-PV and lukF-PV) [45], and SCCmec types I-IV [46] were performed as described elsewhere. As a control for our PCR assay, we used a well-characterized strain collection from the National Institutes of Health Network for Antibiotic Resistance in Staphylococcus aureus (http:// www.narsa.org) that contained all specified genes that were examined [42]. The number of strains analyzed for virulence factors was determined by initial findings and available resources. Statistical Analysis
Because of the low rate of enrollment of children, the possibly distinct pathogenesis in this population, and the extreme rarity of children to have certain exposures (e.g., incarceration and injection drug use), all pediatric data were excluded from the primary analyses. Two secondary analyses were performed to evaluate the impact of (1) inclusion of pediatric data and (2)
exclusion of adults who had been hospitalized within the past year from the population of CA-infected patients. For bivariate analyses, we examined the association of MRSA with 23 risk factors, which were chosen from a list of nearly 100 factors based on the literature at the time of analysis and on clinical judgment. Differences in the prevalence of risk factors between patients with CA-MRSA infection and those with CA-MSSA infection were compared using a x2 test or Student’s t test, as appropriate. Collinearity was assessed by Pearson correlation matrices and tolerance values for the independent variables. All risk factors with an association of P ! .05 at the bivariate level were included in the multivariate logistic model predicting MRSA. Sensitivity, specificity, positive predictive value, negative predictive value, and positive and negative likelihood ratios of MRSA risk factors were calculated using established formulas [47, 48]. Data analysis was performed using SAS software, version 8.2 (SAS Institute). As a post-hoc analysis, we sought to define the utility of using the 3 strongest risk factors (as measured by positive likelihood ratios) in calculating the posttest probability of an adult being infected with MRSA (using a pretest probability of 60%, which is the probability of MRSA infection in the cohort). In addition, we used our data to develop 4 hypothetical cohorts with various proportions of MRSA infection—specifically, 10%,
Table 2. Antimicrobial susceptibilities of community-associated methicillin-resistant Staphylococcus aureus (MRSA) and methicillin-susceptible S. aureus (MSSA) isolates. No. of resistant strains/total no. of strains tested (% of resistant strains) Antibiotic
MRSA
MSSA
120/120 (100) 120/120 (100)
0/82 (0) 0/82 (0)
Ciprofloxacin Clindamycin
86/101 (85) 5/102 (5)
8/67 (12) 0/79 (0)
Erythromycin Gentamicin
111/120 (93) 0/120 (0)
13/82 (16) 1/81 (1)
Ampicillin/sulbactam Cefazolin
Oxacillin
120/120 (100)
0/82 (0)
Penicillin G
120/120 (100)
71/82 (87)
Rifampin Tetracycline
0/120 (0) 23/120 (19)
0/82 (0) 5/82 (6)
TMP-SMX Vancomycin
0/120 (0) 0/120 (0)
1/82 (1) 0/82 (0)
101/101 (100) 14/120 (12)
0/67 (0) 0/80 (0)
0/19 (0)
0/15 (0)
Amoxicillin-clavulanate Levofloxacin Linezolid
NOTE. Data are resistance to antimicrobial agents among community-associated MRSA strains (n p 108) and MSSA strains (n p 72 ) recovered from adults with community-associated S. aureus infection. Antibiotic isolates that were intermediately susceptible were categorized as resistant. TMP-SMX, trimethoprim-sulfamethoxazole.
CA MRSA and MSSA Are Indistinguishable • CID 2007:44 (15 February) • 475
Table 3. Test characteristics of risk factors for methicillin-resistant Staphylococcus aureus (MRSA) infection among 180 adults with community-associated S. aureus infection.
Characteristic
Sensitivity
Specificity
PPV
NPV
Positive LR
Negative LR
0.79
0.24
0.61
0.43
1.0
0.90
Demographic characteristic Male sex Ethnicity White
0.50
0.50
0.70
0.30
1.0
1.0
African American Hispanic
0.41 0.61
0.58 0.25
0.69 0.55
0.30 0.30
0.98 0.82
1.0 1.5
0.09
0.67
0.30
0.30
0.26
1.4
Hospitalizations within prior 12 months
0.24
0.76
0.60
0.40
1.0
0.99
Non–skin-related infection HIV infection
0.06 0.05
0.83 0.96
0.33 0.63
0.37 0.40
0.33 1.1
1.1 1.0
0.08
0.94
0.69
0.41
1.5
0.97
0.11
0.93
0.71
0.41
1.6
0.95
0.18
0.81
0.58
0.39
0.90
1.0
0.36
0.70
0.64
0.42
1.2
0.91
0.16 0.28
0.93 0.88
0.76 0.78
0.44 0.45
2.3 2.4
0.90 0.82
Incarcerated within prior 12 months Used public spa, pool, shower, and/or gym
0.20 0.23
0.92 0.88
0.79 0.74
0.43 0.43
2.44 1.9
0.87 0.88
Homelessness within prior 12 months Close contact with health care worker High-risk sexual behavior Injection drug use
0.27 0.10 0.25 0.25
0.76 0.88 0.83 0.86
0.63 0.55 0.68 0.72
0.41 0.39 0.42 0.44
1.1 0.81 1.4 1.8
0.96 1.0 0.91 0.88
Close contact with people requiring frequent hospitalizations or antibiotic use Close contact with injection drug user
0.14 0.24
0.93 0.85
0.74 0.73
0.42 0.41
1.9 1.7
0.93 0.89
0.26 0.17
0.68 0.86
0.55 0.64
0.38 0.40
0.81 1.2
1.1 0.97
0.10 0.06
0.96 0.93
0.79 0.58
0.42 0.40
2.4 0.93
0.94 1.0
Other/mixed Clinical characteristic
Antibiotic use a Recent b-lactam antibiotic use Any b-lactam use in the prior 12 monthsb a
Recent antibiotic use Any antibiotic use in prior 12 monthsb Exposure Recent contact with known skin infection Snorted/smoked illicit drugs
Lives with children Lives with children who go to out-of-home day care Visits bars, raves, or clubs frequented by men who have sex with men Plays contact sports
NOTE. LR, likelihood ratio; NPV, negative predictive value; PPV, positive predictive value. a b
Use of antibiotics for the infection that eventually led to hospitalization. Excludes antibiotics used for the infection that eventually led to hospitalization.
20%, 30%, and 40%. Data from each subject from the original cohort were duplicated at varying ratios to achieve the desired percentage of MRSA infection within each new hypothetical cohort. For example, to create the hypothetical cohort in which 30% of subjects are infected with MRSA (and 70% with MSSA), we used 2 copies of data from MRSA-infected subjects (60% of the total enrollment) and 7 copies of data from each MSSAinfected subject (40% of the total enrollment) to achieve a cohort with exactly 30% MRSA cases and 70% MSSA cases (i.e., [60% MRSA] ⫻ [2] p 1.2; [40% MSSA] ⫻ [7] p 2.8; 1.2/ [1.2 + 2.8] p 30% MRSA cases; 2.8/[1.2 + 2.8] p 70% MSSA cases). The posttest probability of MRSA infection for subjects 476 • CID 2007:44 (15 February) • Miller et al.
with 0, 1, 2, and 3 risk factors in each hypothetical cohort was then calculated. RESULTS A total of 409 patients had cultures positive for S. aureus within 72 h after hospital admission; 326 (80%) of these patients were queried for study enrollment (figure 1). The remaining 83 patients (20%) were unavailable to study personnel because they had been discharged from the hospital (n p 74), had died (n p 6), or had transferred hospitals (n p 3). The informed consent process was not completed by 23 of the subjects queried for study enrollment, largely as a result of altered mental status.
patients (24%) had been hospitalized within the prior 12 months, and most patients (89%) presented with skin or softtissue infection. The median duration of symptoms prior to presentation was 5 days. Additional clinical details are outlined in table 1. Antimicrobial susceptibility tests demonstrated a high degree of resistance among CA-MRSA isolates to fluoroquinolones and macrolides, a high degree of susceptibility to clindamycin and tetracycline, and a very high degree of susceptibility to trimethoprim-sulfamethoxazole, rifampin, and vancomycin (table 2). Risk factors for MRSA infection in adults. Adults infected with CA-MRSA were significantly younger than adults infected with CA-MSSA (OR for CA-MRSA, 0.952; 95% CI, 0.928–0.977 for each decade of age; P ! .001). (table 1) Subjects with CAMRSA had a lower Charlson comorbidity index (OR, 0.76; 95% CI, 0.61–0.94; P p .01) and were less likely to present with non–skin/soft-tissue infection (6% vs. 17%; OR, 0.28; 95% CI, 0.10–0.82; P p .015). Patients infected with CA-MRSA were more likely to smoke or snort illicit drugs (28% vs. 12%; OR, 2.9; 95% CI, 1.2–6.8; P p .01) and to have been incarcerated in the previous 12 months (20% vs. 8%; OR, 2.8; 95% CI, 1.1– 7.3; P p .03). Of note, all but 1 subject who had been incarcerated had been in the Los Angeles County jail, the site of an ongoing outbreak of CA-MRSA infection [6]. Patients infected with CA-MRSA more frequently laundered clothes in hot water (P p .05) and more frequently visited bars, raves, or clubs (P p .03). Sensitivity, specificity, predictive values, and likelihood ratios for these risk factors were very limited in their ability to distinguish patients with CA-MRSA infection from those with CA-
Table 4. Multivariate analysis predicting methicillin-resistant Staphylococcus aureus (MRSA) among persons with communityacquired S. aureus infection. Characteristic
OR (95% CI)
P
Demographic characteristic: age, per 10 years of age
0.96 (0.94–0.99) .009
Clinical Non–skin-related infection Charlson comorbidity score
0.58 (0.18–1.82) .11 0.88 (0.69–1.11) .27
Exposure Snorted/smoked illicit drugs Incarcerated in prior 12 months
2.15 (0.84–5.47) .11 1.39 (0.49–3.99) .53
Visits bars, raves, or clubs
1.20 (0.72–2.03) .49
Launders clothes in hot water
a
1.30 (0.93–1.80) .12
NOTE. Shown is a multivariate model predicting MRSA infection among 180 adults with community-associated S. aureus infection (108 had MRSA infection, and 72 had methicillin-susceptible S. aureus infection). Variables associated with MRSA with a P value of ⭐.05 are shown in boldface font. a
Higher value means laundered more often.
Of the 303 patients who completed the informed consent process, 280 subjects (92%) consented and 23 (8%) refused participation. Seventy-eight subjects (28%) were determined to have HCA S. aureus infection, and 202 were determined to have CA S. aureus infection, including 180 adults, 108 (60%) of whom had CA-MRSA infection and 72 (40%) of whom had CA-MSSA infection. The median age of the adult population was 45 years, and most patients (78%) were male (table 1). Fifty-one percent of the patients were Hispanic, 26% were white, 18% were African American, and 6% were of other/mixed race. A minority of
Table 5. Model of strongest risk factors for methicillin-resistant Staphylococcus aureus (MRSA) infection among persons with community-associated S. aureus infection. Prevalence of MRSA among community-associated S. aureus infection (pretest MRSA probability), %
Posttest MRSA probability, by no. of risk factors a for MRSA present from 3 item survey 0
1
2
3
60 50
52 (60/116) 42
71 (32/45) 62
81 (13/16) 74
100 (3/3) 100
40 30
32 23
52 41
66 55
100 100
20 10
15 7
29 15
42 24
100 100
b
NOTE. Data are percentages, unless otherwise indicated. This table shows results of modeling of the utility of the strongest 3 risk factors for infection for predicting MRSA (by positive likelihood ratio; see text) among patients with community-associated S. aureus infection. The utility of a 3-item survey was tested in the population studied, which had a 60% prevalence of MRSA among community-associated S. aureus strains. In addition, using data from our cohort, the utility of a 3-item survey for MRSA in other hypothetical cohorts with lower prevalences of MRSA were modeled (see Methods for details). a
The strongest 3 risk factors are close contact with someone with a skin infection in the past 30 days, snorting drugs in the previous 12 months, and incarceration in the previous 12 months. b Current study, which includes 180 adults with community-associated S. aureus infection, 108 (60%) of whom had MRSA infection and 72 (40%) of whom had methicillin-susceptible S. aureus (MSSA) infection. The numerator is the no. of persons with MRSA infection, and the denominator is the no. of persons with community-associated S. aureus (both MRSA and MSSA) infection.
CA MRSA and MSSA Are Indistinguishable • CID 2007:44 (15 February) • 477
Figure 2. PFGE patterns for selected methicillin-resistant Staphylococcus aureus strains. Digestion of chromosomal DNA was by SmaI for selected strains from study subjects: 1, a USA200 strain; 2, a USA400 strain; 3, a USA100 strain; 4, 5, and 7, methicillin-resistant USA300 strains; and 6, a methicillin-susceptible USA300 strain. The band corresponding to the loss of 24kb of DNA containing the type IV SCCmec is indicated by an asterisk.
MSSA infection. For example, the sensitivities of most risk factors for MRSA were !30%, and specificities ranged from 65% to 90% (table 3). All positive likelihood ratios were !3 (with ⭓5 generally considered to be clinically useful) (table 3) [49]. Secondary analyses of MRSA risk factors. In a secondary analysis of adults (82 of whom were infected with MRSA and 55 of whom were infected with MSSA) with “definite” CA infection (i.e., excluding patients who had undergone surgery within the past 12 months), there were minimal differences from our primary analysis. MRSA infection was associated with snorting/smoking illicit drugs (OR, 5.6; 95% CI, 2.0–15.6; P p .004), close contact with an injection drug user (OR, 3.2; 95% CI, 1.2–8.6; P p .02), frequently hospitalized household contacts (P p .04), more frequent laundering of clothes in hot water (P p .002), and more frequent use of antibacterial soap (P p .03). The results of the secondary model, which included pediatric subjects, were likewise similar to the primary model. Although recent contact with someone with a known skin infection trended toward being more common among adult patients with CA-MRSA infection (OR, 2.5; 95% CI, 0.9–7.2; P p .08), the 478 • CID 2007:44 (15 February) • Miller et al.
trend became significant when children were included (OR, 3.5; 95% CI, 1.3–9.7; P p .01). We found no other significant differences between this model and our primary model. Multivariate analyses. In a multivariate model of adults with CA S. aureus infection (table 4), only younger age was associated with MRSA infection (OR per decade of age, 0.96; 95% CI, 0.93–0.99; P p .02). The results of a model that included children also found only younger age to be an independent predictor of MRSA (P p .02). A multivariate model of definite CA S. aureus infection found that younger age (P p .01) and snorting or smoking illicit drugs (OR, 5.0; 95% CI, 1.4–18.0; P p .01) were independently associated with MRSA. Modeling for populations with a lower prevalence of MRSA infection. We modeled the 3 strongest risk factors for MRSA in adults (specifically, close contact with someone with a skin infection, snorting or smoking illicit drugs, and incarceration). Overall, patients who had 0, 1, 2, or 3 of these risk factors had 52% (60 of 116), 71% (32 of 45), 81% (13 of 16), or 100% (3 of 3) probabilities of being infected with MRSA, respectively (table 5). To determine the clinical utility of our findings in populations with a lower prevalence of MRSA infection, we modeled these risk factors in hypothetical cohorts with 10%, 20%, 30%, and 40% pretest probabilities of MRSA infection. In these cohorts, the posttest probability of MRSA infection was only marginally changed (e.g., from 10% to 7%) if no MRSA risk factors were present, regardless of the pretest probability. Molecular characteristics of strains. Of the 123 S. aureus isolates examined, 62 were CA-MRSA, 40 were CA-MSSA, 13 were HCA-MRSA, and 8 were HCA-MSSA. Among the CAMRSA strains, 57 (92%) of 62 were associated with ST8/ USA300 clonal complex (figure 2). Of note, the USA300 strain also accounted for 6 (46%) of 13 HCA-MRSA isolates. The remaining 7 HCA-MRSA isolates belonged to 5 distinct strain types (ST5, ST36, ST72, ST1, and ST8C [USA500]), 3 of which also carried the SCCmec type IV element. Of the 48 MSSA isolates, 13 (27%) demonstrated closely related PFGE profiles (designated USA500), which also belong to the ST8 clonal complex. No other MSSA strain type accounted for 12 isolates. The PFGE profiles of 2 MSSA isolates showed striking similarity to the profiles for the prototypical USA300 MRSA strain (figure 2). The spontaneous excision of the methicillin resistance–encoding determinant—the type IV SCCmec element (∼25 kb in size)—from the chromosome of the methicillinsusceptible USA300 strain can be observed as a shift to a lower molecular weight SmaI band (figure 2). The predominance of ST8/USA300 and ST8/USA500 strains among patients with CA S. aureus infection suggested that the ST8 clonal complex may carry genetic features that facilitate the persistence of these strains in the community. To test this
hypothesis, we further analyzed the distribution of 26 virulence genes in a representative set of 32 MRSA and MSSA isolates. All 17 USA300 and USA500 isolates tested positive for the bsa gene encoding a bacteriocin (table 6), compared with only 3 of 15 isolates from other PFGE strain types. USA300 differed from USA500 and other PFGE strain types with regard to its carriage of the genes encoding Panton-Valentine leukocidin and 2 recently described enterotoxins, Q and K—2 pyrogenic toxin superantigens that lack emetic activity [50, 51]. DISCUSSION Given the increasing number of CA-MRSA outbreaks and infections [5–12, 14–16], clinicians increasingly face a challenging clinical dilemma: should empirical therapy for CA S. aureus infection include b-lactam regimens, which are traditionally effective against only MSSA, or should therapy against MRSA be included? One approach is to utilize the presence or absence of risk factors for MRSA to determine the empirical treatment regimen. However, we found that, in our population of patients hospitalized with CA S. aureus infection, clinical and epidemiologic factors lacked the predictive capacity to distinguish patients with CA-MRSA infection from those with CA-MSSA infection. Our results have important clinical implications. First, our findings suggest that, even if patients with CA S. aureus infection lack exposures associated with MRSA in contemporary outbreaks of infection, the probability of MRSA infection is still sufficiently high to warrant administering empirical therapy against MRSA. For example, in our population, lacking the top 3 MRSA risk factors would change the pretest probability of MRSA from 60% to a posttest probability of 52%. Modeling our data to a population in which MRSA comprise 10% of CA S. aureus isolates, lacking the top 3 risk factors only decreases the posttest MRSA probability to 7%. Thus, risk factors and exposures should be neither used to justify inclusion or omission of empirical antibiotics targeting MRSA nor used to exclude infected patients from receiving infection-control measures for controlling MRSA spread. A second important implication of our investigation relates to findings from our molecular studies. We identified a combination of 4 genes uniquely associated with USA300 that included the genes encoding bacteriocin, Panton-Valentine leukocidin, and enterotoxins Q and K. This 4-gene combination was found only in 1 non-USA300 isolate in our strain population, the USA400/MW2 strain (figure 2) [52], consistent with findings of Orwin et al. [53]. The presence of seq and sek in the 2 predominant CA-MRSA strains found in this country, USA300 and USA400/MW2, suggests that production of enterotoxins may be sufficient to promote S. aureus colonization and transmission [54]. Production of bacteriocin may further facilitate the spread of USA300 and USA400/MW2 in com-
munity settings because of its antimicrobial activity against normal bacteria flora [55]. Finally, the spontaneous excision of the type IV SCCmec element from the chromosome of the methicillin-susceptible USA300 strain (figure 2) suggests that some USA300 isolates may have lost their type IV SCCmec methicillin-resistance determinant and, thus, become MSSA. Alternatively these strains may represent the methicillin-susceptible progenitors of the USA300 strain. This phenomenon has also been seen with the USA400/MW2 strains [56]. There are limitations to our study. First, our investigation was performed at a single health care center. Therefore, the findings may not be generalizable to other areas where circulating strain types and patient exposures differ. Nevertheless, Los Angeles County has been affected by outbreaks of CAMRSA infection in well-defined populations (e.g., athletes, prisoners, and men who have sex with men) [6, 28, 57, 58]. In areas that lack outbreaks that involve these populations, the link between these risk factors and CA-MRSA infection may be even weaker. Second, our study examined patients hospitalized for CA S. aureus infection. However, because of the limitations of our County health care system, patients treated as outpatients in other settings are frequently hospitalized. Therefore, our findings may be generalizable to some outpatient populations. Finally, our definition of CA infection included patients hospitalized in the past year and differs from the ABC criteria [33]. Nonetheless, in our post-hoc analyses, which used a stricter definition of CA infection, excluding patients who had been hospitalized within the past 12 months, our findings did not significantly differ from our primary analysis. There are strengths to our investigation. First, our study was prospective, and epidemiologic data were obtained quickly (typically 1 day after hospitalization). At the time of interview, the susceptibilities of S. aureus isolates were generally not known. In most other studies, MRSA risk factor surveys were performed as part of an outbreak investigation and conducted long after diagnosis and treatment [7, 15, 22, 25, 26], potentially resulting in recall bias [27]. Second, we obtained data on a wide range of potential risk factors that are generally not recorded in medical records (e.g., hygiene and recent incarceration). Third, the participation rate among eligible subjects (92%) was very high. In conclusion, in this prospective investigation, we found no reliable epidemiological or clinical risk factors that could distinguish patients infected with CA-MRSA from those infected with CA-MSSA. From an infection-control perspective, our data suggest that, in areas where a significant proportion of patients hospitalized for CA S. aureus infection carry MRSA, contact isolation should be given to all patients with suspected CA S. aureus infection and should be continued until pathogens are identified. From a clinical standpoint, given the morbidity and mortality associated with delayed treatment of MRSA in-
CA MRSA and MSSA Are Indistinguishable • CID 2007:44 (15 February) • 479
Table 6. Distribution of virulence factors and methicillin resistance determinants in 32 selected isolates of methicillin-resistant Staphylococcus aureus (MRSA) and methicillin-susceptible S. aureus (MSSA), grouped by strain type. Strain type, by PFGE/MLST designation
Variable
Novel MSSA clones
USA300/ST8
USA500/ST8
USA400/ST1
USA100/ST5
USA200/ST36
No. of isolates MRSA
12 10 (83)
5 1 (20)
2 1 (50)
3 3 (100)
1 1 (100)
9 0 (0)
MSSA Community-associated strains Health care–associated strains Virulence factor or resistance determinant
2 (17) 11 (92) 1 (8)
4 (80) 4 (80) 1 (20)
1 (50) 0 2 (100)
0 (0) 1 (33) 2 (66)
0 (0) 1 (100) 1
9 (100) 8 (89) 1 (11)
SCCmec IV
10 (83)
1 (20)
1 (50)
0
0
0
SCCmec II pvl lukE lukM
0 12 (100) 12 (100) 0
0 0 5 (100) 0
0 1 (50) 2 (100) 0
3 (100) 0 3 (100) 0
1 (100) 0 0 0
0 9 (100) 0
bsaA cna splB ssp
12 (100) 0 12 (100) 12 (100)
5 1 5 5
2 2 2 2
(100) (100) (100) (100)
0 0 3 (100) 3 (100)
0 1 (100) 0 1 (100)
1 3 9 9
fnbA icaA tst sea
12 (100) 12 (100) 0 0
5 (100) 5 (100) 0 1 (20)
2 (100) 2 (100) 0 1 (50)
3 (100) 3 (100) 1 (33) 0
1 (100) 1 (100) 0 0
9 (100) 9 (100) 0 0
(100) (20) (100) (100)
(11) (33) (100) (100)
seb sec
0 0
0 1 (20)
0 1 (50)
0 1 (33)
0 0
0 0
see sed sej sei
0 0 0 0
1 (20) 0 0 0
0 0 0 0
0 1 (33) 1 (33) 3 (100)
0 0 0 1 (100)
1 1 1 2
seg sem
0 0
0 0
0 0
3 (100) 3 (100)
1 (100) 1 (100)
2 (22) 2 (22)
sen seo seh sel
0 0 0 0
0 0 0 1 (20)
0 0 2 (100) 1 (50)
3 (100) 3 (100) 0 1 (33)
1 (100) 1 (100) 0 0
2 (22) 2 (22) 0 0
1 (20) 0 0
0 1 (50) 1 (50)
1 (33) 0 0
0 0 0
2 (22) 0 0
sep seq sek
0 12 (100) 12 (100)
(11) (11) (11) (22)
NOTE. Data are no. (%) of virulence genes among 32 strains of interest (see text).
fection [59, 60], it is prudent to include MRSA coverage in empirical antibiotic regimens in settings where a significant proportion of patients hospitalized for CA S. aureus infection have MRSA. Acknowledgments We would like to thank Claudia Morales for her critical contributions with subject recruitment and Kelli Matayoshi and James Lee for their assistance with data management. We wish to thank Dr. Robert Yoshimori and the Clinical Microbiology Laboratory at Harbor-UCLA for their tremendous assistance and patience. We also thank Elizabeth Bancroft and Nolan Lee for their valuable contributions to the survey. We are indebted
480 • CID 2007:44 (15 February) • Miller et al.
to the patients who participated in the study and took the time to speak with us about their illnesses. Financial support. Centers for Disease Control and Prevention (R01/ CCR923419 to L.G.M.); National Institute of Health (K23AI0183 to L.G.M.); and Short Term Training Program at the David Geffen School of Medicine at the University of California, Los Angeles to N.T. Potential conflicts of interest. All authors: no conflicts.
References 1. National Nosocomial Infections Surveillance (NNIS) System Report, Data Summary from January 1992–June 2001, issued August 2001. Am J Infect Control 2001; 29:404–21. 2. Waldvogel F. Staphylococcus aureus (including Toxic Shock Syndrome).
3.
4. 5. 6.
7.
8.
9.
10.
11. 12.
13.
14.
15.
16.
17.
18.
19.
20.
21. 22.
In: Mandell GL, Bennett JE, Dolin R, eds. Mandell, Douglas, and Bennett’s principles and practice of infectious diseases. 5th ed., vol. 1. Philadelphia: Churchill Livingstone, 2000:2072–83. Carratala J, Roson B, Fernandez-Sabe N, et al. Factors associated with complications and mortality in adult patients hospitalized for infectious cellulitis. Eur J Clin Microbiol Infect Dis 2003; 22:151–7. Brook I, Frazier EH. Clinical features and aerobic and anaerobic microbiological characteristics of cellulitis. Arch Surg 1995; 130:786–92. Chambers HF. The changing epidemiology of Staphylococcus aureus? Emerg Infect Dis 2001; 7:178–82. Centers for Disease Control and Prevention. Outbreaks of communityassociated methicillin-resistant Staphylococcus aureus skin infections— Los Angeles County, California, 2002–2003. MMWR Morb Mortal Wkly Rep 2003; 52:88. Baggett HC, Hennessy TW, Rudolph K, et al. Community-onset methicillin-resistant Staphylococcus aureus associated with antibiotic use and the cytotoxin Panton-Valentine leukocidin during a furunculosis outbreak in rural Alaska. J Infect Dis 2004; 189:1565–73. Begier EM, Frenette K, Barrett NL, et al. A high-morbidity outbreak of methicillin-resistant Staphylococcus aureus among players on a college football team, facilitated by cosmetic body shaving and turf burns. Clin Infect Dis 2004; 39:1446–53. Sattler CA, Mason EO Jr, Kaplan SL. Prospective comparison of risk factors and demographic and clinical characteristics of communityacquired, methicillin-resistant versus methicillin-susceptible Staphylococcus aureus infection in children. Pediatr Infect Dis J 2002; 21:910–7. Fridkin SK, Hageman JC, Morrison M, et al. Methicillin-resistant Staphylococcus aureus disease in three communities. N Engl J Med 2005; 352:1436–44. Weber JT. Community-associated methicillin-resistant Staphylococcus aureus. Clin Infect Dis 2005; 41(Suppl 4):S269–72 Fang YH, Hsueh PR, Hu JJ, et al. Community-acquired methicillinresistant Staphylococcus aureus in children in northern Taiwan. J Microbiol Immunol Infect 2004; 37:29–34. Murray RJ, Lim TT, Pearson JC, Grubb WB, Lum GD. Communityonset methicillin-resistant Staphylococcus aureus bacteremia in Northern Australia. Int J Infect Dis 2004; 8:275–83. Moran GJ, Amii RN, Abrahamian FM, Talan DA. Methicillin-resistant Staphylococcus aureus in community-acquired skin infections. Emerg Infect Dis 2005; 11:928–30. Young DM, Harris HW, Charlebois ED, et al. An epidemic of methicillin-resistant Staphylococcus aureus soft tissue infections among medically underserved patients. Arch Surg 2004; 139:947–51; discussion 951–3. Frazee BW, Lynn J, Charlebois ED, Lambert L, Lowery D, PerdreauRemington F. High prevalence of methicillin-resistant Staphylococcus aureus in emergency department skin and soft tissue infections. Ann Emerg Med 2005; 45:311–20. King MD, Humphrey BJ, Wang YF, Kourbatova EV, Ray SM, Blumberg HM. Emergence of community-acquired methicillin-resistant Staphylococcus aureus USA 300 clone as the predominant cause of skin and soft-tissue infections. Ann Intern Med 2006; 144:309–17. Mulligan ME, Murray-Leisure KA, Ribner BS, et al. Methicillin-resistant Staphylococcus aureus: a consensus review of the microbiology, pathogenesis, and epidemiology with implications for prevention and management. Am J Med 1993; 94:313–28. Naimi TS, LeDell KH, Boxrud DJ, et al. Epidemiology and clonality of community-acquired methicillin-resistant Staphylococcus aureus in Minnesota, 1996–1998. Clin Infect Dis 2001; 33:990–6. Salgado CD, Farr BM, Calfee DP. Community-acquired methicillinresistant Staphylococcus aureus: a meta-analysis of prevalence and risk factors. Clin Infect Dis 2003; 36:131–9. Breen JD, Karchmer AW. Staphylococcus aureus infections in diabetic patients. Infect Dis Clin North Am 1995; 9:11–24. Kazakova SV, Hageman JC, Matava M, et al. A clone of methicillinresistant Staphylococcus aureus among professional football players. N Engl J Med 2005; 352:468–75.
23. Methicillin-resistant Staphylococcus aureus infections in correctional facilities---Georgia, California, and Texas, 2001–2003. MMWR Morb Mortal Wkly Rep 2003; 52:992–6. 24. Purcell K, Fergie JE. Exponential increase in community-acquired methicillin-resistant Staphylococcus aureus infections in South Texas children. Pediatr Infect Dis J 2002; 21:988–9. 25. Lee NE, Taylor MM, Bancroft E, et al. Risk factors for communityassociated methicillin-resistant Staphylococcus aureus skin infections among HIV-positive men who have sex with men. Clin Infect Dis 2005; 40:1529–34. 26. Morin CA, Hadler JL. Population-based incidence and characteristics of community-onset Staphylococcus aureus infections with bacteremia in 4 metropolitan Connecticut areas, 1998. J Infect Dis 2001; 184: 1029–34. 27. Rothman K, Greenland S. Case-control studies. In: Rothman K, Greenland S, eds. Modern epidemiology. 2nd ed. Philadelphia: Lippincott Williams and Wilkins, 1998:93–114. 28. Methicillin-resistant Staphylococcus aureus infections among competitive sports participants—Colorado, Indiana, Pennsylvania, and Los Angeles County, 2000–2003. MMWR Morb Mortal Wkly Rep 2003; 52:793–5. 29. Graffunder EM, Venezia RA. Risk factors associated with nosocomial methicillin-resistant Staphylococcus aureus (MRSA) infection including previous use of antimicrobials. J Antimicrob Chemother 2002; 49: 999–1005. 30. Eady EA, Cove JH. Staphylococcal resistance revisited: communityacquired methicillin resistant Staphylococcus aureus: an emerging problem for the management of skin and soft tissue infections. Curr Opin Infect Dis 2003; 16:103–24. 31. Shahin R, Johnson IL, Jamieson F, McGeer A, Tolkin J, Ford-Jones EL. Methicillin-resistant Staphylococcus aureus carriage in a child care center following a case of disease. Toronto Child Care Center Study Group. Arch Pediatr Adolesc Med 1999; 153:864–8. 32. CDC’s Gerberding Warns of Anti-Microbial-Resistant Infections. Available at: http://www.nih.gov/news/NIH-Record/03_02_2004/story01 .htm. Accessed 29 October 2004. 33. Minnesota Department of Health. Community-associated methicillinresistant Staphylococcus aureus in Minnesota. Minnesota Department of Health Disease Control Newsletter 2004; 32:61–72 34. Knaus WA, Draper EA, Wagner DP, Zimmerman JE. An evaluation of outcome from intensive care in major medical centers. Ann Intern Med 1986; 104:410–8. 35. Charlson ME, Pompei P, Ales KL, MacKenzie CR. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis 1987; 40:373–83. 36. Charlebois ED, Bangsberg DR, Moss NJ, et al. Population-based community prevalence of methicillin-resistant Staphylococcus aureus in the urban poor of San Francisco. Clin Infect Dis 2002; 34:425–33. 37. 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. 38. Diep BA, Perdreau-Remington F, Sensabaugh GF. Clonal characterization of Staphylococcus aureus by multilocus restriction fragment typing: a rapid screening approach for molecular epidemiology. J Clin Microbiol 2003; 41:4559–64. 39. Carleton HA, Diep BA, Charlebois ED, Sensabaugh GF, Perdreau-Remington F. Community-adapted methicillin-resistant Staphylococcus aureus (MRSA): population dynamics of an expanding community reservoir of MRSA. J Infect Dis 2004; 190:1730–8. 40. Tenover FC, Arbeit RD, Goering RV, et al. Interpreting chromosomal DNA restriction patterns produced by pulsed-field gel electrophoresis: criteria for bacterial strain typing. J Clin Microbiol 1995; 33:2233–9. 41. McDougal LK, Steward CD, Killgore GE, Chaitram JM, McAllister SK, Tenover FC. Pulsed-field gel electrophoresis typing of oxacillin-resistant Staphylococcus aureus isolates from the United States: establishing a national database. J Clin Microbiol 2003; 41:5113–20.
CA MRSA and MSSA Are Indistinguishable • CID 2007:44 (15 February) • 481
42. Jarraud S, Mougel C, Thioulouse J, et al. Relationships between Staphylococcus aureus genetic background, virulence factors, agr groups (alleles), and human disease. Infect Immun 2002; 70:631–41. 43. Diep BA, Carleton HA, Chang RF, et al. Roles of 34 virulence genes in the evolution of hospital- and community-associated strains of methicillin-resistant Staphylococcus aureus. J Infect Dis 2006; 193: 1495–503. 44. Peacock SJ, Moore CE, Justice A, et al. Virulent combinations of adhesin and toxin genes in natural populations of Staphylococcus aureus. Infect Immun 2002; 70:4987–96. 45. Lina G, Piemont Y, Godail-Gamot F, et al. Involvement of PantonValentine leukocidin-producing Staphylococcus aureus in primary skin infections and pneumonia. Clin Infect Dis 1999; 29:1128–32. 46. Oliveira DC, de Lencastre H. Multiplex PCR strategy for rapid identification of structural types and variants of the mec element in methicillin-resistant Staphylococcus aureus. Antimicrob Agents Chemother 2002; 46:2155–61. 47. Deeks JJ, Altman DG. Diagnostic tests 4: likelihood ratios. BMJ 2004; 329:168–9. 48. Altman DG, Bland JM. Diagnostic tests 1: sensitivity and specificity. BMJ 1994; 308:1552. 49. Goodman SN. Toward evidence-based medical statistics 2: the Bayes factor. Ann Intern Med 1999; 130:1005–13. 50. Orwin PM, Leung DY, Tripp TJ, et al. Characterization of a novel staphylococcal enterotoxin-like superantigen, a member of the group V subfamily of pyrogenic toxins. Biochemistry 2002; 41:14033–40. 51. Orwin PM, Leung DY, Donahue HL, Novick RP, Schlievert PM. Biochemical and biological properties of staphylococcal enterotoxin K. Infect Immun 2001; 69:360–6. 52. Centers for Disease Control and Prevention. Four pediatric deaths from
482 • CID 2007:44 (15 February) • Miller et al.
53.
54.
55.
56.
57.
58. 59.
60.
community-acquired methicillin-resistant Staphylococcus aureus: Minnesota and North Dakota, 1997–1999. MMWR Recomm Rep 1999; 48:707–10. Available at: http://www.cdc.gov/mmwr/preview/mmwr html/mm4832a2.htm. Baba T, Takeuchi F, Kuroda M, et al. Genome and virulence determinants of high virulence community-acquired MRSA. Lancet 2002; 359:1819–27. Ferens WA, Bohach GA. Persistence of Staphylococcus aureus on mucosal membranes: superantigens and internalization by host cells. J Lab Clin Med 2000; 135:225–30. Scott JC, Sahl HG, Carne A, Tagg JR. Antibiotic-mediated anti-lactobacillus activity of a vaginal Staphylococcus aureus isolate. FEMS Microbiol Lett 1992; 72:97–102. Mongkolrattanothai K, Boyle S, Kahana MD, Daum RS. Severe Staphylococcus aureus infections caused by clonally related community-acquired methicillin-susceptible and methicillin-resistant isolates. Clin Infect Dis 2003; 37:1050–8. Ornstein C, Klein G. Outbreak of staph hits USC team: seven football players have been infected with a drug-resistant form of the bacteria. Los Angeles Times. 12 September 2003:B1. Leonard J, Chong F. Staph cases an “emerging epidemic.” Los Angeles Times. 15 February 2005:B1. Blot SI, Vandewoude KH, Hoste EA, Colardyn FA. Outcome and attributable mortality in critically Ill patients with bacteremia involving methicillin-susceptible and methicillin-resistant Staphylococcus aureus. Arch Intern Med 2002; 162:2229–35. Cosgrove SE, Sakoulas G, Perencevich EN, Schwaber MJ, Karchmer AW, Carmeli Y. Comparison of mortality associated with methicillinresistant and methicillin-susceptible Staphylococcus aureus bacteremia: a meta-analysis. Clin Infect Dis 2003; 36:53–9.
